Abstract
The bouba/kiki effect is a naturally occurring association between abstract shapes, spikey or round, and nonsense words, /kiki/ or /baba/, respectively, found across cultures, languages, and different senses (e.e., Spence, 2011). Previously, we found 6-8 year-olds showed weaker audio-tactile (AT) associations than adults between heard nonsense sounds and felt abstract shapes, even after explicit instruction on how to optimally explore shapes or twice as much tactile exposure (Chow et.al, in review). Interestingly, AT associations were strengthened if participants first matched the same non-sense sound to complementary seen shapes (audio-visual) or the same felt shape to complementary seen shapes (visuo-tactile). Given that both these conditions provide prior visual exposure, and that early-blind-adults show abnormal AT associations (i.e., Fryer et.al., 2014), we considered if visual experience is critical or if what matters is highlighting relevant shape features, something more automatic in vision than touch. We tested if prior tactile-only (TT) exposure or prior imagery of shapes (IT) could enhance AT associations. In TT, children matched a smaller shape to one of two larger shapes. Felt shapes differed in 2D contour, not texture or material properties. In IT, children imagined if a cloud or star, matched a felt shape. Following 4 trials of TT or IT exposure, children judged which of two felt shapes, one round/one spiky, matched a sound (16 trials). Association strength was quantified as the proportion of trials a round shape was chosen for /a/ sounds. We found prior TT and TI exposure, which provide no direct visual experience, enhanced AT associations in 6-8 year olds. These results suggest that direct visual experience of abstract shapes may be sufficient, but not necessary, in forming the abstract AT associations tested here. Such audio-tactile associations are weaker and generally emerge later than visuo-tactile associations (e.g., Gori et.al., 2008; Streri & Spelke, 1979).