Abstract
According to the computational conflict monitoring model of Botvinick et al. (2001), cognitive control recruitment in conflict tasks should be modulated on a trial-by-trial basis. Based on this, two hypotheses concerning conflict adaptation can be derived: Conflict interference should not only be reduced in blocks with high conflict frequency compared to blocks with low conflict frequency, but should also show gradual modulations on a trial-by-trial basis if conflict probability is parametrically manipulated. In addition, conflict interference should not only be reduced on trials immediately following a conflict trial, but should also show a gradual reduction over a sequence of multiple conflict trials. We empirically tested those predictions in two visual search experiments, in which the presence of a salient but irrelevant distractor may generate a conflict as to where focal attention should be allocated. Unlike common practice, we analyzed non-aggregated RT data by mixed model (multilevel) regression, which allows for trial-by-trial predictions to be tested on a trial-by-trial level. In Experiment 1 distractor probability (for a given trial) was parametrically manipulated: It varied sinusoidally from low to high and back to low over the course of the experiment. Distractor interference developed inversely proportional to conflict probability, i.e. distractor interference decreased with increasing distractor probability and increased again with decreasing distractor probability. In Experiment 2 the number of consecutive conflict encounters (i.e., distractor-present trials) was parametrically manipulated: Twenty sequences of ten consecutive distractor-present trials were presented over the course of the experiment. Distractor interference during those sequences showed a linear decrease with an increasing number of conflict encounters. Overall, those results are in line with the predicted trial-by-trial adjustments we derived from the conflict monitoring model of Botvinick and colleagues. Additionally, they further support the notion that conflict adaptation effects observed for response conflict can be generalized to attentional conflict.
Meeting abstract presented at VSS 2012