Extending Levelt’s Propositions to perceptual multistability involving interocular grouping

Levelt’s Propositions have been a touchstone for experimental and modeling studies of perceptual multistability. We asked whether Levelt’s Propositions extend to perceptual multistability involving interocular grouping. To address this question we used split-grating stimuli withcomplementary halves of the same color. As in previous studies, subjects reported four percepts in alternation: the two stimuli presented to each eye (single-eye percepts), as well as two interocularly grouped, single color percepts (grouped percepts). Most subjects responded to increased color saturation by more frequently reporting a single color image, thus increasingthe predominance of grouped percepts (Levelt’s Proposition I). In these subjects increased predominance was due to a decrease in the average dominance duration of single-eye percepts, while that of grouped percepts remained largely unaffected. This is in accordance with generalized Levelt’s Proposition II which posits that the average dominance duration of the stronger (in this case single-eye) percept is primarily affectedbychanges in stimulus strength. In accordance with Proposition III, thealternation rate increased as the difference in the strength of the percepts decreased. To explain the mechanism behind these observations, we introduce a hierarchical model consisting of low-level neural populations, eachresponding to input at a visual hemifield, and higher-level populations representing the percepts. The model exhibits the changes in dominance durationobserved in the data, and conforms to all of Levelt’s Propositions.