Fig 1.
Multiple object tracking (MOT) task trial overview.
(A) Objects (total: 18) are shown. (B) A subset of objects (i.e., the “targets”, varied between zero and five) turn gray for two seconds. Note that in the passive viewing condition (zero targets), no objects turn gray in this phase. (C) The target indication is removed. Targets are now indistinguishable from the other objects. (D) Objects randomly move across the screen and participants need to track the movements of the targets. Arrows indicate the current movement direction of the objects. (E) The objects stop moving and participants select the targets using the mouse. To keep motor demands constant, participants always need to select five objects but only the first selections count to performance (e.g., if two targets need to be tracked, then only the first two selections count to performance). After five objects are selected, feedback is given (i.e., correct selections turn green).
Fig 2.
(A) MOT task performance (i.e., fraction of correctly selected targets) as a function of attentional load and task experience. (B) Pupil size changes relative to the passive viewing condition as a function of attentional load. (C) Pupil size changes relative to the passive viewing condition as a function of task experience. (D) MOT task performance plotted against the pupil size changes, separately for each day. In addition, the number of targets in the MOT task is coded by the sizes of the circles (i.e., with increasing circle size, the number of targets increases). Error bars in all panels are standard error of the mean.
Fig 3.
The ordinate shows the model fit in form of the reported Bayesian information criterion (BIC). The abscissa indicates the number of predictors in the model. Green arrows indicate model additions that significantly explained additional variance to the variance explained by previous predictors. Red arrows indicate that the added predictors did not explain additional variance.
Fig 4.
(A) MOT task performance (i.e., fraction of correctly selected targets) as a function of attentional load and task experience. (B) Pupil size change relative to the passive viewing condition as a function of attentional load. Error bars in all panels are standard error of the mean.