# Recurrent connections facilitate occluded object recognition by explaining-away

## TL;DR

The paper shows how recurrent brain connections help recognize partially hidden objects by using knowledge of the occluder to fill in missing details.

## Contribution

The study introduces a novel computational mechanism where recurrent networks 'explain-away' occlusions using occluder information.

## Key findings

- Recurrent networks, unlike feedforward ones, use occluder information to improve recognition of occluded objects.
- A human psychophysics experiment supports the 'explain-away' mechanism in occluded object recognition.
- A recurrent model inspired by the findings successfully recovers fine-grained features of occluded stimuli.

## Abstract

Despite the ubiquity of recurrent connections in the brain, their role in visual processing is less understood than that of feedforward connections. Occluded object recognition, an important cognitive capacity, is thought to rely on recurrent processing of visual information, but it remains unclear whether and how recurrent processing improves recognition of occluded objects. Using convolutional models of the visual system, we demonstrate how a distinct form of computation arises in recurrent–but not feedforward–networks that leverages information about the occluder to “explain-away” the occlusion—i.e., recognition of the occluder provides an account for missing or altered features, potentially rescuing recognition of occluded objects. This occurs without any constraint placed on the computation and is observed both across a systematic architecture sweep of convolutional models and in a model explicitly constructed to approximate the primate visual system. In line with these results, we find evidence consistent with explaining-away in a human psychophysics experiment. Finally, we developed an experimentally inspired recurrent model that recovers fine-grained features of occluded stimuli by explaining-away. Recurrent connections’ capability to explain-away may extend to more general cases where undoing context-dependent changes in representations benefits perception.

Objects in natural scenes are often partially occluded. Here, the authors show that recurrent processing can use knowledge of the occluder to explain away missing features, improving recognition of occluded objects in models and humans.