The results for this experiment are very similar to those of
Experiment 1, as can be seen in
Figure 6. Again, a 2 × 2 ANOVA revealed a significant interaction between the motion conditions and the target conditions,
F(1, 14) = 16.1,
p = 0.001. Subsequent within-subject
t-tests revealed a significant difference between the predictable and unpredictable motion conditions for the two target case,
t(14) = 5.97,
p < 0.001, but not for the four target case,
t(14) = 0.14,
p = 0.89, showing that observers could utilize motion information when tracking two targets but less so, if at all, when tracking four targets.
In the above analysis, a trial was counted as correct only if the observer was able to correctly identify both of the disks that were highlighted at its end. Alternatively, one could quantify tracking performance by measuring the percentage of the highlighted disks that were correctly identified at the end of the trial. Adopting this new measure did not change our findings: A 2 × 2 ANOVA still revealed a significant interaction between motion conditions and target conditions, F(1, 14) = 10.7, p = 0.006. Subsequent within subject t-tests revealed a significant difference between the predictable and unpredictable motion conditions for the two target case, t(14) = 5.64, p < 0.001, but not for the four target case, t(14) = 0.81, p = 0.43. Thus, our findings do not depend on using a particular definition for tracking accuracy.