Memory for mental models of spatial descriptions: an episodic-construction-trace hypothesis

Area of Science:

• Cognitive Psychology
• Human Memory
• Mental Representation

Background:

• Mani and Johnson-Laird's (1982) study proposed that mental models are preserved in episodic memory.
• This theory explained a crossover effect where gist memory favored determinate descriptions and verbatim memory favored indeterminate ones.
• Replication of this crossover effect has been inconsistent, necessitating further investigation into memory for spatial descriptions.

Purpose of the Study:

• To test the replicability of the Mani and Johnson-Laird (1982) crossover effect in memory for spatial descriptions.
• To investigate alternative hypotheses for memory encoding of spatial information beyond simple mental model preservation.
• To explore the role of mental operations in the episodic memory of spatial descriptions.

Main Methods:

• Three experiments were conducted, each involving two phases: diagram-description matching and description recognition.
• Participants judged the congruence between diagrams and verbal descriptions (determinate vs. indeterminate).
• Recognition memory for the studied descriptions was subsequently tested, with variations in description overlap and sentence order.

Main Results:

• All three experiments failed to replicate the previously reported crossover effect.
• Recognition performance was influenced by the overlap in construction-trace between study and test descriptions (Experiments 1 & 2).
• Altering the sentence order between study and test significantly impaired performance (Experiment 3).

Conclusions:

• The findings challenge the hypothesis that mental models are directly preserved in episodic memory.
• An alternative hypothesis posits that episodic memory encodes the mental operations used to construct mental models.
• This construction-trace hypothesis successfully accounts for the observed experimental results, offering a new perspective on spatial description memory.