The role of visual imagery and verbal coding in short-term memory conjunctive binding: evidence from aphantasia

Abstract

Conjunctive binding − as opposed to relational binding − refers to our short-term memory for integrated representations of a series of features into a complex − unitized − object. The mechanisms necessary to encode conjunctive representations in memory are, to date, unknown. In this study, we tested whether mental visual imagery and verbal coding are necessary to encode conjunctive representations in short-term memory. We also explored the memory processes supporting retrieval of these representations. To explore specifically the role of visual imagery, we tested a population of aphantasic participants (N = 29) that we compared with typical imagers (N = 49), as assessed by the VVIQ. To assess the role of verbal coding, we had participants memorise colour-shape conjunctive representations while manipulating the ease with which labels could be used to define the shapes. Results showed a robust effect of the use of verbal coding on performance and confidence ratings, accompanied by an increase in both automatic and controlled memory processes. There was however no difference between aphantasics and typical imagers on these measures. In addition, even in the difficult-to-label condition, aphantasics performed well above chance, suggesting that neither visual imagery nor verbal coding are necessary to perform conjunctive binding. This supports the idea that binding is automatic, and the existing hypothesis according to which the first stage of conjunctive binding is implicit. It is possible that aphantasic participants used implicit representations, unavailable to their phenomenological experience, to solve the task, or non-visual spatial or sensorimotor strategies.