1. The latency between the appearance of a popout search display and the eye movement to the oddball target of the display varies from trial to trial in both humans and monkeys. The source of the delay and variability of reaction time is unknown but has been attributed to as yet poorly defined decision processes. 2. We recorded neural activity in the frontal eye field (FEF), an area regarded as playing a central role in producing purposeful eye movements, of monkeys (Macaca mulatta) performing a popout visual search task. Eighty-four neurons with visually evoked activity were analyzed. Twelve of these neurons had a phasic response associated with the presentation of the visual stimulus. The remaining neurons had more tonic responses that persisted through the saccade. Many of the neurons with more tonic responses resembled visuomovement cells in that they had activity that increased before a saccade into their response field. 3. The visual response latencies of FEF neurons were determined with the use of a Poisson spike train analysis. The mean visual latency was 67 ms (minimum = 35 ms, maximum = 138 ms). The visual response latencies to the target presented alone, to the target presented with distractors, or to the distractors did not differ significantly. 4. The initial visual activation of FEF neurons does not discriminate the target from the distractors of a popout visual search stimulus array, but the activity evolves to a state that discriminates whether the target of the search display is within the receptive field. We tested the hypothesis that the source of variability of saccade latency is the time taken by neurons involved in saccade programming to select the target for the gaze shift. 5. With the use of an analysis adapted from signal detection theory, we determined when the activity of single FEF neurons can reliably indicate whether the target or distractors are present within their response fields. The time of target discrimination partitions the reaction time into a perceptual stage in which target discrimination takes place, and a motor stage in which saccade programming and generation take place. The time of target discrimination occurred most often between 120 and 150 ms after stimulus presentation. 6. We analyzed the time course of target discrimination in the activity of single cells after separating trials into short, medium, and long saccade latency groups. Saccade latency was not correlated with the duration of the perceptual stage but was correlated with the duration of the motor stage. This result is inconsistent with the hypothesis that the time taken for target discrimination, as indexed by FEF neurons, accounts for the wide variability in the time of movement initiation. 7. We conclude that the variability observed in saccade latencies during a simple visual search task is largely due to postperceptual motor processing following target discrimination. Signatures of both perceptual and postperceptual processing are evident in FEF. Procrastination in the output stage may prevent stereot...

To find a target object in a crowded scene, a face in a crowd for example, the visual system might turn the neural representation of each object on and off in a serial fashion, testing each representation against a template of the target item. Alternatively, it might allow the processing of all objects in parallel but bias activity in favor of those neurons that represent critical features of the target, until the target emerges from the background. To test these possibilities, we recorded neurons in area V4 of monkeys freely scanning a complex array to find a target defined by color, shape, or both. Throughout the period of searching, neurons gave enhanced responses and synchronized their activity in the gamma range whenever a preferred stimulus in their receptive field matched a feature of the target, as predicted by parallel models. Neurons also gave enhanced responses to candidate targets that were selected for saccades, or foveation, reflecting a serial component of visual search. Thus, serial and parallel mechanisms of response enhancement and neural synchrony work together to identify objects in a scene. To find a target object in a crowded scene, a face in a crowd for example, the visual system might turn the neural representation of each object on and off in a serial fashion, testing each representation against a template of the target item. Alternatively, it might allow the processing of all objects in parallel but bias activity in favor of those neurons that represent critical features of the target, until the target emerges from the background. To test these possibilities, we recorded neurons in area V4 of monkeys freely scanning a complex array to find a target defined by color, shape, or both. Throughout the period of searching, neurons gave enhanced responses and synchronized their activity in the gamma range whenever a preferred stimulus in their receptive field matched a feature of the target, as predicted by parallel models. Neurons also gave enhanced responses to candidate targets that were selected for saccades, or foveation, reflecting a serial component of visual search. Thus, serial and parallel mechanisms of response enhancement and neural synchrony work together to identify objects in a scene.