Abstract
Oculomotor kinematics and visual behavior vary as a function of both task-set (Castelhano et al., 2009; Henderson et al., 1999; Mills et al., 2011; Smith & Henderson, 2009; Yarbus, 1967) and task-switching (Mills et al., 2015). One parameter that seems to be influenced by task and switch type is the speed by which individuals shift from a global (long saccades, short fixations) to a local (short saccades, long fixations) viewing mode. An earlier shift in viewing mode may reflect a differential emphasis for general spatial information processing vs. more specific object encoding. Velichkovsky et al. (2002, 2005) demonstrated local-type fixations to be positively related to object identification on a recognition task, suggesting that memory accuracy may be influenced by viewing mode. Though memory tasks are often included in task-set and task-switching studies, memory performance is not typically collected nor analyzed in relation to visual behavior, making it unclear as to whether task-related variation in eye movements may be reflective of a task-dependent shift from spatial-to-identity processing. To examine this issue, subjects viewed scenes while performing either a memorization, search, or evaluation task. When the task was memorization, each trial was followed by a forced-choice recognition test in which scenes were identical save for a single global (e.g., entire image flipped) or local (e.g., object removed/added) change. When collapsed across all participants, memory performance was unrelated to kinematics, however, there were considerable individual differences in memory performance and variability as a function of change type. Specifically, the earlier participants shifted to a local viewing mode, the more likely they were to detect a local change whereas a longer sustained global viewing mode led to superior performance for detecting global changes.
Meeting abstract presented at VSS 2017