Busy Doing Nothing: Evidence for Nonaction-Effect Binding

Article excerpt

Research on voluntary action has focused on the question of how we represent our behavior on a motor and cognitive level. However, the question of how we represent voluntary not acting has been completely neglected. The aim of the present study was to investigate the cognitive and motor representation of intentionally not acting. By using an action-effect binding approach, we demonstrate similarities of action and nonaction. In particular, our results reveal that voluntary nonactions can be bound to an effect tone. This finding suggests that effect binding is not restricted to an association between a motor representation and a successive effect (action-effect binding) but can also occur for an intended nonaction and its effect (nonaction-effect binding). Moreover, we demonstrate that nonactions have to be initiated voluntarily in order to elicit nonaction-effect binding.

Daily life involves various incidents of intentional nonaction. Imagine a wife demanding to talk with her spouse about their relationship and him ignoring her request. Presumably this not-acting is a voluntary act. And very often intentional nonactions1 have foreseeable consequences. If, for instance, you decide not to catch a ball flying straight at your face, you can anticipate what will happen. Along the same lines, the Taoist concept of wei wu wei, meaning "acting by not acting" (Loy, 1985), considers nonaction to be dynamic and clearly separate from inaction. In the legal domain, human societies acknowledge nonaction (namely negligence) as an intentional act by considering it to be punishable under the precondition of purposefulness.

Surprisingly, psychological research on human performance focuses entirely on the investigation of action, while neglecting intentional nonaction almost completely. This is presumably due to the fact that in experiments on intentional nonaction, one loses the typical dependent measures of experimental psychology-namely, reaction times (RTs) and error rates. In order to fill this gap, we aim to investigate the hypothesis that intentional nonaction shares properties with intentional action, by testing whether both are coded in a similar way.

An influential theory of motor control, ideomotor theory, states that actions are represented in the form of sensory feedback they produce (Greenwald, 1970; Prinz, 1997). In other words, we control our actions by anticipating the sensory consequences of these actions. In accordance with the ideomotor principle, Elsner and Hommel (2001) demonstrated that participants indeed form action-effect associations. In their experiments, participants freely chose between two actions (pressing a right or left key) that were followed by specific but irrelevant effect tones (high- and low-pitched tones). In a test phase, participants were required to respond to the effect tones by choosing spontaneously which button to press. In accordance with ideomotor theory, participants preferred to choose actions that had previously been associated with the tone (consistent mapping), rather than actions that were associated with the other tone (inconsistent responses).

In the present study, we use action-effect binding to investigate whether intentional nonaction shares essential coding properties with intentional action. At the same time, we want to test whether action-effect binding consists of an association between motor patterns and the sensory effect or higher level representations and the subsequent effect.

For that purpose, we use the same setup as did Elsner and Hommel (2001, Experiment 2A), with the response option to press the right or left response button and the additional option to press no button. On the basis of the assumption that intentional nonactions resemble intentional actions in that they both involve a voluntary intention, we expect to find action-effect binding-namely, more consistent as opposed to inconsistent choices in a free-choice test phase-for voluntary nonactions as well. …