Do area lip neurons encode the probability that a stimulus will serve as a saccad. ic target?

Purpose. To determine whether LIP neurons encode the probability that a particular stimulus will serve as a saccadic target. Methods. Single neurons in Area LIP were studied in 2 headfixed, awake-behaving rhesus monkeys using standard physiological techniques. In the first experiment, neuronal responses were recorded during 2 blocks of a standard delayed saccade task. In the first block, each of 200 uniformly distribuled LEDs was equally likely to serve as the saccadic target on a given trial. In the second block, one particular LED served as the saccadic target on 50% of trials while all remaining LEDs were equally probable. Effects on spatial tuning were assessed by comparing response fields measured under these two different probability conditions. In a second experiment, neuronal responses were recorded while animals performed 5 blocks of a cued saccade task. On these trials, the color of the fixation stimulus determined which of two eccentric yellow LEDs would serve as the saccadic target. One eccentric LED was fixed inside, and the other eccentric LED was fixed outside, the neuronal response field. On each block of cued saccade trials, the probability that the LED located in the neuronal response field would serve as the saccadic target was either 0%, 20%, 50%, 80%, or 100%. For analysis, trials were sorted as a function of whether the LED in the response field served as either the target or the distractor. Target and distractor-associated responses were then plotted as a function of the probability of target occurrence. Results. On delayed saccade trials, most neurons showed no changes in maximum firing rate or spatial tuning when the probability that a saccadic target would appear at a particular location was altered. Population data suggest that on average there was no systematic shift in response field centers towards the high probability target location and no systematic change in spatial tuning breadth. On cued saccade trials, neurons typically fired more strongly in association with an LED identified as a saccadic target than in association with the same LED when it served as a distractor. For some cells, both target- and distractor-associated responses (although different) increased as the probability that the stimulus inside the response field would serve as a saccadic target increased.