Intermodal matching by human neonates

Abstract

Normal human adults judge two identical objects to have the same shape even when they are perceived through different modalities, such as touch and vision. The ontogenesis of man's capacity to recognise such intermodal matches has long been debated. One hypothesis is that humans begin life with independent sense modalities and that simultaneous tactual and visual exploration of shapes is needed to learn to correlate the separate tactual and visual sense impressions of them^1–3. A second hypothesis is that the detection of shape invariants across different modalities is a fundamental characteristic of man's perceptual–cognitive system, available without the need for learned correlations^4–7. Recent research has shown that 6–12-month-old infants can recognise certain tactual–visual matches^8–11. However, such data cannot help resolve the classic theoretical debate. Infants of this age repeatedly reach out and touch objects they see, and such simultaneous bimodal exploration presumably offers ample opportunity for learning to correlate tactual and visual sense impressions. The experiments reported here show that humans can recognise intermodal matches without the benefit of months of experience in simultaneous tactual–visual exploration. We demonstrate that 29-day-old infants can recognise which of two visually perceived shapes matches one they previously explored tactually, thus supporting the second hypothesis listed above.