Do synaesthesia and mental imagery tap into similar cross-modal processes?

Abstract

Synaesthesia has previously been linked with imagery abilities, although an understanding of a causal role for mental imagery in broader synaesthetic experiences remains elusive. This can be partly attributed to our relatively poor understanding of imagery in sensory domains beyond vision. Investigations into the neural and behavioural underpinnings of mental imagery have nevertheless identified an important role for imagery in perception, particularly in mediating cross-modal interactions. However, the phenomenology of synaesthesia gives rise to the assumption that associated cross-modal interactions may be encapsulated and specific to synaesthesia. As such, evidence for a link between imagery and perception may not generalize to synaesthesia. Here, we present results that challenge this idea: first, we found enhanced somatosensory imagery evoked by visual stimuli of body parts in mirror-touch synaesthetes, relative to other synaesthetes or controls. Moreover, this enhanced imagery generalized to tactile object properties not directly linked to their synaesthetic associations. Second, we report evidence that concurrent experience evoked in grapheme-colour synaesthesia was sufficient to trigger visual-to-tactile correspondences that are common to all. Together, these findings show that enhanced mental imagery is a consistent hallmark of synaesthesia, and suggest the intriguing possibility that imagery may facilitate the cross-modal interactions that underpin synaesthesic experiences. This article is part of a discussion meeting issue 'Bridging senses: novel insights from synaesthesia'.

Keywords: cross-modal correspondences; mental imagery; mirror-touch; somatosensation; synaesthesia; touch.

Figure 1.

The mean ratings provided to the (a) TMI and (b) SMI tasks by each of the control, non-MTS (non-MTS) and MT synaesthesia (MTS) groups in experiment 1. Individual participant mean ratings are also shown (as filled circles) within each group. Error bars represent 95% confidence intervals. (Online version in colour.)

Figure 2.

Plot showing the interaction between ratings provided to materials and object properties found in experiment 1. Error bars represent 95% confidence intervals. (Online version in colour.)

Figure 3.

Participants mean ratings to the SMI task for each of the control, non-MTS and MTS groups (adapted from [48]). Groups are colour coded in different shades as follows: control (light), non-MTS (intermediate) and MTS (dark). The five main body regions are colour coded as follows: head (purple), upper body (yellow–orange), lower body (red), foot (blue) and hand (green). (Online version in colour.)

Figure 4.

(a) Example of a stimulus in the colour patch test used in experiment 2. (b) Mean weight ratings provided across both groups of participants to colour patches. Positive (negative) weight ratings indicate that the test stimulus was judged to be heavier (lighter) than the standard stimulus. The dotted line indicates when both stimuli were considered equal in weight. Error bars represent 95% confidence intervals. (Online version in colour.)

Figure 5.

‘Weight’ responses in synaesthete group to letters and colour patches. Error bars represent 95% confidence intervals. Positive (negative) weight ratings indicate that the test stimulus was judged to be heavier (lighter) than the standard stimulus. The dotted line indicates when both stimuli were considered equal in weight.