Biology subjects

Loke, J.

Publications and source records attributed to Loke, J..

3 recordsLinked to original sources

bioRxiv2025

Scene segmentation processes drive EEG-DCNN alignment

Visual processing in biological and artificial neural networks has been extensively studied through the lens of object recognition. While deep convolutional neural networks (DCNNs) have demonstrated hierarchical feature extraction similar to biological systems (DiCarlo and Cox, 2007; Yamins and DiCarlo, 2016), recent findings reveal a growing discrepancy: DCNNs with higher object categorization accuracy paradoxically show worse performance at predicting neural responses (Xu and Vaziri-Pashkam, 2021; Linsley et al., 2023). Using a large-scale human electroencephalography (EEG) dataset (n=10, 82,160 trials), we investigate whether this discrepancy arises because human neural EEG signals predominantly reflect scene segmentation processes rather than high-level, category-specific object representations. We trained DCNNs to perform object recognition using visual diets ([~]1 million training images across 292 object categories) with systematically varying scene segmentation demands: objects-only (pre-segmented), background-silhouette (explicit boundaries), original/background-only images (requiring full segmentation). Despite substantial differences in categorization accuracy (27-53%), all trained models showed remarkably uniform encoding performance, with peak correlations with neural data at [~]0.1s post-stimulus. Layer-wise analysis revealed a significant negative correlation between categorization accuracy and encoding performance, with earlier network layers better predicting EEG responses than deeper layers specialized for object categorization. This dissociation suggests that EEG signals primarily reflect fundamental scene parsing mechanisms rather than object-specific representations, explaining the growing discrepancy between DCNNs increasing categorization performance but deteriorating neural prediction performance. Significance StatementThis research provides a novel perspective on human electroencephalography (EEG) signals during visual processing through systematic manipulation of scene segmentation demands in deep neural networks. Using a large-scale dataset of 82,160 EEG trials and 20 trained DCNNs, we demonstrate that EEG responses primarily reflect early visual processing involved in breaking down and organizing visual scenes (scene segmentation/parsing) rather than high-level object recognition. This finding helps explain previously observed discrepancies between DCNNs categorization performance and neural prediction accuracy, suggesting that improving models ability to segment scenes, rather than simply recognizing isolated objects, may better align artificial and biological visual processing.

neuroscience↗

bioRxiv2023

Human visual cortex and deep convolutional neural network care deeply about object background

Deep convolutional neural networks (DCNNs) are able to predict brain activity during object categorization tasks, but factors contributing to this predictive power are not fully understood. Our study aimed to investigate the factors contributing to the predictive power of DCNNs in object categorization tasks. We compared the activity of four DCNN architectures with electroencephalography (EEG) recordings obtained from 62 human subjects during an object categorization task. Previous physiological studies on object categorization have highlighted the importance of figure-ground segregation - the ability to distinguish objects from their backgrounds. Therefore, we set out to investigate if figure-ground segregation could explain DCNNs predictive power. Using a stimuli set consisting of identical target objects embedded in different backgrounds, we examined the influence of object background versus object category on both EEG and DCNN activity. Crucially, the recombination of naturalistic objects and experimentally-controlled backgrounds creates a sufficiently challenging and naturalistic task, while allowing us to retain experimental control. Our results showed that early EEG activity (<100ms) and early DCNN layers represent object background rather than object category. We also found that the predictive power of DCNNs on EEG activity is related to processing of object backgrounds, rather than categories. We provided evidence from both trained and untrained (i.e. random weights) DCNNs, showing figure-ground segregation to be a crucial step prior to the learning of object features. These findings suggest that both human visual cortex and DCNNs rely on the segregation of object backgrounds and target objects in order to perform object categorization. Altogether, our study provides new insights into the mechanisms underlying object categorization as we demonstrated that both human visual cortex and DCNNs care deeply about object background. Author summaryOur study aimed to investigate the factors contributing to the predictive power of deep convolutional neural networks (DCNNs) on EEG activity in object recognition tasks. We compared the activity of four DCNN architectures with human neural recordings during an object categorization task. We used a stimuli set consisting of identical target objects embedded in different phase-scrambled backgrounds. The distinction between object backgrounds and object categories allows us to investigate the influence of either factor for human subjects and DCNNs. Surprisingly, we found that both human visual processing and early DCNNs layers dedicate a large proportion of activity to processing object backgrounds instead of object category. Furthermore, this shared ability to make object backgrounds (and not just object category) invariant is largely the reason why DCNNs are predictive of brain dynamics in our experiment. We posit this shared ability to be an important solution for object categorization. Finally, we conclude that DCNNs, like humans, care deeply about object backgrounds.

neuroscience↗

bioRxiv2020

On the necessity of recurrent processing during object recognition: it depends on the need for scene segmentation

While feed-forward activity may suffice for recognizing objects in isolation, additional visual operations that aid object recognition might be needed for real-world scenes. One such additional operation is figure-ground segmentation; extracting the relevant features and locations of the target object while ignoring irrelevant features. In this study of 60 participants, we show objects on backgrounds of increasing complexity to investigate whether recurrent computations are increasingly important for segmenting objects from more complex backgrounds. Three lines of evidence show that recurrent processing is critical for recognition of objects embedded in complex scenes. First, behavioral results indicated a greater reduction in performance after masking objects presented on more complex backgrounds; with the degree of impairment increasing with increasing background complexity. Second, electroencephalography (EEG) measurements showed clear differences in the evoked response potentials (ERPs) between conditions around 200ms - a time point beyond feed-forward activity and object decoding based on the EEG signal indicated later decoding onsets for objects embedded in more complex backgrounds. Third, Deep Convolutional Neural Network performance confirmed this interpretation; feed-forward and less deep networks showed a higher degree of impairment in recognition for objects in complex backgrounds compared to recurrent and deeper networks. Together, these results support the notion that recurrent computations drive figure-ground segmentation of objects in complex scenes.

neuroscience↗