Neural representation of translucent and opaque objects images in macaque inferior temporal cortex

Translucency and opacity are optical properties of materials characterized by the extent of light passing through them. Although recent psychophysical studies have revealed the visual properties associated with the perception of translucency, the neural substrates involved in its perception remain unknown. In this study, we conducted electrophysiological experiments using visual stimuli of objects with various shapes and varying degrees of translucency. To regulate the visual attributes systematically, we used the Translucent Appearance Generation (TAG) model, an unsupervised artificial neural network designed to synthesize images that represent the material appearance. The latent variables of this model encode human-interpretable visual attributes, such as object color in the fine-scale layer and shape in the coarse-scale layers. Notably, variables in the middle layers of the model encode feature information related to whether they appear translucent or opaque. We created visual stimuli by generating object images from sampled latent variables (27 points from coarse-scale layers and 7 points from middle layers) using the TAG model and converting them to grayscale. Mean luminance, Michelson contrast, and the area of the object region were equalized to minimize the effect of these factors on the experimental results. We recorded neural responses using four multi-electrode arrays (each consisting of 128 channels) from the inferior temporal cortex of one macaque monkey. The arrays were placed in three sites in the TE area, and one was in the TEO area. The results of representational similarity analysis showed higher correlations between semi-translucent images than between translucent and opaque images after averaging responses across different shapes. The results of unit-level response analysis indicated that some neurons exhibited selectivity to translucent or opaque objects. These findings suggest that neurons in the inferior temporal cortex represent differences in the object's appearance related to translucency or opacity at both population and unit levels.

Authors

Institutions

Publication Details

Journal
Journal of Vision
Published
2024-09-15
DOI
https://doi.org/10.1167/jov.24.10.461
Primary Topic
Neural Networks and Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Neural representation of translucent and opaque objects images in macaque inferior temporal cortex

Ryusuke Hayashi, Daiki Nakamura, Hoko Nakada
Journal of Vision
Neural Networks and Applications
article

Neural representation of translucent and opaque objects images in macaque inferior temporal cortex

Ryusuke Hayashi, Daiki Nakamura, Hoko Nakada
article en

Abstract

Translucency and opacity are optical properties of materials characterized by the extent of light passing through them. Although recent psychophysical studies have revealed the visual properties associated with the perception of translucency, the neural substrates involved in its perception remain unknown. In this study, we conducted electrophysiological experiments using visual stimuli of objects with various shapes and varying degrees of translucency. To regulate the visual attributes systematically, we used the Translucent Appearance Generation (TAG) model, an unsupervised artificial neural network designed to synthesize images that represent the material appearance. The latent variables of this model encode human-interpretable visual attributes, such as object color in the fine-scale layer and shape in the coarse-scale layers. Notably, variables in the middle layers of the model encode feature information related to whether they appear translucent or opaque. We created visual stimuli by generating object images from sampled latent variables (27 points from coarse-scale layers and 7 points from middle layers) using the TAG model and converting them to grayscale. Mean luminance, Michelson contrast, and the area of the object region were equalized to minimize the effect of these factors on the experimental results. We recorded neural responses using four multi-electrode arrays (each consisting of 128 channels) from the inferior temporal cortex of one macaque monkey. The arrays were placed in three sites in the TE area, and one was in the TEO area. The results of representational similarity analysis showed higher correlations between semi-translucent images than between translucent and opaque images after averaging responses across different shapes. The results of unit-level response analysis indicated that some neurons exhibited selectivity to translucent or opaque objects. These findings suggest that neurons in the inferior temporal cortex represent differences in the object's appearance related to translucency or opacity at both population and unit levels.

Journal of VisionVol. 24(10)
National Institute of Advanced Industrial Science and Technology (JP)
Openalex Percentile: Top 86%
Neural Networks and Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.