Symmetry Processing in the Macaque Visual Cortex

Pauline Audurier, Yseult Héjja‐Brichard, Vanessa De Castro, Peter J. Kohler, Anthony M. Norcia, Jean‐Baptiste Durand, Benoit R. Cottereau

Abstract

Symmetry is a highly salient feature of the natural world that impacts a large array of behaviors, such as partner selection or food choice. In humans, the cerebral areas processing symmetry are well-identified, but an animal model for exploring underlying neural mechanisms has been lacking. Previous studies suggested that macaque functional magnetic resonance imaging (fMRI) responses to symmetry were substantially weaker than those in humans. This study re-examines symmetry processing in macaques using both rotation and reflection symmetry, with an experimental design directly derived from human fMRI studies. Highly consistent responses were found in a large network of areas (notably V3, V3A, V4, V4A, and PITd). Response properties in areas V3 and V4 were strikingly similar to their human counterparts, suggesting that the cortical networks processing symmetry in humans and macaques are much more similar than previously reported.

Introduction

While symmetry is fundamental to visual perception and biological decision-making, the specific neural "circuitry" behind its detection remains poorly understood. Mapping these processes in humans has identified the relevant brain regions, but characterizing the behavior of individual neurons requires an animal model. Earlier research indicated that macaques—our closest research relatives—exhibited significantly weaker neural responses to symmetry than humans, casting doubt on their utility as a model. This paper investigates whether this discrepancy was biological or a byproduct of experimental design, specifically the choice of visual stimuli.

Methods and Stimuli

The researchers conducted fMRI on two rhesus macaques (M01 and M02) during a passive fixation task. To ensure a direct comparison with human data, they moved away from simple random dot patterns and utilized "wallpaper patterns"—complex, regular textures based on mathematical groups.

Wallpaper pattern stimuli for rotation (A) and reflection (B) experiments, including their phase-scrambled controls.

The experiments were divided into two main categories:

  1. Rotation Symmetry: Stimuli featured 2-, 3-, 4-, or 6-fold rotation (P2, P3, P4, P6).
  2. Reflection Symmetry: Stimuli featured horizontal, vertical, or orthogonal (PMM) axes of reflection.

For every symmetric stimulus, a "phase-scrambled" control was created. These controls maintain low-level properties like luminance and orientation but lack global symmetry, allowing researchers to isolate the brain's specific response to structural organization. Visual areas were precisely defined for each subject using independent retinotopic mapping and probabilistic atlases.

Key Results

The study revealed a robust and widespread network of symmetry-sensitive areas, including V2, V3, V4, V3A, and higher-order regions like V4A and PITd. Statistical analysis showed consistent T-scores greater than 3 across these regions in both subjects.

The Parametric Response in V3 and V4

A primary finding was the "parametric" nature of the response in areas V3 and V4. As the order of rotation symmetry increased (from 2-fold to 6-fold), the BOLD signal increased linearly. This identical trend is observed in human studies, suggesting a shared computational logic between the two species.

Percent Signal Change (PSC) across visual areas, showing the parametric increase in response to higher rotation symmetry orders, particularly in V3 and V4.

Sensitivity to Reflection

The primate visual system also showed heightened sensitivity to the complexity of reflection. Stimuli with two axes of reflection (PMM) elicited significantly stronger responses than those with only a single axis (PM) in areas V3, V4, and V3A.

Discussion and Future Directions

The results suggest that the primate visual system is specifically tuned to detect symmetry when embedded in complex, naturalistic textures. This highlights the importance of stimulus choice in comparative neuroscience. While V3 is often overlooked in standard models of the ventral stream, this study positions it as a critical early stage for symmetry detection.

The striking homology between human and macaque cortical networks establishes the macaque as a relevant model for symmetry processing. Future work should leverage this by using electrophysiological recordings in V3 and V4 to decode the single-cell mechanisms of symmetry perception.

Reference: Audurier, P., et al. (2021). Symmetry processing in the macaque visual cortex. bioRxiv preprint.

Summary updated

Abstract

Symmetry is a highly salient feature of the natural world that is perceived by many species. In humans, the cerebral areas processing symmetry are now well identified from neuroimaging measurements. Macaque could constitute a good animal model to explore the underlying neural mechanisms, but a previous comparative study concluded that functional magnetic resonance imaging responses to mirror symmetry in this species were weaker than those observed in humans. Here, we re-examined symmetry processing in macaques from a broader perspective, using both rotation and reflection symmetry embedded in regular textures. Highly consistent responses to symmetry were found in a large network of areas (notably in areas V3 and V4), in line with what was reported in humans under identical experimental conditions. Our results suggest that the cortical networks that process symmetry in humans and macaques are potentially more similar than previously reported and point toward macaque as a relevant model for understanding symmetry processing.
BibTeX
@inproceedings{audurier2021symmetry,
  author = {Pauline Audurier and Yseult Héjja‐Brichard and Vanessa De Castro and Peter J. Kohler and Anthony M. Norcia and Jean‐Baptiste Durand and Benoit R. Cottereau},
  title = {{Symmetry Processing in the Macaque Visual Cortex}},
  booktitle = {Cerebral},
  year = {2021},
  doi = {10.1093/cercor/bhab358},
}
Figure