Is the head-fake effect in basketball robust against practice? Analyses of trial-by-trial adaptations, frequency distributions, and mixture effects to evaluate effects of practice.

Reactions to the pass of a basketball player performing a head fake are typically slower than reactions to a basketball player who passes without a head fake (i.e., head-fake effect). The present study shows that extensive practice reduces the head-fake effect in basketball. Additional analyses were conducted to explore the mechanism behind the reduced head-fake effect. First, we analyzed whether or not participants developed some control over the processing of irrelevant gaze direction, as indicated by specific trial-to-trial adaptations (i.e., congruency sequence effect). Second, we fitted the individual frequency distributions of RTs to ex-Gaussian distributions, to evaluate if practice specifically affects the Gaussian part of the distribution or the exponential part of the distribution. Third, we modeled individual RT distributions as the so-called mixture effects to examine whether the way irrelevant gaze direction impacts performance (either occasionally but massively or continuously but moderately) changes with practice. The analyses revealed that the effect of practice could not be explained with an increasing congruency-sequence effect. Also, it could not be found in the ex-Gaussian distributional analyses. The assumption that residual failure to inhibit the processing of the gaze direction in contrast to continuous failures to do so might favor mixed effects over uniform effects at later courses of practice could not be validated. The reduced head-fake effect thus is argued to source in participants' general increasing ability to inhibit the processing of the task-irrelevant gaze direction information and/or in a priority shift of gaze processing to a processing of the pass direction.