Recurrent computations for visual pattern completion

TL;DR

This study provides evidence that recurrent neural computations are crucial for visual pattern completion, enabling recognition of occluded objects through psychophysics, physiology, and improved computational models.

Contribution

It demonstrates that recurrent processing, rather than purely feed-forward mechanisms, underpins the ability to recognize partially visible objects.

Findings

Recognition persists with less than 15% visibility but is impaired by backward masking.

Physiological responses show delayed, selective responses to partial objects, linked to masking effects.

Recurrent neural network models outperform feed-forward models in recognizing occluded objects.

Abstract

Making inferences from partial information constitutes a critical aspect of cognition. During visual perception, pattern completion enables recognition of poorly visible or occluded objects. We combined psychophysics, physiology and computational models to test the hypothesis that pattern completion is implemented by recurrent computations and present three pieces of evidence that are consistent with this hypothesis. First, subjects robustly recognized objects even when rendered <15% visible, but recognition was largely impaired when processing was interrupted by backward masking. Second, invasive physiological responses along the human ventral cortex exhibited visually selective responses to partially visible objects that were delayed compared to whole objects, suggesting the need for additional computations. These physiological delays were correlated with the effects of backward masking. Third, state-of-the-art feed-forward computational architectures were not robust to partial visibility. However, recognition performance was recovered when the model was augmented with attractor-based recurrent connectivity. These results provide a strong argument of plausibility for the role of recurrent computations in making visual inferences from partial information.