Neural Selectivity for Visual Motion in Macaque Area V3A

Abstract

The processing of visual motion is conducted by dedicated pathways in the primate brain. These pathways originate with populations of direction-selective neurons in the primary visual cortex, which projects to dorsal structures like the middle temporal (MT) and medial superior temporal (MST) areas. Anatomical and imaging studies have suggested that area V3A might also be specialized for motion processing, but there have been very few studies of single-neuron direction selectivity in this area. We have therefore performed electrophysiological recordings from V3A neurons in two macaque monkeys (one male and one female) and measured responses to a large battery of motion stimuli that includes translation motion, as well as more complex optic flow patterns. For comparison, we simultaneously recorded the responses of MT neurons to the same stimuli. Surprisingly, we find that overall levels of direction selectivity are similar in V3A and MT and moreover that the population of V3A neurons exhibits somewhat greater selectivity for optic flow patterns. These results suggest that V3A should be considered as part of the motion processing machinery of the visual cortex, in both human and non-human primates.

Keywords: V3A; electrophysiology; macaque monkey; visual dorsal pathway; visual motion.

Figure 1.

Electrophysiological recording locations and progression of receptive fields across the recording channels. A, Receptive field heat maps for neurons recorded by consecutive sites on the linear probe (indicated by green line in B). Neurons mapped from channels 5–8 are located in V3A, while those on channels 24–27 are located in MT. Yellow and bright orange indicate stimulus locations that produced the highest firing rates, while blue indicates that the neuron did not fire above baseline at those locations. B, Location of V3A in a standard macaque monkey brain atlas (Saleem and Logothetis, 2012), with the electrode trajectory indicated by a green line. Red arrows in B, D indicate V3A, while blue arrows denote MT. C, D, Coronal and sagittal view of the MRI from one monkey. The orange dots and lines are the tracks of the recording probe from a CT scan done on the same monkey and registered with the MRI. Dotted white lines are placed to highlight the locations of the recording probe.

Figure 2.

Receptive field sizes and eccentricities in V3A and MT. A, MT cells (blue asterisks) have on average larger receptive fields than V3A cells (orange circles) for the same eccentricities. Linear fits are shown for both populations; orange for V3A and blue for MT. B, Receptive fields from an example recording session. Relationship between RF eccentricity and size with tuning properties are shown in Extended Data Figures 2-1, 2-2.

Figure 3.

RDK stimuli used in the study. A, Visual stimuli presented to the monkey during recording sessions. The central dot indicates the fixation point at which the monkey maintained gaze during stimulus presentation. The dashed white circles show the nine possible locations on the screen where the stimulus could be presented on each trial. B, Geometric representations of the different directions for each motion type displayed around a circle.

Figure 4.

Response latencies for MT and V3A. The distributions of latencies are similar between areas, although the median is slightly highger in V3A. Median values are indicated by arrows on top of the histograms; black for MT and gray for V3A. For data from each animal shown separately, please see Extended Data Figures 4-1, 4-2.

Figure 5.

Sensitivity to motion signal strength. Cumulative distribution functions for the DSI in V3A (orange) and MT (blue) at different constrasts (for grating stimuli) and coherences (for RDKs). The insets show the distribution of DSIs for each population at each contrast or coherence. Note that negative values can occur when the cells lose direction selectivity, in which case the DSI is driven by random fluctuations in the response.

Figure 6.

Tuning Curves for two example V3A cells. Within each motion type, there are nine different tuning curves, corresponding to the nine different stimulus positions. At each position, each circle (tuning wheel) represents the firing of the neuron for each of eight motion directions; orange and red correspond to high firing rates, white to baseline, and blue to firing rates lower than baseline. The green dashed squares indicate the position and the motion type of the stimulus that elicited highest firing. The tuning curve inside the green square Cartesian coordinates to the right of the tuning wheels (the error bars indicate the standard deviation). A, Tuning curves for a translation selective V3A cell. The color bar indicates the range of firing rates in Hz. B, Same as A but for another neuron that is selective for spirals. Average tuning curves for MT and V3A are shown in Extended Data Figure 6-1.

Figure 7.

V3A responses to optic flow stimuli. A, Scatter plots for V3A DSIs (cyan) and MT DSIs (gray) for spirals versus translation (left) and deformation versus translation (right). The unity line is shown by a dashed gray diagonal line. The red triangles refer to the example neurons shown in Figure 6. B, Probability density function of the CMI for sprials (left) and and for deformation (right) for V3A (upper panels) and MT (lower panels). A value of zero for each CMI indicates no preference for complex optic flow over translation motion, while positive numbers indicate a preference for complex optic flow and negative numbers a preference for translation. The red arrows indicate the median for each distribution. For data from each animal shown separately, please see Extended Data Figures 7-1, 7-2.