Base on balls for the Chapman strategy: reassessing Brouwer, Brenner, and Smeets (2002)

A true understanding of skilled behavior includes the identification of the information that underlies the perception-action cycle at work. Often, observers' sensitivity to perceptual variables is established in laboratory-situated simulation-based psychophysical experiments. The observers' sensitivity thus determined is then used to draw conclusions that will generalize the findings to natural behavior. Focusing on the example of running to catch fly balls, the present contribution takes the study of Brouwer, Brenner, and Smeets (Perception & Psychophysics 64:1160-1168, 2002) to illustrate how common assumptions in the steps from psychophysical experiments to natural behavior can result in ungrounded conclusions. These authors built an argument to reject the use of the Chapman strategy of zeroing out optical acceleration. For this argument, they determined the sensitivity of the visual system to acceleration, assuming that acceleration is detected as a velocity ratio. Next, they showed that catchers started running earlier than could be expected on the basis of sensitivity thresholds for this velocity ratio, concluding that running initiation could not have been based on optical acceleration. In the present study, we argue that important assumptions in the Brouwer et al. (Perception & Psychophysics 64:1160-1168, 2002) line of argument are incorrect. First, we show how the assumption of parabolic ball flight trajectories, although convenient, biased Brouwer et al.'s (Perception & Psychophysics 64:1160-1168, 2002) conclusion. Next, we present an experiment revealing that observers do not base their judgments of acceleration on the velocity ratio. Thus, we demonstrate that Brouwer et al.'s (Perception & Psychophysics 64:1160-1168, 2002) argument that optical acceleration cannot serve as the information for running to catch fly balls does not hold.