Dokument

Summary:
Learning bias has been prominently studied in the past few decades. The relationship between the cue and outcome in learning, and the nature of their properties have been observed to have an impact on future learning. One example is the outcome predictability effect. Outcome predictability effects describe better learning about outcomes with a history of greater predictability in a similar but unrelated task compared to outcomes with a history of unpredictability. It was first described in Griffiths et al. (2015) using an allergy task. Despite the prominent observation in various studies, questions about how this effect could be brought about remain open. One of the possible hypotheses states that the outcome predictability effect is driven by a change of motivation. This hypothesis puts the outcome predictability effect to a similar category as the learned helplessness effect, which is also a certain type of learning bias using an instrumental learning paradigm. The present thesis aims to investigate the connection between outcome predictability effect and motivation, and how it is related to learned helplessness. Three studies are presented in this thesis. Study I investigated whether learning about unpredictability decreases outcome-specific motivation to learn. A modified version of the allergy task involving an active learning paradigm was used as a tool to measure participants’ motivation to learn about a certain outcome. Study II also utilized the allergy task to investigate the connection between outcome predictability and motivation. Specifically, Study II investigated whether manipulating extrinsic rewards for a correct prediction could affect subsequent learning and whether it counteract or diminish the outcome predictability effect. Study III investigated this hypothesis in instrumental learning, expecting to find both learned helplessness and outcome predictability effects in the same experiment. We created a new computer-based task in which the participants had to stop several tones. The results from Study I confirmed the hypothesis that outcome unpredictability indeed decreased the motivation to learn about that specific outcome in a new situation. Furthermore, the effect of extrinsic rewards indeed affects subsequent learning. Both the results from Study I and II showed the relationship between outcome predictability effect and motivation. However, this reward effect also appeared to have a significant impact on the outcome predictability effect itself. Using the instrumental task in Study III, we indeed observed a learning bias in a new situation. Implications of the present thesis to learning and motivation, as well as clinical implications, are discussed.

Das Dokument ist im Internet frei zugänglich - Hinweise zu den Nutzungsrechten