Updating mental models in predictive reasoning

Abstract

A mental model account for predictive judgement is proposed. According to this view, solving a predictive task involves the foundation ad updating of a mental he work as the relevant data are provided. The final model state determines the subject's predictive outcome. Two experiments examine the temporal course of reasoning in novices (Experiment 1) and experts (Experiment 2). Each task provided in succession a quantitative source (e.g. “Last year, of the vehidts that stopped at the cafe, 80% were cars and 20% were trucks”), a diagnostic source (e.g. “This vehide is mksy”) and a categorical choice (e.g. “What kind of vehicle was it?”). Two factors were manipulated: The order of sources aimed at analysing the contextual dependence of the updating processes, and the between-sources congruence (congruent or incongruent) in order to explore the integrative processes in model building. The pattern of the sources reading times, the choice times and the categorical choices suggested that: (1) the processing of a source depends on the previous one, (2) subjects try to integrste both sources into a single mental framework, and (3) experts and novices have a similar performance, although experts give more importance to the quantitative source in the diagnostic-quantitative order. These results are better accounted for by the mental model framework than both the heuristic and the formal view of prediction.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.