An expanded model for perceptual norming: insights from Japanese ideophones

Iconicity is inherently grounded in sensory experience, yet few studies investigate how sensory information is packaged in iconic words. We present perceptual strength ratings for Japanese ideophones, to ask how sensory information is encoded in this word class. While previous studies employ a five or six-sense model, we expanded the visual and interoceptive domains to incorporate a total of thirteen perceptual dimensions. Extra dimensions were drawn from asymmetries in the cross-linguistic attestation of ideophones, to explore whether these reflect underlying structural differences in the encoding of these perceptual domains. We replicate previous findings regarding visual dominance, but show that this is primarily driven by the dimension MOVEMENT, with other visual dimensions receiving comparatively lower ratings. As to relations between the senses, we find that these can differ even within a single modality. For example, within vision, MOVEMENT was most connected to SOUND and INTEROCEPTION; SHAPE to TEXTURE; and COLOUR to TASTE and SMELL. The results speak to the value of an expanded model of the senses for perceptual norming studies, with an expanded set of linguistic signs as its object.