Is the construction of spatial models multimodal? New evidences towards sensory-motor information involvement from temporary blindness study.

Using new developments of interference paradigm, this paper addresses the raising question of the involvement of sensory-motor information in the construction of elaborate spatial models (Johnson-Laird in Mental models: towards a cognitive science of language, inference, and consciousness Cambridge University Press Cambridge, 1983). In two experiments, 112 participants had to explore and memorize the spatial arrangement of 12 objects, disposed on 3 tables. Participants were either sighted or blindfolded, leading to a visual or a more sensory-motor based exploration of the room. During exploration, participants were required to perform a classical verbal, a visuo-spatial dual task or none. In the second experiment, more exploratory, we draw on interference paradigm literature and its recent development in the embodied field to develop two original dual tasks meant to interfere directly with the acquisition of sensory-motor information (haptic and action). After this learning phase, five tasks addressing spatial memory and reasoning used in the construction of spatial models were performed. Results showed classical effects for both verbal and visuo-spatial tasks for sighted participants, but not for blindfolded sighted ones, suggesting that a temporary visual deprivation led participants to use other way to build their spatial models. Our second experiment confirmed this point by showing effect of both sensory-motor dual tasks, especially for blindfolded sighted participants. Taking together, our results support a multimodal view of spatial models, and that exploration modality will influence the information used to construct them. Moreover, this challenges the Baddeley's dualist view of working memory as a reference to theorize the construction of spatial models and provide new experimental evidences towards an embodied view of spatial models.