Pink polygon or polygon and pink? Dissociating conjunctions and relations in synaesthesia

In grapheme-colour synaesthesia, words, letters, and/or numbers evoke specific colours. The conceptual mediation hypothesis proposes that the link between graphemes and colours is a conceptual association akin to binding an object with its typical colour. Synaesthetic associations, then, may reflect unusually strong associative learning and memory systems relative to non- synaesthetes. Previous studies report inconsistent findings regarding a synaesthete associative memory advantage. These studies, however, typically do not distinguish between object-colour (conjunctions) and object-object associations (relations) or consider how different retrieval methods may impose different demands on associative memory. Potentially, a synaesthete associative memory advantage depends on the type of association to be remembered and how it is probed. This thesis presents two experiments to test whether grapheme-colour synaesthetes remember novel polygon-colour associations more accurately than non-synaesthetes, and if so, whether this advantage generalises across both association types or is specific to either conjunctions or relations. Chapter 1 presents a general introduction to the background literature on object-colour knowledge, some major theories of synaesthesia, and methods for assessing associative memory in synaesthetes and the general population. Chapter 2 presents an experiment testing for a synaesthete associative memory advantage using an associative recognition task (Experiment 1). Chapter 3 presents a complementary experiment using an associative reconstruction task (Experiment 2). Chapter 4 then brings together the results in a general discussion of the implications in the context of the broader literature. Across the two experiments, there was no evidence for a synaesthete associative memory advantage for the non-inducing associations tested. There was a robust memory advantage for conjunctions relative to relations in both synaesthetes and non-synaesthetes. This study rules out several factors as modulators for the inconsistent synaesthete advantage and opens avenues for investigating whether the synaesthete associative memory advantage reported previously is driven by enhanced associative learning rather than memory per se.