Construction of a spatial mental model from a verbal description or from navigation in a virtual environment

Abstract

A very common everyday experience is the construction of spatial mental representations from navigation, inspection of maps, or even from verbal descriptions. The construction of a spatial representation from a verbal description has been extensively studied, and the characteristics of the representation as well as the processes involved in this construction have been clarified. These spatial representations are usually called spatial mental models, in reference to the Johnson-Laird theory of mental models (1983). Over the last few years, a number of studies have investigated the processes and cognitive abilities involved in the construction of these mental models. Many studies have shown that these representations preserve the spatial relationships between the elements of the described environment (e.g. Schneider and Taylor 1999). One question of recent interest has been to investigate the involvement of working memory in the construction of such representations, and a specific involvement of the visuo-spatial working memory has been evidenced, as well for survey as for route descriptions, even if to a different extend (e.g. De Beni et al. 2005). It has also been shown that imagery instructions help subjects to form a spatial model, and that they then specifically rely on their visuo-spatial working memory to construct their model (Gyselinck et al. 2006). There are many other means to construct a spatial representation of an environment, and the development of new techniques of virtual reality provides new tools to explore the spatial representations. Some studies have compared the spatial mental representations constructed in various ways; learning from a map and reading a verbal description (e.g. Tlanka et al. 2005); navigating in a real environment and learning from a map (Richardson et al. 1999); navigating in a real environment and navigating in a virtual environment (Waller 2000; Chabanne et al. 2003). Results usually show that the representations constructed are comparable. A question is to characterize the mental representations individuals construct when they are immerged in such a virtual environment, and then to examine the processes and cognitive abilities involved in the construction of a spatial model from virtual navigation. The aim of the study reported here is to compare the representation constructed from navigation in a virtual environment with the representation constructed by processing a verbal description of the same environment. Given that gender differences have been found to play a role in many visuo-spatial tasks (Voyer et al. 1995), this variable has been taken into consideration in this study. V. Gyselinck (&) AE S. Nicolas AE P. Piolino C.N.R.S., University R. Descartes, Paris, France e-mail: valerie.gyselinck@univ-paris5.fr