EMULATING THE VISUAL RECEPTIVE-FIELD PROPERTIES OF MST NEURONS WITH ATEMPLATE MODEL OF HEADING ESTIMATION

We have proposed previously a computational neural-network model by wh ich the complex patterns of retinal image motion generated during loco motion (optic flow) can be processed by specialized detectors acting a s templates for specific instances of self-motion. The detectors in th is template model respond to global optic flow by sampling image motio n over a large portion of the visual field through networks of local m otion sensors with properties similar to those of neurons found in the middle temporal (MT) area of primate extrastriate visual cortex. Thes e detectors, arranged within cortical-like maps, were designed to extr act self-translation (heading) and self-rotation, as well as the scene layout (relative distances) ahead of a moving observer. We then postu lated that heading from optic flow is directly encoded by individual n eurons acting as heading detectors within the medial superior temporal (MST) area. Others have questioned whether individual MST neurons can perform this function because some of their receptive-field propertie s seem inconsistent with this role. To resolve this issue, we systemat ically compared MST responses with those of detectors from two differe nt configurations of the model under matched stimulus conditions. We f ound that the characteristic physiological properties of MST neurons c an be explained by the template model. We conclude that MST neurons ar e well suited to support self-motion estimation via a direct encoding of heading and that the template model provides an explicit set of tes table hypotheses that can guide future exploration of MST and adjacent areas within the superior temporal sulcus.