Understanding and Motivating Human Control: Outcome and Reward Information in Action

Abstract

In this thesis motivated goal-directed behavior is investigated from a self-emergent process perspective. This perspective supposes that motivation for goal-directed behavior can be the result of a pattern of relatively simple interactions between reward and outcome information. In this thesis three basic interactions were examined to demonstrate the potential self-emergence of motivated goal-directed behavior. In a first line of research it is shown that prospective monetary gains encourage people to invest effort and mobilize control processes for better task performance. Importantly, an interaction was found such that reward induced effort mobilization only occurred when task demands were high in terms of control processing, even when these control demands were not explicitly cued beforehand. In other words, this process of motivated control, which was governed by a principle of conservation, seems to follow a self-emergent property. In a second line of research it was investigated how reward information and outcome information interact in controlling goal-directed behavior. It was found that the interaction between reward and outcome information spontaneously influenced the way people controlled their goal-directed actions. In a third line of research participants first acquired knowledge about action-outcome relationships in a rewarding context, and it was found that this interaction produced motivated goal-directed behavior on its own accord. This finding again suggests that motivated goal-directed behavior can be the result of a self-emergent process. Taken together, these three main findings fit well with recent mechanistic views on goal-directed behavior and offers insight into the potential building blocks that renders human action goal-directed. It is important to note that these building blocks are not essentially different from those proposed by other approaches to understand goal-directed behavior, such as the traditional expectancy-value approach. This approach also takes the value of an outcome into account to understand goal-directed behavior. Importantly, the traditional expectancy-value approach considers goal-directed behavior to result from people’s ability to explicitly reflect on and compute the value of an outcome. The current thesis tries to offer a different perspective to understand how goal-directed behavior emerges and is maintained over time, and that answering this question could be facilitated by looking at basic interactions between specific types of information that have been argued to form the basis of motivated goal-directed behavior, such as information about demands, outcomes and rewards.