Covert orienting of attention in macaques. I. Effects of behavioral context

1. A task was used by Posner (1980) to measure shifts of attention that occurred covertly, in the absence of an eye movement or other orienting response. This paradigm was used here to assess the nature of covert attentional orienting in monkeys to develop an animal model for neurophysiological studies. Shifts of attention were measurable in monkeys and were consistent across a variety of experimental conditions.

2. The paradigm required that monkeys fixate and release a bar at the appearance of a target, which was preceded by a cue. Reaction times to targets that followed peripheral cues at the same location (validly cued) were significantly faster than those that followed cues in the opposite visual field (invalidly cued). This difference was defined as the validity effect, which as in humans, is used as the measure of a covert attentional shift.

3. When the proportion of validly to invalidly cued targets was decreased, no change was seen in the validity effect of the monkeys. This is in contrast to humans, for whom the ratio of validly to invalidly cued targets affected the magnitude of the validity effect. When 80% of the targets were preceded by cues at the same location, the validity effect was greatest. The effect was reversed when the proportions were reversed. From this result, it is concluded that cognitive processes can affect covert orienting to peripheral cues in humans, whereas in trained monkeys, performance was automatic.

4. To test whether cognitive influences on attention could be demonstrated in the monkey, an animal was taught to use symbolic, foveal signals to covertly direct attention. The magnitude of this validity effect was greater than that obtained with peripheral cues.

5. The effects of motivational and perceptual processes were tested. Although overall reaction times could. be modified, the facilitating effects of the cues persisted. This constancy across motivational and perceptual levels supports the notion that the monkeys were performing the task in an automatic way, under the exogenous control of peripheral cues.

6. Most visual cuing has been tested with visual landmarks at the locations of cues and targets. These monkeys were trained with such landmarks, and when tested without them, the attentional effect of the cues was nearly abolished. These data suggest that local visual features can be important for covert orienting.

7. To determine the spatial extent of the effect of the cue, monkeys and humans were tested with four cue-target distances (0-60-degrees). Reaction times were fastest at a cue-target distance of 0-degrees (validly cued) for the earliest temporal interval for both monkeys and humans. For the monkey, the validity effect peaked at 20-degrees and declined more eccentrically. For humans, the validity effect was qualitatively similar for distances from 20 to 60-degrees. 8. The present studies demonstrate that peripheral visual cues result in attentional shifts in monkeys, similar to those described in humans. This approach provides an animal model for exogenously elicited attentional shifts. The effects of peripheral cues are reliable in trained monkeys and are minimally susceptible to endogenous or experimental influences. However, like humans, monkeys can exert endogenous control over the covert orienting of attention when symbolic cues are used.