Biology subjects

Kamienkowski, J. E.

Publications and source records attributed to Kamienkowski, J. E..

8 recordsLinked to original sources

bioRxiv2026

Beyond completion time: predicting individual differences in executive functions using cursor trajectory features in an online Trail Making Test

BackgroundThe Trail Making Test (TMT) is a widely used instrument for assessing executive functions due to its sensitivity. But, its traditional scoring, based solely on total completion time, limits its specificity by losing the rich behaviour required to complete the task, involving integrating visual search, motor planning and task switching. A computerised implementation of the TMT (cTMT) allows high-resolution cursor trajectories to be recorded, offering access to fine-grained features of the trajectory, while an online administration improves accessibility and statistical power. ObjectiveThis study aimed to extract high-resolution cursor trajectory features from an online cTMT and evaluate their capacity to predict individual differences in core executive functions, such as visual working memory (VWM), inhibitory control and age, using machine learning, as well as validating the feasibility of a fully online data acquisition pipeline as a basis for digital biomarkers. MethodsParticipants completed an online battery comprising the cTMT and three validation tasks: the Change Detection Task (CDT), the Stop-Signal Task (SST) and the Go/No-Go task (GNG). A total of 104 features (26 features x Part A/B x whole-trial/first-10-targets) were extracted from cursor trajectories, including trajectory profiles and segmentation into latent states: Search, Travel and Hesitation. Regression models were trained within a nested Leave-One-Out cross-validation framework with inner 10-fold hyperparameter tuning, feature selection and standardisation applied strictly within folds. Performance was assessed via mean absolute error (MAE), normalised error (MAE/SD) and permutation testing; SHapley Additive exPlanations (SHAP) were used to characterise feature importance. ResultscTMT features strongly predicted age across all models (p < .001), with MAEs roughly one-third lower than the targets dispersion, driven predominantly by distance-based "circuitousness" metrics from both parts. GNG accuracy and c-coefficient were both significantly predicted, but with a dissociated pattern: accuracy was best explained by Part B (alternation) metrics, whereas the c-coefficient was almost exclusively predicted by Part A (simple sequencing) metrics. SST response time (SSRT) was not significantly predicted by any model. VWM, characterised by the mean Cowan s K, was significantly predicted mainly by the more complex models, and involving state transitions and search phases in both Part A and B. ConclusionsMoving beyond completion time, cursor trajectory dynamics from an online cTMT provide a rich behavioural signal for predicting age, VWM capacity and distinct aspects of inhibitory control. The dissociation between Part A and Part B predictors supports differentiated cognitive processes within the TMT and positions the cTMT as a scalable, portable digital biomarker with promise for computational psychiatry and personalised neuropsychology.

neuroscience↗

bioRxiv2026

Time space signatures of hybrid search resolution using EEG and eye movements concurrent recordings

Understanding how the brain supports visual search in naturalistic environments--where attention and memory must work together to find targets among distractors--requires analysing neural signals where responses overlap in time and multiple environmental variables simultaneously interact. Conventional event-related methods cannot disentangle these overlapping signals, creating a fundamental bottleneck for studying cognition in ecologically valid settings. Here, we seek to isolate activation patterns during a hybrid visual and memory search task in naturalistic scenarios. We show that our deconvolution-based approach applied to coregistered EEG and eye-tracking data resolves this problem, capturing fine-grained activation patterns in the temporal response functions (TRFs) for main effects and their interactions. Starting from hypothesis-driven models, we replicated established components for visual processing and target detection in a Hybrid Search task with unrestricted eye movements. Moreover, extending our approach to hierarchically larger data-driven models enabled us to explore interactions between the effects that have otherwise been studied separately. We showed that the TRF estimates remained stable with increasing model complexity, supported by improved model performance (Pearsons correlation coefficient) and controlled by the variance inflation factor (VIF). We identified a late activation consistent with the P300 component for target detection, and revealed that missed detections elicited similar but weaker responses, suggesting a more nuanced role than simple detection. These findings demonstrate how deconvolution methods, complemented with robust measures of model performance that support its expansion in features space, can uncover the dynamic interplay of attention and memory processes underlying free-viewing behavior.

neuroscience↗

bioRxiv2025

Mapping neural activity during naturalistic visual and memory search

In everyday life, individuals often search for one of several items stored in memory. This cognitive process, known as hybrid search, is critical for tasks like navigating using landmarks. While the behavioral aspects of hybrid search have been extensively studied, the underlying neural mechanisms remain less understood. In this study, we combined concurrent magnetoencephalography (MEG) and eye movement recordings to investigate the oscillatory and evoked neural dynamics supporting hybrid search in naturalistic settings. Twenty-one participants (12 males, 9 females) performed a free-viewing task involving visual search for targets embedded in memory (hybrid search) across naturalistic scenes. Time-Frequency analyses revealed specific neural signatures during memory encoding, retention, and visual search. During encoding and retention, posterior alpha-band power decreased with memory load, reflecting heightened perceptual and mnemonic demands. During visual search, frontoparietal beta-band activity scaled with memory load, suggesting increased cognitive control. By aligning MEG signals to eye movement events and applying source reconstruction, we identified an early visually evoked lambda response, localized to V1, followed by a distributed P3m component, with maximum activation in the right inferior parietal lobe, that discriminated target from distractor fixations. Together, these findings demonstrate how oscillatory and evoked responses dynamically support hybrid search in naturalistic settings, revealing how memory, attention, and visual processing interact during active vision. Significance statementUnderstanding how the brain supports everyday tasks, like searching for familiar objects, requires studying cognition under realistic conditions, yet most knowledge comes from highly controlled lab tasks with simple stimuli. Here, we combined MEG and eye-tracking to examine how the brain supports free-viewing hybrid search -looking for any of several remembered items-across natural scenes. We found memory load modulations in posterior alpha oscillations during encoding and retention, as well as in frontoparietal beta oscillations during search. By time-locking neural signals to eye movements, we identified a robust marker of target detection. These findings provide direct evidence of how attention, memory, and vision interact in realistic settings, helping bridge the gap between controlled experimental paradigms and real-world cognition.

neuroscience↗

bioRxiv2025

Uncertainty during visual search: Insights from a computational model and behavioral experiment in natural stimuli

Visual search, driven by bottom-up and top-down processes, offers a unique framework for investigating decision-making. This study examines individuals awareness of their own visual search by combining computational modeling with behavioral experiments. Fifty-seven participants performed a classical visual search task in which the goal was to find an object in a natural scene. Crucially, in some trials, the search was interrupted by clearing the screen before the gaze reached the target object. Participants had to report their best guess of the targets location and the uncertainty on their response. We show that a modified version of the Entropy-Limit Minimization (ELM) model captures scanpaths and perceived target locations, while also revealing that uncertainty is influenced by scanpath length, the distance between the perceived and true target location, and the entropy of decision maps. These findings highlight the models capacity to reflect cognitive processes underlying response selection and uncertainty judgment.

neuroscience↗

bioRxiv2024

Integrating Ideal Bayesian Searcher and Neural Networks Models for Eye Movement Prediction in a Hybrid Search Task

Visual search, where observers search for a specific item, is a crucial aspect of daily human interaction with the visual environment. Hybrid search extends this by requiring observers to search for any item from a given set of objects. While there are models proficient at simulating human eye movement in visual search tasks within natural scenes, none are able to do so in Hybrid search tasks within similar environments. In this work, we present an enhanced version of the neural network Entropy Limit Minimization (nnELM) model, which is based on a Bayesian framework and decision theory. We also present the Hybrid Search Eye Movements (HSEM) Dataset, comprising several thousands of human eye movements during hybrid search tasks in natural scenes. A key challenge in Hybrid search, absent in visual search, is that participants might search for different objects at different time points. To address this, we developed a strategy based on the posterior probability distribution generated after each fixation. By adjusting the models peripheral visibility, we made early search stages more efficient, aligning it closer to human behaviour. Additionally, limiting the models memory capacity reduced its success in longer searches, mirroring human performance. To validate these improvements, we compared our model against participants from the HSEM dataset and against existing models in a visual search benchmark. Altogether, the new nnELM model not only successfully explains Hybrid search tasks, but also closely replicates human behaviour in natural scenes. This work advances our understanding of complex processes underlying visual and Hybrid search while maintaining model interpretability.

neuroscience↗

bioRxiv2023

Investigating the effects of contextual information, visual working memory, and inhibitory control in hybrid visual search

In real-life scenarios, individuals frequently engage in tasks that involve searching for one of various items stored in memory. This combined process of visual search and memory search is known as hybrid search. To date, most hybrid search studies have been restricted to average observers looking for previously well-memorized targets in blank backgrounds. Here we investigated the effects of context and the role of memory in hybrid search by modifying the tasks memorization phase to occur in all-new single trials. Additionally, we aimed to assess how individual differences in visual working memory capacity and inhibitory control influence performance during hybrid search. In an online experiment, 110 participants searched for potential targets in images with and without context. A change detection and go/no-go task were also performed to measure working memory capacity and inhibitory control, respectively. We show that, in target present trials, the main hallmarks of hybrid search remain present, with a linear relationship between reaction time and visual set size, and a logarithmic relationship between reaction time and memory set size. Context affected search efficiency in different ways. In target-absent trials we found large differences between context-present and absent conditions, suggesting participants adoption of adaptive strategies. Finally, working memory capacity did not predict most search performance measures. Inhibitory control, when relationships were significant, could account for only a small portion of the variability in the data. This study provides insights into the effects of context and individual differences on search efficiency and termination.

animal behavior and cognition↗

bioRxiv2023

Web-based eye-tracking for remote cognitive assessments: The anti-saccade task as a case study

Over the last years, several developments of remote webcam-based eye tracking (ET) prototypes have emerged, testing their feasibility and potential for web-based experiments. This growing interest is mainly explained by the possibility to perform tasks remotely, which allows the study of larger and hard-to-reach populations and potential applications in telemedicine. Nevertheless, a decrease in the quality of the camera and a noisier environment bring new implementation challenges. In this study, we present a new prototype of remote webcam-based ET. First, we introduced improvements to the state-of-the-art remote ET prototypes for cognitive and clinical tasks, e.g. without the necessity of constant mouse interactions. Second, we assessed its spatiotemporal resolution and its reliability within an experiment. Third, we ran a classical experiment, the anti-saccade task, to assess its functionality and limitations. This cognitive test compares horizontal eye movements toward (pro-saccades) or away from (anti-saccades) a target, as a measure of inhibitory control. Our results replicated previous findings obtained with high-quality laboratory ETs. Briefly, higher error rates in anti-saccades compared to pro-saccades were observed, and incorrect responses presented faster reaction times. Our web-ET prototype showed a stable calibration over time and performed well in a classic cognitive experiment. Finally, we discussed the potential of this prototype for clinical applications and its limitations for experimental use.

neuroscience↗

bioRxiv2022

Active search signatures in a free-viewing task exploiting concurrent EEG and eye movements recordings

Tasks we often perform in our everyday lives, such as reading or looking for a friend in the crowd, are seemingly straightforward but they actually require the orchestrated activity of several cognitive processes. Free-viewing visual search requires a plan to move our gaze on the different items, identifying them, and deciding on whether to continue with the search. Little is known about the electrophysiological signatures of these processes in free-viewing since there are technical challenges associated with eye movement artifacts. Here we aimed to study how category information, as well as ecologically-relevant variables such as the task performed, influence brain activity in a free-viewing paradigm. Participants were asked to observe/search from an array of faces and objects embedded in random noise. We concurrently recorded EEG and eye movements and applied a deconvolution analysis approach to estimate the contribution of the different elements embedded in the task. Consistent with classical fixed-gaze experiments and a handful of free-viewing studies, we found a robust categorical effect around 150 ms in occipital and occipitotemporal electrodes. We also report a task effect, more negative in posterior central electrodes in visual search compared to exploration, starting at around 80 ms. We also found significant effects of trial progression, and an interaction with the task effect. Overall, these results generalise the characterisation of early visual face processing to a wider range of experiments and show how a suitable analysis approach allows to discern among multiple neural contributions to the signal, preserving key attributes of real-world tasks.

neuroscience↗