Better together? The cognitive advantages of synaesthesia for time, numbers, and space

Abstract

Synaesthesia for time, numbers, and space (TNS synaesthesia) is thought to have costs and benefits for recalling and manipulating time and number. There are two competing theories about how TNS synaesthesia affects cognition. The “magnitude” account predicts that TNS synaesthesia may affect cardinal magnitude judgements, whereas the “sequence” account suggests that it may affect ordinal sequence judgements and could rely on visuospatial working memory. We aimed to comprehensively assess the cognitive consequences of TNS synaesthesia and distinguish between these two accounts. TNS synaesthetes, grapheme–colour synaesthetes, and nonsynaesthetes completed a behavioural task battery. Three tasks involved cardinal and ordinal comparisons of temporal, numerical, and spatial stimuli; we also examined visuospatial working memory. TNS synaesthetes were significantly more accurate than nonsynaesthetes in making ordinal judgements about space. This difference was explained by significantly higher visuospatial working memory accuracy. Our findings demonstrate an advantage of TNS synaesthesia that is more in line with the sequence account.

Keywords:

• Synaesthesia
• Time
• Number
• Sequence–space
• Working memory

Joanna Hale is now at the Institute of Cognitive Neuroscience, University College London. Marinella Cappelletti is also at the Department of Psychology, Goldsmiths, University of London.

The research was conducted at the University of Oxford.

Roi Cohen Kadosh is supported by the Wellcome Trust, which funded this work under Grant WT88378. Marinella Cappelletti is sponsored by the Royal Society.

Notes

¹ Twenty of the 23 TNS synaesthetes in this study had both time forms and number forms. Two TNS synaesthetes reported only time forms, and one reported only number forms. The synaesthetes with time forms did not significantly differ in task performance from the participant with number forms, and the exclusion of these three participants from the main analyses did not change the results. Therefore, the results reported here include all 23 TNS synaesthetes.

² It is possible that participants would intentionally or unintentionally estimate the magnitude of time duration before and after the target in order to complete the sequence manipulation of the Time task. To reduce this possibility, it was emphasized to participants that they should judge the target's sequential position and that the duration of time is an unreliable indicator of the number of circles in the sequence.

It is also possible that participants would attempt to estimate the numerosity of stimuli before and after the target in order to complete the sequence manipulation of the Time task. Even if this were the case, their responses must also reflect a temporal perception of when the target occurred in relation to the rest of the sequence of stimuli.

³ Box's test and Levene's test were insignificant. Mauchly's test was significant for load [W(5) = 0.82, approx. χ² = 12.34, p = .03], so the Greenhouse–Geisser values are reported.