Abstract
Numerous studies claim that optimal decision-makers in sensorimotor tasks use knowledge of both their variability and the task's utility function (how each state of the world relates to a gain). In situations requiring interaction with moving objects, people can base their decision on temporal or spatial information: therefore, utility can be defined either as a function of time or space. We aimed at knowing, in such contexts, to what extent performance depends on the domain of the utility function. Our participants completed one of two conditions. A target which could vary in speed (19.5, 25, 32cm/s) moved toward a line. Participants stopped the target and were rewarded according to the target's temporal (time condition) or spatial (space condition) proximity to the line. Responding after the target crossed the line was penalized. We identified whether response strategies were based on spatial (monitoring the target's position) or temporal (pure velocity) cues. Almost all participants in the space condition displayed a spatial strategy, while participants in the the time condition distributed evenly between temporal and spatial strategies. For both conditions, despite similar mean responses and variability in both strategies, participants using temporal cues were more stable across target speeds, thus winning more than those using spatial cues, who responded more after the line in faster speeds and were more penalized. We suggest that this was caused by the limited resolution of the visual system while tracking the spatial position of a moving object, so spatial uncertainty was larger for faster speeds. Our results show how, in environments with more speed variability, actions based on temporal information may be more adaptive, since it brings stability across different speeds. We also evidence the need to consider possible system constraints underlying response strategies when studying sensorimotor decision-making.
Meeting abstract presented at VSS 2017