Global warming is driving increases in ambient temperatures, posing a risk to mental and physical health. Heat exposure has been associated with alterations in physiological stress systems, and stress in turn is another major risk factor for a plethora of diseases. Yet, the precise impact of more moderate temperature differences on stress reactivity remains unclear. This study investigated how moderate room temperature variations (17°C-28°C) affect physiological and subjective stress reactivity in 73 healthy adult participants (38 men, 35 women; age M=26.63, SD=6.00) exposed to a psychosocial stressor, the Trier Social Stress Test (TSST). Our results first replicated previous findings, demonstrating that the TSST successfully induced subjective and physiological stress responses, resulting in elevated heart rate, blood pressure, alpha-amylase activity (sAA), and cortisol concentration. More importantly, our data suggest that moderate room temperature variations are correlated with the physiological, but not subjective stress reactivity. Specifically, we found heightened reactivity in both the hypothalamic-pituitary-adrenal axis (r=.313, p=.008), and the autonomic nervous system (sAA: r=.438, p<.001, mean arterial blood pressure: r=.279, p=.017, heart rate: r=.276, p=.018) under higher temperatures. These findings were supported by sensitivity analyses regarding statistical outliers and potential confounders (e.g., baseline stress levels, smoking, oral contraceptive use). Moreover, findings were specific to room temperature and not driven by general outside temperature variations. Together, these observations suggest that even mild temperature differences within the thermoneutral zone are associated with physiological stress responses, highlighting a potential mechanism through which rising temperatures may exacerbate stress-related health vulnerabilities.

Recognizing voices is an important skill in our daily lives. Research has shown that voice identities are formed based on the abstract representation of their mean within a multidimensional voice space (Latinus et al., 2013; Lavan et al., 2019). Moreover, the variability of a voice has been identified as an inherent feature that distinguishes one voice from another, as some speakers exhibit greater variability in their utterances than others (Kreiman et al., 2021). However, it remains unknown whether both the mean and variability of a voice are represented and used for voice identity recognition. In an online study (n=108), we explored whether participants learn and incorporate the variability of a voice into their identity representation. Specifically, we presented three different speakers with low, intermediate, and high variability in their voice features (i.e., pitch and formant variability), respectively. We tested whether the identification of ambiguous voice samples would be shifted to the highly variable voice, in line with Bayesian processing. Behavioral responses revealed that, in addition to the mean, the variability of the speaker influenced voice identification. The more variable a voice, the greater the likelihood of attributing an ambiguous voice sample to that voice, and vice versa. Thus, our findings suggest that people can learn and apply the variability of voices to guide their perceptual decisions, in line with the Bayesian processing framework, in which perception results from a combination of the sensory input with the precision-weighted prior.

Adaptive eye movements are crucial in threatening situations, for example to support the preparation of adequate defensive responses. Gaze patterns can, however, also become maladaptive, such as in anxiety disorders, leading to attentional biases that can hinder fear extinction. We used eye movement data from a larger study in which healthy participants (control group only; N = 40) underwent contextualized differential fear conditioning on three consecutive days. Fear acquisition training took place in context A, extinction training in context B, and retrieval in context A, B, and a novel context C. Conditioned responding was measured by skin conductance responses (SCRs) and pupil dilation. We explored the eye movement data using heat maps and calculated three outcome measures: Proportion of total scanned area and scan path length as measures of visual exploration, and cumulative relative fixation time on the conditioned stimulus (CS). We will conduct statistical analyses with these outcome measures for all fear conditioning phases. We hypothesize that visual exploration of the extinction context leads to a more context-specific extinction memory and therefore to more return of fear in context A and C during retrieval (i.e., fear renewal), whereas fixation on the CS may have the opposite effect. Accordingly, we will test whether eye movements can predict fear renewal. These analyses will provide insights into the role of eye movements in contextual emotional learning and memory processes and may contribute to improving exposure therapy.

In rapidly changing contexts, individuals use decision-making strategies that differ in efficiency, cognitive cost, and optimality. A simple yet inflexible approach learns rewards through trial and error, while a more computationally demanding yet adaptive strategy leverages environmental structure to infer rewards (i.e., map-based inference).While humans are known to use both strategies in simple tasks, it remains a question how individuals utilize each of these strategies to avail of reward structures that rapidly change according to environmental structure.

To investigate how individuals balance these competing strategies, we developed a novel experimental paradigm that dissociates map-based generalization from model-free reinforcement learning. Participants first formed a cognitive map by learning stimulus relationships, then performed a two-alternative forced-choice task the following day, where rewards changed frequently but remained tied to the environment, with the highest reward location shifting randomly and other rewards consistently determined by proximity to it. Participants could solve the task by learning stimulus–reward associations or exploiting environmental knowledge for faster reward inference. Computational modeling contrasted behavioral predictions from model-free reinforcement learning and map-based generalization, showing participants used both strategies with varying reliance. Combining behavioral modeling with fMRI data, we aim to investigate the neural basis of this trade-off, focusing on how cognitive maps are represented in the brain and guide flexible behavior. This paradigm offers new insights into the neurocomputational mechanisms of decision-making and promises a better understanding of how biased map-based inference may lead to psychopathology, for examples in anxiety and depression.

Unambiguous facial expressions elicit predictable social responses such as approaching sad faces (perceived as vulnerable) or avoiding angry faces (seen as threatening). However, ambiguous expressions, that is, morphs between sad and angry expressions, may induce approach-avoidance (AA) conflict. This study examines prosocial behaviour as an approach response, which remains underexplored for ambiguous faces. We expect that higher facial ambiguity will lead to stronger AA-conflict, influencing prosocial behaviour and decision times. Additionally, we investigate whether state empathy moderates the relationship between AA-conflict (induced by facial ambiguity) and prosocial behaviour.

Participants perform a binary Dictator game, distributing points between themselves and the recipient, displaying various ambiguous or unambiguous facial expressions. Allocations are prosocial (equal distribution) or non-prosocial (more to self). State empathy is manipulated through imaginative narratives, inducing high, low, or neutral empathy before the task. We apply Drift Diffusion Modeling (DDM) to analyse decisions and response times to investigate the cognitive mechanisms underlying AA-conflict and prosocial behaviour, estimating drift rate (evidence accumulation rate to reach a decision) and decision bias (prior decision tendencies).

We expect unambiguous sad faces to elicit more prosocial behavior than angry faces across all empathy conditions. Ambiguous faces should increase response time due to heightened AA-conflict compared to unambiguous faces, with a smaller drift rate. We anticipate that state empathy moderates the effect of facial ambiguity on prosocial behavior and the relationship between facial expressions and DDM parameters, with higher empathy leading to a stronger prosocial decision bias.

To date, numerous standardized emotion elicitation databases (SEEDs) are available for use in affective neuroscience and emotion research. These databases are invaluable in inducing emotions in controlled laboratory settings. However, they often provide only valence and arousal ratings, neglecting crucial stimulus features and semantic content that may confound experimental results. Emotional images frequently elicit mixed emotions, vary in visual complexity, and differ in perceptual properties—yet only

a few studies systematically control for these aspects.

To address this gap, we screened the most commonly used SEEDs (IAPS, OASIS, NAPS, EmoPicS) and developed a new image database comprising over 2,800 emotional images. Each image was categorized using both human judgment and AI-based content labeling. Furthermore, we conducted detailed analyses of key stimulus features, including luminance, colour, contrast, complexity, spectral power, saturation, colour temperature, symmetry, resolution and colour dominance.

Based on this multidimensional approach, we constructed 696 image sets, each consisting of at least three to four images matched not only for valence and arousal but also for content and perceptual features. This resource enables more fine-grained experimental designs and supports the development of more ecologically valid emotion induction paradigms, ultimately opening new avenues for exploring how emotional experiences emerge from the interplay of content, context, and perception.

Background: Resilience—the ability to adapt positively in the face of adversity—is shaped by the interplay of biological and environmental risk and protective factors. Previous magnetic resonance imaging (MRI) studies on resilience have predominantly focused on single factors, often operationalizing resilience dichotomously as the maintenance of mental health despite adversity.

Methods: In this longitudinal MRI study, we defined resilience as “better-than-expected” depressive symptom severity (Hamilton Depression Scale) relative to cumulative risk. Using ridge-regularized regression in 1804 participants (955 healthy controls, 849 depressed patients) from the Marburg-Münster Affective Cohort Study, we predicted symptom severity and derived residuals as a measure of resilience. Based on these residuals, we identified 208 resilient, 251 vulnerable, and 200 “as-expected” individuals. Complementing group comparisons, residuals were used to predict gray matter volume (GMV) and cortical thickness at T1 and 2-year follow-up (T2; N=808), using voxel- and surface-based analyses.

Results: Cumulative risk factors explained 51.4% of variance in symptoms at T1 and 44.2% at T2. Residual scores showed moderate stability over time (r=0.31, p<0.001). Cross-sectionally, region-of-interest and whole-brain analyses revealed no morphometric differences in GMV or cortical thickness. Longitudinally, higher resilience at T1 predicted lower GMV in the left inferior orbitofrontal gyrus at T2 (lIOFG, k=172, pFWE=0.006, x/y/z=–48/20/–12), with no changes in cortical thickness.

Conclusion: Resilience to cumulative risk—defined as better-than-expected depressive symptom severity—was not associated with immediate brain structural differences. However, longitudinal analyses revealed reduced GMV in the lIOFG over time, suggesting that adaptive coping may carry a delayed neural cost.

The Dortmund Vital Study is a large-scale longitudinal endeavor that intends to investigate determinants and mechanisms of healthy aging in a working population. Participants, initially between 20 and 70 years old, are tested every 5 years amongst others in several neurocognitive tasks with the measurement of 64-channel EEG. Now, we are in the second measurement phase and more than 300 participants have already been tested twice. Here, we present data from a go/nogo-Simon task, focusing on both the aging effects within the cohort and their development over the five years between recordings.

Behaviorally, only response times showed the same age-related slowing in the cross-sectional and longitudinal analyses. Accuracy showed practice effects longitudinally, which could mask potential cross-sectional aging effects.

In the ERPs, similar bifurcations with coherent longitudinal age effects were observed only for some components. Even at the relatively short interval of five years, measures like N1-latency, nogo-P3 amplitude or go-P3 latency showed reliable effects in the same direction both cross-sectionally and longitudinally. For most of the other ERP components, cross-sectional age effects were not mirrored in the change between the two recordings.

As with the behavioral data, practice effects and familiarization may overwrite the age-related changes. Some cross-sectional age effects may also be due to cohort effects. In addition, a low signal-to-noise ratio may affect the outcome of the analyses. Overall, however, the results may call into question previous findings on intra-individual changes in neuro-cognitive processes with age.

Theoretical accounts of emotion regulation (ER) distinguish between different cognitive strategies for the voluntary modification of emotional experiences, which are assumed to differ in the time course of their effects. In this study, we investigated the time course of neural responses to emotional stimulation for the regulation strategies of distraction and reappraisal. In a sample of 279 participants (age: 18 to 70 years, M = 28.94, SD = 12.26), we used EEG to measure event-related potentials (ERP) in reference to a pre-stimulus baseline in an ER implementation task with high-intensity emotional pictures. For the two regulation strategies, we compared the self-reported negative affect and amplitudes of the ERP components N1, P2, N2, P3, and LPP to a control condition with unregulated viewing. While both strategies reduced self-reported negative affect, distraction significantly increased N1 and N2 amplitudes and decreased P2, P3, and LPP amplitudes. Reappraisal increased P2 and P3 amplitudes while having no significant effects on N1, N2, and LPP amplitudes. These findings align with previous research in suggesting that distraction and reappraisal take effect at different stages of the emotion generation process. Distraction modulates perceptual and attentional processes early on, leading to a lasting reduction of the neural response to an emotional stimulus. In contrast, reappraisal increases attentional resources assigned to stimulus processing and is not always useful to achieve a sustained reduction of the neural response. The difference in LPP effects suggests that when encountering particularly intense negative stimuli for the first time, distraction is more effective than reappraisal.

Aims: Music listening has emerged as an effective tool to manage pain and stress. Prior studies indicate that participant-selected (PS) and researcher-selected (RS) music may have distinct effects on health-related outcomes. This study investigates whether these selection strategies, along with gender, impact pain and stress. Furthermore, we examine possible underlying mechanisms, such as emotional responses and specific personality traits.

Methods: Over three days, 61 healthy participants (age: M=24.2, SD=3.89, 30 men, 31 women) listened to auditory stimuli – PS music, RS music, sound of lapping water (LW) – in randomized order during a cold pressor test (CPT). Repeated measurements were taken for pain intensity, pain tolerance, subjective acute stress, heart rate (HR), RMSSD, salivary alpha-amylase (sAA), and salivary cortisol (sCort). Data were analysed using multilevel modeling, mediation and moderation analyses.

Results: While music selection did not directly influence pain intensity, pain was tolerated the longest for PS music. The effect of gender on pain intensity varied depending on the music listening style. Despite no direct effects of selection strategy or gender on stress responses, an interaction effect emerged between selection strategy, gender and time for subjective stress and HR. Stimulus evaluation, induced emotions, and emotion regulation strategies partially modulated these effects.

Conclusion: Our findings indicate that music selection and gender can influence specific pain and stress responses. Pain tolerance appears to be most affected by the selection strategy. Stimulus evaluation, stimulus-induced emotions, music listening style, and emotion regulation strategies might represent important psychological mechanisms in music-based pain and stress management.

The error-related negativity (ERN) and N2 in the event-related potential in a Go / No-Go task are considered a possible correlate of the subject's political orientation. The present study investigated whether other socio-political attitudes and behavioral aspects can also be associated with these ERP components. In particular, conspiracy mentality and heuristic thinking should be associated with lower error management and lower inhibition and thus a lower amplitude of the ERN/N2. The relationship of conspiracy mentality, heuristic thinking, media use, doom-scrolling and self-efficacy with performance and ERPs in a Go/No-Go task was assessed in 44 German-speaking subjects. The results show clear N2 amplitudes for the No-Go stimuli and clear ERN amplitudes for No-Go errors. Beyond that, however, with the exception of self-efficacy, there are no correlations with the other constructs. The results of our study indicate that biological markers such as ERN and N2 may be insufficient predictors of complex sociopolitical aspects such as political orientation or media consumption.

Through our actions we interact with our environment, causing perceivable changes (effects). Learning the relation between actions and effects is an important prerequisite for goal-directed action. Yet, how such action-effect relations are represented in memory is the subject of an ongoing debate. In four experiments, we investigated the contribution of explicit and implicit memory processes to the learning of action-effect relations. Participants produced object images as effects through discrete keypress actions under either incidental or intentional learning conditions, and under either full or divided attention, with each action-effect episode occurring three times. In a later test phase, images were re-presented and participants had to either reproduce the corresponding action or alternate it. Results obtained through multinomial processing tree modeling suggested the learning of action-effect relations to be primarily supported by explicit memory. Contributions of implicit memory were minimal. Memory was better under intentional compared to incidental learning conditions and worse under divided compared to full attention. These findings provide insights into the representational nature of action-effect relations in memory, suggesting that, at least in early learning stages, they are primarily represented in explicit memory.

Speech perception is shaped by both sensory input and prior knowledge. Healthy aging impacts these factors differently: hearing acuity gradually declines, while verbal knowledge continues to accumulate. Real-world, noisy listening conditions further complicate successful speech perception. To compensate, older adults tend to rely on prior knowledge to aid perception. To address the question of what drives lexical bias in speech perception, we report results from three behavioural experiments (N=78, age range: 19–75) examining the effects of age, hearing ability, and background noise on lexical bias in speech perception using the Ganong paradigm (Ganong, 1980).

In two studies (N1 = 41; 22 younger, 19 older; N2 = 8; 3 younger, 5 older), normal-hearing participants categorized the initial phoneme of speech sounds varying in voice-onset time that were embedded in real or non-word contexts (Bunkt/Punkt–Bund/Pund). Older adults’ perception was more strongly drawn towards real words than in younger adults. However, the degree of this lexical bias varied independently of individual hearing ability.

To further assess the influence of acoustic degradation on biased speech perception, a third experiment presented identical phoneme continua in three different noise conditions (clear speech, 10 dB, 0 dB SNR) to a sample of younger adults (N=29). We found a progressive increase in lexical bias with noise intensity, with the strongest bias observed in the high-noise condition.

These findings suggest that both aging and acoustic degradation amplify the influence of lexical knowledge on speech perception, underscoring the adaptive nature of top-down processing in challenging listening conditions.

Brain activity is shaped and constrained by the underlying anatomy and organization, which form the foundation for human cognition and behavior. However, the precise nature of this relationship remains unclear. Here, we present a nuanced approach and ask to what extent biological factors shape resting-state activity. What are the most predictive features, and which aspects of brain signals do they change?

Using openly available data from the Cambridge Center for Ageing and Neuroscience, we computed power spectra (1-60 Hz) for 350 individuals (18-88 years) measured using magnetoencephalography (MEG). 5 minutes of data was used after regressing out the effects of age, sex, and intracranial volume. We applied partial least squares regression and 10-fold cross-validation to predict the averaged power spectrum in 200 brain areas from 70 brain maps (neuromaps: Markello et al., 2022, Nat Meth; enigma toolbox: Larivière et al., 2021, Nat Meth). Spherical rotations of these maps (spin-test with 1000 permutations) were used to generate a null distribution and assess statistical significance.

Our model predicted regional differences in power spectra with high accuracy (R2 = 0.89, p < 0.05). Key predictors included mean neuronal density, opioid, acetylcholinergic, serotonergic, and GABAergic receptors, cerebral blood flow, and the principal component of gene expression.

These results suggest that a substantial portion of cortical variation in brain activity during rest is governed by specific microstructural and neurochemical distributions. Critically, this contributes to a growing effort to understand the fundamental principles linking brain structure and function and the implications for cognitive and clinical neuroscience.

Neuronal replay—the reinstatement of previously encoded activity patterns—plays a crucial role in memory consolidation and planning. While extensively studied in animal models using invasive techniques, measuring replay in humans remains challenging due to limitations in non-invasive neuroimaging methods. This study aimed to advance replay detection by comparing existing approaches that leveraged the spatial precision of fMRI or the temporal resolution of MEG.

In Wittkuhn & Schuck (2021), it was shown that images presented in fast succession (down to 32ms interval) are decodable with fMRI, even at TRs of 1.25 seconds. Additionally, Temporally Delayed Linear Modelling (Liu et al., 2021) was demonstrated to decipher fast sequences of stimuli in MEG recordings. In the current study, we adapted the task of Wittkuhn & Schuck for MEG and recorded 30 participants. We then use the two datasets (existing fMRI & new MEG) to cross-compare the two methods for detecting replay in humans. Our results show that both methods are applicable to either modality, but need fine-tuning and changes in parameters to work accurately. Finally, the resulting dataset and methods will be published open access. Future researchers developing sequence detection methods can validate their approach by showing it detects sequences against the known ground truth contained in the dataset.

Cognitive control requires mental effort, so we must weigh up against an outcome whether it is worth it. The Expected Value of Control (EVC) theory postulates that reward and efficacy are integrated to determine that outcome. Adult-focused studies found participants intensify control allocation if reward and efficacy were high. How this might change throughout development is currently unclear. Thus, the current study examines age-related differences in control allocation according to EVC-theory. N=178 participants aged 8-30 years completed a Stroop task. On each trial a cue announced the money that could be won (high-/low-reward) and whether the reward was contingent on performance or random (high-/low-efficacy). Task performance was evaluated via GLMM (accuracy) and GLM (response time). Hierarchical drift-diffusion models (hddm) are used to examine underlying cognitive processes.

Accuracy was best for high-reward and high-efficacy across age (b=1.18,p<.001), while the solitary reward and efficacy effects decreased with age (rew.:b=0.92,p<.001;eff.:b=-0.87,p=.001). Response times were faster for high- vs. low-reward and efficacy (rew.:b=-5.59,p=.004;eff.:b=- 11.83,p<.001), showing a decrease for the efficacy effect with age (b=-10.72,p<.001). Results show that reward and efficacy are used consistently across age to determine the allocation of cognitive control, while the magnitude of the effects change. Hddm-models are currently being processed, and parameters of the best fitting model will be analyzed: As an indicator for speed-accuracy trade-off, “boundary separation” is expected to vary with age and condition. Variations in the drift rate with age would indicate differences in evidence accumulation as a function of the interaction of reward and efficacy with age.

Background: Society and early childhood professionals have raised concerns about the potential adverse effects of child screen use. However, research yields mixed findings on the associations between time spent using digital media and child wellbeing, which might be due to most research focusing on duration of time spent on digital devices without considering the content or context of use. Therefore, this study aimed to examine associations between different aspects of child screen exposure, including parental phubbing (phone snubbing of parents during interactions with their children), with internalizing and externalizing behavioral difficulties.

Methods: A sample of 529 parents (82% mothers, M=36.8 years, SD=4.7 years) of at least one child aged 0-5 years (48.4% girls, M=3.3 years, SD=1.3 years) and living in Switzerland completed an online survey as part of the larger SWIPE project. The survey included a section on demographics, child digital media use (screentime, content, and context of screen use), child behavioral difficulties, and parental phubbing.

Results: Data is currently being preprocessed for further downstream analysis. Results on associations of child screen use – including time spent with digital media, digital media content, and context of usage – and parental phubbing with child internalizing and externalizing problems will be presented.

Discussion: This will be an important step to understand if and how different aspects of early child screen use beyond screen time might be linked to child developmental outcomes.

The hippocampus volume is susceptible to experiences early in life but also adapts to influences in adulthood. We sought to replicate influences of parental bonding and chronic stress, and the neuroprotective impact of physical activity. These influencing factors are suggested to act on the hippocampus via hypothalamic–pituitary–adrenal axis functioning in a sex-dependent manner. Aerobic capacity was proposed as a mediating pathway for physical activity.

The sample consisted of 71 healthy adults (37 female). Hippocampus volume, VO2max and the cortisol awakening response (CAR) were assessed. Parental bonding, chronic stress, and habitual physical activity were measured with questionnaires. Moderation analyses were used to assess sex-dependent associations and mediation analyses to investigate the mechanistic role of CAR and VO2max.

We observed a negative association between paternal overprotection and left hippocampus volume, and a partially sex-dependent negative association with CAR that was mainly apparent in women. Vigorous physical activity showed a positive association with left hippocampus volume on a trend level. Aerobic capacity emerged as a mediator for moderate physical activity and exploratively for paternal overprotection in women.

We replicated a positive impact of physical activity on hippocampus volume but neither CAR nor aerobic capacity were robust mediators, thus suggesting an importance of additional pathways. For CAR, our results underline the importance of addressing sex-dependent impacts on neural structure. A deleterious effect of disadvantageous parental bonding was evident, but this association was not mediated by CAR. Instead, our study provides hypothesis-generating evidence for a sex-dependent pathway via physical activity and aerobic capacity.

Resilience, the dynamic ability to adapt and recover from stress, may influence responses to repeated stressor exposure. Habituation, an adaptation to recurrent stressors that reduces allostatic load, could be a key protective mechanism related to improved resilience. This study examined whether self-reported trait and state resilience contributes to variability in stress reactivity and habituation across stress response systems: the hypothalamic-pituitary-adrenal (HPA) axis, autonomic nervous system (ANS), and subjective stress experience.

We applied the Trier Social Stress Test (TSST) to 120 healthy participants (61 women; mean age M = 25.66 ± SD = 5.98 years) across two sessions one week apart. Resilience was measured using the Connor-Davidson Resilience Scale (CD-RISC-25), Resilience Scale 25 (RS 25), and State and Trait Assessment of Resilience Scale (STARS). The TSST successfully induced increases in salivary cortisol concentration (sCort), salivary alpha-amylase activity (sAA), heart rate (HR), blood pressure (BP) and subjective stress.

Correlation analyses with stress reactivity indices revealed no association between trait resilience and physiological stress markers on Day 1. On Day 2, there was a positive association with systolic BP, but no associations with habituation. Small negative associations between trait resilience and subjective stress reactivity were found on Day 1, but not on Day 2 or with habituation indices.

These findings do not support a strong association between self-report measures of resilience and physiological markers of stress reactivity or habituation. However, resilience was related to reduced subjective stress. Findings will be discussed and methodological and conceptual challenges highlighted.

Rumination, a persistent but poorly understood cognitive vulnerability of major depressive disorder (MDD), develops through temporally evolving neural interactions characterized by latent non-linear brain dynamics. Understanding these dynamics are vital to unveiling the brain processes behind such disorders, requiring approaches that move beyond the conventional neuroimaging analysis methods of brain activation.

In this study, we used fMRI time series data from a negative mood induction task to model the temporal dynamics of rumination between individuals with remitted MDD (rMDD) and matched Healthy Controls (HCs). Participants were shown personalized negative words in the MRI scanner, prompting recall of previously recounted distressing experiences. The paradigm consisted of counterbalanced phase blocks of resting, negative mood induction, and either rumination or distraction.

To capture the neural dynamics during these states, we trained a piecewise linear recurrent neural network (PLRNN) on the time series data of a custom network of 22 brain regions of interest and subsequently applied dimensionality reduction to extract the key components that represent their network activities. Through PLRNN modelling, we aim to characterize how rumination is temporally expressed in the brain.

We hypothesize that the latent dynamics will show meaningful differences between resting and rumination, and individuals with rMDD and HCs. Such differences would highlight the temporal organization of ruminative thought processes through distinct brain activation patterns, while demonstrating the potential of such computational approaches to bridge the gap between cognitive theory and temporally resolved fMRI data analysis, thus offering new insights into mood dysregulation in depression.

The ability to learn and select appropriate behavior in a given context is essential for navigating everyday life. The Context-Specific Proportion Congruency (CSPC) effect serves as a measure of context-dependent control learning. In this study, participants performed a double-registration task, first selecting between two contexts and then performing an Eriksen-Flanker task. By manipulating the proportion of congruent trials (PC) within each context (e.g., task location), we investigated context-specific control.

We could replicate key findings like the reduction of congruency effects in low PC conditions compared to high PC ones. Additionally, larger CSPC effects emerged with increased context repetition, suggesting that participants rely more on contextual control policies when experiencing a context consecutively. Furthermore, we examined how participants’ context choices—whether to repeat or switch—modulated CSPC effects. Lastly, in addition to reaction time (i.e., time from stimulus onset to response), we analyzed decision time (i.e., time from context choice presentation to context selection) to explore the impact of preparation time on task performance.

Social cognition encompasses several cognitive and affective processes essential for successful social interaction and communication (e.g. empathy, mentalizing, compassion). The interplay of the various processes necessary for understanding the thoughts and feelings of others is incredibly complex, requiring smooth interaction through efficient connections between various brain areas. Previous work has evidenced bidirectional associations between social cognitive deficits and deficient structural connectivity, suggesting that structural connectivity and white matter (WM) integrity might be an essential foundation for social cognitive abilities.

The systematic review and meta-analyses aims to integrate the growing body of literature on associations between WM integrity and metric measures of social cognitive abilities across cohorts. Quantitative meta-analysis of diffusion weighted whole-brain imaging data is aimed to reveal the WM tracts most strongly associated with the investigated social cognitive construct. Meta-analyses of ROI-based studies will grant insights into the relevance of frequently investigated WM tracts. Meta-regression and subgroup analysis will differentiate between investigated socio-cognitive constructs, DTI metrics, clinical diagnoses, and age groups to investigate potential category-specific effects.

The study has been designed as a registered report and received in principle acceptance. It has the potential to reveal associations between global WM integrity and social cognitive abilities. Moreover, the location specific findings would lay the basis for future ROI-based investigations while socio-cognitive construct-, diffusion-metric-, and diagnosis-specific effects would allow for insights into the potentially diverging relevance of different tracts and WM properties for distinct social cognitive concepts and in different populations.

Olfactory and gustatory perception are expected to be not only sensitive to external chemosensory stimuli (volatile odors, soluble tastants), but also to the internal metabolic state, detected by local receptors for gastrointestinal hormones, neuropeptides (e.g., insulin, orexin), and nutrients (glucose and its transporters) on olfactory and partly gustatory sensory neurons and their subsequent brain pathways. The underlying mechanisms are insufficiently understood for humans. To address the effect of metabolism on olfactory and gustatory perception and the role of accompanying metabolic changes, we manipulated nutritional status: After a 12-h overnight fast, 162 healthy subjects (81 males, 81 females, body-mass index [BMI] 18.8-30 kg/m2, 22.8± 2.6 years) were randomly assigned to either remain fasted or consume an equicaloric carbohydrate-rich or fat-rich breakfast. Capillary blood glucose, insulin in saliva, and hunger were repeatedly measured. Olfactory performance was assessed via the Sniffin'-Sticks test, covering threshold, and the suprathreshold subtests discrimination and identification. Gustatory performance (sweet, salty, bitter, sour, umami) was measured via the Taste-Strips test. As predicted, olfactory sensitivity (threshold) was better when fasted, mainly based on normal-weight (< 25 kg/m2) females under fasting vs. fat-rich intake. The suprathreshold performances, discrimination and identification, tended to be better in the satiated carbohydrate-rich vs. fasted state. Nutritional status affected taste recognition for saltiness and bitterness. First regression analyses reveal olfactory threshold and identification to be predicted by specific blood-glucose changes. Insulin analyses are pending. Chemosensory performance varied with nutrient-type, and differentially for sex, BMI-class, and type of olfactory task (sub- vs. suprathreshold). Further analyses are ongoing.

Hintergrund: Cue-Reaktivität und Craving gelten als zentrale Faktoren bei der Entwicklung und Aufrechterhaltung der Pornografie-Nutzungsstörung (PNS; Starcke et al., 2018), die der Störung mit zwanghaften Sexualverhalten (ICD-11) zugeordnet wird. Während frühere Studien nahelegen, dass sexuelle Reize bei Personen mit problematischem Pornografiekonsum eine stärkere Aktivierung von Belohnungsarealen des Gehirns hervorrufen, ist die differenzierte Rolle von proximalen (z. B. pornografischen Videos) und distalen (z. B. mit Pornografie assoziierten) Reizen weniger gut verstanden. Dem ventralen und dorsalen Striatum könnte in diesem Zusammenhang eine besondere Rolle zukommen. Wir hypothetisieren, dass sich ein Zusammenhang zwischen der vorliegenden PNS-Symptomatik und der Reaktion auf distale Reize in einer Aktivierung des ventralen Striatums widerspiegelt, während sich eine Assoziation mit einer Aktivierung auf proximale Reize im dorsalen Striatum finden lässt.

Methode: Unsere Stichprobe wurde aus der PornLoS-Therapiestudie (Stark et al., 2024) rekrutiert, in der PGS mittels eines klinischen Interviews diagnostiziert und die PGS-Symptome mit den Skalen PPCS-6 und CSBD-19 erfasst wurden. Vor Beginn der Psychotherapie führten die Patienten die "Sexual Incentive Delay Task" (SIDT; Markert et al., 2021) im MRT-Scanner durch. Erfasst wurden neuronale Aktivierung und Reaktionszeiten.

Ergebnisse: Abschließende Ergebnisse werden auf der Konferenz vorgestellt.

Schlussfolgerungen: Unsere Ergebnisse können das Wissen um die neuronalen Korrelate der PNS erweitern und neue Erkenntnisse zur laufenden Debatte über die Klassifikation von PNS als Verhaltenssucht oder als Impulskontrollstörung beitragen. Eine erhöhte Reaktion auf distale Reize in Abhängigkeit von PNS-Symptomen würde auf verstärktes Craving hindeuten und die Bedeutung dieses Symptoms für das Störungsbild untermauern.

Background: Adults with former residential care placement have an increased risk for early life stress (ELS) and stress-related mental health conditions, including mood disorders. Altered cellular aging, as for example indexed by telomere length (TL), may mediate this effect, and individuals with mood disorders tend to display shortened TL. Despite regulating TL, telomerase activity (TA) has rarely been studied in mood disorders or samples with high rates of ELS. This study thus investigated links between TL, TA, and mood disorders in previously institutionalized adults in Switzerland.

Method: One-hundred-seventeen young adults (32% female, mean age = 26.3 years, SD = 3.6) completed the Structured Clinical Interview for DSM-V (SCID-5), with 28 fulfilling the diagnosis for a mood disorder. ELS was assessed using the Maltreatment and Abuse Chronology of Exposure (MACE) scale and whole blood was drawn to determine TA and TL. We analyzed associations between mood disorder status, TA, TL and ELS.

Results: Participants with a mood disorder displayed higher ELS than their counterparts. TA was not significantly related to TL. TA and sex predicted mood disorder status, yet a subset of individuals with mood disorder and very high TA seemed to drive this effect. TL was not significantly associated with mood disorder status.

Conclusions: These results point towards altered TA in a subset of individuals with mood disorders. The exploration of determinizing factors is highly warranted, as they could inform targeted early intervention strategies for high-risk populations.

Environmental adversity is a well-established risk factor for the development of mental illness, including major depressive disorder (MDD). Chronic low-grade inflammation, reflected by persistently elevated levels of C-reactive protein (CRP), has been proposed as a potential biological pathway linking environmental stress to depression. However, the molecular mechanisms driving sustained inflammation in response to adversity remain poorly understood. Using a subsample of the Marburg-Münster Affective Disorders Cohort Study (MACS; N=579; n_healthy controls=320; n_MDD=259), we investigated associations between CRP-related DNA methylation (CRPm), early life adversity (Childhood Trauma Questionnaire), and recent life stress (Life Events Questionnaire). We further examined whether CRPm was associated with depression (Hamilton Depression Scale) both cross-sectionally and longitudinally at a two-year follow-up measurement (n_HC=222; n_MDD=185). Genome-wide methylation profiles from whole blood were used to calculate CRPm scores based on available summary statistics, which explained 25.6% of the variance in high-sensitivity CRP levels in our sample. Early life adversity (β=0.100, SE=0.003, p=0.016) and recent life stress (β=0.121, SE=0.003, p=0.003) were significantly associated with elevated CRPm, even after adjustment for age and sex. Elevated CRPm also predicted greater depressive symptom severity at baseline (β=0.105, SE=0.274, p=0.013) and even more strongly at follow-up (β=0.158, SE=0.249, p=0.002). These findings remained robust after adjustment for cell-type composition, although associations were attenuated when controlling for smoking status and body mass index. Our findings suggest that epigenetic modifications associated with low-grade inflammation may mediate the relationship between environmental adversity and depression severity. Moreover, lifestyle and health-related factors appear to play a significant role in these associations.

Graphs, which represent pairwise relationships topologically, may structure relational knowledge

in a computationally efficient way. However, some evidence suggests that much of human cognition

is inherently spatial.

In two behavioral experiments, we tested whether a graph learned from visual input is spatially

represented, even if space is unnecessary to solve the task. In both experiments, participants

learned a small, sparsely connected, planar graph (8 nodes, 7 edges) that was displayed in a 2D

spatial layout. The next day, participants completed a 2AFC task in which they recalled the

graph and selected the shorter of two presented paths based on the number of edges traversed.

A subsequent arena task, in which participants positioned and connected the nodes, allowed to

assess the extent to which their responses were spatially biased.

In Experiment 1, the graph layout rotated between learning trials, keeping Euclidean distances

between nodes and angles between edges constant. In Experiment 2, the graph layout changed

randomly between trials, causing Euclidean distances between nodes and angles between edges to

vary unpredictably. When tested the next day, we find that participants’ responses were signif-

icantly influenced by spatial properties in both experiments. Notably, this influence persisted in

Experiment 2, suggesting that participants imposed a subjectively constructed spatial structure

onto their representation of relational knowledge in the absence of any consistent spatial experience

during learning.

This finding suggest that the influence of space on cognition is remarkably strong—possibly

indicating that space functions as a fundamental (or default) scaffold of human memory.

The relationship between empathy and Theory of Mind (ToM) as mental capacities enabling dynamic information processing in social interactions is complex and situation-specific. While there is a long-standing research tradition studying the two capacities as independent, recent evidence from naturalistic social tasks shows that under certain task demands, neural networks underlying empathy and ToM-processing do interact.

In this exploratory graph-theoretic network analysis, we investigate how neural network organization is dynamically modulated when processing affective and cognitive stimuli in isolation and interaction. In a sample of 143 adults, we employed a naturalistic social fMRI video task, the EmpaToM, to explore the modulation of neural network organization in empathy, ToM, and their interaction. We will use generalized psycho-physiological interaction (gPPI) to investigate task-context dependent reorganization of neural network organizational patterns.

We expect distinct patterns of neural network organization related to empathy and ToM, as well as their interaction. Specifically, we expect to identify network hubs in the default mode and salience network (such as inferior frontal gyrus/ anterior insula, temporoparietal junction), as nodes within these networks show extensive brain-wide patterns of functional coupling, and have been implicated in both domain-specific and –general processing, such as attentional control. Especially bottom-up attentional switching might be an important mechanism driving cross-network interaction in complex social information processing, in turn enabling behavioral adaptation to changing environmental demands in naturalistic social cognition.

Interozeption – die Wahrnehmung und Integration körperinterner Signale – spielt eine zentrale Rolle für das emotionale Erleben. Obwohl theoretische Überlegungen und empirische Befunde auf eine enge Verbindung zwischen Interozeption und Emotion hindeuten, sind die zugrundeliegenden neuronalen Mechanismen nicht abschließend untersucht. Da bei einer Vielzahl psychischer Störungen eine Beeinträchtigung interozeptiver Fähigkeiten sowie Veränderungen im emotionalen Erleben beobachtet werden, ist ein vertieftes Verständnis dieser Wechselwirkung essenziell für die Entwicklung innovativer therapeutischer Ansätze.

Ziel dieser Studie ist daher, interozeptive und emotionale Prozesse in einem multidimensionalen fMRT-Design zu untersuchen. 44 gesunde junge Männer absolvierten eine umfassende Testbatterie, die Selbstberichtsmaße, objektive behaviorale Instrumente, sowie die Erfassung der neuronalen Aktivität während der interozeptiven und emotionalen Verarbeitung kombiniert. Neben klassischen und neuartigen interozeptiven Aufgaben wurden innovative experimentelle Paradigmen mit methodisch starker exterozeptiver Kontrollbedingung eingesetzt, um die neuronalen Grundlagen dieser Prozesse präzise zu erfassen.

Erste Analysen der Pilotphase zeigen eine erhöhte Aktivierung der Insula in der kombinierten Interozeption-Emotion-Bedingung im Vergleich zu allen anderen Bedingungen (Exterozeption-Emotion, Exterozeption-Neutral, Interozeption-Neutral). Dies deutet darauf hin, dass die neuronale Verarbeitung interozeptiver Signale durch eine gleichzeitige emotionale Aktivierung verstärkt werden könnte.

Die vollständige Datenanalyse der Gesamtstichprobe wird in diesem Poster präsentiert. Implikationen der Befunde für das Verständnis der Interaktion zwischen Interozeption und Emotion, sowie deren Bedeutung für die weiterführende Entwicklungen von Ansätzen zur Intervention bei interozeptiven und emotionalen Beeinträchtigungen werden diskutiert.

Previous studies revealed that individuals with greater interoceptive awareness may use their own bodily signals to make affective evaluations of external stimuli. However research has rarely focused on the effect of interoception on the decision-making process related to veracity assessment. This study uses HEP analysis to explore the relationship between interoceptive awareness and veracity assessments accuracy, which provides a novel approach to testing this phenomenon. We hypothesized that veracity assessment accuracy would be positively related to interoceptive awareness expressed in HEP amplitude and self-report - the Multidimensional Assessment of Interoceptive Awareness (MAIA) questionnaire.

Participants (N = 66) assessed the veracity of 20 statements (10 truthful, 10 deceptive) while their EEG and ECG data were collected and they completed the MAIA questionnaire. Data was collected using a 32-channel EEG system and ECG electrodes placed on the wrists.

The ongoing analysis will assess HEP amplitude across different brain regions (frontal, central, parietal, occipital) during veracity assessment performed by good and poor detectors. Furthermore, the study will explore the neural markers associated with interoceptive awareness and its role in decision-making, potentially advancing our understanding of the body-mind interaction in lie detection.

Background: Neural variability, a key feature of neural systems that supports adaptability and efficiency, has recently gained attention due to its association with cognition across different levels of brain signaling and its high reliability. Since cognition can be considered impaired in many mental disorders, its investigation in mental disorders and their treatment seems promising. Here, we will explore whether and how BOLD-variability during resting-state fMRI changes with exposure therapy, a form of cognitive behavioral therapy (CBT).

Methods & expected results: We will analyze fMRI data from the multi-center study PROTECT-AD, which examined exposure therapy across various anxiety disorders, including panic disorder, agoraphobia, social anxiety disorder, and multiple specific phobias. Our sample includes around 200 patients with pre- and post-treatment resting-state fMRI scans and 100 healthy controls measured at similar time points.

A whole-brain analysis will be conducted to assess psychotherapy-related changes in BOLD-variability between pre- and post-treatment. Linear mixed-effects modeling will be employed, incorporating time and group as key-factors along with additional variables such as age, sex, site, and scanner. BOLD-variability changes in significant clusters will be correlated with symptom improvement on the Hamilton Anxiety Rating Scale in the patients group.

Based on the idea of emotion regulation as one main transdiagnostic mechanism of CBT and drawing from previous research, we expect BOLD-variability changes in brain regions typically involved in emotion regulation (e.g., ventromedial prefrontal cortex, dorsolateral prefrontal cortex, anterior cingulate cortex, insula).

In line with open science and transparent research practices, we will preregister our analyses before conducting them.

Fatigue is a common and disabling symptom in individuals with Post-COVID Syndrome. Interoceptive dysfunction may contribute to its development. This study examined whether interoceptive accuracy (IAc), derived from a heartbeat counting task (HCT), and heartbeat evoked potentials (HEP), a neurophysiological marker of interoceptive processing, predict fatigue. We addressed three questions: (1) Does IAc predict fatigue in Post-COVID individuals? (2) Is the change in HEP amplitude from rest to HCT (ΔHEP) related to fatigue or IAc? (3) Is greater IAc associated with larger HEPs during rest or HCT?

We measured fatigue using the Fatigue Severity Scale (FSS) in 23 adults with Post-COVID Syndrome (mean age = 42.39, SD = 11.62; 61% female). IAc negatively correlated with fatigue (ρ = – 0.44, p = .035), suggesting that greater interoceptive ability may buffer against fatigue symptoms. However, IAc did not significantly correlate with HEP amplitudes at centroparietal sites (Cz, CP1, CP2, Pz) during rest or HCT. Moreover, ΔHEP was unrelated to both IAc and fatigue.

These findings indicate a dissociation between behavioral and neurophysiological measures of cardiac interoception in Post-COVID fatigue. While accurate perception of bodily signals (IAc) may protect against fatigue, stronger cortical responses to heartbeats (HEP) may not be necessary. This suggests that IAc and HEPs reflect distinct processes: interoceptive accuracy versus interoceptive attention.

Introduction: Studies have demonstrated that Approach-Avoidance Task (AAT) training can enhance the effectiveness of therapy for patients with alcohol dependence¹. However, the underlying neural mechanisms mediating these effects remain insufficiently understood. A deeper understanding of these neuronal processes could serve as a foundation for further therapeutic developments, such as the combination of transcranial direct current stimulation (tDCS) with AAT. This study investigates the EEG correlates of the AAT in patients with alcohol dependence, with a specific focus on mid-frontal theta activity and alpha asymmetry.

Methods: EEG data were recorded while participants performed the AAT. The study included 18 patients diagnosed with alcohol dependence at the University Hospital of Würzburg and a control group of 20 participants recruited from the general population. In the AAT, participants were presented with images of alcoholic and non-alcoholic beverages on a computer screen. Using a joystick, they were instructed to either pull (zoom in) or push (zoom out) the images. EEG recordings were obtained using Brain Vision Recorder 2.3, and data analysis was conducted with Brain Vision Analyzer 2.3. Time-frequency analysis was performed using wavelet transformation, and statistical analyses were carried out with SPSS.

Preliminary Results: Reaction time analyses revealed a significant interaction of condition*stimulus*group (F(1,36) = 4,397, p = 0.043). For the EEG Data, we expect a higher midfrontal theta activity in patients for approach-avoidance conflict. Furthermore, we expect a higher alpha-asymmetry during alcohol approach conditions compared to the control group.

Autism spectrum conditions are characterized by a detail-oriented visual processing style. Within the framework of predictive processing, this has been attributed to an imbalanced precision between top-down predictions (priors) and bottom-up sensory processing (sensory precision). However, it remains unknown whether this imbalance arises from attenuated priors, increased sensory precision, or both. In this study, we aimed to investigate the relative contributions of priors and sensory precision to visual perception of social and non-social cues, as reflected in the visual Mismatch Negativity (vMMN) of the event-related potential. Non-autistic participants (N = 40) with varying levels of autistic traits (measured via the Autism Spectrum Quotient [AQ]) performed a visual oddball task including sequences of Gabor patches and fearful and neutral faces, which were orthogonally manipulated for sensory processing discriminability (small or large deviation from standard) and stimulus ratios (high or low percentage of deviants). For Gabor patches, the precision-dependent vMMN was modulated by the AQ; post-hoc correlation analyses revealed that higher levels of autistic traits were associated with larger vMMN amplitudes. For emotional faces, the size of the vMMN was not modulated by the AQ. The results indicate that the detail-oriented processing style associated with higher autistic traits may rely on enhanced sensory precision, rather than attenuated top-down predictions, but only for non-social input. Our findings underscore the importance of considering autism-related processing style in combination with the type of visual input in order to understand the dynamics of visual processing across the autistic spectrum.

Our previous fMRI study (Scheliga et al., 2022) revealed diminished activation of the temporoparietal junction (TPJ) and aberrant connectivity patterns in individuals with schizophrenia (SCZ), indicating disrupted bottom-up sensory processing and enhanced top-down modulation. Since functional deficits have also been linked to brain structural changes, it is crucial to further investigate such alterations in SCZ patients. Building on the findings from our previous fMRI study, this study investigated brain structural alterations in patients with SCZ using voxel-based (VBM) and surface-based morphometry (SBM). Sixteen SCZ patients and 23 healthy controls (HC) underwent structural MRI. We found significantly reduced gray matter volume (GMV) in SCZ patients in bilateral insula, temporal, frontal, parietal, and cerebellar regions, and reduced cortical thickness (CT) in the left superior parietal lobule and middle temporal gyrus. Partial correlation analyses within the SCZ group showed a significant negative correlation between left insula GMV and symptom severity (PANSS total score), and a positive correlation with illness duration. These findings suggest that structural alterations in SCZ—particularly in regions involved in sensory processing, cognition, and emotional regulation—may be linked to core symptoms and reflect disease progression. The insula’s role as part of the salience network and its consistent implication in SCZ highlights its potential as a neural marker. Future studies with larger samples are needed to validate these findings and assess their diagnostic and prognostic value. Structural MRI could aid in early detection and inform personalized treatment strategies.

Paradigms that induce ambivalence about the correctness of actions are rare. The fan task, an innovative paradigm, proposes the idea of ambivalent instructions and stimuli to elicit correct, error, and ambivalent outcomes open to subjective interpretations. On this basis, the study explores performance outcome evaluation and the effects of conscientiousness and perfectionism.

Participants were instructed to stop a spinning pendulum on the light areas of a fan, which has white, gray, and black areas, i.e., correct, ambivalent, and error areas. The study examines differences in feedback-related negativity (FRN, 150-250 ms after feedback onset) amplitudes between fan areas. Traits are measured beforehand via self-reports. 150 participants will perform the task while connected to a 64-electrode EEG system. Differences in mean amplitudes will be analyzed on FCz with a linear mixed model. Preliminary results show an inverted U-shape when comparing white, gray, and black areas. Results indicate that performance outcome evaluation with correct, ambivalent, and error outcomes differs highly from brain activity in tasks where only correct and error trials exist. The analysis will also cover differences between adaptive and maladaptive perfectionists, facets of conscientiousness and their differentiated effects, and interactions.

The fan task offers a unique opportunity to investigate medial frontal negativities by extending classical paradigms that only allow the investigation of correct or error responses. The faceted response design also allows us to investigate the effects of adaptive and maladaptive perfectionism.

The ability to briefly maintain information for upcoming tasks in working memory (WM) is a central aspect of human cognition. While signatures of WM information have been disclosed with a variety of neural recording techniques, similar correlates have recently been found in miniature gaze patterns. However, the precise temporal dynamics of WM information in eye movements remains unclear. To address this, we investigated whether human gaze patterns exhibit encoding of multiple items statically or by rhythmic alternation, and how the ocular activity relates to concurrently recorded EEG signals. Our findings indicate that eye movements capture WM information of up to three items at a time. Contrary to evidence for rhythmic WM replay in neural recordings, we found that on a single-trial level, eye movements appeared to reflect the orientations of multiple items rather statically. These preliminary results corroborate that eye tracking provides a complementary window into WM processes not directly captured with EEG.

Mechanisms underlying task prioritization when cognitive and motor demands compete for limited resources remain poorly understood. This study investigates how varying motivational incentives influence behavioral and neural performance during simultaneous task execution while maintaining constant cognitive-motor demands. Participants completed a dual-task paradigm combining a continuous precision stepping task (requiring foot placement into calibrated illuminated target boxes) with a cued auditory stock evaluation task (requiring manual responses to cued stock-value targets). Task demands remained identical across four experimental conditions with systematically varied monetary reward structure: low cognitive-low motor, low cognitive-high motor, high cognitive-low motor, and high cognitive-high motor incentivization. Behavioral performance, gait kinematics, and neural activity were assessed using synchronized EEG and motion capture systems.

For both tasks, we predict improved performance in high versus low reward conditions with specific trade-offs when reward conditions are imbalanced. For the motor task, we predict increased stepping accuracy, while for the cognitive task, we assume enhanced response accuracy. Neural signatures should reveal prioritization patterns: (1) enhanced sensorimotor mu/beta-band power during motor incentivization reflecting increased motor planning; (2) contingent negative variation amplitude increases during cognitive task incentivization, indicating target-related preparatory resource allocation.

This study provides novel insights into how motivation shapes cognitive-motor resource allocation, with implications for understanding real-world multitasking under varying motivational contexts and applications in rehabilitation and performance optimization.

When making decisions, humans typically rely on a combination of model-free learning—based on stimulus–reward associations—and model-based learning, a more flexible but cognitively demanding strategy that uses internal models of environmental structure. Fear-induced avoidance may disrupt this balance, biasing behavior toward less flexible, model-free control.

To investigate this, we developed a two-day decision-making task involving nine visually distinct stimuli arranged in a ring structure. On Day 1, participants learned the relationships between stimuli via exposure to sequences generated from the ring. On Day 2, they completed a two-alternative forced-choice task in which reward magnitudes shifted regularly, but followed the learned structure. This design encouraged both model-free and model-based strategies, which we captured using computational modeling.

To examine the impact of fear, one monster was paired with an aversive sound on 50% of occasions. We tested whether participants generalized the aversive association by avoiding not only the paired stimulus but also its neighbors—despite the potential for higher rewards. We hypothesized that such avoidance would reflect reduced model-based behavior, either due to disrupted access to the task structure or prioritization of threat avoidance over reward. We also explored whether individual differences in anxiety moderated these effects.

By examining how aversive outcomes shape decision-making in structured environments, this study aims to clarify the cognitive processes underlying avoidance and how fear may shift the balance between model-free and model-based control.

Sex differences in brain structure and function have been widely studied, with evidence suggesting that anatomical differences, such as brain size, contribute significantly to classification accuracy. However, it remains unclear whether functional measures of brain activity provide additional discriminatory power. In this preliminary study, we investigate whether MEG-derived time-series features can enhance sex classification beyond brain size, here, intracranial volume (ICV).

Using source-reconstructed MEG resting-state data from 350 participants (ages 18-88 years) (Cam-CAN study) we extracted a comprehensive set of ~6000 time-series features, including both traditional power spectrum-related features and novel descriptors (Fulcher et al., 2013, 2017). A linear classifier (SVM) with 10-fold cross-validation was used to predict sex based on ICV alone or combined with single time-series features.

The analysis revealed a 12% difference in ICV between males and females. ICV alone, with just one data point per participant, predicted sex with 75% accuracy. Including individual time-series features resulted in a small, significant improvement up to 78%. Approximately one-third of the top-performing features were Fourier Spectrum-related, suggesting a strong link between brain size and spectral characteristics. Further analyses indicated that ICV substantially affects the power spectrum, primarily by modulating the intercept.

These results demonstrate that brain size is a strong predictor of sex and has a significant influence on the MEG-derived power spectrum. The modest improvement in classification accuracy through time-series features suggests that sex differences in MEG signals are largely driven by structural factors, emphasizing the need to control for ICV as a potential confounder in MEG studies.

It has already been shown that the neural response of patients with obsessive-compulsive disorder (OCD) is altered after errors during response tasks. Mostly, preliminary studies found an increased neural signal following errors in OCD patients, often measured by the error-related negativity (ERN), an event-related potential occurring after errors. Another way to investigate error-monitoring are EEG frequency bands such as theta (4-7 Hz) and delta (1-4 Hz). In this study, EEG data from a modified version of the Eriksen flanker task have already been collected from 169 participants, i.e. 59 OCD patients, 50 unaffected first-degree relatives and 60 control subjects, and are currently being analyzed. The first aim is to investigate whether results from ERN studies with OCD patients can be transferred to frequency-based analyses, as time-frequency and frequency coupling analysis, i.e. medio-lateral theta/delta coupling in the prefrontal cortex as a measure of resource acquisition in the cognitive control network. In doing so, we hypothesize that we will find increased theta and/or delta power and increased frequency coupling after erroneous responses in OCD patients compared to controls, similar to increased ERN amplitudes. The second research question is whether oscillatory alterations after errors are also found in unaffected first-degree relatives of OCD patients. A study by Riesel et al. (2011) indicated ERN differences between unaffected first-degree relatives of OCD patients and healthy controls. Can these results also be found in error-related theta and/or delta activity?

Funding: Data collection was funded by DFG, GRK 2753/1 (449640848). Current study is funded by DFG-Grant RI2853/2-2.

Aims: Inflammation, a key immune response to pathogens or tissue damage, may also be shaped by psychosocial factors. Low-grade systemic inflammation has been reported for adolescents exposed to severe stress and might be associated with psychopathology. However, the developmental associations between inflammation, social stressors and psychopathology from childhood to adulthood remain unclear.

Self-harm, a transdiagnostic marker of psychopathology, has been linked to both, altered inflammatory responses and exposure to social stressors. We examined whether developmental patterns of inflammation (C-reactive protein, CRP) from childhood to early adulthood are associated with previous and concurrent victimization experiences and subsequent risk of self-harm (suicidal or non-suicidal self-injury, NSSI).

Methods: Data came from a general-population birth cohort (n = 7124). CRP was assayed using blood samples at ages 9, 15, 17 and 24. Victimization was self-reported several times starting at age 8. Self-harm was self-reported at age 24.

Results and Conclusions: Results revealed age-related increases in CRP-levels; females had higher levels than males (p < 0.05) at ages 9, 17 and 24. Elevated CRP at ages 15 and 17 was associated with a higher risk of suicide attempts at 24 (p < 0.05), but not with NSSI. Latent trajectory models identified distinct developmental patterns of CRP from childhood until early adulthood. Further analyses investigate links between these trajectories and victimization experiences and subsequent self-harm.

Our findings offer new perspectives on the interaction between biological and social factors in the development of psychopathology. We discuss key targets and optimal time frames for intervention and prevention efforts.

Sleep is widely acknowledged to benefit memory, but even within classical paradigms such as word-pair learning, findings are mixed. One reason behind this might be that semantic relatedness of word pairs determines to what degree sleep improves memory. However, findings on the interaction between sleep and relatedness of studies items are also mixed, which could be due to lack of operationalization of “relatedness”. To remedy this, we let N = 80 participants learn optimized word-pair lists with objective semantic relatedness ratings (based on co-occurrence in a Wikipedia corpus) in a preregistered online study. Immediately after learning and after a 12 h retention interval containing wake or sleep (within-subjects AM-PM-PM-AM design), participants’ memory was tested in a cued recall procedure. We find evidence for less forgetting of lowly related word pairs in the sleep group. We also demonstrate high correlations between subjective relatedness ratings and objective ratings (Wikipedia co-occurrence, German FastText, BERT), which implies that post-hoc evaluations of semantic relatedness can be used to compare existing and future studies (e.g., meta-analytic estimates of the effect of semantic relatedness).

Eye contact is a fundamental component of social bonding. However, in digital video communication, genuine eye contact is prevented, as one can look either into the interaction partner’s eyes or at the webcam. AI-driven gaze enrichment, which adjusts the eye gaze direction in a natural way, offers a solution, but its impact on social dynamics - particularly whether restoring eye gaze allows individuals to better control their social skills (e.g., empathy) for social bonding - remains unknown. Here, we aimed to close this research gap by analyzing the impact of gaze enrichment on perceived eye contact, physiological arousal (i.e., heart rate), and social dynamics in digital dyadic interactions between N=55 healthy male participants. Their interaction was based on the Fast Friend Procedure, which enables rapid social bonding by taking turns in answering predetermined personal questions. Participants were randomly assigned to an AI-enriched eye gaze condition (n = 25), or to a standard eye gaze condition (n = 30). We were amazed to find that gaze enrichment had no effect on perceived eye gaze behavior or heart rate. However, it did moderate social dynamics. Emotional empathy fostered feelings of interpersonal closeness only when the gaze was enriched. In sum, while AI gaze correction does not directly affect subjective perception or arousal, it subtly influences social interactions by restoring their natural dynamics.

Background:

Interpersonal synchronization of heart rate variability (HRV) has emerged as a promising physiological marker of social interactions. While most research has focused on individual HRV patterns, this study investigates dynamic HRV synchronization between individuals during a trust-based economic interaction (Investment Game - IG).

Methods:

Dyadic pairs took part in repeated rounds of IG under two conditions: remote and face–to-face. The study included phases of physical separation, a first in-person encounter, a shared baseline without interaction, two face-to-face IG sessions and a series of perceived synchronization tasks (e.g. rating how in sync participants felt with each other on a scale). HRV data were continuously recorded using Polar heart rate sensors and synchronization was quantified using cross-wavelet power analyses. Data collection is ongoing, current analyses are based on 60 participants (15 men; M = 23.52 years, SD = 6.71).

Results:

HRV synchronization was minimal during physical separation and increased significantly upon initial face-to-face encounter. Synchronization remained stable during the shared baseline, suggesting that co-presence alone does not enhance physiological synchronization. Notably, synchronization increased during the interactive investment game sessions and peaked during the perceived synchronization tasks, indicating that social interaction and intentional co-regulation amplify physiological coupling.

Conclusion:

Preliminary findings suggest HRV synchronization is not merely a byproduct of physical proximity, but rather a dynamic, socially responsive marker influenced by experimental interaction paradigms. Ongoing analyses will explore how individual differences, particularly early life adversity, influence physiological synchronization and trust-related decisions.

Event-related analysis of EEG signals provides insights into cognitive processing by improving signal-to-noise ratio compared to continuous signal analysis. While conventional ERP methodology relies on repeated presentation of artificial stimuli, recent research demonstrates that blinks represent a naturally occurring alternative that doesn't interfere with the investigated situation. The advantages of the ERP method remain, allowing for interference-free assessment of cognitive states in naturalistic situations.

This study investigates whether expectation of task-relevant stimuli is reflected in blink-related EEG activity of 590 participants during the cue-target interval of the CPT-AX task. In this task, participants press a button in response to X-stimuli that follow A-cues. A-cues indicate likely target appearance, whereas B-cues indicate no-go trials.

Blink-locked ERP analysis shows significantly more pronounced CNV amplitude in response to blinks in A compared to B trials. Time-frequency decomposition reveals significantly stronger blink-related desynchronization of μ-rhythm at left centro-parietal recording sites in A versus B trials. Comparison of inter-trial-phase coherences demonstrates stronger phase synchronization across theta, alpha, and beta bands at parieto-occipital electrodes in A-trials.

These results collectively indicate increased task-preparation in terms of motor-readiness when task-relevant stimuli are expected. The time-frequency measures highlight specific advantages of blink-related analysis, showing that task-related information is reflected in blink-evoked activity and suggesting increased sensory sensitivity that couldn't be observed without blink-locked analysis.

Most decisions, such as what to invest in, where to spend one’s holiday, or whether to adopt a novel medical treatment, involve uncertainty. In fact, often even the probabilities of the outcomes are unknown. We make such decisions not only when we are rested but also when we are tired and exhausted. In an increasingly uncertain world with an increasingly fatigued population, understanding the interplay between fatigue and decision-making under uncertainty becomes ever more important but currently remains limited. Here, participants made decisions involving risk or ambiguity before and after experimentally induced cognitive fatigue. Each participant performed the experiment twice, once with induction of strong fatigue and once with induction of weak fatigue. Preliminary results suggest that with high levels of self-reported fatigue, participants found it harder to make decisions and were less confident in their choices. Moreover, induced fatigue affected women more strongly than men in terms of how hard they found the decisions. A better characterisation of the role of fatigue in decision-making under uncertainty, with a focus on the underlying subjective processes, will not only inform scientific theories but might also help to understand and improve decisions, behaviour and well-being in daily life.

Borderline Personality Disorder (BPD) severely impacts emotion regulation, social interactions, and cognition. Most individuals with BPD have experienced childhood maltreatment (CM), which affects neural development, leading to alterations in the default mode network (DMN), a key network for self-referential processing and mentalizing. The current study investigates DMN connectivity differences in BPD versus healthy controls (HC) and explores the effects of CM on DMN connectivity.

Resting-state fMRI data were analyzed using seed-based connectivity (SBC) and ROI-to-ROI analysis, focusing on the medial prefrontal cortex (mPFC), bilateral parietal cortex (LP), posterior cingulate cortex (PCC), and bilateral hippocampus. Data were preprocessed and analyzed using the CONN toolbox in MATLAB. The sample includes n = 49 BPD and n = 36 HC (M = 27.84 years, 82% female).

Both SBC and ROI-to-ROI analysis did not yield any significant group differences. The pattern of the DMN was present in both groups. Exploratory analysis revealed decreased functional connectivity between mPFC and pre- and postcentral gyri with increasing exposure to childhood maltreatment.

The current study does not support the hypothesis of altered functional connectivity in the DMN in BPD compared to HC. However, reduced connectivity between the mPFC and somatosensory and primary motor cortices may be linked to altered pain perception in BPD, warranting further investigation.

Studies have demonstrated a correlation between the temporal resolution of perception and neuronal oscillations. Specifically, the frequency with the most prominent peak in the spectrum, i.e., the peak frequency (PF), is thought to define a temporal binding window for perception. Most of the supporting evidence comes from research in the visual domain, particularly examining parieto-occipital PF. However, evidence from other sensory modalities remains limited. Furthermore, the existing evidence linking PF to perception is primarily correlative in nature.

The present study aimed to investigate a potential causal relationship between PF and perception. To achieve this, we sought to modulate PF of neuronal oscillations through the administration of lorazepam, a GABAa-receptor agonist known to reduce PF in neuronal oscillations. In a double-blind protocol, participants completed three sessions in which they were administered either 0.5 mg, 1.5 mg lorazepam, or a placebo. During these sessions, participants performed a tactile temporal discrimination task while neural activity was recorded using magnetoencephalography (MEG). We hypothesized that lorazepam impairs tactile temporal perception and decreases PF in somatosensory areas.

Preliminary results indicate that – in line with the hypothesis -lorazepam impaired participants' performance on the tactile discrimination task. In contrast to our hypothesis, however, a negative correlation between temporal resolution and PF was observed in the alpha band (~8-12 Hz) in parieto-occipital regions, but not in somatosensory regions.

In summary, these findings provide preliminary evidence supporting a causal link between PF and temporal perception. Future analyses will study the impact of lorazepam on PF across the entire brain.

One of the most bacis functions of the central nervous system is to detect unexpected changes in the environment. In humans, this function is often examined using oddball sequences, in which a series of frequent standard stimuli is interrupted by rare deviants, i.e., oddballs. A large body of evidence indicates that deviants elicit a mismatch negativity (MMN) response in the event-related potential from approximately 100 ms to 300 ms after stimulus onset. Despite its thorough characterization in both healthy and clinical populations, the temporal evolution of the MMN remains poorly understood, particularly in the visual modality. Using high-density EEG recordings in healthy volunteers, we tested whether visual mismatch responses can be traced back to the earliest stages of processing in primary visual cortex (V1). To minimize stimulus-specific adaptation, we used arrays of high-contrast bars whose orientations varied randomly across trials. Our findings reveal that mismatch responses can be detected at the level of the C1, the earliest cortical component of the visual evoked potential (VEP). This indicates that mismatch detection affects initial stimulus-evoked activity in V1 and suggests that information integrated over several seconds can modulate the C1 component in a top-down fashion.

Effective regulation of approach-avoidance behaviors is critical for adaptive decision-making. This study examined whether competing outcomes—reward and threat—facilitate overcoming automatic behavioral tendencies towards affective stimuli. Fifty participants completed a two-session Free vs. Forced Approach-Avoidance Conflict task. During the task, CSs were presented in an anticipation phase followed by a response phase where participants used a joystick to either approach and obtain CS-specific outcome or avoid without receiving anything. In Session 1, five conditioned stimuli (CSs) predicted appetitive (monetary reward, appCS+), aversive (threat via electrical stimulation, avCS+), or neutral (neuCS−) outcomes. In Session 2, motivational conflict was introduced by reassigning one reward-paired CS to predict both reward and threat (app2confCS+) and one threat-paired CS to include both threat and reward (av2confCS+), while others retained original associations. Behavioral, subjective, and physiological measures were analyzed. Results revealed that conflicting outcomes disrupted automatic tendencies: app2confCS+ reduced approach to reward, while av2confCS+ diminished avoidance of threat. Effects diverged across domains—valence ratings were preferentially influenced by reward conflict, whereas avoidance behavior was more sensitive to threat conflict. Physiologically, in Session 1 aversive CSs elicited the largest pupil dilation and avoidance-oriented gaze fixation, reflecting heightened arousal, while appetitive CSs evoked relatively smaller dilation and approach-oriented gaze. In Session 2, conflicting CSs produced intermediate pupillary responses and balanced gaze distributions, suggesting moderated arousal and attentional reallocation. These findings suggest that competing outcomes can disrupt habitual action tendencies and facilitate behavioral inhibition in approach-avoidance conflicts, providing translational insights for psychiatric interventions targeting maladaptive approach-avoidance behaviors.

Sensorimotor integration processes are crucial for everyday activities, such as grasping objects or driving a car. Theta band activity (TBA) in distributed brain networks is likely essential to perform sensorimotor integration successfully. Directed communication in these brain networks is shaped by brain maturation during adolescence. This study investigates how age-related effects attributable to brain maturation influence directed communication in a theta-associated sensorimotor integration network. We conducted an EEG study with a continuous pursuit-tracking task performed by an adult group (n = 41) and an adolescent group (n = 30). Both groups show elevated theta-band activity during higher sensorimotor demands. Yet, the adult group outperformed the adolescent group, particularly during higher demands. Further analyses revealed that this is likely due to enhanced directed connectivity between frontal areas and the ventral processing stream in adults, which likely enables effective integration of visual and motor information. Adolescents rely on frontal TBA signaling surprise and prediction error, with no input from ventral stream areas. This might cause lower performance during higher demands. Across age, TBA appears to serve distinct functions during sensorimotor integration. Age-related processes alter the neural processes underlying complex sensorimotor integration.

Hintergrund: Erhöhte selbstfokussiere Aufmerksamkeit bei gleichzeitig verminderter externer Aufmerksamkeitslenkung gilt als zentraler Aufrechterhaltungsfaktor der Sozialen Angststörung (SAD). Die Veränderung des Aufmerksamkeitsfokus ist ein zentrales Element kognitiv-verhaltenstherapeutischer Interventionen. Erste Ansätze nutzen Virtual Reality (VR) für Aufmerksamkeitsübungen, jedoch wurde bislang kein VR-gestütztes Training mit blickbasiertem Feedback von Aufmerksamkeitsprozessen im Kontext öffentlicher Vortragssituationen untersucht.

Fragestellung: In einer randomisiert-kontrollierten Studie wurden die Effekte eines VR-Aufmerksamkeitstrainings mit blickbasiertem Feedback bei hoch sozial ängstlichen Personen untersucht. Verglichen wurden zwei Trainingsversionen mit Aufmerksamkeitsfokus auf soziale Reize mit variabler vs. invarianter Valenz. Es wurde angenommen, dass sich die allgemeine sowie die aktuelle Angst, die Aufmerksamkeitslenkung und die Selbstwahrnehmung im Posttest signifikant verbessern und die variable Gruppe einen Vorteil ggü. der invarianten Gruppe aufweist.

Methode: Teilnehmende richteten ihren Blick gezielt auf virtuelle Publikumspersonen, anschließend erfolgte ein Feedback via Eye Tracking. Die variable Bedingung beinhaltete ein virtuelles Publikum mit unterschiedlich valenter Mimik (z. B. gelangweilt, interessiert), die invariante Bedingung nur Personen mit positiven Gesichtsausdrücken. Erfasst wurden subjektive, physiologische und eye-tracking-basierte Maße vor und nach dem Training.

Ergebnisse: Erste Ergebnisse zeigen eine signifikante Reduktion der Angst sowie physiologischer Reaktionen in Redesituationen – unabhängig von der Valenz der Gesichtsausdrücke. Auch die selbst eingeschätzte Wirkung auf andere sowie das eigene Auftreten verbesserten sich signifikant.

Fazit: VR-basiertes Aufmerksamkeitstraining mit Feedback kann kurzfristig soziale Ängste reduzieren. Die Langzeiteffekte sollten in zukünftigen Studien geprüft werden.

Available cognitive resources modulate the processing of emotional stimuli. Dual-task paradigms suggest that under low cognitive workload, negative emotional distractors impair task performance, whereas positive distractors can enhance it. In contrast, under high workload, attentional control mechanisms facilitate goal-directed processing, inhibit distraction, and attenuate emotional processing. Prefrontal regions have been implicated in top-down inhibitory control. However, despite evidence from tightly controlled laboratory studies, the neurophysiological basis of such mechanisms in complex, naturalistic environments remains poorly understood.

Using whole-head magnetoencephalography (N = 48), we investigated how negative and positive naturalistic speech and visuospatial workload modulate neural processing in a multisensory driving simulation. Although subjective ratings supported our hypothesis of top-down inhibition of emotional processing, we found no interaction between emotion and workload at the neurophysiological level using mass-univariate analysis. To increase sensitivity and circumvent assumptions about inter-individual shared anatomical and functional neural origins, we employed multivariate pattern analysis using linear discriminant analysis and common spatial patterns in sensor space. Emotional speech and workload conditions were clearly discriminated (F1 > 0.7), and contributing physiologically interpretable patterns were transformed into source space. This approach identified the left inferior frontal gyrus, orbitofrontal gyrus, and left anterior temporal lobe as key regions underlying the interaction between emotional speech processing and attentional control during visuo-spatial workload. We conclude that these regions are involved in top-down mechanisms that support flexible up- and down-regulation of emotional speech processing, depending on available cognitive resources. We advocate using multivariate statistics to study brain function in complex, naturalistic environments.

In visual search, incidental learning of statistical regularities – such as repetitions of some or all display configurations over blocks – facilitates performance. For example, in contextual cueing (cc), observers search faster in repeated than novel displays. Recently, however, it was shown that the mixture of novel and repeated contexts in cc led to slower search compared to search in all repeated or, surprisingly, even in all novel displays (Vaskevich & Luria, 2018, Cognition). Presumably, the low predictiveness of display repetition in mixed repeated/novel presentations slowed down search. Here, we re-evaluated this claim. To this end, groups of participants performed visual search tasks with all repeated displays, all novel displays, and a mix of 50% novel and 50% repeated displays. First and foremost, we found that the mixed regularities in cc did not result in longer search times, higher number of fixations or more inefficient scan-pattern ratios in comparison to visual search in solely repeated or solely novel contexts. Thus, we could not replicate the costs of low predictiveness of display repetition on visual search. One difference between Vaskevich & Luria's experiment and ours was that we constrained randomness of novel displays with respect to target eccentricity and the distribution of the search items across display quadrants. We hypothesize that this may limit the applicability of repetition predictiveness effects.

Aims: Emotion dysregulation (ED) may arise as maladaptive processing of emotional stimuli, leading to emotion lability and altered emotional expression. Additionally, ED has been associated with early life adversities exposure, ranging from prenatal stress to bullying experiences. While previous studies have separately explored brain changes associated with ED and lifespan adversities, their interplay remains unclear. The following study aims to investigate how different risk factors and neurobiological mechanisms shape ED across development, contributing to a better understanding of its underlying mechanisms. Methods: A multivariate data-driven approach was employed to explore associations between ED, lifespan adversities and structural brain features in a population-based sample (IMAGEN, n=880). Results: A shared latent representation linking ED, adversity and cortical thickness suggests ED being influenced by environmental adversity, in particular traumatic experiences, family stress, prenatal smoking and stressful life events during adolescence, and cortical thickness of regions implicated in emotion regulation, including the anterior cingulate and orbitofrontal cortex. Further analyses will elucidate multivariate associations with additional structural brain features. Conclusions: Our results emphasize ED as being associated with both environmental adversity and cortical structural alterations. Clarifying the developmental origins and neurobiological correlates of ED provides valuable insights into its underlying mechanisms and predictive value for psychopathology, identifying potential targets for early prevention and intervention strategies.

Objective

The therapeutic alliance is widely recognized as a key predictor of psychotherapy outcome, though it is predominantly assessed via subjective self-reports. This study investigated whether interpersonal movement and physiological synchrony, specifically skin conductance (SC) and heart rate (HR), could serve as objective biomarkers for therapy effectiveness in naturalistic cognitive behavioral therapy (CBT) settings.

Method

Twenty-five patient-therapist dyads participated in up to two CBT sessions. Physiological signals were continuously recorded using wrist-worn wearable sensors. Head and body movement were video-recorded. After each session, both dyad members completed alliance questionnaires. Patients also provided standardized symptom ratings. Synchrony was quantified using lagged-windowed cross-correlations and the calculation of a single session index (SSI).

Results

Skin conductance response (SCR) synchrony significantly predicted changes in patients' global severity index (GSI). High positive SCR concordance was linked to a reduction in patients' GSI, while negative or small positive SSI values were associated with symptom aggravation. Preliminary findings suggest a similar predictive pattern for HR synchrony, with HR synchrony outperforming subjective affiliation ratings in explaining outcome variance. *Final HR synchrony results will be presented at the conference.

Conclusion

The findings demonstrate the presence of interpersonal synchrony in clinical interactions. Interpersonal physiological synchrony appears to be a meaningful, unobtrusive marker of therapeutic engagement and effectiveness. Incorporating wearable biosensing into psychotherapy research offers a scalable, objective approach to understanding and enhancing treatment effectiveness. Future studies with larger samples should refine synchrony-based prediction models and explore their use in personalizing treatment choices and improving patient-therapist matching in evidence-based psychotherapy.

Event-related potentials (ERPs), i.e., brain responses to stimuli presented during electroencephalography (EEG) measurements, are a key tool in neuropsychology research. A standard EEG experiment is conducted in a highly controlled, static environment where the participants remain still, and the stimuli are presented on a 2D computer screen. However, these artificial conditions may affect the generalizability of findings. Integrating virtual reality (VR) technology into EEG/ERP research increases ecological validity while maintaining experimental rigour. Given the high sensitivity of EEG to external interference, VR head-mounted displays (HMDs) may impact EEG signal quality. The present study investigates the influence of VR HMDs on P3 (ERP component related to attention) measured by a 32-channel EEG system (ANT Neuro). EEG in two conditions, CS (the stimuli were presented on the computer screen without VR) and VR (the stimuli were presented using HMD), was recorded using an oddball paradigm (in which the participants react to rare (target) and ignore the frequent (standard) stimuli) with visual and auditory stimuli. The preliminary results show that P3 ERP was elicited in both conditions with similar amplitudes and latencies. The good agreement between the conditions in the ERPs obtained suggests the suitability of EEG/VR experiments using HMD. Such an evaluation of the impact of VR headsets on EEG signal quality during neuropsychological experiments is essential for correct data interpretation in attention research that uses tools to simulate real-world neuropsychological processes.

Social hierarchies are central to human interaction, and navigating them requires adaptive decision-making. During adolescence and early adulthood, developmental changes in cognitive control, stress reactivity, and hormonal systems shape how individuals learn from social and non-social feedback. Previous research suggests that stress modulates reward learning by altering dopaminergic responses, but its impact on decision-making across developmental stages and contexts is less clear. Computational reinforcement learning models offer a powerful framework to examine how individuals process information and update their beliefs under uncertainty. We conducted two cross-sectional studies comparing social hierarchy and non-social reinforcement learning under stress in 60 girls aged 15–18 and adults aged 18–25 (50 women, 25 men). Girls and women were assigned to equally sized groups: half used hormonal contraceptives, and half were naturally cycling and tested in the mid-luteal phase. All participants completed two testing sessions, one including the Trier Social Stress Test and one neutral control session, in balanced order. After the manipulation, they completed a social hierarchy learning task—framed as a competition against peers—and a matched non-social reinforcement learning task. The adolescent sample completed both tasks during fMRI scanning to assess neural activation during learning. We collected salivary cortisol, gonadal hormones, and cardiovascular measures. Decision-making strategies were analysed using model-based Bayesian reinforcement learning, comparing six models based on Bayesian Information Criterion. We examined the modulatory effects of stress, age, pubertal status, and hormones. Preliminary findings regarding developmental learning processes and their relevance for social functioning and mental health will be discussed.

Diese Studie untersuchte Herzrate (HR) und Herzratenvariabilität (HRV) als physiologische Marker für Arousal und die Aktivität des autonomen Nervensystems (ANS) und ihre Beziehung zu Suizidgedanken und Suizidversuchen im prospektiv multizentrischen Studienprojekt APOS (https://osf.io/axnws). Methoden: Es wurden 150 psychiatrische Patient:innen untersucht, die aufgrund akuter Suizidalität stationär aufgenommen wurden. 96 Patient*innen berichteten von mindestens einem Suizidversuch in der Lebensgeschichte (34 Patient*innen hiervon gaben einen Suizidversuch in den letzten 4 Wochen an), 54 Patient*innen berichteten ausschließlich akute Suizidgedanken. Neben der Erhebung der physiologischen Marker mittels einer 8-minütigen Polar H10-Brustgurt Messung, wurde der SSEV-Fragebogen und das SITBI-Interview erfasst. Ergebnisse: In einem t-test für unabhängige Stichproben ergaben sich signifikant höhere HR-Werte in der Subgruppe von stationär-psychiatrischen Patient:innen mit Suizidversuch(en) in der Lebensgeschichte (M = 81,34, SD = 12,54, N = 97) verglichen mit stationär-psychiatrischen Patient:innen ohne Suizidversuch (M = 76,33, SD = 12,59, N = 48; t (143) = 2,37, p = .019). Zudem wurden auch bzgl. zwei der drei HRV-Parameter signifikante Unterschiede zwischen beiden Patient:innengruppen gefunden. Es konnte u.a. ein positiver Zusammenhang zwischen der Stärke der Suizidgedanken in den letzten 4 Wochen (SSEV) und dem physiologischen Arousal kontrolliert für das Alter der Proband*innen gefunden werden, (r = .316*, p = .005, N = 62). Schlussfolgerung: Vorläufige Daten der APOS-Studie zeigen, dass physiologisches Arousal (HR) positiv mit Suizidgedanken korreliert ist. Patienten mit und ohne Suizidversuch unterscheiden sich signifikant bezüglich des physiologischen Arousals und auch einzelner Selbstregulationsfähigkeitskennwerte (HRV). Nach Abschluss der Studie werden diese Zusammenhänge prospektiv und in hoher zeitlicher Auflösung analysiert werden.

Background: Holotropic breathwork, characterized by prolonged rhythmic hyperventilation, can induce altered states of consciousness that may have psychiatric therapeutic potential (Fincham et al, 2024). Even though this breathing technique bears clinical interest, there is scant evidence on the neurobiological effects of holotropic breathwork (Fincham et al., 2023; Evan et al., 2024). This study aims to investigate the neurobiological effects of holotropic breathing on global brain perfusion using functional magnetic resonance imaging (fMRI).

Methods: In a non-randomized cohort of 30 healthy participants, perfusion-weighted fMRI based on pseudo-continuous arterial spin labeling (pcASL) is conducted over the course of a holotropic breathwork session (Kilroy et al., 2014). Imaging acquisition takes place before, during and after the holotropic breathwork. Participants rate the intensity of the acute psychedelic experience every 12 minutes using a keyboard-based Likert-scale.

Conclusion: 10 participants have already been included in the study and preliminary data analysis is currently underway. Hypotheses include a global reduction in cerebral blood flow (CBF) during acute breathwork sessions compared to baseline, and that changes correlate with the intensity of the subjective psychedelic experience. These findings may provide critical insights into both the neurobiological effects and potential therapeutic mechanisms of holotropic breathing techniques.

Framing effect is when people's decisions are influenced by how options are presented, even if the options are identical. While widely studied, how neutrality interacts with this bias remains unclear, as research typically relies on forced binary choices that overlook preferences for neutral stances.

We conducted an online experiment with over 690 participants using a modified classic framing task with three options: sure, risky, and neutral. If participants selected the neutral option, the scenario was represented in the classic binary format for a second time.

Results showed participants were more likely to choose the neutral option under the loss frame in two of three scenarios. Notably, individuals who initially selected neutral option still exhibited the classic framing effect—risk aversion in gain frames and risk-seeking in loss frames—when later forced to make binary choices.

These findings support our hypothesis that neutrality is meaningful in framing tasks, with loss frame triggering greater decisional ambivalence. The persistence of framing effect when neutral responders were forced to decide suggests neutrality in our experimental context represents a temporary deferral of decision-making rather than frame immunity.

This nuanced understanding enriches theoretical decision-making models by demonstrating how preference abstention interacts with classic framing effects. Including neutral options in framing, research may capture a broader response range and offer deeper insights into framed decision processing.

Fast attentional orienting towards fearful faces has been repeatedly confirmed by many studies. It is assumed that fear relevance is rapidly extracted from low-spatial frequency (LSF) visual information coming from subcortical magnocellular signals. However, it is unclear whether these dynamics operate similarly for newly acquired threat cues, i.e., without the presence of perceptual threat cues by fearful faces (i.e., open mouths, wide eyes) that are ontogenetically well-established. Furthermore, how such attentional shifts in visual processing would parallel peripheral orienting responses, such as the galvanic skin response (GSR), has been rarely investigated so far. In this study, we tested whether rapid attentional shifts towards fear-relevant faces favor LSF information of newly learned threat cues. We manipulated fear-relevance by using a differential fear conditioning paradigm in which 2 out of 4 neutral faces were paired (CS+) with a white noise burst. Central and peripheral electrophysiological signals of 32 healthy participants were recorded. Preliminary data show selective enhancement of the left-hemispheric P1 event-related potential in response to LSF CS+. GSR peak amplitude will be evaluated in the future, but a first data exploration shows a similar pattern as the P1. Data suggest that rudimentary magnocellular information facilitates rapid fear retrieval for newly learned threat cues. As the amygdala is assumed to be crucial for fear acquisition, these data might confirm fear-specific pathways from the amygdala to the visual cortex innervated by magnocellular signals.

Autism is a neurodevelopmental disorder characterised by atypical communication, social interaction, and stereotyped-repetitive behaviours. Using ABIDE1 data, we examined early structural brain differences through cortical folding measures: local gyrification index (GI), surface ratio, and fractal dimension. As folding tends to be associated with functional connectivity, we hypothesised reduced folding in frontal and parietal-temporal areas in autistic individuals. Given the heterogeneity in autism, we complemented this with three analytical approaches: (1) ADOS -based subtyping, (2) covariate analysis, and (3) cluster analysis. Our results revealed reduced folding for autistic brains in rostral middle frontal, pars opercularis, and pars triangularis areas, alongside increased folding in occipital areas. GI specifically indicated reduced folding in parietal-temporal regions. ADOS-based comparisons showed that R-autistic individuals (with more stereotyped-repetitive behaviours) exhibited lower degrees of folding in precentral and superior frontal areas than C-autistic individuals (with more communication difficulties). K-means and hierarchical cluster analyses identified three data-driven subgroups that transcended ADOS domain classifications. These clusters suggest that verbal and full-scale intelligence, in interaction with symptom severity, may additionally capture meaningful variation within the autism spectrum. Intelligence accounted for approximately 20% of group variance, while combined ADOS and ADI-R severity scores explained around 30%. Our findings support a dimensional perspective on autism, and highlight the value of combining neuroimaging with behavioural data-driven stratification. Future research may explore the clinical relevance of such clusters, and assess – e.g., through participatory approaches – the degree to which these subgroupings reflect autistic individuals’ lived experiences.

Relational integration (RI) is a core feature underlying higher-order cognition and is required when multiple relations must be jointly considered to make inferences. As higher cognition is known to decline in older age, our study addressed the role of RI in neurocognitive aging focusing on age-related differences in conditions of increasing RI demand, the nature of age deficits in RI, and the influence of RI on age effects in other higher cognitive tasks.

N=153 healthy adults (20-80 years, n=76 female) completed two RI tasks and tasks measuring working memory capacity (WMC), interference control, processing speed, fluid intelligence, and planning. ANOVAs revealed a significant age group x RI demand interaction in both RI tasks, indicating that older adults are more affected by increases in RI demands than younger and middle-aged adults. Mediation analyses assessed whether age-related declines in RI were driven by WMC, interference control, or processing speed. Across the two RI tasks, WMC was the most consistently found mediator. To test whether RI drives age effects in higher cognitive tasks, models with RI as mediator and planning and fluid intelligence as outcomes were calculated. Planning ability was fully mediated by performance in both RI tasks, while fluid intelligence was partially mediated through one RI task.

These findings support the assumption of RI as a key mechanism in neurocognitive aging, with RI itself being primarily influenced by WMC and in turn significantly contributing to age effects in other domains of higher cognition such as planning and fluid intelligence.

Stressful life events (LE) are well researched in their influence on adverse (mental) health outcomes. Less is known about their effect on acute stress reactions, with ambiguous results discussing sensitizing vs. blunting effects on biological stress reactions.

To further explore the relationship between previous experiences and acute stress reactivity, we measured subjective and cortisol stress responses to the Trier Social Stress Test (TSST) and 55 LE typically relevant for student samples in N = 124 undergraduates.

As expected, the results show significantly higher subjective and cortisol stress responses in the TSST- compared to the control group.

We also observed a significant effect of LE on cortisol reactivity, with predominantly negative LE being associated with a higher cortisol reactivity in both the TSST- and control group.

With respect to the time of occurrence of LE, a significant interaction of LE and TSST responses were observed for childhood (under age 13), with differences in cortisol reactions only in the TSST-stress group (higher for predominantly negative, lower for predominantly positive LE). LE occurring after childhood show no significant main or interaction effect on stress reactions. Moreover, LE effects did not relate to subjective stress responses in both time frames.

The results indicate an influence of LE on endocrine stress reactions, with childhood events being the main influence.

Differences in subjective vs. endocrine reactions, the possibility of sensitization towards anticipated threats through more recent events, as well as protective effects of positive experiences are discussed.

Background: Social interaction is a primary aspect of communicating how others judge us. It allows us to update ourselves and our expectations about others. While humans generally exhibit self-related positive biases in their updating behavior, theoretical accounts propose that this biased processing is attenuated, absent, or negatively biased in participants with depressive symptoms. The process of aligning and integrating social evaluative feedback in realistic interaction scenarios that would test this assumption is, however, lacking. We provide an Event-Related Potential (ERP) study that combines neuronal (Feedback-Related Negativity, FRN, and Late Positive Potential, LPP) and behavioral measures of evaluative feedback processing and updating behavior. Methods: We selected healthy adults (N = 62) with depression scores across the range of low and high values, measured with the Beck Depression Inventory (BDI). Participants received feedback from supposed experts and peer senders, with feedback manipulated to be worse, congruent, or better than the self-ratings of participants. Results: Participants with higher depression scores exhibited more negative initial self-ratings and developed a more negative feedback expectation across the experiment. In addition, we found that higher depression scores led to more negative updating towards worse expert feedback and less positive updating after better peer feedback. Concerning ERPs, unexpected but not self-incongruent feedback increased the FRN, while both types of incongruence increased the LPP. Finally, BDI scores correlated with LPP amplitudes to all feedback. Conclusions: The results contribute to a deeper understanding of how individuals process and integrate social evaluative feedback and its relation to depressive symptoms.

Understanding how brain network architecture supports individual differences in behaviour is a key challenge in neuroscience. While diffusion MRI-based structural connectomes have become a standard tool for modelling white matter organization, their utility for phenotype prediction remains limited. Here, we propose percolation-based structural eigen vector centrality networks (PSENs) as a novel network representation that emphasizes globally influential nodes under simulated network degradation. PSENs are derived by iteratively thresholding individual structural connectomes (i.e., normlaized streamline count) and computing eigenvector centrality at each percolation level, capturing structural co-vulnarability between nodes. Using data from the Human Connectome Project – HCP1200 – we extracted structural connectivity and PSENs for each individual and trained ridge regression models to predict 58 behavioural phenotypes across cognition, emotion, and personality. PSENs significantly outperforms SC in significantly predicting the behavioural variables (59% variables predicted versus 40% variables predicted). Especially in state-like behavioural variables across emotion, these networks have demonstrated 55% significant predictions in comparison to SC model with 23%. These results position PSENs as a powerful biologically informed representation of brain structure for individual-level predictions.

A theory is presented on how specific structural design features of gambling products may foster the emergence of erroneous gambling-related beliefs via dopaminergic mechanisms. Multi-line electronic gambling machines (EGMs) are strongly associated with problem gambling. Dopamine (DA) plays a central role in substance-use disorders, which share clinical and behavioral features with disordered gambling. The structural design features of multi-line EGMs likely lead to the elicitation of various dopaminergic effects within their nested anticipation-outcome structure, including effects related to prediction error, reward anticipation under uncertainty, and cue saliency. The present theory draws an analogy between EGM gambling and latent state inference accounts of conditioning. In this account, maladaptive gambling-related beliefs and expectancies emerge from a process of erroneous latent state inference under reward uncertainty, which may be exacerbated by EGM design features and associated dopaminergic processes. Over the course of repeated exposure to gambling, these processes may in some individuals foster the emergence of maladaptive state priors, which clinically manifest as gambling-related cognitions, beliefs, and expectancies.

Humans experience seasonal variations in sleep and circadian processes, likely driven by photoperiod length. Anecdotally, many people in Switzerland, Germany, and Austria report increased fatigue in spring ("spring fatigue"), but scientific evidence is lacking.

Thus, we here investigate seasonal variations in fatigue, daytime sleepiness, and sleep quality via an ongoing online survey (since April 2024), with repeated assessments every six weeks over one year (up to nine measurements). We hypothesised higher fatigue (Fatigue Severity Scale, FSS) and daytime sleepiness (Epworth Sleepiness Scale, ESS) and lower sleep quality (Bernese Sleep Health Questionnaire, BSHQ) with shorter photoperiods in winter. Insomnia severity (Insomnia Severity Index, ISI) was also expected to increase. Bayesian linear mixed models were used for analysis. The study was pre-registered on Open Science Framework.

As of March 25 2025, 405 adults (median age 32, range 18-74; 50 men, 329 women) completed two to eight assessments. Of these, 46% reported suffering from spring fatigue. Analyses provided strong evidence (Bayes Factors [BFs] < 0.1) against photoperiod-related variations in fatigue and insomnia, and moderate evidence against effects on sleepiness and sleep quality (BF ≤ 0.13). Extreme evidence was found against monthly (BFs ≤ 0.002) and seasonal variations (BFs ≤ 0.01), except for sleepiness, where evidence was anecdotal (BF = 0.34).

Preliminary findings indicate no systematic seasonal variations in fatigue, sleepiness, or sleep quality. Limitations include incomplete data (eight of nine assessments) and a non-representative convenience sample. At the meeting, updated analyses will be presented.

Unser Leben wird zunehmend globalisiert und Reisen via Flugzeug, ob nun geschäftlich oder privat, ist mittlerweile Standard geworden. Allerdings kann Fliegen Stress verursachen (ein erhöhtes Stresslevel wird von 17-20 % der Fluggäste berichtet), und einige Personen leider sogar unter einer ausgeprägten Flugangst (geschätzte Punktprävalenz von 2,5 %). Daher ist es wichtig, Faktoren zu identifizieren, die die Erregung und den Stress verringern können. Verschiedene Studien bestätigen, dass das Hören bestimmter Musikgenres Entspannung fördern kann. Beispielsweise kann Entspannungsmusik mit langsamen Rhythmen die erlebte und physiologische Erregung reduzieren, indem der parasympathische Nervensystemanteil aktiviert wird. Physiologische Indikatoren dafür sind bspw. eine erhöhte Herzratenvariabilität (HRV) oder reduzierte elektrodermale Aktivität (EDA). Hingegen gibt es Hinweise, dass schnellere Musikstile, wie Rockmusik, häufig das sympathische Nervensystem stimulieren. Das Ziel der Studie ist es, zu untersuchen, ob das Abspielen von Entspannungsmusik (vs. Rockmusik) während eines simulierten Fluges mit Bewegungssimulation und v.a. bei den angstbesetzten Flugphasen (Start- und Landephase sowie Flugphasen mit Turbulenzen) mit einer reduzierten subjektiven Anspannung sowie einer erhöhten HRV und einer reduzierten EDA einhergehen. Erste Analysen zeigen hypothesenkonforme Unterschiede zwischen Personen, die Entspannungs- vs. Rockmusik während des simulierten Fluges hörten. Das Abspielen entspannender Musik könnte demnach eine unkompliziert durchführbare Maßnahme sein, um Passagiere an Bord von Flugzeugen zu entspannen und ggf. Angstzustände bei flugängstlichen Personen insbesondere während Flugturbulenzen zu reduzieren.

While neuroticism has become a benchmark trait in behavioral genetics—yielding numerous associated variants and substantial genetic correlations with major depressive disorder—other personality dimensions remain comparatively underexplored. One such trait is Hypomanic Personality, proposed as subclinical expression of vulnerability to bipolar disorder and schizophrenia. A persistent challenge in studying the genetics of psychological traits is limited sample sizes and statistical power. Here, we capitalize on previous large-scale genome-wide association studies (GWAS)—particularly from the Psychiatric Genomics Consortium—to enhance variant discovery through pleiotropy-informed analyses. We conducted a GWAS of Hypomanic Personality in 7,004 individuals of European ancestry, identifying one genome-wide significant locus at 2p13.3 (rs79707808; p=7.8e-10), and estimating SNP-based heritability at 12% (SE=0.066). Summary statistics from bipolar disorder and schizophrenia were then incorporated through ‘pleioFDR’ to enhance variant discoverability. LD Score Regression supported this strategy, revealing substantial genetic correlations with schizophrenia (rg=0.40, SE=0.15), and bipolar disorder (rg=0.24, SE=0.12). Consequently, pleiotropy-informed analyses identified 10 loci, including an intronic region in CACNA1C, a well-established psychiatric risk gene. To validate and extend our findings, we computed polygenic scores for Hypomanic Personality in up to 341,335 UK Biobank participants. Phenome-wide analyses across >20,000 traits revealed associations with behavioral (e.g., risk-taking: p=2.6e-11), affective (tiredness/lethargy: p=3.6e-07), cognitive (fluid intelligence: p=6.5e-08), substance use (smoking: p=1.7e-08), and physical traits (BMI: p=3.4e-11). Together, our findings position Hypomanic Personality as biologically grounded construct linking subclinical personality variation to major psychiatric disorders. Polygenic associations with cognitive, affective, and behavioral traits underscore its relevance as cross-cutting marker of shared genetic liability.

The study of attentional processes has typically focused on stimuli that are salient due to either bottom-up or top-down factors, such as threat or other physical factors or goal-directed behavior, respectively. However, little research has been made on the topic of subjective preference, i.e. the question of why certain stimuli inherently hold higher value to individuals. For instance, people may favor certain colors or flavors over others, without this having any apparent reasons in terms of the other attentional dimensions. In the present study, we aim to examine this specific kind of attention by measuring it in a binary choice task while simultaneously deriving 64-channel EEG. Participants are presented with stimuli depicting snack items which they previously rated based on their subjective liking. We hypothesize that preferred items, when contrasted with neutral or non-preferred items, will elicit an N2pc component that encodes an attentional bias towards these items. This result would demonstrate that the N2pc component is not only sensitive due to the commonly measured goal directed or stimulus driven attentional processes, but also to those caused by participants' subjective preference. We will present the study design as well as preliminary data collected up to this point.

Electroencephalography (EEG) is a widely used method for investigating human brain dynamics. However, EEG analyses are frequently conducted with limited a priori knowledge regarding exact effect sites or latencies. This lack of predefined spatial or temporal effect boundaries poses challenges for identifying meaningful condition differences. To address this, mass univariate analyses have become a valuable complement to traditional approaches, such as analyzing predefined time window or electrode group averages. These methods have emerged as powerful tools to correct for multiple comparisons, while mitigating the risk of false positives, and enhancing the robustness of statistical inferences in high-dimensional EEG data. Here, we review different approaches to delineate spatial and temporal effect boundaries in three different datasets. Specifically, we focus on permutation approaches (cluster-based permutation, tmax procedure) and Bayesian alternatives to address condition differences in i) event-related potentials, ii) steady-state evoked responses, and iii) time-frequency data. Overall, our simulations indicate that cluster-based permutation tests provide a relatively liberal approach to correct for multiple comparisons across datasets, with high sensitivity for detecting large effects. In contrast, the tmax procedure presents the most conservative method among the analyzed approaches. We highlight the impact of different significance thresholds on statistical outcomes across a range of simulated effect sizes in both empirical and simulated data.

The recent introduction of organic light-emitting diode (OLED) monitors with refresh rates of up to 480 Hz opens new possibilities for their use as precise stimulation devices in cognitive neuroscience. These affordable high-speed monitors, targeted at video gamers, promise several advantages over the cathode ray tube (CRT) and liquid crystal display (LCD) monitors used in many labs. Unlike LCDs, OLED displays have self-emitting pixels that can show true black, resulting in superior contrasts, a broad color gamut, and wide viewing angles. More importantly, their high refresh rates and rapid transition times may allow us to implement new paradigms that previously required prohibitively expensive projector systems, like Rapid Invisible Frequency Tagging (RIFT).This study characterized two recently introduced 240 Hz and 480 Hz OLED monitors in terms of timing properties, spatial uniformity, viewing angles, warm-up times, and other properties. We also compared their responses to those of CRTs and LCDs. Results confirm excellent temporal properties, with CRT-like transition times (around 0.25 ms), wide viewing angles, and decent spatial uniformity. We then illustrate the possibilities offered by these monitors in two time-critical paradigms: (1) The gaze-contingent presentation of stimuli during saccadic eye movements and (2) the implementation of RIFT with affordable consumer hardware. Our findings suggest that the newest OLED monitors are precise and cost-effective stimulation devices for visual and cognitive neuroscience that offer key advantages over both CRTs and LCDs.

Singing is a complex skill integrating auditory, somatosensory, and motor systems to ensure accurate vocal performance. Vocal training has been shown to induce adaptive changes in neural circuits supporting auditory processing, motor control, and multisensory integration. The developmental trajectories of such changes in children remain underexplored, despite evidence that neural adaptations are most pronounced when training begins early in life.

We plan a longitudinal study to investigate how structured choir singing influences neural and behavioral plasticity in young children. Over five months, three groups of 30 six-year-old girls will participate: one in choir practice, one in dance classes, and a no-contact control group. Each child will undergo behavioral and neuroimaging assessments at two time points, evaluating executive functions, music perception and production (e.g., tonal, rhythmic), and dance-related skills. Neuroimaging will include structural MRI, diffusion-weighted imaging, and functional MRI (resting-state and task-based).

We expect choir training to enhance music perception and production skills. Structurally, we anticipate changes in auditory and motor somatotopic representation areas and strengthened white matter connectivity, particularly in the arcuate fasciculus. Functionally, we predict increased connectivity in singing-related and multisensory integration regions, such as the insula. We also expect the dance group to show improved rhythm perception, production, and dance-related skills.

Efficient response control relies on coordinated activity across cortical and subcortical circuits. Within these circuits, the subthalamic nucleus (STN) plays a critical role in integrating internal goals with external demands to gate behavioral output. Moreover, the STN has been shown to be involved in cognitive functions such as interference resolution and attention. Parkinson’s disease (PD) is characterised by a pathological shift towards inhibition in subcortical response control circuits, associated with increased low beta oscillations in the STN. Deep brain stimulation (DBS) can modulate this imbalance, proposedly by suppressing low beta activity. This study aimed at further deciphering the role of the STN within subcortico-cortical networks. To this end, we used a novel task combining motor response reprogramming and flanker interference in PD patients with STN DBS (n = 23). Ten of the included patients had a Percept PS (MedtronicTM) implantable pulse generator (IPG), which allowed for the recording of STN LFPs alongside the cortical EEG ON and OFF DBS. Our results show that frontal and STN low beta power ~300 ms pre-response predicted errors in the novel task. In particular, trials requiring both reprogramming and interference resolution showed disproportionate error increases with higher STN low beta power ON and OFF DBS. Furthermore, we found DBS-specific increases in errors driven by flanker interference at higher STN low beta power. These results provide further evidence for a detrimental impact of low beta oscillations for flexible, domain-general inhibitory control via the STN and reveal a specific adverse effect of DBS on interference inhibition.

Distraction from pain via cognitive engagement is known as an effective pain modulation strategy. While inhibition of pain relies on the prefrontal cortex, a structure that is affected by age, little is known how cognitive distraction from pain is influenced by age and age-related cognitive decline. In addition, age has been shown to alter the stress response, which itself is assumed to have a negative impact on cognitive modulation of pain.

To investigate the impact of age and acute stress on distraction from pain, healthy young and older participants (60+ years) underwent the Trier Social Stress Test (TSST) and a respective control condition in a two-session design. Before and afterwards, they performed a pain distraction task, a n-Back working memory task with low and high cognitive load, during which participants received individually adjusted transdermal electrical pulse trains in non-painful and moderately painful intensities to the inner forearm. Participants rated them regarding their intensity and unpleasantness. Pain-related evoked potentials were recorded with a 64-channel EEG and saliva was collected for cortisol analysis.

Preliminary behavioral and electrophysiological results revealed a pain relief by distraction irrespectively of the cognitive load and age. While older and younger participants were equally stressed by the TSST, showing similar elevated cortisol levels, young participants tend to be more affected by acute stress in their pain-relief than older participants.

The findings suggest that healthy older adults showing only a mild age-appropriate cognitive decline benefit from psychological strategies to reduce pain and acute stress does not diminish its efficacy.

Facial emotion recognition (FER) biases refer to systematic errors that people make when processing facial expressions. In youths with conduct disorder (CD), who are characterized by highly impairing antisocial behavior, most research on FER biases has been conducted on hostile attribution biases and found that CD youths perceive ambiguous social cues as more negative (i.e., as angry). However, CD youths may not only show biases towards anger, which is why we investigated FER biases in CD youths towards six basic emotions. We analyzed data from 610 youths with CD (60% female) and 818 typically developing controls (TDCs; 68% female), aged 9 to 18 years (M = 14.1, SD = 2.41 years), collected in the European FemNAT-CD multisite study. FER biases were assessed using the Emotion Hexagon Task by showing morphed emotional expressions and asking participants to choose the predominant emotion. Biases were calculated as misrecognition of or tendency towards an emotion shown at 0%, 10%, 30%, or 50% intensity. Our findings from hierarchical linear modelling indicate that CD youths exhibited stronger FER biases than TDCs across all emotions, meaning that they misclassified each emotion more often. However, this difference varied by intensity, with CD youths displaying weaker biases at higher intensity levels and a smaller increase in bias with increasing intensity level. Our findings indicate that youths with CD do not only show a hostile attribution bias but rather misclassify emotions as predominant when they are present at low intensity, regardless of type of emotion.

In social interactions, the decisions of an individual should be influenced by their current state, the given environment, and the (past and expected) behavior of the interaction partners. A decision-maker thus faces the strategic challenge of elaborating the relative importance of environmental and social components.

Here, we developed a novel sequential game that was framed as a hunter-gatherer foraging task and that follows the well-established 2x2 framework, where two players have two identical options to simultaneously choose from. Given our experimental design, we can develop a number of hypothesis driven strategies, which focus to varying degrees on the player’s state (e.g., rich or poor energy level), environmental features (e.g., good or bad weather), and and the other player’s behavior (defection or cooperation).

In a sample of 28 healthy adults, we tested each strategy as a latent variable for explaining participants’ behaviors across trials and against different artificial agents. Our results suggest that individuals rely their choices mainly on environmental features or the other player’s behavior with different levels of sophistication. Furthermore, some people seem to have a clear strategy, while others are less consistent.

Differences in strategic tendencies may essentially be shaped by psychological factors like a person’s personality or psychiatric dispositions. Therefore, the implementation of different strategies and their adherence levels has potential to quantify psychological traits and clinical conditions in future work.

Advances in neuroimaging have increasingly focused on understanding brain activity during everyday behaviors. EEGManySteps is an international initiative investigating how variations in experimental design influence the measurement of brain dynamics during gait. By uniting laboratories, our project leverages archival and newly recorded EEG data to explore the interplay between experimental conditions, data processing choices, and the reliability of results in studying brain function during gait.

By standardizing acquisition and analysis protocols, we seek to harmonize datasets obtained from different hardware systems and study designs, establishing robust benchmarks for gait-related brain research. This collaborative effort addresses the complexity inherent in studies recording EEG while walking and aims to identify key factors— gait or task related neural dynamics or movement-related artifacts—that modulate EEG responses. The project is structured into Work Packages (WPs), each focusing on distinct aspects. The main WPs are: WP1: Acquisition Protocols (identifying and accessing existing datasets), WP2: Data Collection (designing standardized protocols, assisting with setups), WP3: Data Curation (converting datasets to BIDS format, parameterizing covariates), WP4: Analysis (developing analysis pipelines), and WP5: Dissemination (paper writing, tutorials, workshops).

EEGManySteps aims to establish a framework for reproducible research and Open Science, specifically engineered to be used across laboratories with differing resource constraints. Three contribution tracks allow to A) integrate additional measures into existing studies, B) contribute existing data sets or C) conduct full dedicated recording sessions. Through these efforts, we aim to advance the understanding of gait-related EEG recordings, particularly how cognitive and motor processes interact during walking.

The MNS is thought to automatically react to social stimuli. However, previous studies have also shown that the MNS responds more strongly when observing actions with motivational salience – whether due to intrinsic behavioral/social relevance or extrinsic incentives. However, the distinct contributions of these salience mechanisms to MNS engagement remain unclear. In this study, we dissociated two sources of motivational salience: 1. external reward (monetary incentive), and 2. task-driven social relevance (emotion recognition). By comparing these conditions, we investigated how reward-based and relevance-dependent salience differentially modulate MNS responses, shedding light on the neural dynamics of action observation under varying motivational contexts.

In our study, 74 participants completed an fMRI task. The participants were shown emotional facial expressions (joy and fear), which they either had to observe or to recognize. Half of the observe and half of the recognize trials were rewarded.

Both, observation and recognition of the facial expressions was associated with strong process-specific activation across the whole brain. Specifically, recognition of the emotion was related to insula and inferior frontal gyrus activation and parietal cortex activation, independent of reward. Monetary reward did not result in enhanced activation for either condition.

Our findings suggest that monetary reward does not increase activation in the MNS. Instead, it is the social relevance of stimuli that drives involvement of MNS regions. Our findings challenge the MNS theory as a system that automatically responds to social stimuli and emphasizes the role of social relevance.

Dopamine plays a crucial role in psychological and physiological processes, although its involvement in distinct cognitive and affective processes remains partially unclear. For mental imagery (MI) - sensory representations without external stimuli - evidence suggests dopaminergic involvement, shown by dopaminergic hyperactivity linked to hallucinations. In affective processing (AP), findings are heterogeneous: some studies report emotion recognition deficits with dopamine deficiency, others a negativity bias. To examine dopamine's role in MI and AP, we conducted a case-control study. Using a pre-post design with 42 male participants, dopamine synthesis was temporarily reduced in the experimental group (n=21) using an amino acid mixture. Additionally, Parkinson's patients (n=15) served as an extreme group with dopamine deficiency. MI was measured via mental imagery priming and AP through binocular rivalry with emotional faces. Results show that acute dopamine depletion significantly reduced MI strength in the experimental group. These participants showed MI performance comparable to Parkinson's patients, whose MI strength was significantly decreased relative to healthy individuals' baseline. For AP, there were no significant differences between the experimental and control groups, suggesting that the experimental dopamine reduction did not affect emotion perception. However, Parkinson's patients showed significant deficits in the perception of happiness and fear, and more often struggled to discriminate between emotions. Our findings confirm dopaminergic involvement in MI, while the suspected negativity bias from dopamine depletion was not observed in either condition. Future research in our laboratory will analyse prolactin levels and include persons with ADHD as another group with dopaminergic dysregulation.

Neocortical memory engrams are thought to rely on a distributed network of brain regions. However, much remains unknown about how individual components of a complex memory are organized. In the present study, we investigate the formation of memory engrams for basic visual features using a novel behavioral memory task.

Here, healthy young adult participants (planned n=80) are shown line drawings of inanimate objects containing uniformly colored and moving dots. They are asked to learn the association between the object and either the color (color group) or the motion direction (motion group) over five encoding-recall repetitions. To assess the evolution of memory with continued learning, half of the participants complete a daily online training with adaptive task difficulty for one week to achieve precise color/motion memory. We test memory performance and acquire functional and diffusion-weighted MRI data during encoding, after 24 h and after 7 days.

Preliminary results (n=8) indicate clear learning effects across repetitions, reflected in decreased distance to the correct associated color/motion (range:0–15) (rep 1: 3.27±1.24 [M±SD]; rep 5: 1.42±0.36), with comparable difficulty between groups (color: 1.53±0.74; motion: 1.55±0.96). Memory precision increased during the online training to a mean distance of 0.97±0.64 and a color space refined by 27.67±10.59%, similarly in both groups.

Our findings demonstrate that already one session of repeated learning induces basic visual feature memory that can be maintained and refined with continued training, enabling us to track its evolution with multimodal MRI.

Research findings referring to the relationship of sleep duration and memory measurements in school-aged-children are inconsistent. In this study, we used two different sleep measures, a subjective as well as an objective one, to investigate and compare potential relationships between memory performance in a vocabulary-learning task and sleep duration in school-aged children.

Sleep and memory data from 125 participants (eight to 14 years; Mage: 11.19, SD: 1.58) was analyzed. Participants reported their daily sleep duration subjectively via sleep diary (one week). Additionally, a fitness tracker (fitbit Alta HR) was used to estimate sleep duration objectively. At day one, all children learned ten word-pairs (one German, one fictional word) within a word list or embedded in a short story. Memory was tested three times: directly after learning (t1), one day after learning (t2) and one week after initial learning (t3). Memory performance was constant over time (Mt1: 5.89, SD: 3.17; Mt2: 5.94, SD: 3.15; Mt3: 5.98, SD: 3.16). The subjective and objective measures of sleep duration were highly correlated (n = 86; Spearmans ρ = .718, p < .001). Overall, there were no correlations between memory performance at t2 and both subjective and objective sleep duration. By calculating the memory difference scores, however, it was shown that there is a positive relationship between Mdiff (t3-t1) and total sleep time (Night 1 sleep diary; Spearmans ρ = .220, p = .02). This relationship between memory improvement and subjective sleep duration as well as potential reasons will be further discussed.

Weltweit ist ein Anstieg von Autoritarismus und damit verbundenen Repressionsmaßnahmen zu verzeichnen. Die gesundheitlichen Folgen politischer Repression auf Betroffene sind jedoch weitestgehend unerforscht. In der Deutschen Demokratischen Republik (DDR) setzten die Behörden verdeckte (leise) Repressionsmaßnahmen gegen wahrgenommene politische Oppositionelle ein, darunter Überwachung, Rufschädigung und gezielt herbeigeführtes berufliches oder soziales Scheitern. Ziel war es, die psychosoziale Integrität der Betroffenen systematisch zu untergraben und Angst, Isolation sowie Verunsicherung hervorzurufen. Die vorliegende Studie untersucht die Gesundheitsfolgen leiser politischer Repression in der ehemaligen DDR, mit einem Fokus auf psychische Belastung, Schutzfaktoren, systemische Inflammation und zelluläre Alterung. Insgesamt wurden 101 Personen im Alter von 50 bis 78 Jahren rekrutiert. Darunter 50 Personen, die mindestens zwei staatlich organisierte Maßnahmen leiser Repression in der DDR erlebt haben, sowie 51 Personen in einer Kontrollgruppe, parallelisiert auf Alter, Geschlecht und Herkunft. Erfasst wurden unter anderem Depressivitäts-, Angst- und Trauma-Symptomatik, soziale Unterstützung, sowie die Entzündungswerte Interleukin-6 (IL-6) und C-reaktives Protein (CRP) und die Telomerlänge als Indikator für zelluläre Alterung. Bei Teilnehmenden mit Repressionserfahrung zeigte sich im Vergleich zur Kontrollgruppe eine höhere psychische Belastung sowie geringere Werte auf Schutzfaktoren. Zudem wiesen sie signifikant erhöhte Entzündungswerte (IL-6) auf. Im CRP und der Telomerlänge wurden keine signifikanten Gruppenunterschiede festgestellt. Allerdings war innerhalb der Repressionsgruppe weniger wahrgenommene soziale Unterstützung assoziiert mit kürzeren Telomeren. In dieser Studie wurden erstmals psychobiologische Gesundheitsfolgen leiser politischer Repression untersucht. Die Ergebnisse unterstreichen die langfristigen gesundheitlichen Konsequenzen leiser Repression und heben die potenzielle protektive Rolle sozialer Unterstützung bei zellulärer Alterung hervor.

Der Abruf des autobiografischen Gedächtnisses (AuG) aktiviert ein dem Default Mode Netzwerk (DMN) ähnliches Gehirnnetzwerk. Personen mit mesialer Temporallappenepilepsie (PmTLE) weisen Veränderungen der DMN-Konnektivität auf. Ein Zusammenhang mit AuG-Leistungen ist bislang nicht hinreichend untersucht, insbesondere ist unbekannt, ob die DMN-Konnektivität einen Einfluss auf etwaige AuG-Veränderungen nach Resektion des Temporallappens, der Teil des erweiterten DMN ist, hat. Ziel der Studie war die Untersuchung der postoperativen Veränderung der AuG-Leistungen nach Temporallappenresektion in Abhängigkeit der präoperativen funktionellen DMN-Konnektivität.

Es wurden präoperative resting-state fMRT-Daten (Ruhephasen aus geblocktem Aufgaben-fMRT) sowie prä- und sechs Monate postoperative AuG-Leistung (erhoben mit dem Autobiografischen Gedächtnisinterview) von 16 PmTLE nach Temporallappenresektion analysiert. Auf Konnektivitätsebene wurden die Ergebnisse einer Seed-Based-Konnektivitäts-Analyse (Kovariate Alter) von vier DMN-Kernregionen (Medialer Präfrontalcortex [MPFC], Posteriorer Cingulärer Cortex [PCC] sowie linkem und rechtem lateralen Parietallappen [LP]) mit der Veränderung der AuG-Leistung korreliert.

Die Konnektivitätsanalyse ergab zwei marginalsignifikante Cluster in Subsystemen des DMN, die eine Korrelation mit der Veränderung der AuG-Leistung aufwiesen: linker Frontalpol (Clustergröße p-FDR= .082, Voxel p-unc.= <.001) und rechter inferiorer frontaler Gyrus (Clustergröße p-FDR= .088, Voxel p-unc.= <.001). Post-hoc-Analysen ergaben negative Zusammenhänge zwischen MPFC (β= -.12), linkem (β= -0.14) und rechtem LP (β= -.09) und dem linken Frontalpol sowie zwischen linkem LP (β= -.13) und rechtem inferioren frontalen Gyrus. Dabei war höhere Konnektivität mit größerer postoperativer AuG-Verschlechterung assoziiert.

Die Ergebnisse deuten darauf hin, dass die Konnektivität der DMN-Kernregionen mit frontalen Subregionen des DMN relevant für die Vorhersage der Veränderung der AuG-Leistung nach Temporallappenresektion ist. Diese sollte zukünftig unter Berücksichtigung weiterer klinischer Kovariaten beleuchtet werden.

In the domain of action-perception interface, behavioral analyses using conventional methods have yielded highly variable and inconclusive results, with no consistent patterns clearly emerging. Perceptual decisions based on self-induced versus passively presented sensory outcome were reported to be either facilitated or inhibited, thus hindering the theorization of the forward-model mechanisms. To provide a unified perspective on these findings, we conducted a large-scale re-analysis of 14 behavioral datasets examining active versus passive conditions in tasks ranging from intensity judgments to delay detection, across both auditory and visual modalities. Using drift-diffusion modeling, we found that the drift rate parameter during the perceptual decision process across tasks clearly distinguished active from passive trials, particularly for auditory stimuli. While the effect in visual tasks was more variable, meta-analytic aggregation across all studies confirmed a robust, overall advantage in drift rate under active conditions.

We further analyzed three independent datasets comparing patients with schizophrenia to matched healthy controls. While healthy participants consistently exhibited the “active drift-rate advantage,” patients consistently failed to demonstrate this effect, suggesting disruptions in generating or utilizing motor predictions for evaluating feedback of self-initiated actions. These findings indicate potential impairments in sensorimotor predictive mechanisms in schizophrenia.

Together, our results provide a computationally grounded and unified account of the action-perception interface, clarifying previously ambiguous behavioral findings. They also highlight important clinical implications for disorders characterized by disturbances of agency. This integrative approach underscores how drift-diffusion modeling can illuminate both basic cognitive mechanisms and deficits in clinical populations.

Recent epidemiological data show that 2.6% of adolescents in Germany meet diagnostic criteria for cannabis use disorder, with higher prevalence in older adolescents and males. Ongoing brain development during this stage makes adolescents especially vulnerable to cognitive impairments linked to SUDs—such as difficulties in memory, attention, inhibition, and decision making—that may interfere with learning and daily functioning. This study examined cognitive functioning in adolescent chronic cannabis users (CCU), aiming to: (1) Examine whether verbal episodic memory, working memory, verbal comprehension, processing speed, and overall cognitive functioning differ between controls and chronic cannabis users (CCU); (2) determine whether group differences remain after accounting for socioeconomic factors; and (3) explore whether age of onset and duration of cannabis use predict cognitive outcomes within the CCU group. Participants included 40 controls (mean age 15.4) and 27 CCU adolescents (mean age 16.0) from the DELTA study. Groups did not differ significantly in age or sex. Cannabis users were identified using specific cut-off criteria based on usage frequency. All participants completed the VLMT and WISC/WAIS subtests. No significant differences were found between groups in VLMT scores. However, one-way MANOVA revealed a significant group effect on combined WAIS/WISC indices (Pillai’s Trace = 0.30, F (4, 49) = 5.26, p = .001). Post-hoc tests showed lower scores in the CCU group for Verbal Comprehension (p = .001), Processing Speed (p = .024), and General IQ (p < .001). These findings suggest selective cognitive impairments in adolescent cannabis users, despite intact verbal episodic memory.

Oxytocin spielt eine Rolle bei sozialen Prozessen, wodurch ein Einfluss auf Stress und Empathie denkbar ist. Ein Singlenukleotid-Polymorphismus (SNP) im Oxytocin-Rezeptor-Gen (rs53576) (OXTR) könnte diese Beziehung beeinflussen. Es wird angenommen, dass homozygote G-Allel-Tragende sensibler auf soziale Stimuli reagieren, was einerseits zu einer stärkeren Stressreaktion (Anstieg von negativem Affekt) und andererseits zu einer stärkeren Zunahme von Empathie im Vergleich zu A-Allel-Tragenden führen könnte.

An der Untersuchung nahmen N = 140 Probanden teil, davon 82 weiblich und 58 männlich, im Alter von 18 bis 43 Jahren. Zur Stress-Induktion wurde der TSST verwendet, während Empathie durch eine Filmsequenz aus dem Film „Bambi“ induziert wurde. Die Genotypisierung beruhte auf Epithelzellabstrichen. Zur Messung des negativen Affekts, wurde die Positive and Negative Affect Schedule verwendet. Diese wurde um vier Items ergänzt, um Empathie zu bestimmen. Die Manipulation-Checks für Stress- und Empathie-Induktion waren erfolgreich.

Es zeigte sich in Bezug auf den negativen Affekt ein signifikanter Interaktionseffekt (Geschlecht, OXTR SNP, Stress- vs. Kontrollbedingung). Männer mit dem GG-Genotyp wiesen in der Stressbedingung höhere Werte im negativen Affekt auf, während sich bei Frauen dieser Effekt nicht zeigte.

In Bezug auf die Empathie-Werte zeigte sich ebenfalls ein signifikanter Interaktionseffekt (Geschlecht, OXTR SNP, Stress- vs. Kontrollbedingung). In der Kontrollbedingung wiesen Frauen mit dem GG-Genotyp höhere Empathie-Werte auf als Frauen mit A-Allel, während sich dies bei Männern in der Stressbedingung zeigte.

Hauptlimitationen der Untersuchung waren Varianzeinschränkungen aufgrund einer relativ homogenen Stichprobe und eines stark wirkenden Induktionseffekts von Empathie. Die Ergebnisse werden vor dem Hintergrund komplexer Interaktionen zwischen Genpolymorphismen, Umwelterfahrungen (Stress) und Geschlechtsdifferenzen diskutiert.

Transcranial alternating current stimulation (tACS) is a promising non-invasive approach for alleviating tinnitus symptoms. By delivering alternating currents between two electrodes, tACS targets the auditory system to reduce the persistent phantom sound experienced by patients. However, its clinical application remains in its early stages and is hindered by stimulation-induced side effects, including phosphenes, muscle twitching, and tingling sensations on the skin.

The tACSitus project investigates various electrode configurations, including a conductive earplug placed in the ear canal, to optimize stimulation tolerability and effectiveness. We systematically assess two key experimental variables: the magnitude of side effects and the occurrence of a hearing sensation. The hearing sensation—a soft pure tone perceived near the stimulation frequency—serves as an indicator that the electrical current successfully reaches auditory structures. In our study, normal-hearing participants receive stimulation at different frequencies with stepwise increases in amplitude. Side effects are evaluated at the first occurrence of a hearing sensation, enabling a controlled comparison of tolerability and efficiency across electrode montages.

Preliminary findings indicate that an ear-to-ear montage using two earplugs significantly reduces side effects compared to conventional configurations, while still reliably inducing a hearing sensation. These results support the feasibility of ear-to-ear tACS as a refined stimulation approach that minimizes discomfort while maintaining precise auditory engagement. This represents an important step toward developing a clinically viable stimulation protocol for tinnitus treatment. Future studies will further investigate its therapeutic potential in tinnitus patients.

The seminal theory of motivational conflicts distinguishes three types of conflict: approach–approach (AP-AP), where a decision is made between desirable alternatives; avoidance–avoidance (AV-AV), where a decision is made between undesirable alternatives; and approach-avoidance (AP-AV), where a single choice carries both positive and negative outcomes. While behavioural differences between these conflicts are well-documented, their neural underpinnings remain largely unexplored. Here we contrast these three conflicts using a novel paradigm and EEG data. Both of our behavioural and neural results show that AV-AV conflicts are hardest to resolve. Specifically, we show that midfrontal theta, a well-established neural marker of conflict, differentiates between these conflict types. AV-AV conflicts trials have highest mid-frontal theta response while AP-AP conflicts are greater than AP-AV. We also find similar trend in beta suppression activity across three conflicts. These findings provide neural evidence that AP-AP, AV-AV, and AP-AV conflicts are distinct and establish midfrontal theta as a marker of conflict intensity in subjective decision-making.

Previous studies indicate that emotion regulation (ER) strategies differ in their effectiveness at reducing various components of an emotional response. In this study, we compare the effects of three different emotion regulation strategies (reappraisal, acceptance, and expression suppression) on self-reported affect and facial electromyography (EMG) data from a non-clinical sample of young adults in an ER implementation task. Participants were instructed to implement either of the strategies while being presented with negative emotional stimuli of differing intensity (low and high intensity pictures from the Nencki Affective Picture System, NAPS). We expected all three emotion regulation strategies to lead to a reduction in (a) negative affect and (b) corrugator activity compared to a no-regulation control condition (“look”). While data collection is ongoing (target sample size N = 200), preliminary findings (n = 137) suggest that only reappraisal improved affect ratings, whereas suppression and acceptance lead to an aggravation of negative affect. In contrast, we observed no significant effects of ER strategies on corrugator activity. The poster will discuss how the intensity of the stimuli used for emotion induction and the specific instructions given for the implementation of the regulation strategies may shape ER outcomes across subjective and physiological measures.

The thalamus is a central generator for sleep rhythms, such as sleep spindles, that facilitate stimulus processing and support memory consolidation during sleep. In this study, we examined the effect of thalamic lesions on spatial learning, sleep architecture and EEG correlates indicating memory processing during sleep. For that, we recruited patients suffering from essential tremor who underwent unilateral focused ultrasound thalamotomy (FUS). The experiment involved an adapted virtual version of the Morris Water Maze, developed to assess both hippocampus-associated learning and consolidation mechanisms as well as striatum-associated reversal learning. The experimental procedure was conducted pre- and post-surgery including learning, immediate recall and reversal learning in the evening as well as a recall phase in the morning after a night of sleep, respectively.

Preliminary data analysis indicates that patients showed differences in hippocampus-associated place learning while reversal learning performance remained the same between conditions. Despite the lesion, patients consolidated the newly learned reversal information during sleep. Importantly, EEG analysis showed a significant decrease in both the percentage of spindles coupled to slow oscillations and the overall number of spindles during the night following surgery.

In summary, our findings show that unilateral thalamic lesions do not seem to influence behavioral performance of hippocampus- and striatal-associated spatial memory formation whereas sleep exhibits a supporting role on both memory systems. In contrast to our behavioral findings, lesions of the thalamus impact electrophysiological correlates of thalamic contributions to memory-associated signal processing.

Pain assessment traditionally relies on self-reporting, yet physiological signals offer potential for objective measurement. Current research predominantly addresses intensity estimation, neglecting temporal dynamics of pain fluctuations. Developing methods to objectively detect pain dynamics, especially decreases, could create an illusion of control. By prompting patients to initiate treatments precisely when pain is naturally decreasing, they can attribute relief to their action rather than spontaneous fluctuation. This enhanced perception of control could help prevent pain chronification. Our study presents a multi-modal approach using physiological signals to identify pain decreases.

We recorded multiple psychophysiological signals (electrodermal activity, pupillometry, photoplethysmography, electroencephalography, and facial expressions) from 42 healthy participants (23 females, mean age: 26.2 years ± 5.1 SD). To evoke pain fluctuations, participants experienced calibrated heat pain (0-70 on visual analog scale) with unpredictable temperature curves. Temperature increases/decreases lasted 5-20 seconds each. For future real-time applications, we employed minimal preprocessing. Using 5-second time windows, we framed the problem as a binary classification differentiating major pain decreases from increases (2826 samples total).

Comparing state-of-the-art deep learning architectures, a transformer-based model achieved the highest classification accuracy (78.4%) on our held-out test set (738 samples). Electrodermal responses and pupillometry provided the strongest predictive signals, consistent with previous research on autonomic correlates of pain. This work demonstrates the potential for detecting pain decreases from physiological signals with good accuracy. However, important limitations include the laboratory setting, healthy participant sample, and focus on experimentally induced heat pain rather than clinical pain.

Pupillometry is commonly used to assess cognitive states, but the comparability of results is limited by variability in data-processing methods. It is currently unclear which pre-processing method yields the most accurate results, partly because of a lack of large benchmark data sets. To address these challenges, we introduce (a) the PupilFear corpus, a curated and standardized fear-conditioning dataset with pupil size recordings (N = 399), and (b) the CaliBench engine, a tool for automated benchmarking of data pre-processing methods. In a calibration approach, we evaluate commonly used preprocessing techniques on the PupilFear corpus by their retrodictive validity, i.e. their ability to reproduce the experimental manipulation. We find that data trimming, excluding loss of fixations, and filtering implausible samples, improves retrodictive validity. In a psychophysiological modelling (PsPM) approach, condition-wise modeling yielded better results than trial-wise modeling, due to non-estimable parameters when trials contain missing data due to blinks or loss of fixation. For both trial-wise and condition-wise estimations, not including the CS-derivative in the model yielded consistently better results. With this, we give updated recommendations for pupillometry in cognitive experiments. Furthermore, we anticipate that the standardized PupilFear dataset and the CaliBench engine will be useful tools in further methodological investigations.

Effective treatments exist for anxiety disorders, but relapse is common, especially after life adversity. Relapse can be modeled using differential fear conditioning, in which the unconditioned stimulus reoccurs unsignaled after extinction (reinstatement, RI). Previous studies suggest that recent life adversity reduces the ability to discriminate between danger and safety cues (i.e., generalized RI), but longitudinal data are still lacking. One hundred and twenty healthy participants, selected for the absence of childhood adversity, underwent fear acquisition, 24-hour delayed extinction, and reinstatement (T0). After six months, 95 participants returned to complete the same procedure (T1). Skin conductance responses (SCRs), fMRI data, fear ratings, and reports of recent life adversity (from the past six months) were collected at both time points. Our analyses indicate that participants exposed to life adversity during the T0-T1 interval displayed generalized reinstatement (RI) in skin conductance responses (SCRs), while unexposed participants showed differential RI (i.e., CS-specific). A similar pattern was observed for fear recall. These group differences were associated with corresponding activation differences in brain regions involved in fear processing, such as the insula and the amygdala, but were not reflected in fear ratings. As the inability to maintain discrimination between danger and safety cues is linked to pathological fear, our findings - when translated to clinical practice - highlight that recent life adversity may serve as a significant risk factor for clinical relapse. Identifying inter-individual differences that influence relapse could ultimately enable the development of personalized interventions, promoting long-term remission.

Social anxiety disorder (SAD) is a prevalent mental health condition in children, characterized by excessive anxiety, social avoidance, and fear of negative evaluation. Cognitive models of adult SAD suggest that internal (heightened self-focused attention) and external (hypervigilance-avoidance-pattern of threat cues) attention biases play a key role in the disorder’s maintenance. Studies on children, however, are limited with inconsistent results. In addition, research has focused either on internal or external attention biases, whereby an interplay can be assumed.

Our aim was to investigate internal and external attentional biases in children with SAD simultaneously in anticipation of a social stress task by using a multimodal assessment.

We assumed that children with SAD show greater self-focused attention, as indicated by faster reaction times to internal stimuli, and exhibit initial hypervigilance towards a potentially threatening stimulus (audience with neutral facial expression), and avoidance of the stimulus as the task progresses, as assessed through gaze patterns.

Participants aged 9-14 years with SAD (n = 46) and healthy controls (n = 46) were exposed to a speech task to induce anxiety while attentional biases were assessed through an reaction time experiment and mobile eye-tracking measures in anticipation of the task.

No significant differences were found between the groups in their reaction times and eye-tracking data. Having found increased attention to internal and external cues in previous studies using self-reports, the results should be discussed for future research to determine whether the use of reaction time paradigms and eye-tracking is relevant in the context of child SAD.

Variations in cardioafferent traffic are relayed to the brain and have been shown to modulate sensorimotor processing. Stress potentially influences these effects via both humoral and neuronal pathways altering cardiovascular activity and modulating areas of the central nervous system. We investigated the interaction of cardiac-cycle and stress effects by employing a visual change detection task to disassociate response and attentional inhibition processes. 75 participants completed four experimental blocks, responding to lateral luminance changes while ignoring simultaneous orientation changes that occurred either isolated (catch condition), in the same (facilitation condition) or opposing hemifield (conflict condition). Stimulus presentation was aligned to the ECG, targeting systole versus diastole i.e., phases of high versus low cardioafferent activity. Subjects were randomly assigned to a cold pressor test (CPT) or control condition and were exposed before each experimental block. The CPT increased all stress response parameters. We found the expected task effects i.e., higher response times and percentage errors in the conflict versus all other conditions. In the CPT group, systole trials showed more errors and decreased misses in conflict trials. For unilateral luminance changes, systole trials showed more misses which was accompanied by frontal alpha/beta lateralization. Our results are in line with previous findings reporting inhibitory effects of cardioafferent traffic on perceptual processes as well as facilitation of response related processes but suggest that these cardioafferent effects depend on the visual context and possibly act via attentional mechanisms. Stress seems to influence these effects potentially via altered autonomic modulation.

Sleep provides an optimal environment for memory reprocessing, free from external interference. Research has shown that through targeted memory reactivation (TMR), we can harness the brain's sleep-dependent mechanisms for processing of new memories. However, it remains unclear which of the oscillations involved in nighttime memory reprocessing are essential, or sufficient, for this process to unfold. To investigate this, we designed a visual memory task using images from two different content categories (faces and houses), each associated with distinct tones that were repeatedly presented during the learning phase. TMR was then applied during the night to bias the preferential reprocessing of one content category, providing a temporal framework to study the dynamics and oscillatory activity patterns involved in neural reactivation. Using a support vector machine (SVM) classification approach, we successfully decoded content category reactivation during deep NREM sleep (stage 4). We then examined the distinct roles of slow waves and spindles in this process by dividing TMR trials into those with high and low power of spindles and slow waves, respectively. Both slow waves and spindles enabled content classification. To assess whether either of these oscillatory hallmarks alone is sufficient for reactivation, we further analyzed all trials with low power, high power, and, finally, those with exclusively high slow wave or spindle power. We found that TMR successfully reactivated memory content only when both slow waves and spindles were present, highlighting that the interplay between these sleep oscillations is essential for memory reactivation during sleep.

Every time we learn a new skill or store an experience into memory, our brain undergoes functional and structural changes. Advances in neuroimaging, such as diffusion-weighted MRI (DW-MRI), enables to non-invasively track microstructural changes by analyzing the motion of water molecules within human brain tissue. Recent studies have demonstrated that repeated encoding and retrieval of an object–location association task leads to the rapid formation of a physical memory trace in the parietal cortex. Notably, this trace fulfills all engram criteria within just 90 minutes post-learning, highlighting the dynamic and rapid nature of memory formation. If microstructural memory traces differ depending on learned information remains an unanswered question.

In this study, 79 participants completed an object-location learning task with repeated encoding and retrieval of image pairs depicting either objects (n=40) or scenes (n=39). DW-MRI was measured immediately after the learning task until up to 60 minutes later.

We identified material-specific brain regions with an independent functional localizer task. Material-specific activity was found in early visual areas, lateral occipital cortex, parahippocampal areas and regions of the cuneus and precuneus. We find a general decrease in mean diffusivity (MD) in the cerebellum and bilaterally in the superior division of the lateral occipital cortex. Preliminary analyses suggest object-specific MD decreases in the precuneus region, and scene-specific decreases in the lingual gyrus, the occipital pole, right superior lateral occipital cortex and angular gyrus.

While it is well established that sleep benefits memory encoding and, particularly, memory consolidation, the role of sleep in memory retrieval remains unclear. According to the Synaptic Homeostasis Hypothesis (SHY), down-selection of weak synapses during sleep desaturates the brain’s ability to encode new information and improves memory consolidation by increasing signal-to-noise ratios. Based on SHY, we hypothesized that sleep-dependent changes in synaptic strength continue after the consolidation period, potentially contributing to processes improving memory retrieval following sleep. To investigate the role of sleep in memory retrieval, healthy participants took part in two conditions in a within-subjects design. Following encoding of new information, they spent a one-week consolidation period at home and then returned to the sleep laboratory for retrieval testing, which took place either following an afternoon nap or an equivalent period of daytime wakefulness. Additionally, we recorded polysomnography and wake electroencephalography during and following the sleep/wake manipulation to investigate electrophysiological correlates of sleep-dependent memory retrieval. Our preliminary behavioral results (n = 20) indicate a beneficial effect of sleep on memory retrieval in both declarative and procedural memory tasks. Particularly, we found an effect of daytime napping on retrieval of word-pair associations and their accessibility from long-term memory. Furthermore, we found a significant sleep-dependent improvement in correctly typed sequences in a procedural finger-tapping task. Together, our preliminary results provide first evidence for an active role of sleep in retrieval of previously encoded memories.

Individuals suffering from obesity show increased risk taking behaviour and increases in impulsivity, traits shared with individual suffering from disordered gambling as well as other forms of addiction. In the current study, we tested obese individuals (N=42) and a lean control group (N=40) on a realistic slot machine simulation (Paliwal et al., 2014), to investigate potential behavioural differences linked to metabolic effects. Trait impulsivity was quantified via the Barratt Impulsiveness Scale (eleventh revision; BIS-11). To account for behaviour during task performance (e.g. bet increases, machine switches, casino switches, response times), we use computational modelling. Specifically, we compare the hierarchical Gaussian filter (HGF) of Daunizeau and colleagues (2010), a model that proved to be successful in previous related work, to the volatile Kalman filter (VKF) of Piray and Daw (2020) using hierarchical Bayesian modelling. We use model comparison to identify the best-fitting combination of perceptual (encoding of the outcome) and response variables, both in terms of fit to the observed data, and in terms to their accuracy in predicting participants’ trait impulsivity via a general linear model with the models’ estimated parameters as predictors and the BIS-11 scores as dependent variable. Additional modelling analyses will focus on group differences in post reinforcement pauses, as a behavioural measure of reward sensitivity in gambling.

Background: Interventions for anxiety disorders have been implicated with changes in neural activity in frontal structures associated with salience processing during cognitive or affective tasks. The extent to which an intervention alters brain function when task-based salience mapping is not required is largely unclear. The aim of this study was to investigate intrinsic changes in the task independent resting state before and after a therapeutic intervention.

Methods: N = 72 individuals with a DSM-5 diagnosis of spider phobia were assessed using resting-state functional magnetic resonance imaging before and after a single exposure treatment of three hours duration. A data-driven, voxel-to-voxel-based functional connectivity multi-variate pattern analysis (fc-MVPA) was used to address the question: Where in the brain does connectivity differ before and after treatment?

Results: The fc-MVPA identified a cluster in the right occipital fusiform gyrus (rOFus) as a region of significant change between pre- and post-treatment. Subsequent seed-based analysis revealed stronger connectivity of the rOFus with numerous other visual regions and stronger anticorrelation with regions in the frontal pole, paracingulate cortex and angular gyrus in the pre- to post-treatment contrast.

Discussion: Results are consistent with previous research using voxel-based measures identifying the OFus in anxiety disorders. Reduced connectivity after treatment may indicate desensitisation in early perceptual regions. A decrease in anticorrelation to frontal regions may facilitate top-down control of visual perception. However, the pre-post design lacks a control group, so non-specific pre-post effects cannot be ruled out.

Transcutaneous auricular vagus nerve stimulation (taVNS) is an emerging neuromodulation technique with promising applications in psychophysiological research and clinical interventions. By targeting the vagus nerve, taVNS is thought to influence autonomic regulation, emotion, and cognition, yet the optimal stimulation parameters remain debated. Identifying the most effective configurations is critical for maximizing its therapeutic potential. This study systematically examined how taVNS parameters affect autonomic responses and perceptual experience. Forty-two participants received taVNS at the cymba conchae and sham stimulation at the earlobe across two lab session scheduled at least one week apart. They were stimulated with nine distinct stimulation protocols that varied in frequency (low: 10Hz, medium: 25Hz, high: 50Hz) and pulse width (100µs, 200µ 300µs). Results revealed an immediate drop in heart rate and an increase in skin conductance following stimulation onset in both taVNS and sham conditions. Crucially, pulse width significantly modulated these autonomic changes whereas frequency effects were less pronounced. Moreover, higher pulse widths were consistently perceived as more intense and painful. These findings highlight pulse width as a key factor shaping both physiological and perceptual responses to taVNS. Future studies should further explore how stimulation parameters optimize therapeutic efficacy and patient comfort.

Positive peer group dynamics in childhood are a vital part of healthy development. However, little is known about how group dynamics emerge and whether physiological factors play a role in the development of peer relationships. Within interactions, social connectedness is characterized by interdependent physiological processes, specifically physiological synchrony. Measuring physiological synchrony in peer interactions may thus relate to the emotional and social connections that children experience. To test this, we collected heart rate data using Polar H10 sensors from N=141 5th- and 6th-graders (age range 10–13 years) and determined heart rate synchrony (in N=251 dyads). Children from the same classroom attended experimental sessions in small groups (4–6 individuals). Following a baseline measurement of heart rate, they listened to a story about social exclusion at school and discussed questions of morality and fairness exemplified through the story characters. Multilevel models related synchrony (calculated as cross-wavelet power across time and high and low frequency bands) to dyadic friendship status, feelings of closeness, and momentary emotional state. Results indicate that friends exhibited increased heart rate synchrony compared to non-friends or one-sided friendships (β=0.25; p<0.01 for high frequencies; β=0.26; p<0.01 for low frequencies). Further results will be presented at the conference. A physiological connection between friends could give insights into the regulatory benefits of friendship. Monitoring how synchrony changes over time could indicate how peer group dynamics develop and inform interventions in fostering a positive classroom climate.

While the role of the neurotransmitter acetylcholine in arousal, attention and memory functions is well-established, how the molecular and cellular architecture of this system relates to variation in macroscale grey matter morphology remains poorly understood. The current study combined cytoarchitectonic maps of the basal forebrain - the primary source of cortical acetylcholine - with normative PET radioligand binding maps for the vesicular acetylcholine transporter, M1-muscarinic and α4β2-nicotinic receptors, to assess the correspondence in grey matter variation between cholinergic seed and target regions. T1-weighted MRI data of three independent cohorts were preprocessed using the Computational Anatomy Toolbox to derive normalized grey matter probability maps. Basal forebrain grey matter was extracted based on cytoarchitectonic masks for medial septum/diagonal band of Broca (Ch1-3) and basal nucleus of Meynert (Ch4) subregions. Structural covariance was examined using general linear models, while controlling for effects of age, sex and total intracranial volume. The strongest voxel-wise effects for basal forebrain structural covariance were in basal ganglia, thalamus, hippocampus and amygdala. Correspondence between subregional (Ch1-3/Ch4) grey matter covariance and (muscarinic/nicotinic) receptor subtype topographies showed both, between- and within-region effects, which validates seed and target region markers for cholinergic system integrity. Explorative analyses on ageing effects and nicotine dependence further quantify involvement and variability of cholinergic neurotransmitter system markers. Overall, our results show that readily indexed grey matter covariance patterns reflect underlying molecular architecture, suggesting that grey matter metrics may serve as a valuable and accessible proxy for assessing cholinergic integrity in studies of health and disease.

Background: Schizotypy is a multidimensional construct of schizophrenia-like personality traits, indicative of psychosis proneness. Previous studies reported reduced gray matter volume (GMV) in hippocampal regions in schizophrenia and depressive disorder, but also in association with schizotypy in healthy individuals. Here, associations between schizotypy dimensions and GMV in the hippocampus, its subfields and additional regions were investigated in a healthy population, hypothesising association patterns similar to those seen in schizophrenia.

Method: In N=928 healthy participants (mean age=34.28, 64.2 % female) from the FOR2107 cohort, partial correlations were conducted (controlled for age, sex, total intracranial volume, site characteristics) between facets of the Schizotypal Personality Questionnaire Brief (SPQ-B, positive, negative, disorganised, total score) and extracted regions of interest GMV. Furthermore, interaction effects between these dimensions were tested in multiple regression models.

Results: We found a significant positive correlation between right mammillary body GMV and negative (r=0.09,p=0.005,pcorr=0.047) and total (r=0.09,p=0.007,pFDR=0.047) schizotypy. Additionally, a nominally significant (not reaching FDR correction thresholds) negative association was found between left fornix GMV and positive schizotypy (r= -0.07, p=0.0352, pFDR=0.123) and left alveus GMV and disorganised schizotypy (r=-0.07, p=0.0301, pFDRr=0.12). None of the interaction effects between schizotypy dimensions were statistically significant.

Conclusion: Dimension- and subfield-specific associations between schizotypy and parts of the (extended) hippocampal system are reported. This indicates the importance of analysing correlates of complex personality constructs such as schizotypy on a facet level, as well as considering morphologically distinct subregions. Such patterns, if disregarded, might also contribute to null-findings and underestimate associations between subclinical phenotypes and brain structure.

Have you ever seen someone visibly disappointed about a gift? Affective reactions offer cues about how others anticipate and interpret events. From a computational perspective, these reactions reflect prediction errors—a core reinforcement learning concept that refers to the gap between expected and actual outcomes.

Here, we use a reinforcement learning lens to investigate how people infer hidden expectations from observed affective reactions. We further probe whether these affect-based inferences shape broader social impressions of the observed individual.

In two preregistered online studies (N = 553), participants observed two hypothetical demonstrators reacting to lottery outcomes. The demonstrators were designed to have higher and lower hidden expectations, respectively. Their affective reactions deterministically corresponded to prediction errors, with trial-by-trial updating governed by a fixed learning rate. On each trial, participants viewed both the demonstrator’s outcome and their reported affective reaction. Based on this information, participants then estimated what the demonstrator had expected to receive on that trial.

As hypothesized, participants systematically inferred higher expectations for the high-expectation demonstrator. Moreover, model comparisons suggest these inferences reflect dynamically updated—rather than static—expectations.

In an ongoing study using the same task, we examine whether affective responses also shape social impressions. To this end, participants additionally rate each demonstrator on attributes such as social status, competence, and wealth.

Together, these studies provide computational insights into the relationship between observed affect and observers’ inferences. As such, they hold promise for a more granular understanding of how we form beliefs about others in everyday social interactions.

According to Aaron Beck’s cognitive model, depressed individuals show a negative bias. To find electrophysiological correlates of this bias, we conducted a facial oddball task where participants saw a stream of happy, neutral, and sad faces, and they had to rate the valence of the faces. Unknown to the participants, 80% of the faces where in one gender, while 20% were in the other gender. In a previous study with 300 trials, we found that participants with high depressive symptoms showed larger N170 amplitudes after rare sad faces, while healthy participants showed larger N170 amplitudes after rare happy and neutral faces, compared to their frequent counterparts. In the current study, we recruited 121 participants to replicate this result in a German sample. Contrary to the previous study, we used 375 instead of 300 trials, and we took the happy, neutral, and sad pictures from five female and five male persons (instead of one female and one male). We present both behavioral and electrophysiological results regarding the N170.

Early life stress is known to be associated with both stress system dysregulation and emergence of depressive symptoms later in life. Here, we investigated to what extent aberrations in stress processing mediate the increased susceptibility to depression due to early life stress. A total of 350 participants (~50% female) were repeatedly exposed to the Cold Pressor stress test. Salivary cortisol levels, cardiovascular responses, and subjective ratings were assessed to quantify acute stress reactions and recovery. Early life stress, perceived chronic stress burden, and depression symptoms were assessed via questionnaires using the CTQ, TICS, and BDI, respectively. Mediation and moderation analyses were conducted to assess how early life stress contributes to symptoms of depression. Consistent with prior research, our results showed that early life stress was associated with blunted cortisol reactivity and reduced cardiovascular responses to acute stress as well as greater depressive symptoms. Importantly, we found that chronic stress mediated the impact of early life stress on depression indicating that increased susceptibility to depressive symptoms was strongly dependent on perceived chronic stress. Moreover, moderation analyses revealed that blunted physiological reactivity to acute stress significantly influenced the strength of this association. Furthermore, altered acute stress response patterns mediated the link between early life stress and chronic stress. These findings highlight the enduring impact of early life stress and suggest that changes in how we deal with stress are an important mechanism that can explain how childhood adversity contributes to psychopathology later in life.

To map the signals that arrive in our senses to perceptual objects our brain makes use of predictions: our prior beliefs on the content of the sensory world. Predictive coding hypothesises that sensory neurons encode prediction error: the difference between the predictions and the observations. By combining the predictions with the errors, our brain can generate an internal representation of what is out there.

There are two functionally different types of prediction errors: positive errors add unpredicted objects to the representations; negative errors remove incorrectly predicted elements from the representations.

Here we studied, for the first time, the encoding of negative prediction error in the human auditory pathway. We presented participants with predictable sequences of tones. We introduced rare omissions of one tone to elicit negative prediction error and use abstract rules to manipulate the expectations of the participants on the omissions.

The results show that unexpected omissions elicit robust activation in auditory thalamus and cortex; and that the omission responses are inversely related to the expectations of the participants on the omissions, confirming that the responses encode negative prediction error.

Most importantly, the results show that auditory midbrain does not robustly encode negative prediction error, although several previous studies had shown that general prediction error is robustly encoded in the human auditory midbrain.

We conclude that positive and negative prediction error are encoded by physiologically distinct circuits in the human auditory pathway. This functional distinction may be key for the study of hallucinations in psychosis.

Alcohol use disorder (AUD) is a major contributor to global disability and mortality. Cross-sectional studies have linked AUD to reduced cognitive control, heightened risky decision-making, and increased stress. However, little is known about the interplay of these variables in real life.

Here, we deployed several smartphone-based, gamified tasks in a one-year longitudinal ecological momentary assessment study of N = 288 participants diagnosed with mild to moderate AUD. Tasks measured cognitive control (working memory capacity [WMC]; response inhibition) and decision-making (risky decision-making; information sampling). Participants completed tasks monthly and reported alcohol consumption and stress daily.

We found that monthly fluctuations in risky decision-making predicted subsequent alcohol consumption. Specifically, increased risk-taking and decreased information sampling were related to higher alcohol consumption in subsequent months (b = 1.07, p = .008; b = –0.90, p = .017). Between-participant differences in WMC moderated these effects (b = –1.03, p = .022), with higher WMC mitigating the impact of risky decision-making on drinking. WMC further moderated the adverse effects of stress on drinking (b = –0.95, p = .018)—a moderation effect that was also reflected in differences in dorsolateral prefrontal cortex (dlPFC) grey matter volume, as higher dlPFC volume similarly mitigated stress-driven drinking (b = –1.38, p = .005).

Our findings offer novel insights into the cognitive-behavioral forces driving changes in alcohol consumption: decision-making alterations precede changes in alcohol consumption, and WMC as well as dlPFC volume act as protective factors. This suggests that smartphone-based gamified tasks can be used to identify periods of heightened risk, paving the way for mechanism-based, just-in-time interventions in AUD.

The hippocampal ripple is one of the cardinal oscillations that facilitate sleep-dependent memory consolidation. The numerous methodologies used to detect ripples across species have reported highly variable ripple characteristics. This variability sparked an ongoing debate that calls for a consensus on what constitutes a “true” ripple, thus, stressing the necessity to standardize ripple characteristics and detection methods. Importantly, ripple detection frequently utilizes a fixed amplitude threshold criterion for the signal envelope of band-pass filtered data (~80-150 Hz) performed against background activity. This methodology may introduce a detection bias, as background activity - frequently termed aperiodic activity – can be affected by changes in brain state. Background activity can be estimated by the spectral exponent of the power spectrum in log-log space, and delineates sleep stages as well as cognitive engagement. Combined, these observations raise the question whether ripple detection is affected by underlying changes in the spectral exponent values of the data, thus, potentially confounding high-frequency events for “true” ripples. In this study, we use both experimental and simulated data to demonstrate that ripple detection varies across brain states as a function of the spectral exponent of the data, and show that ripple detection is biased towards data with the highest spectral exponent. These findings offer a parsimonious explanation why ripple density increases as a function of cognitive engagement. Crucially, these findings emphasize the importance of conducting simulation-based inference to determine if a change in ripple density can be explained by a change in the spectral characteristics of the data.

Bei Personen mit Temporallappenepilepsie (PmTLE) kann es nach einem epilepsiechirurgischen Eingriff, insbesondere nach Resektion in der sprachdominanten Hemisphäre, zu Verbalgedächtnisverschlechterungen kommen. Bislang wurde unzureichend untersucht, inwiefern sich die Gedächtnisleistungen nach initialer postoperativer Verschlechterung weiterentwickeln. Aufgrund diskrepanter vorangegangener Studienergebnisse zu geschlechtsabhängigen Veränderungen nach Epilepsiechirurgie ist insbesondere die Frage nach Geschlechtseffekten auf die postoperative Funktionsentwicklung von Interesse. Ziel der Studie ist eine Analyse der Geschlechtseffekte auf die postoperative Entwicklung der Gedächtnisfunktionen nach initialer Verschlechterung.

Retrospektiv wurde die verbale Langzeitgedächtnisleistung (Verbaler Lern- und Merkfähigkeitstest) von 37 PmTLE zu drei verschiedenen Zeitpunkten (präoperativ (T1), 6 (T2) und 24 Monate postoperativ (T3)) ausgewertet. Es wurden ausschließlich PmTLE mit initialer Verschlechterung (T1-T2) nach Temporallappenresektion (Epilepsie-Zentrum Bethel; 2014-2021) in der sprachdominanten Hemisphäre berücksichtigt. Das Geschlechterverhältnis wurde in Subgruppen (weitere, spätere Verschlechterung: ja (n=5) vs. nein (n=32); spätere Verbesserung: ja (n=8) vs. nein (n=29)) analysiert . Mithilfe einer Repeated-Measures-ANOVA wurden Effekte von Zeitpunkt und Geschlecht auf die Verbalgedächtnisleistung analysiert.

PmTLE mit weiterer, späterer Verschlechterung waren ausschließlich männlich (5/5 vs. 17/32, p=0.05), während die Gruppe mit späterer Verbesserung überwiegend weiblich war (5/8 vs. 10/29, p=0.15). Die Repeated-Measures-ANOVA ergab eine Interaktion zwischen Zeitpunkt und Geschlecht (p<.001). Weibliche PmTLE zeigten bessere Gedächtnisleistungen zu T3 (p=0.04). Zu T1 oder T2 gab es keine Geschlechtsunterschiede.

Die Ergebnisse beschreiben erstmalig einen Geschlechtseffekt nach initialer Verbalgedächtnisverschlechterung bei sprachdominant-operierten PmTLE. Es zeigte sich eine deutliche Funktionserholung im postoperativen Verlauf bei weiblichen PmTLE, welche auf geschlechtsspezifische dynamische Prozesse neuronaler Erholung hindeutet und interessante Hinweise auf eine höhere Neuroplastizität bei weiblichen PmTLE liefert.

Introduction: Individuals with autism spectrum condition (ASC) exhibit atypicalities in perceptual category acquisition, i.e. categorizing stimuli based on holistic visual appearance, which could contribute to social symptoms. Suggested mechanisms might be problems abstracting a category’s central tendency ("prototype"), or decreased processing of unattended stimuli.

Methods: We investigated visual category acquisition in ASC compared to neurotypical controls (NT) using an A/hidden-B paradigm. The stimulus set comprised of polygon Subsets 1 and 2, generated by statistical deviations from a prototype. Assignment of the Subsets to the target category was counterbalanced. Bayesian computational modelling and functional magnetic resonance imaging (fMRI) provided insights into the underlying mechanisms.

Results: In the training phase, a three-way interaction revealed that ASC (n = 30) categorised members less accurately than NT (n = 72) when Subset 1 was the target, and non-members less accurately when Subset 2 was the target. Preliminary fMRI analyses showed stimulus-specific group differences while processing negative feedback. In the training phase, the single-category model (assuming that decisions are based only on distances to the target category) and dual-category model (assuming that decisions are also based on distances to the unattended category) showed almost equal fits. In the transfer phase, fits of the two model types were almost equal in ASC, but the dual-category model was superior in NT.

Conclusions: The data do not suggest difficulties in prototype abstraction but are consistent with findings that ASC-related peculiarities in category learning are dependent on stimulus characteristics. Model-fits might indicate that NT processed the unattended category more than ASC.

While substantial evidence indicates that sleep supports memory consolidation, the question of whether sleep preferentially strengthens memories associated with higher rewards remains unresolved. Previous research has yielded inconsistent findings, largely due to methodological variations and small sample sizes. To address these limitations, we are conducting a large-scale (N=1750) study examining how sleep affects the consolidation of memories associated with varying reward values. Using an AM:PM-PM crossover design, participants study images associated with different reward magnitudes (50-2150 points) and complete recognition tests both immediately after learning and following either sleep or daytime wakefulness. Our stratified sample represents the German population across sex, age (20-59 years), and education levels, allowing for robust analysis of potential moderating factors. Our primary hypothesis predicts that sleep will enhance the retention of high-reward memories compared to wake, resulting in a smaller decline of memory performance for high-reward items following sleep. We further expect that sleep will generally enhance memory retention and that high-reward items will be better remembered than low-reward items regardless of sleep condition. This study will provide crucial insights into whether sleep facilitates selective consolidation of motivationally relevant information. Results will establish benchmark effect sizes to inform sleep-based interventions for psychiatric disorders and advance our understanding of sleep's role in reinforcing adaptive memory processes. Data collection is currently ongoing, with completion expected within 12 months.

Aims: The transition from childhood to adulthood is a vulnerable period marked by biological, psychological, and social changes. Research has documented increases in mental disorders during puberty, which are partly sex-specific (e.g., more depressive symptoms in females than males). Sex hormones (e.g., testosterone, estradiol, progesterone, DHEA), which drive physical maturation and influence brain development, are likely to play a role. However, research in the field is scarce and methodologically diverse, which limits established knowledge. With our review and mini-meta-analyses, we aim to clarify current knowledge and lay the groundwork for future research on the impact of sex hormones on adolescent development of psychopathology.

Methods: A systematic search of PubMed, PsycInfo, and Web of Science identified studies examining the associations between sex hormones and mental health (e.g., internalizing and externalizing symptoms) in young people aged 10 to 19 years. Cross-sectional and longitudinal designs were eligible. Data on sample characteristics, hormone assessment methods (blood, saliva, urine, hair), and potential confounders are extracted.

Results and Conclusions: Data extraction is ongoing; a heterogeneous body of studies with varying designs, mental health indicators, and hormone assessment methods is expected. We will present findings on the associations between hormone levels and mental health by socio-demographics and assessment methods. Specific implications will be discussed (e.g., hair-tests offer insights into cumulative hormone secretion over months, other methods provide evidence of in-the-moment hormone secretion). Our conclusions address the relative impact of biological versus social factors in (sex-specific) adolescent psychopathology. We explore implications for intervention practice targeting adolescent mental health.

Working memory is a fundamental cognitive ability to actively maintain and manipulate information. While traditional research has emphasized the role of oscillatory neural activity, emerging evidence highlights the contribution of irregular, or fractal, brain dynamics characterized by 1/f power distributions. Despite the dominance of fractal activity in background electrophysiological signals, its interplay with oscillatory activity during working memory remains poorly understood. Interestingly, visual stimuli commonly used to probe working memory, such as Gabor gratings and natural images, mirror these neural dynamics when examined in the frequency domain: Gabor gratings exhibit oscillatory features, whereas natural images display fractal properties. Sensory cortices have been shown to track these features accurately, facilitating efficient neural processing. However, whether and how such stimulus dynamics influence working memory in higher-order regions like the prefrontal cortex remains unclear. Here, we directly compared human intracranial recording of pharmaco-resistant epilepsy patients engaging in two delayed-match-to-sample tasks. Critically, we developed novel stimuli that allow precise parameterization of fractal properties. We found that performance is more accurate for fractal stimuli compared to Gabor gratings. Importantly, we show that high frequency activity (70 – 150 Hz), a surrogate marker for neural population activity, is higher during stimulus presentation and delay interval for fractal stimuli compared to Gabor gratings. This is also mirrored in an upward tilt of the power spectrum. These results suggest, that hidden stimulus statistic shape neural activity during working memory processing in higher-order cortical areas beyond the sensory cortex.

Goal-oriented behaviour requires selective attention. Attention models assume enhancement of relevant targets and suppression of distractors. However, the dynamics of attentional enhancement and suppression are not entirely clear. This project is based on the idea that understanding selective attention in the real world requires considering humans as active agents who act according to their goals to shape their sensory environment. Two types of attention filters must therefore be considered. First, neural filters of attention refer to patterns of brain activity related to target enhancement or distractor suppression. Previously, we found that modulation of lateralized alpha (~10 Hz) oscillations in the human electroencephalogram (EEG) signify attentional enhancement and suppression. Second, behavioural filters of attention refer to movements of the body or sensory organs to enhance versus reduce sampling of targets versus distractors, respectively. Here, we used an auditory spatial attention task in which participants (n = 33) were allowed to freely move their head in half of the blocks. A gyroscope-augmented mobile EEG setup recorded neural activity and head rotation. Participants systematically rotated their head in the direction of anticipated target sounds. Better performance in motion-allowed trials demonstrates the power of behavioural filtering. Neural filtering, assessed as the hemispheric lateralisation of alpha activity, was present in the no-motion trials, but also in motion-allowed trials. Critically, neural filtering scaled positively with the extent of behavioural filtering (i.e., head rotation), both within and between subjects. These findings pave the way for the inclusion of behavioural filters in prevailing models of attention.

Virtual reality (VR) is gaining popularity as a tool in psychological and neuroscience research, providing new possibilities for studying human behavior and cognition. However, it can require time-intensive software development to create VR experiments that meet scientific standards. The EDIA toolbox supports scientists to design and conduct experiments in VR using the Unity game engine. A key feature of EDIA is its compatibility with various commercially available VR headsets equipped with built-in eye trackers. It offers tools for logging, visualizing, and streaming eye-tracking data at the device’s native sampling rate, while also enabling experimenters to easily create reusable components. This flexibility facilitates hardware switching and reusing code across experiments and setups. EDIA supports data streaming via the LabStreamingLayer network protocol, enabling synchronization with external data sources such as EEG. Additionally, it offers remote monitoring and control of experimental applications, a feature particularly relevant for mobile VR setups. For experiments implemented using EDIA, researchers can alter relevant study parameters (e.g., the design matrix) using platform-independent configuration files. This allows researchers and their assistants to customize experimental sessions without modifying Unity code. We demonstrate EDIA in a study evaluating eye tracking performance in five state-of-the-art VR headsets: three mobile models (Meta Quest Pro, PICO 4 Enterprise, HTC Vive Focus 3) and two tethered devices (HTC Vive Pro Eye, Varjo Aero). With a sample of 30 participants, we compare the headsets' spatial accuracy, precision, and latency (relative to concurrent electrooculography) under conditions with and without head movements.

Language is at the center of human cognition, setting us apart from other species through its complexity, flexibility, and ability to convey abstract concepts. Studying language processing, therefore, lies at the intersection of diverse research fields. Currently, a major challenge for interdisciplinary research is the variability in experimental setups and stimulus materials, making it difficult to generalize findings across fields. The goal of the Munich Sentence (MuSe) Database project is to address this challenge by providing carefully validated German sentence materials to promote reproducible research. We collected completion norms for 619 German sentences from 226 native speakers (Meanage = 25.88, SDage = 9.44). All responses were reviewed by human raters to reduce the influence of spelling and typing errors. In addition to cloze probability, entropy, and word frequency measures, we provide demographic data (age, gender, education, number of second languages) and subclinical trait scores (schizotypal and autistic traits) to obtain a broader representation of the general population. Moreover, our database includes professionally recorded sentence beginnings with high-cloze final words for 480 sentences and low-cloze final words for a subset of 320 sentences to facilitate research on semantic violations and prediction errors. The MuSe Database is accessible via OSF and GitHub and provides raw data and customizable scripts for selecting tailored stimulus subsets. Taken together, MuSe aims to support interdisciplinary research and the development of ecologically valid language processing paradigms.

In our preregistered study (OSF), we investigated the role of action planning in working memory (WM) input gating. In the primary WM task, participants stored the color of an object while the object’s shape indicated which hand they should use to report the color. During the storage phase, participants performed a secondary task requiring a left or right button press, which either corresponded or did not correspond to the response hand for the WM task. Additionally, the secondary task involved either a non-colored object (motor interference) or an object with a distracting color (visuomotor interference). Our results showed that WM was more susceptible to visual interference when the interfering stimulus required the same response as the actual WM target item. This was reflected in a higher attraction bias, where the WM representation was more influenced by the interfering color when the motor responses of both tasks corresponded. Furthermore, EEG analysis revealed that frontal theta activity was higher in the corresponding action condition, and greater frontal theta power predicted lower attraction bias. We interpret this as evidence of an automatic conflict mitigation mechanism, wherein action correspondence mediates the gating of WM input to counteract interference.

Ultra high field imaging at 7 Tesla is a promising method to investigate the function of small brain structures that were hitherto difficult to access in humans. Here we conducted a functional magnetic resonance imag-ing study at 7T at 1 mm isotropic resolution with a monetary probabilistic reward contingency learning task that allows to investigate reward, loss, positive and negative prediction error processing in N=27 young healthy participants. We aimed at characterizing specifically the response of the habenula, a small pair of nuclei in the epithalamus which is top-down connected to midbrain monoaminergic systems and has been been implicated in heterogeneous cognitive, emotional, and motivational processes like reinforcement learn-ing, goal-directed and avoidance behavior (Lecourtier & Kelly, 2007).

We found an activated ROI-corrected significant cluster at the right habenula and cluster-level significant activation in the right anterior insula (rAI) and dorsomedial prefrontal cortex that were associated with the processing of losses. Resting state functional connectivity analyses from the same MRI session with the acti-vated right habenula cluster as seed regions revealed connectivity with a cluster within the activated region in the rAI, suggesting a central role of a rAI-Habenula network in the neural processing of monetary losses.

We further used our high-resolution functional imaging data to investigate the influence of data resolution and different smoothing kernels on the identifiability of the detected effects and found that an isotropic resolution of 1.5 mm seemed optimal to capture the activation, but with different smoothing kernels for the habenula and the cortex.

Investigating ways of promoting healthy aging has never been more pressing than in our rapidly aging society. An emergent field is dedicated to investigating the impact of later life language learning, given the influence of lifelong bilingualism on cognitive reserve (Berkes & Bialystok, 2022). Language learning has been proposed as a particularly effective later life training, as the activated brain regions overlap with areas often affected by age-related cognitive decline (Antoniou et al. 2013).

To date, investigations into later life language learning and its cognitive effects have produced mixed results, pointing to a need for replication to increase comparability across studies (Pot et al., 2019; van der Ploeg et al., 2020). Some studies report enhanced cognition (Meltzer et al., 2021; Pfenninger, & Polz, 2018; Bak et al., 2016), but this has not been robustly found.

Though some studies have started comparing later life language learning to other interventions to enhance cognition, it remains unclear which type of intervention is most beneficial and, critically, also feasible for older adults at different stages of cognitive decline. In our study, we investigate the effect of a language intervention (learning English in a non-anglophone environment) compared to a combined physical and cognitive training (playing cognitively demanding board games and movement) on cognition and well-being in older adults. Data collection through questionnaires, behavioral tasks, and EEG is ongoing. By the time of the conference, three courses (n=21) will be completed. We present the set-up and preliminary findings of behavioral tasks.

Background: Pregnancy is a very vulnerable phase of life due to significant physiological and psychological changes. Mental health disorders during this time can negatively affect the family, including offspring development. Although social support plays a protective role against depression, the specific mechanisms by which different types of social interactions influence maternal mental health remain understudied.

Objective: This study investigates the impact of social interactions, particularly different types of social support, on depressive symptoms in women during the perinatal period. The goal is to identify which types of support are most valuable in order to provide new insights for clinical interventions targeting perinatal depression. Preliminary data collected up to that point will be presented.

Methods: In this prospective longitudinal cohort study with Ecological Momentary Assessment (EMA) pregnant women in their third trimester with mild to moderate depressive symptoms will be recruited from a specialized outpatient clinic. Social interactions and mood will be assessed using EMA techniques, while additional factors (e.g. loneliness, stress, relationship satisfaction) will be measured via standardized questionnaires. A follow-up survey 3-6 months postpartum will assess social support, self-efficacy, parental stress, and mother-child bonding to track changes over time.

Discussion: Effective prevention and treatment of perinatal depression are crucial for maternal mental health and offspring development. As social support is a main factor of prevention and therapy in perinatal depression, identifying the most effective types of support can help to develop targeted prevention strategies to reduce antenatal depression.

Phantom limb pain (PLP) affects up to 80% of amputees and involves central and peripheral changes. Cortical oscillations have been linked to chronic pain, yet their role in PLP is unclear. This study examines cortical oscillations over the somatosensory, motor and prefrontal cortices during evoked PLP using EEG recordings and continuous pain ratings.

Nine amputees, capable of inducing PLP through movements of their phantom limb (evoked PLP), participated in the study. To obtain time courses of brain activity in specific frequency ranges, EEG data were time-frequency transformed and fitted in regression models. The power of the alpha, beta, and gamma frequency bands within specific regions of interest was analyzed as a predictor of individual pain ratings, considering both group-level and individual-level effects. Cross-correlations accounted for time lags between neurophysiological responses and pain ratings.

Increased beta power recorded over the somatosensory cortex and decreased beta power over the motor cortex, both contralateral to the amputation site, were associated with perceived evoked PLP. Alpha suppression recorded over the somatosensory cortex ipsilateral to the amputation site, and gamma power changes recorded over the prefrontal cortex predicted perceived evoked PLP.

These results highlight the complex nature of pain processing in PLP and indicate that different cortical regions and frequency-specific oscillations influence the perception of evoked PLP. Moreover, cortical oscillations could serve as therapeutic targets for personalized neuromodulation therapies, for example, using transcranial alternating current stimulation.

Adolescence represents a critical period of development characterized by heightened vulnerability to depression (Davey et al., 2008). Behavioral Activation (BA), a therapeutic approach involving the planning and scheduling of reward-related activities to re-establish effort-reward expectations, has demonstrated efficacy in ameliorating reward system dysfunctions and alleviating depressive symptoms among adolescents (e.g., Cernasov et al., 2021; Malik et al., 2021; Tindall et al., 2024; Webb et al., 2023). In our ongoing study, 22 depressed adolescents completed an 8-week group CBT with BA components and performed a physical effort task (button-pressing balloon task) at pre-treatment, post-treatment, and follow-up assessments, choosing between high-effort/high-reward (HEHR) and low-effort/low-reward (LELR) options with different reward/effort magnitudes. Preliminary analyses revealed that depressive symptoms decreased significantly (β=-2.34, p =0.0082). The task worked as intended, as reward magnitude differences (ΔR=high high -low) increased HEHR selection (β=0.386, p<0.001), while effort demand differences (ΔE=high-low) decreased it (β=-0.353, p=0.005). CriticallyInterestingly, at follow-up, patients had a greater preference for HEHR options at smaller ΔE and more frequent rejections of HEHR options when ΔE increased (β = -0.623, p < 0.001). Future work will compare these findings with a control group (planned final N = 31) to control for retest effects and employ computational modeling, including the Drift Diffusion Model (DDM) and Subjective Value Model (SVM), to quantify how BA modulates the manipulated cost-benefit differences (ΔR and ΔE). on the caution-speed balance when choosing HEHR options, as well as on the perceived worth of HEHR options.

Working memory is crucial for the temporary storage and manipulation of information, but is vulnerable to perturbations such as task interruptions that divert attention away from the primary task. While research has explored the effects of interruptions and task switching, less is known about how varying the sensory modality and domain of the interrupting task impacts memory retention and task resumption. In this study, we used a visual working memory task, and asked participants to maintain the orientation of two colored bars out of four presented. During the maintenance phase, they were occasionally interrupted by a discrimination task, in which they either had to decide whether the sum of two successively presented (visual or auditory) digits was greater or less than thirteen, or decide which of two successively presented horizontal bars was larger (visuospatial). We recorded EEG to monitor neural correlates of cognitive processes, e.g. posterior alpha asymmetry, to assess reactivation of visual memory representations after interruptions. We found that interruptions impaired performance of the main task, with visuospatial interruptions having the strongest effect due to sensory and cognitive overlap with the primary task. At the neural level, we observed weakened frontal theta power, reflecting reduced cognitive resources, and weaker posterior alpha suppression, indicating impaired attentional reorientation following interruptions. As predicted, posterior alpha asymmetry was the strongest following visuospatial interruptions, suggesting greater difficulty in refocusing attention. Findings reveal that sensory overlap between primary and interrupting tasks modulate the extent of disruption, providing insight into working memory protection and resumption mechanisms.

A putative primary role of the endocannabinoid signalling system is to restore homeostasis after psychological and physiological threats. While CB2 receptors are mainly expressed in the immune system, CB1 receptors are widespread in the brain. Action of endocannabinoid signalling occurs along the entire HPA-axis and exogenous agonists at low doses inhibit its activity while large doses increase it. Here we used a low dose of Dronabinol (vs. placebo) given in the evening before bedtime in a balanced double-blind within-subject design. Twenty participants slept eight hours and then performed their daily routines as usual. During sleep, blood was collected in a subsample of n = 6 and exploratory analyses revealed an increase in cortisol plasma levels during the second half of the night. In the next afternoon, all participants returned to the lab and performed a socially evaluated cold pressure test (SECPT). The SECPT elicited the expected changes in cortisol levels in saliva, heart rate and blood pressure. However, there was no difference in cortisol levels evident between the Dronabinol and placebo sessions. When adding session (first vs. second) to an exploratory analysis there was a significant three-way interaction. Follow-up analyses indicated that participants, who received Dronabinol in the first session showed a reduced a reduced cortisol response in the SECPT. Surprisingly, this reduced cortisol response remained at the second session. This is indicative of a prolonged change in stress response after a single low dose of synthetic THC possibly due to long lasting changes in stress anticipation.

During the encoding and retrieval of event sequences, episodic and semantic memory systems may interact in generating and reconstructing a memory representation. As a result, recall of event sequences may not always faithfully reflect the actual temporal event structure, but may be systematically influenced by semantic associations between encoded items. This semantic bias in episodic memory retrieval would be particularly evident in cases where temporal and semantic associations would imply differential responses. Previous studies have shown similar effects in spatial episodic memory and we could demonstrate semantic bias effects in a temporal sequence memory task in our behavioral pilot study. In our current neuroimaging project, we aim to investigate the neural correlates of semantic bias effects on temporal sequence memory. Specifically, we are interested in whether neural representational patterns at encoding and retrieval of object image sequences can predict memory bias on a trial-by-trial basis. Behavioral findings from our pilot study suggested that the presence of pronounced semantic structure within the encoding set (realized as clustering of semantically similar items) is critical for semantic bias effects on temporal sequence memory. Thus, our main focus lies on understanding how semantic context information is encoded and how this is associated with semantically biased memory retrieval. Preliminary results from representational similarity analyses of fMRI data acquired during encoding of image sequences and subsequent temporal sequence memory retrieval will be presented and further possible approaches will be discussed.

Recent evidence suggests that open-label placebos (OLPs) may effectively reduce negative affect and stress, but replications with physiological data are needed. Furthermore, the role of personality traits in the OLP response remains unclear. This study examines the effects of OLPs on negative affect and physiological stress markers (salivary cortisol, salivary alpha amylase, and hair cortisol) in the context of an oral university exam - a real-life stressor. We further investigate whether the personality traits from the five-factor model influence OLP responses.

In a randomized controlled design, 200 participants are divided into an OLP or control group (no treatment). The OLP group takes two placebo pills daily for four weeks leading up to the oral exam. Measures of negative affect were assessed continuously. Saliva and hair samples were collected on exam day, representing acute (saliva) and long-term (hair) stress markers.

Preliminary findings (N=134) indicate significantly lower hair cortisol concentrations and less negative affect in the OLP group during the intervention month, as well as lower subjective stress and salivary alpha amylase on exam day. No group differences were found in salivary cortisol yet. In the OLP group, improvements in exam grades are mediated by reduced pre-exam negative affect. Moderated regression analyses revealed that higher conscientiousness is associated with a stronger OLP effect on negative affect but not on physiological parameters. These findings suggest that non-deceptive placebos can effectively reduce negative affect and certain physiological stress markers in a real-life stress scenario, with conscientiousness moderating the psychological impact.

Long-term memory consolidation is critically linked to sleep, but recent findings suggest that hunger may also play a role. In this study, two within-subject experiments with healthy male participants examined how fasting versus a satiated state affects memory consolidation during a 10-hour post-learning period. In Experiment 1, after fasting for 18.5 hours, participants learned three types of memory tasks: word pairs, a visual version of the Deese-Roediger-McDermott task, and a motor task (finger tapping). They then either continued fasting or ate a standard meal, with memory tested 48 hours later in a fed state. Experiment 2 introduced a What-Where-When episodic memory task in place of the DRM task, with recall tested 24 hours later while fasting. Results showed that fasting improved cued recall of word pairs and item memory in the What-Where-When task, suggesting enhanced consolidation of semantic-like information. However, fasting impaired context-related episodic memory (e.g., spatial or temporal details). Finger tapping performance remained unchanged. These findings suggest that fasting may boost memory consolidation in the cortex (semantic representations), possibly by reducing interference from the hippocampus during wakeful consolidation. However, the exact neural mechanisms are still unclear.

Introduction

Accurate witness identification (ID) is vital for criminal investigations, yet lineup procedures remain outdated, typically relying on static 2D photographs. Eyewitnesses frequently make errors, often misidentifying innocent individuals[1,2]. Research shows that interactive lineups significantly improve discriminability compared to static formats, and stereoscopic face presentation enhances accuracy due to richer depth cues, increased attention to volumetric features, and reduced cognitive load[3].

We hypothesize that simultaneous interactive lineups presented in stereoscopic 3D will yield higher ID accuracy than monoscopic presentations when delivered in a virtual reality (VR) environment.

Methods & Results

This study is conducted by an international consortium across four sites: the University of Birmingham, University of Stirling (UK), University of Victoria (Canada), and the Max Planck Institute for Human Cognitive and Brain Sciences in collaboration with Humboldt-University (Germany). Using pre-recorded 2D facial images from an existing dataset[4], we created 3D models via Gaussian splatting[5], which were used as both fillers and perpetrators in immersive VR lineups.

Participants are tested under three conditions: 2D (monoscopic, static), 2.5D (monoscopic, interactive), and 3.5D (stereoscopic, interactive), using a Varjo Aero headset with eye-tracking. A total of 840 participants (70 per condition per site) view 8 mock-crime videos and complete 3x2 lineups. This poster presents pilot data from the 3.5D condition, including ID accuracy, confidence ratings, and gaze behavior during suspect selection.

Discussion

The 3D Interactive Lineup (3DIL) approach provides novel insights into facial processing and aims to inform and enhance forensic ID procedures.

Humans effortlessly predict upcoming language, yet the cognitive costs of these predictions remain debated. In two behavioural studies (N = 175, replication N = 96), participants read short articles presented word-by-word while performing a secondary n-back task on font colours, thereby manipulating cognitive load. Language predictability was quantified using word surprisal estimates derived from a generative pre-trained transformer model (GPT-2). Our results showed that higher cognitive load consistently diminished the impact of word predictability on reading time. Extending these findings, we recorded EEG data from 33 young adults (19–32 years, M = 22.15±2.80 years) performing the same dual-task experiment. To examine the neural dynamics underlying the behavioral effects, we employ multivariate Temporal Response Functions (mTRFs; encoding-model, boosting algorithm; EEG signal filtered to 0.5–30 Hz) to assess the influence of word surprisal between cognitive load conditions on ERP-like components reflecting predictive processing. Preliminary results indicate that increasing cognitive load diminishes neural signatures of language prediction. Specifically, this affects the central N250, which reflects orthographic predictions, and the centro-parietal N400, which scales with the unexpectedness of a word given its context. By integrating behavioral and neural measures, this study aims to clarify the relationship between domain-general executive resources and predictive processing in language comprehension.

Anxiety symptoms and related traits, such as neuroticism, intolerance of uncertainty, and negative emotionality, have been associated with dysregulated fear and safety processing. However, studies vary in their focus on different aspects of fear acquisition and extinction, outcome measures, and anxiety-related traits, making it challenging to integrate findings across research. To bridge this gap, a systematic literature search and meta-analysis will be conducted to examine the relationships between anxiety-related traits and fear acquisition and extinction. Additionally, a content analysis of the questionnaires used in the identified studies will assess the overlap in the constructs measured. The poster will present the overall approach of this study and share preliminary findings from the literature search.

The human auditory system is constantly exposed to a wide variety of sounds. Some sounds are relevant targets, while others are irrelevant distractors. Our understanding of processing distraction is premature, as experimental control conditions to delineate enhancement from suppression have often been lacking. To fill this gap, we introduce irrelevant stimuli that are perceptually salient (i.e., distractors) or non-salient (i.e., controls) into an auditory search task. On every trial, three spatially separate, unique auditory objects (target, distractor, control; german single-syllable spoken digits) were presented simultaneously. Participants had to identify an amplitude-modulated target and report its identity. Behaviorally, we observed distractor suppression as decreased performance in trials where the current target shared features with the distractor from the previous trial (i.e., negative priming). Target enhancement surfaced in increased performance in trials where the current target shared features with the previous target (i.e., positive priming). Over the course of the experiment, performance relatively increased in trials with distractors, indicating a gradual transition from attentional capture by the distractor to its suppression. In the electroencephalogram (EEG), lateral distractors evoked a distractor positivity (Pd) component in the event-related potential (ERP), which has been associated with inhibitory attentional processing in the visual modality. Conversely, lateral targets evoked an N2ac component, associated with attentional focusing. These results suggest that searching for auditory targets involves separable mechanisms of target enhancement and distractor suppression. More successful suppression after repeated exposure supports the notion that auditory distractor suppression is subject to implicit learning.

Delay discounting (DD) and probability discounting (PD) tasks are frequently used to study neural correlates of altered decision-making in addiction. However, real-world decisions often involve outcomes that are both uncertain and delayed. This fMRI study examines brain areas uniquely activated by a combined delay-probability discounting (DPD) task compared to individual DD and PD tasks, focusing on decision-making processes and working-memory involvement. Sixty-five healthy participants will complete adaptive discounting tasks in two parts: outside the MRI to model individual discounting patterns, and inside the MRI across three trial conditions (DPD, DD, PD). Additionally, an n-back task will localize working-memory activation. Participants will also complete questionnaires assessing risk-taking (DOSPERT), impulsivity (BIS-15), perceived stress (PSS), and alcohol-related problems (AUDIT).

Primary analyses will identify unique brain areas activated during combined discounting decisions and attribute-specific evaluations (delay, probability, magnitude), as well as continuous parametric modulation by subjective value and stimulus attributes. Covariance analyses will test associations between individual discounting rates and brain activation. Further analyses will examine whether the DPD task engages working-memory regions more strongly than individual DD and PD tasks and explore correlations between working-memory activation and discounting rates.

At the conference, preliminary results based on collected data will be presented. Ultimately, this study evaluates whether the DPD task could serve as a useful paradigm for future clinical research into impaired decision-making, such as in addiction.

Habits are automatic behaviors facilitating daily activities by reducing cognitive demand and enabling efficient performance. Yet, they can also result in behavioral inflexibility, hindering the ability to adapt to changes in the environment. Since many habitual behaviors rely on motor activity, and motor learning provides a framework for investigating human habit formation, we developed an fMRI-adapted version of our previously developed motor sequence learning task (MSLT) to investigate the neural mechanisms underlying motor automatization. Twenty-nine participants (16 female) completed the MSLT across two consecutive days. Participants implicitly learned a fixed sequence of 12 visual stimuli, each associated with a corresponding spatial motor response. To assess behavioral flexibility, test blocks intermittently introduced deviations from the learned sequence by omitting one item. Participants’ responses to these omissions were used to evaluate their flexibility: executing the omitted response despite its absence–termed action slip–was interpreted as an indicator of behavioral inflexibility.

With prolonged training, reaction times within learning blocks significantly decreased (t(28) = 16.15, p<.001), while action slips within test blocks increased (t(28) = 11.40, p<.001), indicating a shift towards automatic and inflexible responses. fMRI analyses revealed decreased activity in the (pre)motor cortex, anterior insula, occipital gyri, and cerebellum across learning blocks (whole brain corrected, pFWE <.05), suggesting reduced sensory and motor processing. In contrast, the bilateral dorsal putamen showed increased activation over extensive training (ROI, pFWE <.05). These findings support the MSLT as a promising paradigm for investigating the mechanisms underlying the development of motor automaticity related to human habit formation.

The more you repeat information, the better you will remember it later. Fittingly, it has recently been shown that rehearsal can accelerate the process of memory consolidation by rapidly forming stable neocortical memory traces. However, there is limited understanding of how this process affects memory quality. Here, we report preliminary findings from a novel task designed to assess the influence of rehearsal on memory precision and stability, both in behavior and functional brain activity.

Healthy, young adult participants (planned n=60) encode and recall new associations between a cue and a target stimulus in two spatial locations presented on a specific background scene either once (single items) or four times (rehearsed items) during fMRI. We test memory performance both 60 minutes and three days after learning along four dimensions: Memory for target category, target exemplar, target location and scene. Additionally, stimulus material was designed to enable classification and similarity analyses on fMRI data.

Preliminary data of n=8 participants suggest that memory for detailed item features and spatial precision is higher in rehearsed items compared to single items in both delays and that memories for single items degrade significantly over time. Preliminary fMRI results confirm a higher neocortical contribution for rehearsed vs. single items already during the learning session.

Together, our data indicate that rehearsal might rapidly engender not only more stable, but also more precise memory traces in the neocortex, contradicting traditional systems consolidation theories of a slow learning neocortical system, which stores information in a semantic, gist-like format.

Aversive learning processes have been associated with internalizing symptoms that typically appear around adolescence. To shed more light on this association, we contrasted age differences in learning processes between aversive and reward environments. To do so, we used a gamified learning task in which participants learned about the values of different screen zones to accurately position their spaceships. Task conditions required avoiding asteroid crashes and collecting rewarded space treasures. Both environments were stochastic, and the value of zones changed over time. We analysed 202 data sets collected online (age-range: 9-31, M = 20.26 , SD = 6.45). To characterize the learning process, we implemented different variations of hierarchical approximate Bayesian models and two Rescorla-Wagner models in PyMC. In a linear mixed model, we found that task performance, defined as how well participants chose good zones, increased with age specifically in the aversive condition (age * condition: F = 5.51, p = .020, slope [aversive]: β = 0.0014, CI [0.000265, 0.002443], slope [reward]: β = -0.0001, CI [-0.0012, 0.0010]; condition: F = 9.38, p = .003). In our computational modelling analysis, we found an association between age and the variance-weighting bias parameter, which captures how much participants take the model-derived uncertainty estimate into account (age: F = 5.94, p = .016) . The performance increase with age aligns with recent research suggesting age-related differences are more pronounced in aversive contexts. As a next step, we will investigate the associations of model parameters with the expression of psychiatric symptoms across age.