Covert orienting of attention in macaques. II. Contributions of parietal cortex

1. To understand some of the contributions of parietal cortex to the dynamics of visual spatial attention, we recorded from cortical cells of monkeys performing attentional tasks. We studied 484 neurons in the intraparietal sulcus and adjacent gyral tissue of two monkeys. We measured phasic responses to peripheral Visual stimuli while the monkeys attended toward or away from the stimuli or when attention was not controlled. Neurons were tested while the monkeys gazed at a spot of light (simple fixation task), actively attended to a foveal target (foveal attention task), performed a reaction time task. (cued reaction time task), made saccadic eye movements to visual targets (saccade task), or responded to a repetitious peripheral target (probability task).

2. In a previous paper we demonstrated that monkeys, Like humans, responded more quickly to visual targets when the targets followed briefly flashed visual cues (validly cued targets) (Bowman ct al. 1993). It has been hypothesized that the cue attracts attention to its locus and results in faster reaction times (Posner 1980). In the present physiological studies, visual cues consistently excited these neurons when they were hashed in the receptive held. Such activity might signal a shift of attention. Visual targets that fell within the receptive field and that immediately followed the cue evoked relatively weak responses. This response was due to a relative refractory period.

3. Next we tested attentional processes in these tasks that were independent of the visual response to the cue. We placed the cue outside of the receptive field and the target within the receptive field. We found that 23% of these cells had a significant decrease in their firing rate to validly cued targets in their receptive fields under these conditions. Strong responses were evoked by the same target when the cue was flashed in the opposite hemifield (invalidly cued targets). Thus this group of neurons responded best when attention was directed toward the opposite hemifield.

4. For another group of parietal cells (13%) there was an enhanced response to targets in the visual receptive field when the cue was in the same hemifield. For the remaining 64% of the cells there was no significant modulation in this task.

5. The cued reaction time task involved exogenous control of attention; the sensory cue gave spatial and temporal direction to attention. We used several other tasks to test for endogenous control of attention. For some cells, when a monkey simply gazed at a spot of light there was only a modest response to peripheral visual stimuli; when the monkey performed the foveal attention task there was an increase in the intensity of response of the same cell to the same peripheral stimulus. Thus, when attention was directed away from the Visual receptive held by endogenous control, there was a similar augmentation of response.

6. When an animal responded repetitiously to targets outside of the visual receptive field (probability task), there was a strong response evoked when the stimulus appeared unexpectedly within the receptive field. Weak responses were elicited at expected locations. The modulations in the cued reaction time, foveal attention, and probability tasks were quantitatively similar. These observations are also consistent with other data showing that a group of parietal cells responded best when attention was not directed into the visual receptive field. This was true whether attention was manipulated exogenously or endogenously.

7. Approximately 45% of the neurons tested discharged in relation to saccadic eye movements, and the largest number of such cells was located in the posterior bank of the intraparietal sulcus.

8. We conclude from these experiments that parietal cells participate in attentional processes. All respond to the visual cue that directs attention, and this may signal a shift of attention. When the visual cue was positioned near the receptive held, differential activity was produced, all of which may signal attentional shifts. Certain of these cells also had modulations in endogenous tasks that augmented the response when attention was away from the receptive held; such activity could signal a shift of attention to the receptive field. These data provide some mechanisms for contributions of parietal cortex to the dynamics of visual attention.