Construction and Revision of Spatial Mental Models under High Task Demand

Construction and Revision of Spatial Mental Models under High Task Demand Jelica Nejasmic (jelica.nejasmic@psychol.uni-giessen.de)* Leandra Bucher (leandra.bucher@psychol.uni-giessen.de)* Paul D. Thorn (thorn@phil-fak.uni-duesseldorf.de)** Markus Knauff (markus.knauff@psychol.uni-giessen.de)* *Justus Liebig University, Experimental Psychology and Cognitive Science, Otto-Behaghel-Str. 10F, 35394 Giessen, Germany **Heinrich Heine University, Institute for Philosophy, Universitaetsstr. 1, 40204 Duesseldorf, Germany Abstract Individuals often revise their beliefs when confronted with contradicting evidence. Belief revision in the spatial domain can be regarded as variation of initially constructed spatial mental models. Construction and revision usually follow distinct cognitive principles. The present study examines whether principles of revisions which follow constructions under high task demands differ from principles applied after less demanding constructions. We manipulated the task demands for model constructions by means of the continuity with which a spatial model was constructed. We administered tasks with continuous, semi-continuous, and discontinuous conditions as between-subject factor (experiment 1) and as within-subject factor (experiment 2). Construction and revision followed distinct cognitive principles in the changeless conditions of experiment 1. With increased task demands due to switches between different continuity conditions (experiment 2), reasoners adapted the principles they used for model revisions to the principles which they had used during antecedent constructions. Keywords: spatial reasoning, spatial mental models, belief revision, episodic trace, memory effect Construction of spatial mental models To cope with every-day life, humans in almost all situations have to make decisions on the basis of actual given or mentally stored information. Sometimes we are confronted with situations requiring different reasoning abilities. Imagine the following situation: You are talking to a friend who visited Paris and now tells you about his vacations. He says: “The book store is to the left of the Eiffel Tower” “And the cafe is to the left of the Champ de Mars” Given these information it is not possible to determine how the buildings are related to each other. In particular, the two statements do not yet allow you to continuously arrange the named objects. It is assumed that humans process spatial information of this kind by constructing an integrated mental representation, called “mental model”. Mental models represent what is true, given by the premises and in a more restricted sense how reality could be (Craik, 1943; Johnson-Laird & Byrne, 1991; Goodwin & Johnson-Laird, The common notion is that human reasoning relies on the construction and inspection of mental models. We are inclined to integrate related pieces of information into one model and thus you would, most likely, start to think about the arrangement of the buildings (Johnshon-Laird, 1983; Knauff, Rauh, & Schlieder, 1995; Knauff, Rauh, Schlieder, & Strube, 1998; Johnson-Laird et al., 2004; Goodwin & Johnson-Laird, 2005; Krumnack, Bucher, Nejasmic, Nebel, & Knauff, 2011; Nejasmic, Krumnack, Bucher, & Knauff, 2011). The construction of a mental model is influenced by a number of factors, among them the number of arguments; unambiguousness of the arguments, or the premise order (Ehrlich & Johnson-Laird, 1982; Evans, Newstead, & Byrne, 1993; Knauff et al., 1998; Nejasmic et al., 2011). The premise order determines how information is integrated into a spatial mental model. For instance, premises can be presented in a continuous (Ar 1 B, Br 2 C, Cr 3 D), a semi- continuous (Br 2 C, Cr 3 D, Ar 1 B), or a discontinuous order (Cr 3 D, Ar 1 B, Br 2 C), with A, B, C, and D representing objects and r n the relation (e.g. right of) between them. These different premise orders result into different sequentially information integration and thus different cognitive demand. It is easier (faster and less error-prone) to construct models and draw inferences based on these models, when premises are given in a continuous and semi- continuous as opposed to discontinuous orders, known as the “continuity effect” (Ehrlich & Johnson-Laird, 1982; Knauff et al., 1998; Nejasmic et al., 2011). The continuity effect is usually explained as follows: humans process discontinuous information by constructing one temporary mental model which must be modified in order to match the last piece of information provided by the final premise of the discontinuous order. The construction of the temporary model basically equals the construction of a model based on continuous or semi-continuous information. However, the modification of the temporary model is an additional step which a) requires time and b) provides a source of errors. With a view to our introductory example this means that you construct a temporary model (M1) from the first two statements, with a preferred working direction from left to right (Nejasmic et al., 2011): (M1) Book store – Eiffel Tower – Cafe – Champ de Mars