Academic journal article Memory & Cognition

Preferred Mental Models in Reasoning about Spatial Relations

Academic journal article Memory & Cognition

Preferred Mental Models in Reasoning about Spatial Relations

Article excerpt

The theory of mental models postulates that individuals infer that a spatial description is consistent only if they can construct a model in which all the assertions in the description are true. Individuals prefer a parsimonious representation, and so, when a description is consistent with more than one possible layout of entities on the left-right dimension, individuals in our culture prefer to construct models working from left to right. They also prefer to locate entities referred to in the same assertion as adjacent to one another in a model. And, if possible, they tend to chunk entities into a single unit in order to capture several possibilities in a single model. We report four experiments corroborating these predictions. The results shed light on the integration of relational assertions, and they show that participants exploit implicit constraints in building models of spatial relations.

Imagine that a speaker is describing a room to you. The speaker describes what is along one wall, as follows:

A table is between the TV and a chair.

Like most descriptions of a spatial relation, this assertion is compatible with many different arrangements of the entities to which it refers. Yet a theory that we will defend in this article postulates that you build a parsimonious mental model of a simple and typical spatial layout. In particular, you should be likely to imagine an arrangement in which the items of furniture are adjacent to one another and in which they are laid out in the following left-to-right order:

TV table chair.

This diagram denotes a mental model of the layout, but real mental models contain, not words in English, but representations of entities in the world. We use such diagrams to denote mental models throughout the present article, and for simplicity, we often refer to them as though they themselves were models. Our real concern, however, is mental models of spatial relations. Individuals have a natural tendency to construct them in order to make sense of spatial descriptions.

Early research on relational descriptions studied how individuals make inferences such as

Art is taller than Beth.

Cath is shorter than Beth.

Therefore, Art is taller than Cath.

One account of these so-called three-term series problems relied solely on the linguistic structure of the premises (Clark, 1969; see also Rips, 1994), and linguistic factors do affect the difficulty of these problems (Sternberg, 1981). But many results show that the relations are integrated into a linear spatial array, either a vertical one or a horizontal one (e.g., De Soto, London, & Handel, 1965; Huttenlocher, 1968; Potts & Scholz, 1975; for a review, see Evans, Newstead, & Byrne, 1993). An array can be used to derive novel conclusions or to check given conclusions even if the relation is temporal or abstract (Boroditsky, 2000; Goodwin & Johnson-Laird, 2005). The array is a mental model of the situation, and recent evidence for the use of models in reasoning about all sorts of relations comes from studies using secondary working memory tasks (Klauer, Stegmaier, & Meiser, 1997; Vandierendonck & De Vooght, 1997) and neuroimaging (e.g., Fangmeier, Knauff, Ruff, & Sloutsky, 2006; Knauff, Fangmeier, Ruff, & Johnson-Laird, 2003).

In the present article, we examine the theory of mental models and some of the consequences of its principle of parsimony. Three-term series problems, however, do not reveal much about me structure of models, and so, in order to investigate parsimony, we used a novel sort of relational reasoning in which individuals do not draw a conclusion from premises but, instead, decide whether a set of spatial assertions is consistent or inconsistent. This task occurs spontaneously in the comprehension of everyday discourse, especially when individuals are obliged to follow instructions about, say, a stage setting, the assembly of a kit, or the directions to a destination. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.