The starmaze: a new paradigm to characterize multiple spatial navigation strategies

Spatial orientation disorders constitute a major problem in aged subjects. Among the different cognitive strategies used for spatial navigation, the most complex ones seem to be primarily deteriorated during ageing. Being able to dissociate the different cognitive strategies might help us in the early detection of age-related brain dysfunctions. In order to study multiple spatial navigation strategies, a new task was designed and named 'starmaze'. This paradigm was first developed for animal experiments and then adapted to humans using virtual reality. In both cases, up to three different strategies of navigation can be dissociated. The starmaze is composed of a central pentagonal ring from which five alleys radiate. A hidden goal has to be found. To locate the target, the subject can use either distal cues (allocentric strategy), or proximal cues located on the inner walls (guidance strategy), or a sequence of body movements leading to the goal (egocentric strategy). Four versions of the task can be used: (i) the multiple strategies, (ii) the allocentric, (iii) the egocentric, and (iv) the guidance version. In the multiple strategies version, the task has two components: the first assesses the learning capability of the subject; the second permits the identification of the strategy spontaneously used by the navigator to solve the task (multiple strategies can be employed: allocentric, egocentric and guidance). The allocentric version of the starmaze requires the subject to learn to reach the goal from different starting points and therefore necessitates a spatial representation of the environment. In the egocentric and guidance versions of the task there are no distal extra-maze cues (the apparatus is surrounded by black uncued curtains); subjects need to either perform a left-right-left movement sequence or to use intra-maze cues (e.g. to first approach a chessboardlike, then a black, and finally a white wall) to reach the target from different departure points. Data acquisition is performed by means of a video recording system and/or a tracking software. A set of parameters are measured to characterize the spatial behavior of the subject quantitatively (for example the distance travelled during each trial or the number of visited alleys). Data processing is automated via a MATLAB batch program developed in our laboratory.