Purchase this article with an account.
Celine Paeye, Laurent Madelain; Finding the target as a reinforcer of saccadic amplitude variability in a visual search task.. Journal of Vision 2012;12(9):1236. doi: https://doi.org/10.1167/12.9.1236.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
We recently showed that part of saccadic amplitude variability may be controlled by operant learning (Paeye & Madelain, 2011). Saccadic amplitude distributions were reinforced with a tone depending on amplitude variability criteria.
In the present study we designed a new paradigm involving a visual search task to test whether finding a target among distractors could also be effective to reinforce various saccadic amplitude variability levels: the target was visible only if the saccadic amplitude variability reached a specific criterion.
Subjects were instructed to perform a visual search task. At the beginning of each trial 24 distractors (0.3 deg circles with a dash tilted to the left) were displayed. We used a gaze-contingent criterion to display the target (a 0.3 deg circle with a dash tilted to the right) at the item location that was looked at.
In a first experimental condition the target appeared when the current saccadic amplitude was rare (to increase amplitude variability). In a second experimental condition the target was displayed when the current saccadic amplitude was frequent (to decrease amplitude variability). In a control group target appearance was independent of saccadic amplitudes.
U-value (measuring distribution uncertainty) increased from a baseline of 0.57 on average to 0.86 at the end of the first experimental condition and then returned to 0.51 after changing the reinforcement criteria. In the control group averaged U-value first decreased (from 0.56 to 0.49) and then remained at a low level (0.52).
These results were confirmed in a second experiment. Furthermore we found that learning transferred to regular visual search trials (in which the target was displayed independently of subjects behavior).
Seeing the target is a consequence controlling saccadic properties. Moreover these results confirm that an operant learning process can guide saccadic amplitude variability.
Meeting abstract presented at VSS 2012
This PDF is available to Subscribers Only