Attention model of binocular rivalry
PhD Student, Heeger Lab, NYU
When the two eyes are presented with incompatible images, perception alternates between the two images, creating a phenomena known as binocular rivalry. During rivalry, perceptual experience evolves dynamically while the external inputs are held constant. Binocular rivalry thereby offers a gateway for studying intrinsic cortical computations. In conventional theories of binocular rivalry, the competition between the two percepts has been characterized as mutual inhibition between two populations of neurons selective for each of the two stimuli. However, converging experimental evidence has shown that rivalry also depends on attention: rivalry is largely eliminated when attention is diverted away from the stimuli. In addition, the competing image in one eye suppresses the target image in the other eye through a gain change similar to that induced by attentional modulation. These results require a revision of the current theories of binocular rivalry, in which the role of attention is ignored.
We investigated the role of attention in binocular rivalry in a psychophysical and a preliminary MEG experiment. We found that binocular competition is driven by both attention and mutual inhibition, which have distinct selectivity. We developed a new computational model of rivalry, and with a bifurcation analysis, we identified the parameter space in which the model’s behavior was consistent with experimental results. The model provides a parsimonious account of various perceptual dynamics of rivalry for which there was no previous explanation.