Academic journal article Memory & Cognition

Memory for the Process of Constructing an Integrated Mental Model

Academic journal article Memory & Cognition

Memory for the Process of Constructing an Integrated Mental Model

Article excerpt

We report two experiments in which people read descriptions of integrated spatial configurations, together with comparable descriptions that did not describe integrated spatial configurations. The integrated spatial descriptions, but not the comparable descriptions, thus supported the construction of a coherent mental model. In Experiment 1, each sentence of the comparable descriptions described the spatial relation between two objects that were not mentioned elsewhere in the description. In Experiment 2, the comparable descriptions were nonspatial, having been constructed by replacing the spatial relations with nonspatial relations. In both experiments, participants were given a surprise recognition test in which they had to identify each of the original descriptions-of both integrated spatial configurations and nonspatial configurations-among a set of distractors. When the sentences in the original description were reordered (and participants were instructed to ignore sentence order), recognition memory was reliably depressed, but only for the integrated spatial descriptions. Reordering descriptions does not change their prepositional content, nor does it change the described situation; however, it does change the process of constructing a mental model of that situation. These findings thus suggest that memory for the descriptions retains a trace of the process of constructing an integrated mental model and that reordering the sentences disrupts this memory because the reordering reduces the similarity of the processing of the descriptions at recognition.

In previous work on the comprehension and memory of simple spatial and temporal descriptions (Baguley & Payne, 1999, 2000; Payne, 1993), we have defended the suggestion that people comprehend these descriptions by constructing mental models of the premises and proposed that their memory of the descriptions contains a trace of the processing steps of model construction-an episodic construction trace.

This suggestion instantiates a processing view of human memory in a specific cognitive domain-namely, understanding and remembering text. The traditional view of memory for comprehended text distinguishes memory for the text itself (including surface form and propositional content) from memory for the situation described by the text (e.g., Bower & Morrow, 1990; Gernsbacher, 1990; van Dijk and Kintsch, 1983). Our suggestion goes beyond this in proposing that the episodic record of the process of constructing a situation model is also remembered.

Our claim also relates to the important transferappropriate processing tradition (e.g., Morris, Bransford, & Franks, 1977). The essential tenet of this tradition is that the processes that are performed during encoding influence memory in more particular ways than a correlation between "depth" and robustness of the memory (see Craik & Lockhart, 1972). The essential tenet, however, can be interpreted in relatively weaker or stronger versions of the role of mental processes in memory. A weak version would contend that all memory traces have multiple features, and that the processes at encoding merely emphasize some of these features at the expense of others. For example, judging the semantics of a word will emphasize the semantic features in its trace, whereas judging it for rhyme will instead emphasize the trace's phonetic features. At the other extreme, a strong version would contend that the memory trace is simply the process of encoding (e.g., Crowder, 1993; Kolers, 1973; Lansdale, 2005). Under this hypothesis, recognition would result from the awareness that a mental process or parts of that process were being repeated, and thus recognition would not require any explicit comparison. A fundamental distinction between weaker and stronger versions of transfer-appropriate processing is made between theories that propose that merely the end product of processing is retained and theories that propose that intermediate processing is also retained. …

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

Oops!

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.