Biology subjects

Farahat, A.

Publications and source records attributed to Farahat, A..

4 recordsLinked to original sources

bioRxiv2025

Representational drift shows same-class acceleration in visual cortex and artificial neural networks

The neural code is not fixed but shows substantial representational drift over time. It has been proposed that representational drift reflects continuous learning resulting from input-dependent plasticity. Using theoretical analysis and simulations in artificial neural networks, we show that input-dependent plasticity entails a same-class-acceleration principle: Representational drift for a given class of stimuli is predominantly caused by presenting that class of stimuli, rather than presenting other classes of stimuli. We analyze electrophysiological recordings of mouse visual cortex to examine whether within-session representational drift is consistent with this principle. Within-session representational drift was not explained by changes in behavioral state. Instead, it showed a systematic temporal structure that reflected sensory experience rather than behavior. Drift for a given set of stimuli accelerated during blocks presenting that set of stimuli and slowed down during blocks in which other stimuli were presented. Thus, sensory inputs, not elapsed time, organize representational drift in a stimulus-specific manner, consistent with theoretical predictions for training neural networks.

neuroscience↗

bioRxiv2025

Neural responses in early, but not late, visual cortex are well predicted by random-weight CNNs with sufficient model complexity

Convolutional neural networks (CNNs) were inspired by the organization of the primate visual system, and in turn have become effective models of the visual cortex, allowing for accurate predictions of neural stimulus responses. While training CNNs on brain-relevant object-recognition tasks may be an important pre-requisite to predict brain activity, the CNNs brain-like architecture alone may already allow for accurate prediction of neural activity. Here, we evaluated the performance of both task-optimized and brain-optimized convolutional neural networks (CNNs) in predicting neural responses across visual cortex, and performed systematic architectural manipulations and comparisons between trained and untrained feature extractors to reveal key structural components influencing model performance. For human and monkey area V1, random-weight CNNs employing the ReLU activation function, combined with either average or max pooling, significantly outperformed other activation functions. Random-weight CNNs matched their trained counterparts in predicting V1 responses. The extent to which V1 responses can be predicted correlated strongly with the neural networks complexity, which reflects the non-linearity of neural activation functions and pooling operations. However, this correlation between encoding performance and complexity was significantly weaker for higher visual areas that are classically associated with object recognition, such as monkey IT. To test whether this difference between visual areas reflects functional differences, we trained neural network models on both texture discrimination and object recognition tasks. Consistent with our hypothesis, model complexity correlated more strongly with performance on texture discrimination than object recognition. Our findings indicate that random-weight CNNs with sufficient model complexity allow for comparable prediction of V1 activity as trained CNNs, while higher visual areas require precise weight configurations acquired through training via gradient descent.

neuroscience↗

bioRxiv2023

Ventrointermediate thalamic stimulation improves motor learning in humans

Ventrointermediate thalamic stimulation (VIM-DBS) modulates oscillatory activity in a cortical network including primary motor cortex, premotor cortex, and parietal cortex. Here we show that, beyond the beneficial effects of VIM-DBS on motor execution, this form of invasive stimulation facilitates production of sequential finger movements that follow a fixed sequence but not when the sequence is selected at random. These results highlight the role of thalamo-cortical activity in motor learning.

neuroscience↗

bioRxiv2019

Convolutional Neural Networks for Decoding of Covert Attention Focus and Saliency Maps for EEG Feature Visualization

ObjectiveConvolutional neural networks (CNNs) have proven successful as function approximators and have therefore been used for classification problems including electroencephalography (EEG) signal decoding for brain-computer interfaces (BCI). Artificial neural networks, however, are considered black boxes, because they usually have thousands of parameters, making interpretation of their internal processes challenging. Here we systematically evaluate the use of CNNs for EEG signal decoding and investigate a method for visualizing the CNN model decision process.\n\nApproachWe developed a CNN model to decode the covert focus of attention from EEG event-related potentials during object selection. We compared the CNN and the commonly used linear discriminant analysis (LDA) classifier performance, applied to datasets with different dimensionality, and analyzed transfer learning capacity. Moreover, we validated the impact of single model components by systematically altering the model. Furthermore, we investigated the use of saliency maps as a tool for visualizing the spatial and temporal features driving the model output.\n\nMain resultsThe CNN model and the LDA classifier achieved comparable accuracy on the lower-dimensional dataset, but CNN exceeded LDA performance significantly on the higher-dimensional dataset (without hypothesis-driven preprocessing), achieving an average decoding accuracy of 90.7% (chance level = 8.3%). Parallel convolutions, tanh or ELU activation functions, and dropout regularization proved valuable for model performance, whereas the sequential convolutions, ReLU activation function, and batch normalization components, reduced accuracy or yielded no significant difference. Saliency maps revealed meaningful features, displaying the typical spatial distribution and latency of the P300 component expected during this task.\n\nSignificanceFollowing systematic evaluation, we provide recommendations for when and how to use CNN models in EEG decoding. Moreover, we propose a new approach for investigating the neural correlates of a cognitive task by training CNN models on raw high-dimensional EEG data and utilizing saliency maps for relevant feature extraction.

neuroscience↗