Visual and haptic recognition of unfamiliar three-dimensional objects: effects of transfer

We investigated whether the representation of objects is modality-independent or modality-specific for visual and haptic memory. Recent studies have shown that the representation of verbal material is modality-independent (Easton, Srinivas, and Green, 1997 Journal of Experimental Psychology: Learning, Memory, and Cognition 23 153 - 163). Other studies have reported modality-specific representations for familiar objects (Easton, Green, and Srinivas, 1997 Psychonomic Bulletin and Review 4 403 - 410). However, we argue that verbal material may be coded in a lexicon which is shared by the visual and haptic systems and therefore the results are equivocal. Also, property differences between common objects may promote better recognition performance within modalities suggesting differences in strategy not representation. In our experiments objects were constructed from six parts (LegoTM standard bricks) arranged randomly in stacks. All objects were made of the same material with the same overall size and aspect ratio. We used a recognition memory paradigm to test subjects' ability to recognise objects that were studied either haptically or visually. At test, the objects were presented either within or transferred across modalities. A cost of transfer was expected if vision and haptics did not share the same representation. In experiment 1, subjects studied each target object, visually or haptically, for 30 s. We found a cost in transfer on recognition performance. However, this cost was mainly due to the high number of correct responses within the visual modality. In experiment 2, the study time for haptic recognition was increased so that haptic performance was equivalent to visual performance. Again, we found a cost of transfer. We conclude that the visual system and the haptic system do not share the same representations