Cross-modal matching in first school children
This research examines how cross-modal and within-modal matching differ. Three broad classes of difference are considered, that crossmodal matching requires (a) "translation" between modality-specific stores or (b) "transformation" of information when different information is available through each modality or (c) transformation, whatever the information available through each modality, owing to differences in encoding strategy. Visual-kinaesthetic matching of the end-point of lever movements has been investigated. It is argued that adult cross-modal performance depends on information and strategy, not directly on modality. Results with children have been less clear, experiments were therefore undertaken, with subjects aged 6-9 years. The hypothesis was that childrens' performance also would be determined by available information, and strategy. With information differences eliminated, modality conditions did not differ in within-subject variability, with up to 20 second unfilled retention intervals. With visual information enhanced by background cues and emphasis of the body midline, visual matching was superior to kinaesthetic, and within-modal was superior to cross-modal matching. These differences disappeared with practice, together with coding relative to the midline in the cross-modal conditions. Midline-relative coding was the norm with the midline emphasised, and without background cues. With or without variability differences, VV did not differ from KK in bias, but KV resulted in greater overshooting, VK greater undershooting. The most likely explanation is resistance to movement when perception is kinaesthetic, causing overestimation of distance moved. Consideration of the information normally available to subjects, generated the hypothesis that temporal and spatial parameters should interact more with kinaesthetic than with visual perception. This was supported, since movement velocity biased only kinaesthetic judgements. It is concluded that matching performance depends on the information encoded and used as the basis of matching, which depends on strategy; strategy depends on information (a) available during stimulus presentation, (b) normally available in each modality, (c) which it is anticipated will be available during response.