Sensory Polymorphism and Behavior: When Machine Vision Meets Monkey Eyes

TL;DR

This paper reviews recent interdisciplinary research connecting sensory polymorphisms, especially in vision, to behavioral variability, emphasizing the role of computational models and machine vision in understanding these complex relationships.

Contribution

It provides an overview of recent advances in integrating physiological, behavioral, and computational approaches to study sensory polymorphisms and suggests strategies for future research.

Findings

Machine vision approaches reveal links between sensory polymorphisms and behavior.

Interdisciplinary methods enhance understanding of sensory-behavioral relationships.

Strategies for integrating physiological, behavioral, and computational data are proposed.

Abstract

Polymorphism in the peripheral sensory system (e.g., congenital individual differences in photopigment configuration) is important in diverse research fields, ranging from evolutionary biology to engineering, because of its potential relationship to the cognitive and behavioral variability among individuals. However, there is a gap between the current understanding of sensory polymorphism and the behavioral variability that is an outcome of potentially complex cognitive processes in natural environments. Linking peripheral sensor properties to behavior requires computational models of nervous processes transforming sensory representations into actions, based on quantitative data from physiological and behavioral studies. Recently, studies based on machine vision approaches are shedding light on the quantitative relationships between sensory polymorphisms and the resulting behavioral variability. To reach a convergent understanding of the functional impacts of sensory polymorphisms in realistic environments, a close coordination among physiological, behavioral, and computational approaches is required. Aiming at enhancing such integrative researches, this paper provides an overview for the recent progresses in those interdisciplinary approaches, and suggests effective strategies for such integrative paradigms.