Katarzyna Jurewicz
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Katarzyna Jurewicz has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Katarzyna Paluch
Laboratory of Neurophysiology of Mind, Center of Excellence for Neural Plasticity and Brain Disorders: BRAINCITY, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw 02-093, Poland
9 shared publications - Ewa Kublik
Laboratory of Emotions Neurobiology, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland
7 shared publications - Andrzej Wróbel
Nencki Institute of Experimental Biology (Poland)
4 shared publications - Ingrida Antonova
Laboratory of Emotions Neurobiology, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw 02-093, Poland
4 shared publications - Jakub Wojciechowski
Laboratory of Emotions Neurobiology, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland
4 shared publications - Patrycja Dzianok
Laboratory of Emotions Neurobiology, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland
4 shared publications
Latest Publications
- Open Access
Auditory localization improves with aligned saccadic orienting
Animals actively sample the environment through movements of their sensory organs, yet whether these motor behaviors shape perception beyond the modality they directly control remains unresolved. In the auditory system, eye movements modulate neural activity from the auditory periphery to cortex, but their perceptual consequences have remained elusive. Here, we tested whether spontaneous eye movements exert a general influence on auditory perception or selectively interact with auditory computations involved in spatial orienting. Across four psychophysical experiments, participants freely explored natural scenes while performing auditory localization or pitch discrimination tasks. Brief sounds rapidly reorganized spontaneous gaze behavior, preferentially promoting large orienting saccades toward the sound source while suppressing saccades directed away. Critically, auditory localization systematically depended on the direction of spontaneous saccades: sound localization improved when saccades were directed toward the sound source and declined when saccades were directed away. No comparable relationship was observed during pitch discrimination despite closely matched auditory stimulation and visual exploration. These findings identify a selective coupling between saccadic behavior and auditory localization, providing a behavioral counterpart to the widespread oculomotor modulation observed throughout the auditory system. More broadly, our results suggest that the behavioral consequences of eye movements extend beyond vision, selectively engaging sensory computations linked to spatial orienting.
Open AccessAuditory localization improves with aligned saccadic orienting
Animals actively sample the environment through movements of their sensory organs, yet whether these motor behaviors shape perception beyond the modality they directly control remains unresolved. In the auditory system, eye movements modulate neural activity from the auditory periphery to cortex, but their perceptual consequences have remained elusive. Here, we tested whether spontaneous eye movements exert a general influence on auditory perception or selectively interact with auditory computations involved in spatial orienting. Across four psychophysical experiments, participants freely explored natural scenes while performing auditory localization or pitch discrimination tasks. Brief sounds rapidly reorganized spontaneous gaze behavior, preferentially promoting large orienting saccades toward the sound source while suppressing saccades directed away. Critically, auditory localization systematically depended on the direction of spontaneous saccades: sound localization improved when saccades were directed toward the sound source and declined when saccades were directed away. No comparable relationship was observed during pitch discrimination despite closely matched auditory stimulation and visual exploration. These findings identify a selective coupling between saccadic behavior and auditory localization, providing a behavioral counterpart to the widespread oculomotor modulation observed throughout the auditory system. More broadly, our results suggest that the behavioral consequences of eye movements extend beyond vision, selectively engaging sensory computations linked to spatial orienting.
- Open Access
Human-like meaning maps from single-prompt VLM ratings of local scene meaning
Visual attention is shaped both by low-level perceptual features and by the semantic properties of scene regions. However, compared to the salience of low-level image features, the characterization of image properties that lead to scene understanding has been more difficult. The semantic properties of scene regions have been operationalized via meaning maps, proposed by Henderson and Hayes (2017), by having human observers rate the informativeness of scene patches. However, this rating procedure is labor-intensive and limits the method’s versatility. An automated alternative, the DeepMeaning model, reduces rating costs but still requires human training data from a similar class of images, limiting generalizability. Here, we tested whether instruction-following vision-language models (VLMs) can generate human-like ratings in a zero-shot manner: using the original human-rating instructions as a single prompt without any task-specific training. We obtained meaningfulness ratings from both open-weight and proprietary VLMs for scene patches across three meaning-map datasets, and compared the resulting zero-shot AI meaning maps (AIMMs) to human meaning maps and maps obtained from the DeepMeaning model. Zero-shot AIMMs closely approximate human meaning maps, and maps from the task-specific DeepMeaning model, showing that VLMs can replicate human semantic judgments without any training. Zero-shot AIMMs demonstrate utility and validity, matching human meaning maps in their ability to predict human fixations and reproducing the same pattern of relative performance against other gaze-prediction models. We provide open-source code that allows researchers to use this zero-shot approach to quickly and flexibly explore meaning maps across spatial scales, prompts, and scene categories.
Open AccessHuman-like meaning maps from single-prompt VLM ratings of local scene meaning
Visual attention is shaped both by low-level perceptual features and by the semantic properties of scene regions. However, compared to the salience of low-level image features, the characterization of image properties that lead to scene understanding has been more difficult. The semantic properties of scene regions have been operationalized via meaning maps, proposed by Henderson and Hayes (2017), by having human observers rate the informativeness of scene patches. However, this rating procedure is labor-intensive and limits the method’s versatility. An automated alternative, the DeepMeaning model, reduces rating costs but still requires human training data from a similar class of images, limiting generalizability. Here, we tested whether instruction-following vision-language models (VLMs) can generate human-like ratings in a zero-shot manner: using the original human-rating instructions as a single prompt without any task-specific training. We obtained meaningfulness ratings from both open-weight and proprietary VLMs for scene patches across three meaning-map datasets, and compared the resulting zero-shot AI meaning maps (AIMMs) to human meaning maps and maps obtained from the DeepMeaning model. Zero-shot AIMMs closely approximate human meaning maps, and maps from the task-specific DeepMeaning model, showing that VLMs can replicate human semantic judgments without any training. Zero-shot AIMMs demonstrate utility and validity, matching human meaning maps in their ability to predict human fixations and reproducing the same pattern of relative performance against other gaze-prediction models. We provide open-source code that allows researchers to use this zero-shot approach to quickly and flexibly explore meaning maps across spatial scales, prompts, and scene categories.
- Open Access
- 6 June 2026
- Published by Elsevier in NeuroImage
- Vol. 337, 122030
Common but different: An ERP study of single- and multi-source interference processing in MSIT
- 6 June 2026
- Published by Elsevier in NeuroImage
- Vol. 337, 122030
Open AccessCommon but different: An ERP study of single- and multi-source interference processing in MSIT
- Open Access
Common but different: An ERP study of single- and multi-source interference processing in MSIT
Flexible behavior often requires processing of complex, interfering information. Research has investigated conflict-related brain processes mostly using single tasks which hindered direct comparison of different interference types. Thus, the question if they are resolved by a common mechanism or by a set of different, task-specific mechanisms remains open. In this study, we used event-related potentials (ERPs) to examine the spatio-temporal dynamics of cognitive control across Simon, flanker, multi-source and no-conflict conditions. Our findings reveal that all trial types engaged the same sequence of processing stages, as indicated by common ERP waveforms and consistent number and order of microstates across conditions. However, the intensity and duration of these common stages scaled with difficulty of the conflict task (as measured by RTs and accuracy) from Simon to flanker to multi-source interference. Flanker conflict uniquely influenced early ERP components strongly engaging the dorsal attentional system and visual areas, likely due to demands posed by the presence of flanker distractors. Later ERP components (with sources including ventral attention and somatomotor network areas) were affected by both conflicts. Accordingly, when flanker and Simon conflicts were presented together, early processes lineary summed up, but there was an interaction at the later stage of processing paralleling nonlinear drop of accuracy in a multi-conflict condition. Our study provides novel insights into the neural dynamics underlying cognitive control engaged across different conflict types and their interaction. The use of source analysis allowed us to ground ERP-based findings in the wider context of studies, including those using neuroimaging techniques. Highlights: - The same processing stages, in same order are evoked in control and conflict trials - Their intensity/duration scale with conflict behavioral difficulty - Flanker conflict demands enhanced early attentional (DAN) and visual processing - Flanker and Simon interact at the late-stage processing involving VAN and SM networks - Flanker and Simon interaction is sublinear
Open AccessCommon but different: An ERP study of single- and multi-source interference processing in MSIT
Flexible behavior often requires processing of complex, interfering information. Research has investigated conflict-related brain processes mostly using single tasks which hindered direct comparison of different interference types. Thus, the question if they are resolved by a common mechanism or by a set of different, task-specific mechanisms remains open. In this study, we used event-related potentials (ERPs) to examine the spatio-temporal dynamics of cognitive control across Simon, flanker, multi-source and no-conflict conditions. Our findings reveal that all trial types engaged the same sequence of processing stages, as indicated by common ERP waveforms and consistent number and order of microstates across conditions. However, the intensity and duration of these common stages scaled with difficulty of the conflict task (as measured by RTs and accuracy) from Simon to flanker to multi-source interference. Flanker conflict uniquely influenced early ERP components strongly engaging the dorsal attentional system and visual areas, likely due to demands posed by the presence of flanker distractors. Later ERP components (with sources including ventral attention and somatomotor network areas) were affected by both conflicts. Accordingly, when flanker and Simon conflicts were presented together, early processes lineary summed up, but there was an interaction at the later stage of processing paralleling nonlinear drop of accuracy in a multi-conflict condition. Our study provides novel insights into the neural dynamics underlying cognitive control engaged across different conflict types and their interaction. The use of source analysis allowed us to ground ERP-based findings in the wider context of studies, including those using neuroimaging techniques. Highlights: - The same processing stages, in same order are evoked in control and conflict trials - Their intensity/duration scale with conflict behavioral difficulty - Flanker conflict demands enhanced early attentional (DAN) and visual processing - Flanker and Simon interact at the late-stage processing involving VAN and SM networks - Flanker and Simon interaction is sublinear
- Open Access
- 30 July 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-15
Irrational choices via a curvilinear representational geometry for value
We make decisions by comparing values, but it is not yet clear how value is represented in the brain. Many models assume, if only implicitly, that the representational geometry of value is linear. However, in part due to a historical focus on noisy single neurons, rather than neuronal populations, this hypothesis has not been rigorously tested. Here, we examine the representational geometry of value in the ventromedial prefrontal cortex (vmPFC), a part of the brain linked to economic decision-making, in two male rhesus macaques. We find that values are encoded along a curved manifold in vmPFC. This curvilinear geometry predicts a specific pattern of irrational decision-making: that decision-makers will make worse choices when an irrelevant, decoy option is worse in value, compared to when it is better. We observe this type of irrational choices in behavior. Together, these results not only suggest that the representational geometry of value is nonlinear, but that this nonlinearity could impose bounds on rational decision-making.
- 30 July 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-15
Open AccessIrrational choices via a curvilinear representational geometry for value
We make decisions by comparing values, but it is not yet clear how value is represented in the brain. Many models assume, if only implicitly, that the representational geometry of value is linear. However, in part due to a historical focus on noisy single neurons, rather than neuronal populations, this hypothesis has not been rigorously tested. Here, we examine the representational geometry of value in the ventromedial prefrontal cortex (vmPFC), a part of the brain linked to economic decision-making, in two male rhesus macaques. We find that values are encoded along a curved manifold in vmPFC. This curvilinear geometry predicts a specific pattern of irrational decision-making: that decision-makers will make worse choices when an irrelevant, decoy option is worse in value, compared to when it is better. We observe this type of irrational choices in behavior. Together, these results not only suggest that the representational geometry of value is nonlinear, but that this nonlinearity could impose bounds on rational decision-making.
- Open Access
- 9 January 2024
- Published by Wiley in Human Brain Mapping
- Vol. 45 (1) , e26549
Common and distinct BOLD correlates of Simon and flanker conflicts which can(not) be reduced to time‐on‐task effects
The ability to identify and resolve conflicts between standard, well‐trained behaviors and behaviors required by the current context is an essential feature of cognitive control. To date, no consensus has been reached on the brain mechanisms involved in exerting such control: while some studies identified diverse patterns of activity across different conflicts, other studies reported common resources across conflict tasks or even across simple tasks devoid of the conflict component. The latter reports attributed the entire activity observed in the presence of conflict to longer time spent on the task (i.e., to the so‐called time‐on‐task effects). Here, we used an extended Multi‐Source Interference Task (MSIT) which combines Simon and flanker types of interference to determine shared and conflict‐specific mechanisms of conflict resolution in fMRI and their separability from the time‐on‐task effects. Large portions of the activity in the dorsal attention network and decreases of activity in the default mode network were shared across the tasks and scaled in parallel with increasing reaction times. Importantly, the activity in the sensory and sensorimotor cortices, as well as in the posterior medial frontal cortex (pMFC) – a key region implicated in conflict processing – could not be exhaustively explained by the time‐on‐task effects.
- 9 January 2024
- Published by Wiley in Human Brain Mapping
- Vol. 45 (1) , e26549
Open AccessCommon and distinct BOLD correlates of Simon and flanker conflicts which can(not) be reduced to time‐on‐task effects
The ability to identify and resolve conflicts between standard, well‐trained behaviors and behaviors required by the current context is an essential feature of cognitive control. To date, no consensus has been reached on the brain mechanisms involved in exerting such control: while some studies identified diverse patterns of activity across different conflicts, other studies reported common resources across conflict tasks or even across simple tasks devoid of the conflict component. The latter reports attributed the entire activity observed in the presence of conflict to longer time spent on the task (i.e., to the so‐called time‐on‐task effects). Here, we used an extended Multi‐Source Interference Task (MSIT) which combines Simon and flanker types of interference to determine shared and conflict‐specific mechanisms of conflict resolution in fMRI and their separability from the time‐on‐task effects. Large portions of the activity in the dorsal attention network and decreases of activity in the default mode network were shared across the tasks and scaled in parallel with increasing reaction times. Importantly, the activity in the sensory and sensorimotor cortices, as well as in the posterior medial frontal cortex (pMFC) – a key region implicated in conflict processing – could not be exhaustively explained by the time‐on‐task effects.
- Open Access
- 22 August 2023
- Published by Center for Open Science in PsyArXiv
Fast and frequent corrections of saccadic decisions underlie a satisficing strategy when searching everyday scenes
It remains unclear how and how much visual mechanisms integrate visual information across fixations. We analyze three large eye-movement datasets from human observers free-viewing or searching for category-defined targets in photographs of everyday scenes (COCO-Search18, COCO-FreeView and 900 Scenes). During goal-directed search, fixation periods with strikingly short duration (< 125 ms) form over a third of all fixations after the first saccade. Visual information obtained during these short-duration fixations cannot influence the following short?latency saccade, which therefore has to be executed based on target identity and location information carried over from previous fixations. Short-latency saccades are goal-directed: they foveate the search-target more often than other (regular-latency) saccades. Active searching and search-target salience increase the frequency of short-duration fixations. Human searchers use a satisficing strategy based on information transfer across saccades that aids minimizing the cost of making additional saccades to non-target stimuli.
- 22 August 2023
- Published by Center for Open Science in PsyArXiv
Open AccessFast and frequent corrections of saccadic decisions underlie a satisficing strategy when searching everyday scenes
It remains unclear how and how much visual mechanisms integrate visual information across fixations. We analyze three large eye-movement datasets from human observers free-viewing or searching for category-defined targets in photographs of everyday scenes (COCO-Search18, COCO-FreeView and 900 Scenes). During goal-directed search, fixation periods with strikingly short duration (< 125 ms) form over a third of all fixations after the first saccade. Visual information obtained during these short-duration fixations cannot influence the following short?latency saccade, which therefore has to be executed based on target identity and location information carried over from previous fixations. Short-latency saccades are goal-directed: they foveate the search-target more often than other (regular-latency) saccades. Active searching and search-target salience increase the frequency of short-duration fixations. Human searchers use a satisficing strategy based on information transfer across saccades that aids minimizing the cost of making additional saccades to non-target stimuli.
- Open Access
Common and distinct BOLD correlates of Simon and flanker conflicts which can(not) be reduced to time-on-task effects
The ability to identify and resolve conflicts between standard, well trained behaviors, and behaviors required by the current context is an essential feature of cognitive control. To date, no consensus has been reached on the brain mechanisms involved in exerting such control: while some studies identified diverse patterns of activity across different conflicts, other studies reported common resources across conflict tasks or even across simple tasks devoid of conflict component. The latter reports attributed the entire activity observed in the presence of conflict to longer time spent on the task (i.e. to the so-called time-on-task effects). Here we used an extended Multi-Source Interference Task (MSIT) which combines Simon and flanker types of interference to determine shared and conflict-specific mechanisms of conflict resolution in fMRI, and their separability from the time-on-task effects. Increases of activity in the dorsal attention network and decreases of activity in the default mode network were largely shared across the tasks and scaled in parallel with increasing reaction times. Importantly, activity in the sensory and sensorimotor cortices, as well as in the posterior medial frontal cortex (pMFC)–a key region implicated in conflict processing–could not be exhaustively explained by the time-on-task effects. Highlights Flanker and Simon conflicts activate and deactivate largely the same brain regions. Common activity in DAN and DMN is mostly explained by time-on-task effects. Conflict-specific activity emerges mostly in the sensory cortices. pMFC (incl. dACC and pre-SMA) shows both time-on-task and conflict-related activity.
Open AccessCommon and distinct BOLD correlates of Simon and flanker conflicts which can(not) be reduced to time-on-task effects
The ability to identify and resolve conflicts between standard, well trained behaviors, and behaviors required by the current context is an essential feature of cognitive control. To date, no consensus has been reached on the brain mechanisms involved in exerting such control: while some studies identified diverse patterns of activity across different conflicts, other studies reported common resources across conflict tasks or even across simple tasks devoid of conflict component. The latter reports attributed the entire activity observed in the presence of conflict to longer time spent on the task (i.e. to the so-called time-on-task effects). Here we used an extended Multi-Source Interference Task (MSIT) which combines Simon and flanker types of interference to determine shared and conflict-specific mechanisms of conflict resolution in fMRI, and their separability from the time-on-task effects. Increases of activity in the dorsal attention network and decreases of activity in the default mode network were largely shared across the tasks and scaled in parallel with increasing reaction times. Importantly, activity in the sensory and sensorimotor cortices, as well as in the posterior medial frontal cortex (pMFC)–a key region implicated in conflict processing–could not be exhaustively explained by the time-on-task effects. Highlights Flanker and Simon conflicts activate and deactivate largely the same brain regions. Common activity in DAN and DMN is mostly explained by time-on-task effects. Conflict-specific activity emerges mostly in the sensory cortices. pMFC (incl. dACC and pre-SMA) shows both time-on-task and conflict-related activity.
- Open Access
Irrational choices via a curvilinear representational geometry for value
We make decisions by comparing values, but how is value represented in the brain? Many models assume, if only implicitly, that the representational geometry of value is linear. However, in part due to a historical focus on noisy single neurons, rather than neuronal populations, this hypothesis has not been rigorously tested. Here, we examined the representational geometry of value in the ventromedial prefrontal cortex (vmPFC), a part of the brain linked to economic decision-making. We found that values were encoded along a curved manifold in vmPFC. This curvilinear geometry predicted a specific pattern of irrational decision-making: that decision-makers will make worse choices when an irrelevant, decoy option is worse in value, compared to when it is better. Indeed, we observed these irrational choices in behavior. Together, these results not only suggest that the representational geometry of value is nonlinear, but that this nonlinearity could impose bounds on rational decision-making.
Open AccessIrrational choices via a curvilinear representational geometry for value
We make decisions by comparing values, but how is value represented in the brain? Many models assume, if only implicitly, that the representational geometry of value is linear. However, in part due to a historical focus on noisy single neurons, rather than neuronal populations, this hypothesis has not been rigorously tested. Here, we examined the representational geometry of value in the ventromedial prefrontal cortex (vmPFC), a part of the brain linked to economic decision-making. We found that values were encoded along a curved manifold in vmPFC. This curvilinear geometry predicted a specific pattern of irrational decision-making: that decision-makers will make worse choices when an irrelevant, decoy option is worse in value, compared to when it is better. Indeed, we observed these irrational choices in behavior. Together, these results not only suggest that the representational geometry of value is nonlinear, but that this nonlinearity could impose bounds on rational decision-making.