Bistable perception of ambiguous images: simple Arrhenius model

Area of Science:

• Cognitive Neuroscience
• Computational Neuroscience
• Visual Perception

Background:

• Ambiguous images induce perceptual bistability, characterized by random oscillations between two interpretations.
• Traditional models describe neural system evolution using "movement" equations on energy landscapes with stochastic forces.
• Understanding the dynamics of perceptual switching is crucial for cognitive science.

Purpose of the Study:

• To propose a simpler, alternative approach to modeling perceptual bistability using the Arrhenius equation.
• To determine the probabilities of transitions between different perceptions based on energy landscape barriers and noise levels.
• To predict the relationship between perceptual hysteresis and the period of stimulus parameter variation.

Main Methods:

• Utilized the Arrhenius equation to describe the quasi-stationary state of the neural system.
• Modeled perception transitions as overcoming energy barriers influenced by noise (analogous to temperature).
• Estimated mean noise values from experimental data.

Main Results:

• The Arrhenius equation approach successfully describes the probability of perceptual variation.
• Transition probabilities are determined by energy landscape barriers and noise intensity.
• The model predicts a logarithmic dependence of perception hysteresis width on the period of cyclic parameter sweeping, which aligns with experimental findings.

Conclusions:

• The quasi-stationary state model using the Arrhenius equation offers a simpler yet effective framework for understanding perceptual bistability.
• This model allows for the estimation of the mean noise level and the time interval between successive perceptions.
• The findings provide insights into the neural mechanisms underlying visual perception and decision-making processes.