| |
Abstract:
URPOSE: How does performance in multi-element tracking
(Pylyshyn and Storm, 1988, Spatial Vision, 3, 179-197) change when
the scene contains elements at differing depths? METHODS: Observers
tracked a subset of 16 moving elements. The target set size was 2,
4, 6 or 8. In one experiment, all elements were either at the same
depth defined by disparity or at one of two depths. In the latter
case, both targets and distractors were divided equally between the
two depths, so depth was not predictive of the target vs.
distractor distinction. All elements moved smoothly without
intersecting one another. In the query phase, observers marked all
the elements they had tracked. As a control, a similar experiment
was run with elements that were either of the same color or of one
of two colors. Five observers did 30 trials per condition in each
experiment. RESULTS: Performance in the two-depth case was
significantly better than for the one-depth case. No significant
difference was found between the one-color and two-color cases. The
proportion of variance due to depth was much higher than that due
to color. CONCLUSION: If multi-element tracking is performed by
perceptually grouping the targets into a single object
representation (Yantis, 1992, Cognitive Psychology, 24, 295-340),
then it is easier to group them across different depths than across
different colors.
|
