Binocular rivalry between spiral stimuli and linear motion in human observers.

摘要

Binocular rivalry describes the alternating perception of two competing monocular images. It is hypothesized to arise at multiple levels of the visual pathway due to competitions between neuronal populations that represent the displayed images (Freeman, 2005; Wilson, 2003). From our previous findings, we concluded that the stronger represented expanding motion in area MSTd caused a bias for expansion when presented against contraction (Wilcoxon Signed-Rank test, Left Eye (LE):p=0.0156; Right Eye (RE):p=0.0781) and rotation stimuli (LE:p=0.02345; RE:p=0.02345). Moreover periods were significantly longer for competing spiral stimuli than for our controla??rivaling linear motions in the cardinal directions (LE:p=0.0000376, RE:p=0.00000423), even though the latter are more strongly represented in MT and MSTd (Tanaka et al, 1986). We conducted two additional experiments on twelve human observers. The first measured the rivalry periods for combinations of expanding, contracting, or rotating random dot patterns (RDPs) against linear motion RDPs. The RDPs were matched for contrast and speed and were displayed through a mirror stereoscope. We found that a) the bias for expansion ceased and the dominance durations among spiral stimuli were no longer significantly different (LE & RE:p0.1) and b) spiral stimuli continued to have significantly greater rivalry periods than linear motion stimuli (LE:p=0.000142; RE:p=0.00000553), however the linear stimuli were perceived for longer periods than in our initial experiment (p=0.0078). In the second experiment we compared the rivalry periods for combinations of expanding and contracting RDPs with and without the gradient. In general, stimuli with speed gradients dominated over those without. Moreover, the bias for expansion over contraction was maintained (LE:p=0.0095; RE:p=0.00805). Our results support that motion binocular rivalry takes place at various levels of the visual pathway and it is influenced foremost by the types of stimuli rivaling and by the bias in neuronal representations of those stimuli.