Skip to main content
Explore URMC

URMC / Labs / Cognitive Neuroscience Lab / Projects / Neuronal-Integration-of-Visual-Cues

Neuronal Integration of Visual Cues

Combined object motion and optic flow evoke a wide variety of response interactions with landmark and animate objects creating comparable effects. When baseline activity is subtracted, response interactions are distributed around simple additivity with a substantial number of sub-additive and super-additive responses. The non-additive interactions often reflect individual neurons’ sensitivity to the relative local motion from optic flow and object motion at some particular location in the visual field. Other neurons show non-additive interactions that are related to preferences for a particular sequence of optic flow stimuli defining a path of simulated self-movement. Across the sample of neurons, small responses to combined stimuli tend to be additive, whereas larger excitatory responses combine to yield sub-additive firing rates, and larger inhibitory responses combine to yield super-additive firing rates. We conclude that individual MST neurons idiosyncratically combine optic flow and object motion cues to create a population profile that yields an unbiased representation of self-movement in diverse, naturalistic circumstances.

Integration of Optic Flow and Object Motion

Visual cues about self-movement are derived from the visual motion of discrete objects and the patterned motion of optic flow. We recorded neuronal activity in monkey medial superior temporal (MST) cortex during continuous centered fixation while viewing object motion and optic flow simulating self-movement. In the first experiment, 20 stimulus configurations presented naturalistic combinations of objects and optic flow with small responses yielding additive interactions, larger excitatory responses yielding sub-additivity, and larger inhibitory responses yielding super-additivity. In individual neurons, local motion stimulus parameters and the global patterns of simulated self-movement, idiosyncratically influence response interactions.

« back to all projects