Aniol Santo-Angles
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Aniol Santo-Angles has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
Ainsley Temudo
Division of Science and Mathematics, New York University Abu Dhabi, Abu Dhabi, United Arab Emirates
2 shared publications- Grace W. Lindsay
Center for Data Science, New York University
2 shared publications - Kartik K. Sreenivasan
Division of Science and Mathematics, New York University Abu Dhabi, Abu Dhabi, United Arab Emirates
2 shared publications Ying Zhou
Division of Science and Mathematics, New York University Abu Dhabi, Abu Dhabi, United Arab Emirates
2 shared publications- Albert Compte
Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain
1 shared publication - Amalia Guerrero-Pedraza
FIDMAG Germanes Hospitalàries Research Foundation, Av Jordà 8, 08035 Barcelona, Spain
1 shared publication
Latest Publications
- Open Access
- 23 October 2026
- Published by Springer Nature in Nature Communications
Neural subspaces encode sequential working memory without detectable neural sequences
The neural mechanisms of multiple-item working memory are not well understood. In the current study, we address two competing hypotheses about the neural basis of sequential working memory: neural subspaces versus neural sequences. Using broadband MEG data from human participants, we applied dimensionality reduction and multivariate decoding techniques to test whether a sequence of items is maintained through the reactivation of individual items in sequence (neural sequences) or by organizing them into distinct low-dimensional subspaces (neural subspaces). Our results revealed behaviorally relevant, low-dimensional neural subspaces that organized memory representations according to ordinal position in the sequence, supporting the neural subspaces hypothesis. In contrast, we found no evidence of sequential neural replay during the retention period, contrary to the neural sequences hypothesis. In this work, we show that sequential working memory is maintained through structured geometric organization in low-dimensional representational space, rather than through the sequential reactivation of individual items, thus clarifying the mechanisms of multi-item working memory.
- 23 October 2026
- Published by Springer Nature in Nature Communications
Open AccessNeural subspaces encode sequential working memory without detectable neural sequences
The neural mechanisms of multiple-item working memory are not well understood. In the current study, we address two competing hypotheses about the neural basis of sequential working memory: neural subspaces versus neural sequences. Using broadband MEG data from human participants, we applied dimensionality reduction and multivariate decoding techniques to test whether a sequence of items is maintained through the reactivation of individual items in sequence (neural sequences) or by organizing them into distinct low-dimensional subspaces (neural subspaces). Our results revealed behaviorally relevant, low-dimensional neural subspaces that organized memory representations according to ordinal position in the sequence, supporting the neural subspaces hypothesis. In contrast, we found no evidence of sequential neural replay during the retention period, contrary to the neural sequences hypothesis. In this work, we show that sequential working memory is maintained through structured geometric organization in low-dimensional representational space, rather than through the sequential reactivation of individual items, thus clarifying the mechanisms of multi-item working memory.
- Open Access
Working memory operations emerge from dynamic changes in neural subspace geometry
Working memory representations of multiple items are maintained within quasi-orthogonal neural subspaces, a mechanism thought to reduce interference between memory contents. Previous studies in both non-human and human primates have demonstrated the existence of such subspaces, but it remains unclear how cognitive operations are performed on the information they encode. One hypothesis proposes that these operations are implemented through dynamic changes in subspace geometry, yet direct empirical evidence for this idea in the context of working memory is lacking. Furthermore, the neural mechanisms underlying such geometric reconfigurations remain unknown, particularly the role of oscillatory and aperiodic neural activity. Here, we addressed these questions using simultaneous magneto- and electroencephalography (M/EEG) recordings from healthy human participants performing a multi-item, multi-feature delayed match-to-sample visual WM task. Neural subspaces were estimated using dimensionality reduction techniques, and their geometry was characterized by quantifying the orthogonality between subspaces, distances between memory representations and subspace shapes. Participants maintained orientation and shape information simultaneously while task demands required either deprioritizing one feature or updating the representation of the prioritized feature. We found that subspace geometry was flexibly reshaped according to task demands: prioritization expanded representational subspaces, whereas de-prioritization shrunk them. In addition, subspaces representing different feature domains were generally oblique and became more orthogonal only under high cognitive demands. Finally, these task-dependent changes in subspace geometry were largely driven by interactions between aperiodic neural activity and slow oscillations in the delta and theta frequency bands.
Open AccessWorking memory operations emerge from dynamic changes in neural subspace geometry
Working memory representations of multiple items are maintained within quasi-orthogonal neural subspaces, a mechanism thought to reduce interference between memory contents. Previous studies in both non-human and human primates have demonstrated the existence of such subspaces, but it remains unclear how cognitive operations are performed on the information they encode. One hypothesis proposes that these operations are implemented through dynamic changes in subspace geometry, yet direct empirical evidence for this idea in the context of working memory is lacking. Furthermore, the neural mechanisms underlying such geometric reconfigurations remain unknown, particularly the role of oscillatory and aperiodic neural activity. Here, we addressed these questions using simultaneous magneto- and electroencephalography (M/EEG) recordings from healthy human participants performing a multi-item, multi-feature delayed match-to-sample visual WM task. Neural subspaces were estimated using dimensionality reduction techniques, and their geometry was characterized by quantifying the orthogonality between subspaces, distances between memory representations and subspace shapes. Participants maintained orientation and shape information simultaneously while task demands required either deprioritizing one feature or updating the representation of the prioritized feature. We found that subspace geometry was flexibly reshaped according to task demands: prioritization expanded representational subspaces, whereas de-prioritization shrunk them. In addition, subspaces representing different feature domains were generally oblique and became more orthogonal only under high cognitive demands. Finally, these task-dependent changes in subspace geometry were largely driven by interactions between aperiodic neural activity and slow oscillations in the delta and theta frequency bands.
- Open Access
Neural Subspaces Encode Sequential Working Memory, but Neural Sequences Do Not
The neural mechanisms of multiple-item working memory are not well understood. In the current study, we address two competing hypotheses about the neural basis of sequential working memory: neural subspaces versus neural sequences. Using broadband MEG data from human participants, we applied dimensionality reduction and multivariate decoding techniques to test whether sequential items are maintained during the retention period through the reactivation of individual items in sequence (neural sequences), or by organizing them into distinct low-dimensional subspaces (neural subspaces). Our results revealed behaviorally relevant, low-dimensional neural subspaces that organized memory representations during the retention period but not during stimulus encoding, supporting the neural subspaces hypothesis. In contrast, we found no evidence of sequential neural replay during the delay period, contrary to predictions from the neural sequences hypothesis. Together, our findings suggest that sequential working memory is maintained through structured geometric organization in low-dimensional representational space, rather than through the sequential reactivation of individual items.
Open AccessNeural Subspaces Encode Sequential Working Memory, but Neural Sequences Do Not
The neural mechanisms of multiple-item working memory are not well understood. In the current study, we address two competing hypotheses about the neural basis of sequential working memory: neural subspaces versus neural sequences. Using broadband MEG data from human participants, we applied dimensionality reduction and multivariate decoding techniques to test whether sequential items are maintained during the retention period through the reactivation of individual items in sequence (neural sequences), or by organizing them into distinct low-dimensional subspaces (neural subspaces). Our results revealed behaviorally relevant, low-dimensional neural subspaces that organized memory representations during the retention period but not during stimulus encoding, supporting the neural subspaces hypothesis. In contrast, we found no evidence of sequential neural replay during the delay period, contrary to predictions from the neural sequences hypothesis. Together, our findings suggest that sequential working memory is maintained through structured geometric organization in low-dimensional representational space, rather than through the sequential reactivation of individual items.
- Open Access
- 4 March 2024
- Published by Frontiers Media SA in Frontiers in Human Neuroscience
- Vol. 18, 1339728
Effective connectivity of working memory performance: a DCM study of MEG data
Visual working memory (WM) engages several nodes of a large-scale network that includes frontal, parietal, and visual regions; however, little is understood about how these regions interact to support WM behavior. In particular, it is unclear whether network dynamics during WM maintenance primarily represent feedforward or feedback connections. This question has important implications for current debates about the relative roles of frontoparietal and visual regions in WM maintenance. In the current study, we investigated the network activity supporting WM using MEG data acquired while healthy subjects performed a multi-item delayed estimation WM task. We used computational modeling of behavior to discriminate correct responses (high accuracy trials) from two different types of incorrect responses (low accuracy and swap trials), and dynamic causal modeling of MEG data to measure effective connectivity. We observed behaviorally dependent changes in effective connectivity in a brain network comprising frontoparietal and early visual areas. In comparison with high accuracy trials, frontoparietal and frontooccipital networks showed disrupted signals depending on type of behavioral error. Low accuracy trials showed disrupted feedback signals during early portions of WM maintenance and disrupted feedforward signals during later portions of maintenance delay, while swap errors showed disrupted feedback signals during the whole delay period. These results support a distributed model of WM that emphasizes the role of visual regions in WM storage and where changes in large scale network configurations can have important consequences for memory-guided behavior.
- 4 March 2024
- Published by Frontiers Media SA in Frontiers in Human Neuroscience
- Vol. 18, 1339728
Open AccessEffective connectivity of working memory performance: a DCM study of MEG data
Visual working memory (WM) engages several nodes of a large-scale network that includes frontal, parietal, and visual regions; however, little is understood about how these regions interact to support WM behavior. In particular, it is unclear whether network dynamics during WM maintenance primarily represent feedforward or feedback connections. This question has important implications for current debates about the relative roles of frontoparietal and visual regions in WM maintenance. In the current study, we investigated the network activity supporting WM using MEG data acquired while healthy subjects performed a multi-item delayed estimation WM task. We used computational modeling of behavior to discriminate correct responses (high accuracy trials) from two different types of incorrect responses (low accuracy and swap trials), and dynamic causal modeling of MEG data to measure effective connectivity. We observed behaviorally dependent changes in effective connectivity in a brain network comprising frontoparietal and early visual areas. In comparison with high accuracy trials, frontoparietal and frontooccipital networks showed disrupted signals depending on type of behavioral error. Low accuracy trials showed disrupted feedback signals during early portions of WM maintenance and disrupted feedforward signals during later portions of maintenance delay, while swap errors showed disrupted feedback signals during the whole delay period. These results support a distributed model of WM that emphasizes the role of visual regions in WM storage and where changes in large scale network configurations can have important consequences for memory-guided behavior.
- Open Access
Alpha phase coding supports feature binding during working memory maintenance
The ability to successfully retain and manipulate information in working memory (WM) requires that objects’ individual features are bound into cohesive representations; yet, the mechanisms supporting feature binding remain unclear. Binding (or swap) errors, where memorized features are erroneously associated with the wrong object, can provide a window into the intrinsic limits in capacity of WM that represent a key bottleneck in our cognitive ability. We tested the hypothesis that binding in WM is accomplished via neural phase synchrony and that swap errors result from perturbations in this synchrony. Using magnetoencephalography data collected from human subjects in a task designed to induce swap errors, we showed that swaps are characterized by reduced phase-locked oscillatory activity during memory retention, as predicted by an attractor model of spiking neural networks. Further, we found that this reduction arises from increased phase coding variability in the alpha-band over a distributed network of sensorimotor areas. Our findings demonstrate that feature binding in WM is accomplished through phase coding dynamics that emerge from the competition between different memories. Significance: We investigate the neural basis of working memory, focusing on how feature binding is accomplished and how binding or ‘swap’ errors arise. Using magnetoencephalography, we found that stable phase-locking of alpha oscillations supports correct feature binding, while swap errors correlate with reduced alpha phase preservation, localized to specific brain areas. These findings align with a biologically-plausible computational model predicting that temporal synchrony in neuronal firing underpins feature binding. This work advances our understanding of the neural mechanisms of working memory, providing empirical support for theories of time-based binding and demonstrating the utility of biophysically-realistic models in human neuroimaging studies.
Open AccessAlpha phase coding supports feature binding during working memory maintenance
The ability to successfully retain and manipulate information in working memory (WM) requires that objects’ individual features are bound into cohesive representations; yet, the mechanisms supporting feature binding remain unclear. Binding (or swap) errors, where memorized features are erroneously associated with the wrong object, can provide a window into the intrinsic limits in capacity of WM that represent a key bottleneck in our cognitive ability. We tested the hypothesis that binding in WM is accomplished via neural phase synchrony and that swap errors result from perturbations in this synchrony. Using magnetoencephalography data collected from human subjects in a task designed to induce swap errors, we showed that swaps are characterized by reduced phase-locked oscillatory activity during memory retention, as predicted by an attractor model of spiking neural networks. Further, we found that this reduction arises from increased phase coding variability in the alpha-band over a distributed network of sensorimotor areas. Our findings demonstrate that feature binding in WM is accomplished through phase coding dynamics that emerge from the competition between different memories. Significance: We investigate the neural basis of working memory, focusing on how feature binding is accomplished and how binding or ‘swap’ errors arise. Using magnetoencephalography, we found that stable phase-locking of alpha oscillations supports correct feature binding, while swap errors correlate with reduced alpha phase preservation, localized to specific brain areas. These findings align with a biologically-plausible computational model predicting that temporal synchrony in neuronal firing underpins feature binding. This work advances our understanding of the neural mechanisms of working memory, providing empirical support for theories of time-based binding and demonstrating the utility of biophysically-realistic models in human neuroimaging studies.
- 11 April 2021
- Published by Springer Nature in Brain Structure and Function
- Vol. 226 (5) , 1553-1569
Reward and fictive prediction error signals in ventral striatum: asymmetry between factual and counterfactual processing
Reward prediction error, the difference between the expected and obtained reward, is known to act as a reinforcement learning neural signal. In the current study, we propose a model fitting approach that combines behavioral and neural data to fit computational models of reinforcement learning. Briefly, we penalized subject-specific fitted parameters that moved away too far from the group median, except when that deviation led to an improvement in the model’s fit to neural responses. By means of a probabilistic monetary learning task and fMRI, we compared our approach with standard model fitting methods. Q-learning outperformed actor-critic at both behavioral and neural level, although the inclusion of neuroimaging data into model fitting improved the fit of actor-critic models. We observed both action-value and state-value prediction error signals in the striatum, while standard model fitting approaches failed to capture state-value signals. Finally, left ventral striatum correlated with reward prediction error while right ventral striatum with fictive prediction error, suggesting a functional hemispheric asymmetry regarding prediction-error driven learning.
- 11 April 2021
- Published by Springer Nature in Brain Structure and Function
- Vol. 226 (5) , 1553-1569
Reward and fictive prediction error signals in ventral striatum: asymmetry between factual and counterfactual processing
Reward prediction error, the difference between the expected and obtained reward, is known to act as a reinforcement learning neural signal. In the current study, we propose a model fitting approach that combines behavioral and neural data to fit computational models of reinforcement learning. Briefly, we penalized subject-specific fitted parameters that moved away too far from the group median, except when that deviation led to an improvement in the model’s fit to neural responses. By means of a probabilistic monetary learning task and fMRI, we compared our approach with standard model fitting methods. Q-learning outperformed actor-critic at both behavioral and neural level, although the inclusion of neuroimaging data into model fitting improved the fit of actor-critic models. We observed both action-value and state-value prediction error signals in the striatum, while standard model fitting approaches failed to capture state-value signals. Finally, left ventral striatum correlated with reward prediction error while right ventral striatum with fictive prediction error, suggesting a functional hemispheric asymmetry regarding prediction-error driven learning.
- Open Access
- 4 November 2019
- Published by Cambridge University Press (CUP) in Psychological Medicine
- Vol. 51 (1) , 121-128
Autobiographical memory and default mode network function in schizophrenia: an fMRI study
Background: The brain functional correlates of autobiographical recall are well established, but have been little studied in schizophrenia. Additionally, autobiographical memory is one of a small number of cognitive tasks that activates rather than de-activates the default mode network, which has been found to be dysfunctional in this disorder.Methods: Twenty-seven schizophrenic patients and 30 healthy controls underwent functional magnetic resonance imaging while viewing cue words that evoked autobiographical memories. Control conditions included both non-memory-evoking cues and a low level baseline (cross fixation).Results: Compared to both non-memory evoking cues and low level baseline, autobiographical recall was associated with activation in default mode network regions in the controls including the medial frontal cortex, the posterior cingulate cortex and the hippocampus, as well as other areas. Clusters of de-activation were seen outside the default mode network. There were no activation differences between the schizophrenic patients and the controls, but the patients showed clusters of failure of de-activation in non-default mode network regions.Conclusions: According to this study, patients with schizophrenia show intact activation of the default mode network and other regions associated with recall of autobiographical memories. The finding of failure of de-activation outside the network suggests that schizophrenia may be associated with a general difficulty in de-activation rather than dysfunction of the default mode networkper se.
- 4 November 2019
- Published by Cambridge University Press (CUP) in Psychological Medicine
- Vol. 51 (1) , 121-128
Open AccessAutobiographical memory and default mode network function in schizophrenia: an fMRI study
Background: The brain functional correlates of autobiographical recall are well established, but have been little studied in schizophrenia. Additionally, autobiographical memory is one of a small number of cognitive tasks that activates rather than de-activates the default mode network, which has been found to be dysfunctional in this disorder.Methods: Twenty-seven schizophrenic patients and 30 healthy controls underwent functional magnetic resonance imaging while viewing cue words that evoked autobiographical memories. Control conditions included both non-memory-evoking cues and a low level baseline (cross fixation).Results: Compared to both non-memory evoking cues and low level baseline, autobiographical recall was associated with activation in default mode network regions in the controls including the medial frontal cortex, the posterior cingulate cortex and the hippocampus, as well as other areas. Clusters of de-activation were seen outside the default mode network. There were no activation differences between the schizophrenic patients and the controls, but the patients showed clusters of failure of de-activation in non-default mode network regions.Conclusions: According to this study, patients with schizophrenia show intact activation of the default mode network and other regions associated with recall of autobiographical memories. The finding of failure of de-activation outside the network suggests that schizophrenia may be associated with a general difficulty in de-activation rather than dysfunction of the default mode networkper se.