Abstract
Purpose:
Our previous findings show that practice can substantially improve dynamic bisection acuity. However, since the stimulus exposure time (300 ms) was longer than the latency for saccadic eye movements, the improved acuity could have been merely the result of improved tracking eye movements. In the present experiment, we used a shorter stimulus duration (150 ms) in order to isolate the effects of perceptual learning on motion processing.
Methods:
The bisection stimulus comprised of three briefly presented black square dots (duration 150ms) separated by 5 arcmin. The observer’s task was to localize the position of the central dot relative to the adjacent dots. On day 1, the pre-training session consisted of 5 stimulus conditions: [1] static horizontal orientation, [2] static oblique (45 deg) orientation [3] oblique stimulus moving in an orthogonal direction (135 deg) at 1.33 deg/s, and [4-5] oblique stimulus moving along a 45 degree trajectory at two different speeds (1.33 & 2 deg/s). Observers practiced the task with the static oblique stimulus from day 2 to 8, subsequently with the moving oblique stimulus (2 deg/sec) from day 10 to 16. Performance for all five conditions was re-evaluated in two post-training sessions on days 9 and 17.
Results:
Bisection acuity improved dramatically (35%) after practicing with the static oblique stimulus. Learning transferred substantially to the untrained static horizontal stimulus (26% improvement) and the three untrained moving oblique stimuli (1.33 deg/s along 135 deg, 34% improvement; 1.33 deg/s along 45 deg, 31% improvement; 2 deg/s along 45 deg, 17% improvement). Subsequent practice with the oblique stimulus in motion (2 deg/s) resulted in a further 37% improvement in dynamic bisection acuity (a total of 47% from day 1), and the performance for lower speed stimuli improved as well.
Conclusions:
Our findings suggest that perceptual learning enhances tolerance of image motion. We characterized the specificity of learning a bisection acuity task and showed that the learning generalizes substantially, but not completely, from static to moving stimuli, and across different stimulus speeds and orientations.