Biology subjects

Grootswagers, T.

Publications and source records attributed to Grootswagers, T..

At least 19 recordsLinked to original sources

bioRxiv2026

Sustained multidimensionality of object representations as a scaffold for diverse behaviours

The human brain transforms a brief glance into a neural representation flexible enough to serve many behavioural goals. This flexibility may reflect multidimensional visual representations read out selectively across tasks, rather than goal-specific instantiation. Combining a large-scale EEG dataset with a behavioural battery of 12 tasks spanning 200 objects, including perceptual, categorical and social decisions, we tested how behaviours are supported by stimulus-evoked representations. Information relevant to 11 of 12 tasks was accessible in the neural signal, emerging early and overlapping in time. Orthogonal behavioural dimensions showed distinct temporal profiles, indicating a genuinely multidimensional representation. The neural code converged on a human-relatedness axis, while feature-level information remained accessible. Behavioural accessibility was thus broad but structured. Object familiarity alone was reliable in behaviour yet absent from the shared neural signal, revealing a limit on what a glance makes accessible. We propose that sustained, recurrent processing serves this broad behavioural accessibility.

neuroscience↗

bioRxiv2026

The Centroparietal Positivity Traces the Path of Evidence Accumulation During Dietary Decisions

Each day we make dozens of decisions about food. These choices are informed not only by information that we receive from our senses, but also by our memories, values, cultural norms, and goals. Evidence accumulation models have been proposed to explain how we transform these disparate sources of information into dietary choices and their corresponding motor actions. These models describe decisions arising from a noisy integration of decisional evidence until a criterion is reached. We investigated whether the centroparietal positivity (CPP), an event-related potential component proposed as a neural correlate of evidence accumulation during perceptual judgements, also traces accumulation trajectories during dietary decisions. Participants (N = 110) made speeded Yes/No judgements about food images relating to three attributes (healthiness, tastiness, and willingness to eat) while electroencephalography was recorded. They subsequently provided continuous ratings for each attribute. After applying signal deconvolution techniques, CPP waveforms exhibited a close correspondence to hypothesised evidence accumulation trajectories, showing steeper amplitude build-up rates for both faster decisions and continuous ratings more strongly in favour of ones categorical choice. Our findings provide clear support for a neurally instantiated evidence accumulation process that integrates multiple sources of information to produce dietary decisions.

neuroscience↗

bioRxiv2026

Enhanced neural encoding of individual exemplars in the Super-Recogniser brain is specific to faces

Super-Recognisers are known for their exceptional ability to encode and remember faces. Recent work has found robust neural codes for face identity in brain activity. Here we test whether the enhanced individuation of face identity in super-recogniser neural codes extends to exemplars from non-face categories. We recorded 64-channel electroencephalography (EEG) as 22 Super-Recognisers and 20 typical recognisers viewed rapid, randomised sequences of faces, dogs, houses, and cars (10 different images of 10 different exemplars per category). Using a neural decoding approach, we found that Super-Recognisers neural responses contained stronger and more temporally sustained information discriminating individual face identities. Moreover, Super-Recognisers displayed increased inter-individual consistency in the organisation of their face-identity representations, suggesting a more stable and/or canonical representational structure in individuals with superior face recognition abilities. Crucially, there was no evidence to suggest that these encoding advantages generalised to non-face categories: neural discrimination of individual dogs, cars, and houses was comparable across groups, and group-differences in inter-subject consistency were evident only for faces. Together, these findings suggest that enhanced perceptual encoding underlying exceptional face recognition is highly category-specific, supporting the view that expertise in this group reflects specialised mechanisms for processing facial identity rather than broadly enhanced visual discrimination capacities.

neuroscience↗

bioRxiv2026

Robust neural face identity codes in the Super-Recogniser brain

Super-Recognisers show exceptional ability in face recognition, providing a natural model of how perceptual systems optimise for individuating visually similar stimuli in variable viewing conditions. However, the neural representations supporting this extreme perceptual expertise are unknown. Here, we tested whether Super-Recognisers (n = 23) differed from typical recognisers (n = 21) in the dimensional organisation of neural face identity coding. We recorded 64-channel electroencephalography while participants viewed random and rapidly-presented sequences containing 10 naturally varying images of 40 unfamiliar identities. Using time-resolved representational similarity analysis we measured the geometry of identity representations, their consistency across observers, and how clearly they specified face identity. Although neural expression of identity information was robust in both groups, we found three key differences between Super-Recognisers and typical recognisers. First, the geometry of face identity representations differed between groups. Second, Super-Recognisers showed greater inter-individual consistency in representational geometry. Third, Super-Recognisers neural signals discriminated between face identities more strongly than those of typical recognisers. Differences in the coding of broader face categories (sex, age, ethnicity) were notably weaker, suggesting that the observed group differences reflected fine-scale differences in identity coding rather than global reshaping of representational geometry. Strikingly, all three differences emerged within a common mid-latency interval ([~]300-500ms), implicating higher-stages of face processing associated with representations that are sensitive to face familiarity and link between perceptual and semantic domains. Together, these findings indicate that individual differences in face recognition ability reflect higher-level differences in neural identity coding, rather than enhanced early sensory processing. Significance statementSuper-Recognisers have an exceptional ability to recognise unfamiliar faces, yet the neural basis of this advantage has remained unclear. We find brain activity expresses the identity of unfamiliar faces more strongly and consistently in Super-Recognisers than typical recognisers. These advantages were specific to individual identity and could not be explained by enhanced encoding of broader facial categories such as sex, age, or ethnicity. Critically, group differences emerged during a mid-latency stage of processing ([~]300-500 ms) associated with higher-level representations that support stable identity coding across variable images. The findings suggest that exceptional face recognition arises from more robust neural representations of individual identity rather than enhanced early visual processing, providing new insight into the neural basis of perceptual expertise.

neuroscience↗

bioRxiv2026

A multibrain advantage for cooperative human behaviour

Collaboration is a cornerstone of human behaviour, enabling individuals to achieve complex goals that transcend individual capabilities. While previous research has largely focused on the alignment of neural representations across brains during social interactions, real-world cooperation frequently involves complementary roles, where success depends on integrating distinct yet interdependent representations. Here, we combined electroencephalography (EEG) hyperscanning and multivariate pattern analysis (MVPA) methods to investigate neural representations in pairs of participants as they jointly performed a speeded two-dimensional visual matching task with asymmetric controls to encourage task division (i.e., attending to different stimulus features). Our results provide direct neural evidence of the emergence of selective information encoding within individual brains and complementary information encoding across brains during successful cooperation. Specifically, optimally divided task dimensions give rise to a late (> 485 ms) multibrain advantage, wherein the combined neural signals of the two participants carry more target-related information than that of either brain individually. Notably, the magnitude of these information-processing modulations within and across brains robustly predicted pairs collective task performance. These findings have broader implications for future research on collaboration, highlighting the need to move beyond representational alignment to harness diverse human as well as machine capabilities across our societies to unlock collective potential.

neuroscience↗

bioRxiv2025

FOODEEG: An open dataset of human electroencephalographic and behavioural responses to food images

Investigating the neurocognitive mechanisms underlying food choices has the potential to advance our understanding of eating behaviour and inform health-targeted interventions and policy. Large, publicly available neural and behavioural datasets can enable new discoveries and targeted hypothesis tests, yet no such datasets are currently available. We present the FOODEEG dataset containing electroencephalographic (EEG) responses to a diverse array of food images, as well as behavioural measures of food cognition (food categorisation task, food go/no-go task, and food choice task responses), collected from 117 participants. We also provide normative ratings for the food image stimuli with respect to 22 food attributes, including nutritive, hedonic and taste properties, familiarity, and elicited emotions. Our dataset also includes questionnaire-based measures of participants food motivations, dietary styles, and general motivational tendencies. In the validation analyses, we demonstrate that early food-evoked EEG responses in our dataset are consistent with observations in previous work. The FOODEEG dataset will be valuable for accelerating research into the neural substrates of visual food processing, dietary decisions, and individual differences.

neuroscience↗

bioRxiv2025

A comprehensive EEG dataset for investigating visual touch perception

Visual touch perception is fundamental to body awareness and social bonding, helping us perceive ourselves and connect with others. However, understanding its neural basis has been limited by a lack of comprehensive public datasets. To address this, we present an electroencephalography (EEG) and self-report dataset from 80 participants. It includes neural responses from 2,880 trials per participant, during which participants viewed touch interactions between two hands involving either direct skin contact or object-mediated touch. Complementing the EEG data are self-report measures on empathy, perspective-taking, vicarious touch, and mirror-touch synaesthesia, providing a robust resource for exploring individual differences in visual touch perception. Ninety unique videos were presented in four orientations, with trials evenly split between self- and other-perspectives, and showing touch to either the left or right hand. This design supports investigation into how viewing perspective and hand laterality influence touch processing. Technical validation reveals distinct neural patterns associated with visual, sensory, and emotional touch dimensions, highlighting the datasets potential to advance research into diverse aspects of visual touch perception.

neuroscience↗

bioRxiv2025

Robust Neural Decoding with low density EEG

High-density Electroencephalography (EEG) recording enhances spatial resolution for neural signal decoding, yet the relationship between electrode density and decoding performance remains unclear. To address this, we systematically investigated decoding accuracy across electrode configurations of varying densities (16, 32, 64, 96, and 128 electrodes) using visual grating stimuli characterized by orientation, contrast, spatial frequency, and color. As expected, decoding accuracy increased with electrode density. Remarkably, however, reliable above-chance decoding was still achieved with as few as 16 electrodes, highlighting the robustness of decodable neural signals. To test the generalization of these results to more complex natural stimuli, we conducted a similar analysis with a diverse set of naturalistic images categorizable into living/non-living and moving/non-moving. The results consistently showed that effective decoding persists even with a 16-electrode configuration, showing robust decoding efficacy even for complex naturalistic stimuli. These findings demonstrate both the benefits of higher-density EEG and the robustness of neural decoding under sparse spatial sampling, providing new insights into how efficiently and broadly neural signals can be decoded.

neuroscience↗

bioRxiv2025

The neural processes of illusory occlusion in object recognition

Fast and accurate object recognition is crucial for effective behaviour in dynamic visual environments. In some cases, the visual system must overcome ambiguity in visual input during object recognition, such as when an object is partially hidden behind another. Recurrent processing between higher- and lower-order areas is thought to play a role in resolving such ambiguity, enabling the filling in of missing visual information. Here we examined this claim using a novel paradigm in which partial object images appear occluded by an illusory Kanizsa figure, perception of which also depends on recurrent processing by the grouping of Pacmen inducers. If both recognising the partial object and perceiving the illusory shape depend on recurrent processing, object recognition should vary as a function of the presence of illusory shape. Across two behavioural experiments and a separate EEG decoding study, we found no evidence for an interaction between illusion perception and object recognition, and different neural time courses for the illusory figures alone compared to those of the partial objects, which were decoded earlier. Ultimately, our study highlights the robustness of the visual system to solve the identity of the ambiguous object, independently from the processing of different ambiguities occurring at the same time, providing new insights into the mechanisms of recurrence in the early and late stages of information processing and its application to ambiguous object recognition.

neuroscience↗

bioRxiv2025

Multimodal Human Perception of Object Dimensions: Evidence from Deep Neural Networks And Large Language Models

Object recognition in the human visual system is implemented within a hierarchy characterised by increasing feature complexity. Here, we investigated whether human-derived dimensions of object knowledge show a similar progressive emergence across layers in deep neural networks (DNNs), and how this emergence is shaped by architecture, learning objective, and stimulus statistics. To test this, we predicted human-derived dimensions from layer-wise activations of multiple DNNs and transformer models trained on large-scale datasets. Results showed that trained DNNs exhibit emergence profiles resembling theoretical expectations from human vision, with behaviourally relevant object dimensions largely absent in early layers, strengthening across layers, and peaking in later layers. Architectural mechanisms such as recurrence and skip connections amplified this encoding, learning objectives redistributed information across layers, and changes in stimulus statistics confirm that hierarchical emergence is a general principle extending to material perception. These findings demonstrate that the hierarchical emergence of human-derived dimensions is a fundamental property of trained networks and highlight design and input factors that shape layer-wise representational organisation, providing hypotheses for the structure of visual representations in the brain.

neuroscience↗

bioRxiv2025

Characterising the neural time-courses of food attribute representations

Dietary decisions involve the consideration of multiple, often conflicting, food attributes that precede the computation of an overall value for a food. The differences in the speed at which attributes are processed play an important role; however, it is unknown whether different attributes are processed over distinct time windows. We mapped the neural time-courses of 12 choice-relevant food attributes. Participants (N = 110) viewed food images while we recorded brain activity using electroencephalography (EEG). A separate group of participants (N = 421) rated the same images on nutritive properties (healthiness, calorie content, edibility, and level of transformation), hedonic properties (tastiness, willingness to eat, negative and positive valence, and arousal), and familiarity (previous exposure, recognisability, and typicality). Using representational similarity analysis, we quantified differences in patterns of multivariate EEG signals across foods and assessed whether the structure of these differences was correlated with differences in attribute ratings. We observed similar correlation time-courses for many attributes. There was an early window of correlations ([~]200 ms from image onset), followed by sustained windows of correlation from [~]400-650 ms. Using principal components analysis, we identified a set of broader constructs that accounted for variance in ratings across multiple attributes, and were also correlated with the EEG data. Our results indicate that food attributes important for choice are represented rapidly and in parallel, over similar time windows. Furthermore, we reveal that broad dimensions underlying individual attributes are also represented in the neural activity with distinct time-courses, indicating a multilevel structure of food attribute representations.

neuroscience↗

bioRxiv2025

Multiple Neural Modules Orchestrate Conflict Processing

Cognitive conflict is a ubiquitous aspect of our daily life, yet its underlying neural mechanisms remain debated. Competing theories propose that conflict processing is governed by either a domain-general system, multiple conflict-specific modules, or an architecture involving partially overlapping mechanisms. The aim of the current study was to clarify how these mechanisms operate at the neural level using a data-driven decoding approach. We analyzed electroencephalogram (EEG) data from 507 healthy participants (ages 20-70) from the Dortmund Vital Study (ClinicalTrials.gov NCT05155397) who completed three conflict tasks: a Change Detection task, a Simon task, and a Stroop task. Using multivariate decoding to distinguish conflict from non-conflict trials, we observed robust within-task decoding across all tasks. However, cross-task decoding revealed shared conflict representations only in specific task pairs. These findings support an account, in which conflict processing relies on partially overlapping neural mechanisms rather than being fully domain-general or entirely task specific. We argue that conflict-related neural processes might be best conceptualized as a continuum of overlap and differentiation, with domain-generality and domain-specificity representing the endpoints of this spectrum. Research Transparency StatementThe authors declare no conflicts of interest. Funding: This work was supported by the International Visiting Scholar Program at The MARCS Institute for Brain, Behaviour, and Development of Western Sydney University (MS), and the Australian Research Council grants DE230100380 (TG) and DP220103047 (MV). Artificial intelligence: Artificial intelligence assisted technologies were utilized solely for language refinement and grammar corrections, ensuring clarity and readability without altering the original content or ideas. Ethics: The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Ethics Committee of the Leibniz Research Centre for Working Environment and Human Factors, Ardeystra{beta}e 67 44139 Dortmund, Germany (IfADo; code A93-3) as detailed in the published study protocol (Gajewski et al., 2022). Preregistration: The current study was not preregistered. However, the dataset used in this manuscript originates from the Dortmund Vital Study, which is registered on ClinicalTrials.gov (NCT05155397). Data: A sample dataset is publicly available at https://doi.org/10.17605/OSF.IO/EDTNM Analysis Scripts: All analysis scripts are publicly available at https://doi.org/10.17605/OSF.IO/EDTNM.

neuroscience↗

bioRxiv2025

Neural decoding of competitive decision-making in Rock-Paper-Scissors using EEG hyperscanning

Social interactions are fundamental to daily life, yet social neuroscience research has often studied individuals brains in isolation. Hyperscanning, the simultaneous recording of neural data from multiple participants, enables real-time investigation of social processes by examining multiple brains while they interact. Previous hyperscanning research has largely focused on cooperative tasks, with fewer studies examining competitive contexts. Here, we obtained electroencephalography (EEG) hyperscanning data from 62 participants (31 pairs) who played a computerised version of the Rock-Paper-Scissors game, a classic paradigm for studying competitive decision-making. Although the optimal strategy is to be unpredictable and thus act randomly, participants exhibited behavioural biases, deviating from this ideal. Using multivariate decoding methods to measure neural representations within the two players brains in interaction, we found information about decisions made by participants during gameplay, revealing certain strategies. Notably, losers uniquely represented information about prior trials, suggesting this may impair optimal performance. These results reveal how competitive decision-making is shaped by cognitive biases and previous outcomes, highlighting the difficulty of achieving randomness in strategic contexts. This work advances our understanding of decision-making and cognitive dynamics in competitive interactions.

neuroscience↗

bioRxiv2025

Unlocking information alignment between interacting brains with EEG hyperscanning

Social interactions shape our perception of the world, influencing how we interpret incoming information. Alignment between interacting individuals sensory and cognitive processes is key to successful cooperation and communication, but the neural processes underlying this alignment remain unknown. Here, we leveraged Representational Similarity Analysis (RSA) on electroencephalography (EEG) hyperscanning techniques to investigate information alignment in 24 pairs of participants who performed a categorisation task together based on agreed upon rules. Significant interbrain information alignment emerged within 45 ms of stimulus presentation and persisted for hundreds of milliseconds. Early alignment (45-180 ms) occurred in both real and randomly matched pseudo-pairs, reflecting shared sensory responses. Importantly, alignment after 200 ms strengthened with practice and was unique to real pairs, driven by shared representations associated with, and extending beyond, the categorisation rules they formed. Together, these findings highlight distinct processes underpinning interbrain information alignment during social interactions, that can be effectively captured and disentangled with Interbrain RSA.

neuroscience↗

bioRxiv2024

Mapping object space dimensions: new insights from temporal dynamics

How is object information organized in high-level visual cortex? Recently, a comprehensive model of object space in macaques was proposed, defined by the orthogonal axes of animacy and aspect ratio (Bao et al., 2020). However, when using stimuli that dissociated category, animacy, and aspect ratio in humans, no tuning of aspect ratio was observed in fMRI data (Yargholi & Op de Beeck, 2023). This difference could be a result of different stimuli, the limited temporal resolution of fMRI, or information available about the presented stimuli. Here, we asked if and when information about aspect ratio, animacy, and category is available over time. We collected whole-brain electroencephalography (EEG) data while participants (N = 20) viewed the stimulus set used by Yargholi & Op de Beeck (2023). To mask object details and increase reliance on shape information, we also presented silhouette versions of the stimuli. Stimuli were presented in 5Hz streams using rapid serial visual presentation, with intact and silhouette stimuli sets were shown in separate streams. Using standard multivariate decoding pipelines and representational similarity analysis, we found that aspect ratio, category, and animacy were represented during visual object processing. The dominant dimension was modulated by stimulus type, demonstrating that the observable dimensions of object space depend on the nature of the stimuli presented. Taken together, these findings demonstrate that aspect ratio is represented during object processing, however earlier and more transiently than categorical dimensions, such as animacy. By focusing on underlying temporal dynamics, our results provide clear new insights into the contradicting findings reported in previous work and reveal a more nuanced understanding of how object space evolves over time.

neuroscience↗

bioRxiv2024

Rapid visual engagement in neural processing of detailed touch interactions

Touch perception is an inherently multisensory process in which vision plays an essential role. However, our understanding of how vision encodes sensory and emotional-affective aspects of observed touch, and the timing of these processes, is still limited. Here we address this gap by investigating the neural dynamics of visual touch observation by analysing electroencephalographic (EEG) data from participants viewing detailed hand interactions from the Validated Touch-Video Database. We examined how the brain encodes basic body cues such as hand orientation and viewing perspective, in addition to sensory aspects, including the type of touch (e.g., stroking vs. pressing; hand vs. object touch) and the object involved (e.g., knife, brush), as well as emotional-affective dimensions. Using multivariate decoding, information about body cues was present within 60 ms, with sensory and emotional details, including valence, arousal, and pain present around 130 ms, demonstrating efficient early visual processing. Threat was most clearly identified by approximately 265 ms, similarly involving visual regions, suggesting that such evaluations require slightly extended neural engagement. Our findings reveal that bottom-up, automatic visual processing is integral to complex tactile assessments, important for rapidly extracting both the personal relevance and the sensory and emotional dimensions of touch.

neuroscience↗

bioRxiv2023

Optimising analysis choices for multivariate decoding: creating pseudotrials using trial averaging and resampling

Multivariate pattern analysis (MVPA) is a popular technique that can distinguish between condition-specific patterns of activation. Applied to neuroimaging data, MVPA decoding for inference uses above chance decoding to identify statistically reliable condition-specific information in neuroimaging data which may be missed by univariate methods. However, several analysis choices influence decoding results, and the combined effects of these choices have not been fully evaluated. In particular, an increasingly popular approach is to average data from several trials together before training an MVPA classifier, but the decision about how much averaging to do is arbitrary and the effect of varying this parameter has not been documented. Here we systematically assessed the influence of trial averaging and resampling on decoding accuracy and subsequent statistical outcome on simulated data. Although the optimal parameters varied with the classifier and cross-validation approach used, we found that modest trial averaging using up to 5-10% of the total number of trials per condition improved decoding accuracy and associated t-statistics. In addition, a small amount of resampling could improve t-statistics and classification performance, but was not always necessary. We provide code to allow researchers to optimise these analysis choices for the parameters of their data.

neuroscience↗

bioRxiv2023

Minimal condition repetitions required in rapid serial visual presentation decoding paradigms

Rapid Serial Visual Presentation (RSVP) decoding paradigms allow testing a greater number of conditions than was previously possible within short experimental sessions. However, in these designs individual neural responses may be more susceptible to noise due to responses overlapping with adjacent epochs. This study investigates the minimum number of repetitions required for reliable decoding accuracies in RSVP decoding paradigms. We used previously published EEG data and conducted a standard decoding analysis while varying the number of repetitions used. We found that it is possible to obtain reliable decoding accuracies with only around six repetitions of each condition, which has important implications for research questions that require short experiments, particularly for studying populations who may not be able to tolerate longer or more demanding protocols. These findings highlight the potential benefits of using efficient RSVP decoding designs and conducting short experiments and may have far-reaching impacts in cognitive neuroscience, by providing insights into optimizing data collection methods for diverse populations and experimental protocols.

neuroscience↗