Abstract
Research into the visual control of interception has generally assessed overall characteristics of ensemble averages of participant behavior. In contrast, the present contribution applied the novel Qualitative Inconsistency Detection (QuID) method on individual trials of locomotor interception under different target motion conditions. Starting from the assumption that visual control of locomotor interception is based in the nulling of (time derivatives of) pertinent optical angles, this method compares actual changes in locomotor direction with the direction signaled a visuomotor delay earlier by such nulling, on a trial-to-trial basis. In doing so, QuID hunts for inconsistencies as a method to disqualify consistent reliance on candidate informational variables. We considered the target-heading angle THA and the target’s bearing angle BA in two studies of locomotor interception of targets moving along circular trajectories in the agent’s plane of motion. We detected sign changes in participant heading direction (7346 instances in the two data sets) and evaluated if nulling zeroth-, first- or second-order time derivative changes in THA and BA could be responsible for each individual direction change. Two inconsistencies could surface. First, nulling a specific informational variable could lead to a change in the direction opposite to that observed. Second, the nulling of a specific informational variable should lead to a change in heading in a specific direction but no such change was observed. The analyses of the 7346 events ruled out the use of THA and its time derivatives, and implied the use of the first time derivative of BA early (up to 1 s) into the trial and a combination of the first and second time derivatives of BA later on. An alternative for combining variables would be the use of a time derivative of fractional order.