Figure 5. Influence of Spatial Context on the Encoding of Visual Motion by Cortical Neurons.

Data were obtained by stimulating the rfs of neurons in area mt (see Figure 2) with unambiguously moving patterns of stripes (panel A) and with barber-diamond stimuli (panel B; see text and Figure 4 for stimulus description), for which the perceived direction is dictated by context.

(A) The striped pattern was moved in each of six different directions (indicated by small stimulus icons) within the rf of the recorded mt neuron. The motion percepts elicited by these six stimuli are determined solely by the physical motions of the stimulus. The mean neuronal responses are plotted in polar form, where the radius represents the frequency of action potentials (spikes per second) and the polar axis represents the direction that the stimulus moved. Like many mt neurons, this cell was highly selective, with a preference for motion down and to the right.

(B) The same mt neuron shown in (A) was in this case stimulated with four barber-diamond conditions. As detailed in text and in Figure 4, there are only two retinal image motion directions (left and right) present in these four stimuli, which are indicated by the black arrows on each stimulus icon. Because of the different depth configurations in the surround (see Figure 4), these two image motions are perceived to move in each of four unique directions (along the four diagonals), indicated by the gray arrows on each stimulus icon. We predicted that the neuronal response would be largest when the perceived motion matched the neuron’s known direction preference (established from the test shown in panel A), such as down and to the right. This is precisely what was observed. Importantly, the retinal image motions for the two conditions on the right side of the panel are physically identical, but the neuronal response reflects the direction perceived, which is the visual scene motion inferred based upon spatial context.

Source: Robert O. Duncan, Thomas D. Albright, and Gene R. Stoner, “Occlusion and the Interpretation of Visual

Motion: Perceptual and Neuronal Effects of Context,” The Journal of Neuroscience 20 (2000): 5885–5897.