Free choice tasks as random generation tasks: an investigation through working memory manipulations.

Abstract

Free choice tasks are tasks in which two or more equally valid response options per stimulus exist from which participants can choose. In investigations of the putative difference between self-generated and externally triggered actions, they are often contrasted with forced choice tasks, in which only one response option is considered correct. Usually, responses in free choice tasks are slower when compared with forced choice task responses, which may point to a qualitative difference in response selection. It was, however, also suggested that free choice tasks are in fact random generation tasks. Here, we tested the prediction that in this case, randomness of the free choice responses depends on working memory (WM) load. In Experiment 1, participants were provided with varying levels of external WM support in the form of displayed previous choices. In Experiment 2, WM load was induced via a concurrent n-back task. The data generally confirm the prediction: in Experiment 1, WM support improved both randomness and speed of responses. In Experiment 2, randomness decreased and responses slowed down with increasing WM load. These results suggest that free choice tasks have much in common with random generation tasks.