The Effects of Task- and Switch-Predictability on Oculomotor Inhibition of Return During Visual Search.
Bahle B., Mills M., Dalmaijer E., Van der Stigchel S., Dodd M.
Task-switching methods have long been used to study cognitive control processes but only recently have attempts been made to extend these methods to the study of oculomotor control processes. Nonetheless, preliminary examinations have demonstrated that oculomotor control processes are quite sensitive to changing task contexts, thus establishing the basic utility of task-switching methods for investigating oculomotor control processes. A recent study, for example, reported a general effect of task-switching on the expression of oculomotor inhibition-of-return (O-IOR) such that O-IOR was observed on task-repetition trials whereas facilitation-of-return (FOR) was observed on task-switch trials (Mills et al., VSS, 2014 ). From the perspective that O-IOR beneficially services search processes by biasing the eyes away from recently inspected locations and toward novel locations (Klein, 1998), the observation of FOR on task-switch trials represents a considerable cost to visual behavior for switching tasks. It would be instructive, therefore, to determine how the oculomotor system mitigates this cost. Behavioral studies of task-switching indicate that switch-costs can be markedly reduced and even eliminated if the schedule of a switch is predictable (Koch, 2005). The main goal of the present study, therefore, was to investigate effects of predictability on O-IOR switch-costs. Participants viewed scenes while either searching for a target 'N' or 'Z', memorizing the scene in preparation for a memory test, or evaluating scene pleasantness. The critical manipulations were Switch-Predictability (whether or not participants knew when a switch would occur) and Task-Predictability (whether or not participants knew on trial n which task to switch to on trial n+1). When participants were prepared for a task switch to search, their initial fixations were more efficient (e.g. less refixations and a greater degree of O-IOR) compared to when the switch-predictability and task-predictability was random. Meeting abstract presented at VSS 2015.