Tracking down the neurons of perception.

Tracking down the neurons of perception

A simple sketch of a cube can prompt two rivalrous perceptions: A person might see the cube extending forward at one moment and backward the next, switching perceptions with each new glance.

By presenting rivalrous visual stimuli to rhesus monkeys and monitoring the monkeys' silent, inner perceptions -- through behevioral clues and the responses of brain cells known to process visual motion -- two neurophysiologists think they may have observed the cellular activity of perception.

"This is one of the first demonstrations of the activity of single neurons corresponding to a monkey's perceptual state," says Jeffrey D. Schall of Vanderbilt University in Nashville, who works on the ongoing project with Nikos K. Logothetis of the Massachusetts Institute of Technology in Cambridge. They report their progress in the Aug. 18 SCIENCE. "It's the best attempt I know of to look at the neurophysiology underlying rivalrous perception," notes neuroscientist William T. Newsome of Stanford University.

The stimuli in the experiments come from vertically moving gratings of horizontal bars, each visible to one eye through a viewer that fuses the gratings into a composite stimulus. In half the trials, the gratings move in the same direction and the monkey can perceive only up or down motion. In the other half, the gratings move in opposite directions, eliciting rivalrous perceptions of upward and downward movement.

Since monkeys can't describe their perceptions verbally, the scientists trained them to make quick eye movements to the right (signaling upward motion) or left (signaling downward motion) of a central fixation point presented after each trial. And during each trial, researchers tracked vertical movements of each monkey's eyes for a check on the animal's subsequent report of its perception.

At the same time, they used electrodes to study how 59 single neurons in the superior temporal sulcus -- a brain groove roughly located behind each ear -- responded in the different trials. As expected, most of these neurons fired fastest in response to exclusively upward or exclusively downward movement of the gratings. The response of many of the cells didn't change even during trials presenting rivalrous stimuli. That means they responded only to the visual stimulus and did not reflect the internal perception of the monkey. Yet with the same rivalrous stimuli, 13 of the neurons did change their firing behavioral. These cells, therefore, could be related to the internal perceptual state of the monkeys, and not merely to some physical feature of the stimuli, Schall and Logothetis argue.