Associative analyses of reasoning-like behaviour in rats

• Main Author: Burgess, Katy V.
• Published: Cardiff University 2012
• Subjects: 599.35; BF Psychology
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.569934

This thesis examines how rats represent relationships in their environment. There are currently two broad classes of account of how animals learn about such relationships: The associative account offers a relatively simple mechanistic account of behaviour; while the second account proposes that animal behaviour, like human behaviour, is underpinned by the processes of causal and deductive reasoning, that are beyond associative analyses. Chapter 1 identifies three domains in which these two classes of account provide quite different analyses of animal behaviour, which are experimentally investigated in Chapters 2, 3 and 4. Chapter 2 reports three experiments that investigated the accuracy of predictions derived from the claim that rats are capable of forming and using causal models involving their own interactions with their environment (interventions) and external events (Blaisdell, Sawa, Leising, & Waldmann, 2006). The results failed to confirm these predictions and were instead more consistent with the operation of simpler processes. The results from Chapter 2 left open two interpretations: either rats can represent causality but do not use such representations to reason, or they do not represent cause per se. Chapter 3 investigated these alternatives in three experiments using a timing task, which should be sensitive to whether rats are more likely to represent their actions as causal than external events (Buehner & Humphreys, 2009). The results provided no support for the view that causal binding occurs in rats. Chapter 3 examined the possibility that sensory preconditioning might reflect a form of deductive reasoning (Hall, 1990). However, taken together, the results from four experiments provided no support for such an analysis; but instead helped to inform the nature of the associative processes that underlie sensory preconditioning. In summary, while the results reported in this thesis provide no support for analyses of animal behaviour that rely on the processes of causal or deductive reasoning, they do help to inform the nature of the associative processes involved.