Results

We recorded extracellularly from 143 neurons in area MT in 11 macaque monkeys. For each neuron, we initially characterized its response to drifting sinusoidal gratings presented in a standard fashion, for a several second interval with a period of blank screen between each stimulus. We determined the optimal direction, spatial and temporal frequency, and size, and then proceeded with experiments to test the dynamics of each neuron's response over time.

Figure 7-1A shows a schematic of our stimulus. For each cell, we presented a continuous sequence of drifting gratings and plaids. We interleaved gratings at 50% contrast drifting in 12 evenly spaced directions, 12 plaid stimuli created by adding together two gratings separated by 120$^\circ$, and 4 uniform blank screens (mean gray). After each 320 ms stimulus epoch we changed the stimulus, holding spatial and temporal frequency and size constant at the optimal values. The stimuli were presented in random order together in a block. This block was typically repeated 5 times (each time in a new random order), followed by several seconds of blank screen. This entire procedure was usually repeated 5-40 times to attain 25-200 repeats of each stimulus, depending on the reliability of the cell's response.

Figure 7-1: A, We used a stimulus which rapidly changed between gratings and plaids drifting in different directions. For ease of illustration, we have only used gratings in this figure. The stimulus remained on the screen drifting for 320 ms, after which a new stimulus was chosen randomly. This figure shows an example sequence with a spike raster below it. Each cell has a response latency (the time it takes for a change in the stimulus to be reflected in a response change), which is indicated by $\Delta$t. We collected multiple repeats and then parsed out the response for each stimulus to arrive at a detailed response histogram. B, Most cells in MT are tuned for the direction of a drifting grating. An example response is shown here in the polar plot. C, When a plaid stimulus is presented, we might predict two possible responses. The pattern prediction (dotted line) is that the neuron sums the motion signals and responds to the plaid as it does to the individual grating. The component prediction (solid line) is that the neuron responds to individual grating components of the plaid as if they were presented alone.

When using a stimulus which runs continuously without pause, parsing out the spike train is not as simple as with discrete, separately presented stimuli. Below the schematic stimulus in Figure 7-1A, there is a raster plot showing a cell's response. Because each neuron has a response latency, the onset of a high firing rate induced by a preferred stimulus lag behind the transition to that stimulus. For this reason the spike times of the response have to be shifted by some amount in time, shown as $\Delta$t in the figure. We used an automated procedure to calculate the value of $\Delta$t for each cell (see Methods in Ch. 2 and below).

Most cells in MT can be classified as either PDS or CDS based on their responses to plaid and grating stimuli. Figure 7-1 B shows a grating stimulus on the left. On the right, there is a polar plot of a cell's response, where the angle represents the direction of the stimulus and the length of the vector represents the response strength. In Figure 7-1B, a plaid is shown on the left with two possible responses on the right. The solid line shows the component prediction: the cell simply responds in a linear fashion to each of the two gratings which comprise the plaid, producing two peaks on the direction tuning curve. The dotted line shows the pattern prediction: the cell responds to the true direction of the stimulus, and its tuning for the plaid is the same as for a single grating.