Abstract
Smooth pursuit target selection is known to be influenced by bottom-up factors, like stimulus contrast as well as by top-down factors like attention. We investigated how bottom-up salience and top-down value information interact for the dynamic guidance of smooth pursuit. Our pursuit stimulus consisted of two overlapping random-dot kinematograms (RDK) with opposite contrast polarity, which moved in directions differing by 20 deg (dot density 1 dot/deg2, speed 10 deg/s). One RDK had a coherence of 40%; the coherence of the other was varied between 20 and 80% to manipulate the relative salience. In a salience baseline condition, we instructed subjects to simply pursue the stimulus. In a value condition, subjects won points for pursuing one RDK and lost points for pursuing the other RDK. The subjects were instructed to make as many points as possible, which were converted into a monetary reward at the end of the experiment. Both salience and value influenced the pursuit direction. Subjects moved more in the direction of the RDK if it was more salient and rewarded. However, the relative contributions of salience and reward changed over time. In the early phase of pursuit, subjects followed a salience-weighted average direction in both conditions. Later pursuit followed exclusively either the more salient direction in the salience condition or the rewarded direction in the value condition. In the value condition, subjects took about 450 ms until they followed the rewarded direction. Our results show that salience and value information are dynamically integrated for smooth pursuit. The integration of the top-down value information seems to be time-consuming since it is not present in the early phase of pursuit. This integration process seems to be slower for the pursuit of moving stimuli than what we had previously observed for saccadic eye movements to stationary stimuli.
This work was supported by the DFG Forschergruppe FOR 560 “Perception and Action”.