We constantly perform tasks within complex and dynamic environments. Some of these tasks (e.g., road crossing or playing team sports) require predicting future states of the world to decide which action to unfold and when to do so. However, it remains largely unexplored how the variability in a scene influences perceptual decision-making. Here we examine how increasing the scene variability influences our ability to make perceptual judgements and decisions by using a go/no-go decision task in a dynamic scenario mimicking a road-crossing situation with different levels of stimuli variability. Parameters of psychometric functions revealed that differences in variability do not influence judgements about the objects’ time-to-contact, or the difficulty in making such judgements. Nevertheless, increases in the scene variability influence the go/no-go decisions leading people to adopt more conservative criteria. How much the criterion changes across levels of variability is well accounted for by the actual amount of variance in the scene, but the overall criterion is tightly linked to the precision or reliability with which one can estimate perceptual information about the objects’ arrival time. These results suggest that the reliability on our own perceptual estimates modulate our criterion when completing perceptual decision-making tasks under different scene variabilities.
Malla et al. (Mon,) studied this question.