Biology subjects

Schwiedrzik, C. M.

Publications and source records attributed to Schwiedrzik, C. M..

14 recordsLinked to original sources

bioRxiv2025

The Sweet Spot of Plasticity in Perceptual Learning Dependson Training Variability

Perceptual learning improves sensory discrimination, yet the brain must balance specificity and generalization, especially in variable environments. To investigate how variability shapes perceptual learning, we used functional magnetic resonance imaging while human subjects performed an orientation discrimination task with low or high task-irrelevant variability in spatial frequency. Behaviorally, both training regimes improved performance, but only high variability training enabled transfer to a novel location. Retinotopic orientation representations in early visual areas changed similarly across groups, indicating that variability does not alter local sensory representations. In contrast, functional connectivity differed: low variability training enhanced task-dependent coupling between the intraparietal sulcus and V2, while high variability training differentially affected connectivity with hV4. These results support a "sweet spot" model of perceptual learning, in which the locus of learning flexibly depends on the training environment. But rather than modifying sensory representations, variability reconfigures network-level readout, enabling specific or generalizable improvements in perception.

neuroscience↗

bioRxiv2025

Multiple task-demands flexibly optimize neural geometry in human ventral temporal cortex

A hallmark of human intelligence is the ability to perform multiple tasks immediately upon instruction. Yet, the neural processes that implement such flexibility remain unclear. Using intracranial electrophysiology in epilepsy patients, we examined how representational geometries evolve as participants switched among three tasks - individuation, categorization, and conceptualization - on a trial-by-trial basis. Although tasks differed in the required representational geometries, neural representations across cortex were not immediately optimized following task cues. Instead, task-tailored geometries emerged gradually over successive stimulus repetitions within a trial. Ventral temporal cortex was the only region to exhibit task-specific adjustments for all three tasks, dynamically transforming within- and between-category distances in line with task demands. Importantly, these gradual representational changes were behaviorally relevant, predicting trial-level performance. Our results show that behavioral flexibility is supported by incremental, task-dependent refinement of representational geometry already within sensory cortex - far earlier in the processing hierarchy than previously thought.

neuroscience↗

bioRxiv2025

Re-emergence of orientation coding in primate IT cortex and deep networks reveals functional hubs for visual processing

The primate visual system is a hierarchical network of brain areas that transform retinal inputs into rich percepts. According to efficient neural coding principles, basic visual features such as orientation, which are already represented in early visual areas, should not be redundantly encoded in higher areas like V4 and inferotemporal (IT) cortex. We tested this hypothesis by quantifying orientation representation throughout the entire visual stream using functional Magnetic Resonance Imaging (fMRI) and multivoxel pattern analysis (MVPA) in macaque monkeys watching full-field gratings. Contrary to predictions from efficient coding, we found robust orientation information along the entire visual ventral stream from V1 to IT cortex. Orientation information was distributed in cortical and representational space, especially in IT. Voxels with multivariate responsiveness to faces, objects, and bodies carried more orientation information than highly category-selective voxels. They also displayed stronger connectivity within IT and with upstream areas V1-V4. These finding suggests the existence of functional "hubs" that integrate low- and high-level features through enhanced connectivity. This phenomenon generalized to deep convolutional neural network models with high brain similarity, underscoring the importance of hub-like units in hierarchical processing of complex inputs. Together, our results show that orientation re-emerges in IT cortex not through redundancy but as a computational byproduct of functional diversity and connectivity, propounding the existence of "hub" neurons that may support flexible, integrative visual processing beyond object recognition.

neuroscience↗

bioRxiv2025

Correlations of neural predictability and information transfer in cortex and their relation to predictive coding

Predictive-coding like theories agree in describing top-down communication through the cortical hierarchy as a transmission of predictions generated by internal models of the inputs. With respect to the bottom-up connections, however, these theories differ in the neural processing strategies suggested for updating the internal model. Some theories suggest a coding strategy where unpredictable inputs, i.e., those not captured by the internal model, are passed on through the cortical hierarchy, whereas others claim that the predictable part of the inputs is passed on. Here, we addressed which neural coding strategy is employed in cortico-cortical connections using an information-theoretic approach. Our framework allows for quantifying two core aspects of both strategies, namely, predictability of inputs and information transfer, through local active information storage and local transfer entropy, respectively. A previous study on the neural processing of retinal ganglion cells connected to the lateral geniculate nucleus showed a coding for predictable information, captured by an increase in the information transfer with the predictability of inputs. Here, we further investigate predictive coding strategies at the cortical level. In particular, we analyzed LFP activity obtained from intracranial EEG recordings in humans and spike recordings from mouse cortex. We detected cortico-cortical connections with increasing information transfer with the predictability of inputs in recorded channels from frontal, parietal and temporal areas in human cortex. In the mouse visual system, we detected connections exhibiting both an increase and decrease in the information transfer with input predictability, although the former was predominant. Our evidence supports the presence of both predictive coding strategies at the cortical level, with a potential predominance of encoding for predictable information. SummaryThe ability of the brain to infer the hidden causes of sensory experiences has been conceptualized within the computational framework of predictive coding. This framework explains perceptual inference and learning as a process of constantly updating an internal model of the world. Predictive coding describes cortical activity as a communication of sensory evidence and predictions generated from prior expectations. While different views of predictive coding agree on the communication of prior expectations throughout cortex, they differ in how internal expectations are updated. One view states that, to update the internal model, the cortex propagates the mismatch between the expected neural activity and the actual neural response to sensory stimuli. In contrast, another view suggests that the cortex propagates the match between the expected neural activity and the actual neural response. In this work, we were able to tease apart these two views, both in human cortex and the mouse visual system, using information theory. We observed that the brain predominantly propagates expected information, i.e., the match between prior expectations and incoming sensory inputs.

neuroscience↗

bioRxiv2024

Dissociable dynamic effects of expectation during statistical learning.

The brain is thought to generate internal predictions, based on previous statistical regularities in the environment, to optimise behaviour. Predictive processing has been repeatedly demonstrated and seemingly explains expectation suppression (ES), or the attenuation of neural activity in response to expected stimuli. However, the mechanisms behind ES are unclear and various models of the mechanisms supporting ES have been suggested with conflicting evidence. Sharpening models propose that expectations suppress neurons that are not tuned to the expected stimulus, increasing the signal-to-noise ratio for expected stimuli. In contrast, dampening models posit that expectations suppress neurons that are tuned to the expected stimuli, increasing the relative response amplitude for unexpected stimuli. Previous studies have used decoding analyses to examine these effects, with increases in decoding accuracy interpreted in terms of sharpening and decreases related to dampening. The opposing process theory (OPT) has suggested that both processes may occur at different time points, namely that initial sharpening is followed by later dampening of the neural representations of the expected stimulus as learning progresses. In this study we aim to test this theory and shed light on the dynamics of expectation effects, both at single trial level and over time. Thirty-one participants completed a statistical learning task consisting of paired scene categories whereby a "leading" image from one category is quickly followed by a "trailing" image from a different category. Multivariate EEG analyses focussed on decoding stimulus information related to the trailing image category. Within-trial, decoding analyses showed that stimulus expectation increased decoding accuracy at early latencies and decreased decoding accuracy at later latencies, in line with OPT. However, across trials, stimulus expectation decreased decoding accuracy in initial trials and increased decoding accuracy in later trials. We theorise that these dissociable dynamics of expectation effects within and across trials can be explained in the context of hierarchical learning mechanisms. Our single trial results provide evidence for the OPT, while our results over time suggest that sharpening and dampening effects emerge at different stages of learning in the visual domain.

neuroscience↗

bioRxiv2024

Brain Charts for the Rhesus Macaque Lifespan

Recent efforts to chart human brain growth across the lifespan using large-scale MRI data have provided reference standards for human brain development. However, similar models for nonhuman primate (NHP) growth are lacking. The rhesus macaque, a widely used NHP in translational neuroscience due to its similarities in brain anatomy, phylogenetics, cognitive, and social behaviors to humans, serves as an ideal NHP model. This study aimed to create normative growth charts for brain structure across the macaque lifespan, enhancing our understanding of neurodevelopment and aging, and facilitating cross-species translational research. Leveraging data from the PRIMatE Data Exchange (PRIME-DE) and other sources, we aggregated 1,522 MRI scans from 1,024 rhesus macaques. We mapped non-linear developmental trajectories for global and regional brain structural changes in volume, cortical thickness, and surface area over the lifespan. Our findings provided normative charts with centile scores for macaque brain structures and revealed key developmental milestones from prenatal stages to aging, highlighting both species-specific and comparable brain maturation patterns between macaques and humans. The charts offer a valuable resource for future NHP studies, particularly those with small sample sizes. Furthermore, the interactive open resource (https://interspeciesmap.childmind.org) supports cross-species comparisons to advance translational neuroscience research.

neuroscience↗

bioRxiv2024

Task-irrelevant phase but not contrast variability unlocks generalization in visual perceptual learning

Performance on visual tasks can be improved by practice, a process called visual perceptual learning. However, learning-induced performance improvements are often limited to the specific stimuli and visual field locations used during training. Recent research has shown that variability along task-irrelevant stimulus dimensions during training can reduce this specificity. This has been related to higher stages of visual processing that harbor neurons which are invariant to the task-irrelevant dimension. Here, we test whether task-irrelevant trial-by-trial variability in two visual features for which invariances arise at different stages of processing, contrast and spatial phase, results in different degrees of generalization in space in an orientation discrimination task. We find that randomizing spatial phase results in complete generalization of learning to a new spatial location, contrary to randomizing contrast. Our results thus suggest that the neural population undergoing plasticity in visual perceptual learning is determined by the training task, which, in turn, affects generalization. This lends further support to the hypothesis that task-irrelevant variability is an independent factor in determining the specificity of perceptual learning.

neuroscience↗

bioRxiv2023

Predictions enable top-down pattern separation in the macaque face-processing hierarchy

Distinguishing faces requires well distinguishable neural activity patterns. Contextual information may separate neural representations, leading to enhanced identity recognition. Here, we use functional magnetic resonance imaging to investigate how predictions derived from contextual information affect the separability of neural activity patterns in the macaque face-processing system, a 3-level processing hierarchy in ventral visual cortex. We find that in the presence of predictions, early stages of this hierarchy exhibit well separable and high-dimensional neural geometries resembling those at the top of the hierarchy. This is accompanied by a systematic shift of tuning properties from higher to lower areas, endowing lower areas with higher-order, invariant representations instead of their feedforward tuning properties. Thus, top-down signals dynamically transform neural representations of faces into separable and high-dimensional neural geometries. Our results provide evidence how predictive context transforms flexible representational spaces to optimally use the computational resources provided by cortical processing hierarchies for better and faster distinction of facial identities.

neuroscience↗

bioRxiv2022

Visual Perceptual Learning of Feature Conjunctions Leverages Non-linear Mixed Selectivity

Visual objects are often defined by multiple features. Therefore, learning novel objects entails learning conjunctions. Visual cortex is organized into separate compartments, each of which is devoted to processing a single feature. A prime example of this is are neurons purely selective to color and orientation, respectively. However, neurons that jointly encode multiple features (mixed selectivity) also exist across the brain and play critical roles in a multitude of tasks. Here, we sought to uncover the optimal policy that our brain adapts to achieve conjunction learning using these available resources. 59 human subjects practiced orientation-color conjunction learning in four psychophysical experiments designed to nudge the visual system towards using one or the other resource. We find that conjunction learning is possible by linear mixing of pure color and orientation information, but that more and faster learning takes place when pure and mixed selectivity neurons are involved. We also find that learning with mixed selectivity confers advantages in performing an untrained "exclusive or" (XOR) task several months after learning the original conjunction task. This study sheds light on possible mechanisms underlying conjunction learning and highlights the importance of learning by mixed selectivity in such accounts.

neuroscience↗

bioRxiv2022

Input variability taps on invariant representations to enable generalization in visual perceptual learning

Stimulus and location specificity are long considered hallmarks of visual perceptual learning. This renders visual perceptual learning distinct from other forms of learning, where generalization can be more easily attained, and unsuitable for practical applications, where generalization is key. Based on hypotheses derived from the structure of the visual system, we test here whether stimulus variability can unlock generalization in perceptual learning. We train subjects in orientation discrimination, while we vary the amount of variability in a task-irrelevant feature, spatial frequency. We find that independently of task difficulty, this manipulation enables generalization of learning to new stimuli and locations, while not negatively affecting the overall amount of learning on the task. We then use deep neural networks to investigate how variability unlocks generalization. We find that networks develop invariance to the task-irrelevant feature when trained with variable inputs. The degree of learned invariance strongly predicts generalization. A reliance on invariant representations can explain variability-induced generalization in visual perceptual learning, suggests new targets for understanding the neural basis of perceptual learning in high-order visual cortex, and presents an easy to implement modification of common training paradigms that may benefit practical applications.

neuroscience↗

bioRxiv2022

Decision-making processes in perceptual learning depend on effectors

Visual perceptual learning is traditionally thought to arise in visual cortex. However, typical perceptual learning tasks also involve systematic mapping of visual information onto motor actions. Because the motor system contains both effector-specific and effector-unspecific representations, the question arises whether visual perceptual learning is effector-specific itself, or not. Here, we study this question in an orientation discrimination task. Subjects learn to indicate their choices either with joystick movements or with manual reaches. After training, we challenge them to perform the same task with eye movements. We dissect the decision-making process using the drift diffusion model. We find that learning effects on the rate of evidence accumulation depend on effectors, albeit not fully. This suggests that during perceptual learning, visual information is mapped onto effector-specific integrators. Overlap of the populations of neurons encoding motor plans for these effectors may explain partial generalization. Taken together, visual perceptual learning is not limited to visual cortex, but also affects sensorimotor mapping at the interface of visual processing and decision making.

neuroscience↗

bioRxiv2020

Previous motor actions outweigh sensory information in sensorimotor learning

Humans can use their previous experience in form of statistical priors to improve decisions. It is however unclear how such priors are learned and represented. Importantly, it has remained elusive whether prior learning is independent of the sensorimotor system involved in the learning process or not, as both modality-specific and modality-general learning have been reported in the past. Here, we used a saccadic eye movement task to probe the learning and representation of a spatial prior across a few trials. In this task, learning occurs in an unsupervised manner and through encountering trial-by-trial visual hints drawn from a distribution centered on the target location. Using a model-comparison approach, we found that participants prior knowledge is largely represented in the form of their previous motor actions, with minimal influence from the previously seen visual hints. By using two different motor contexts for response (looking either at the estimated target location, or exactly opposite to it), we could further compare whether prior experience obtained in one motor context can be transferred to the other. Although learning curves were highly similar, and participants seemed to use the same strategy for both response types, they could not transfer their knowledge between contexts, as performance and confidence ratings dropped to naive levels after a switch of the required response. Together, our results suggest that humans preferably use the internal representations of their previous motor actions, rather than past incoming sensory information, to form statistical sensorimotor priors on the timescale of a few trials.

neuroscience↗

bioRxiv2019

Pupil diameter tracks statistical structure in the environment

Pupil diameter determines how much light hits the retina, and thus, how much information is available for visual processing. This is regulated by a brainstem reflex pathway. Here, we investigate whether this pathway is under the control of internal models about the environment. If so, this would allow adjusting pupil dynamics to environmental statistics, and hence optimize information transmission. We manipulate environmental temporal statistics by presenting sequences of images that contain internal temporal structure to humans and macaque monkeys. We then measure whether the pupil tracks this structure not only at the rate that immediately arises from variations in luminance, but also at the rate of higher order statistics that are not available from luminance information alone. We find entrainment to environmental statistics in both species during the image sequences. Furthermore, pupil entrainment predicts later performance in an offline task that taps into the same internal models. Thus, the dynamics of the pupil are under control of internal models which adaptively match pupil diameter to the temporal structure of the environment, in line with an active sensing account.

neuroscience↗

bioRxiv2017

An open resource for nonhuman primate imaging

Non-human primate neuroimaging is a rapidly growing area of research that promises to transform and scale translational and cross-species comparative neuroscience.\n\nUnfortunately, the technological and methodological advances of the past two decades have outpaced the accrual of data, which is particularly challenging given the relatively few centers that have the necessary facilities and capabilities. The PRIMate Data Exchange (PRIME-DE) addresses this challenge by aggregating independently acquired non-human primate magnetic resonance imaging (MRI) datasets and openly sharing them via the International Neuroimaging Data-sharing Initiative (INDI). Here, we present the rationale, design and procedures for the PRIME-DE consortium, as well as the initial release, consisting of 13 independent data collections aggregated across 11 sites (total = 98 macaque monkeys). We also outline the unique pitfalls and challenges that should be considered in the analysis of the non-human primate MRI datasets, including providing automated quality assessment of the contributed datasets.

neuroscience↗