Putting the Probability Heuristics Model to the Test

AbstractIn the last decades there was a shift from more logically inspired theories describing human reasoning towards the new paradigm of probabilistic approaches. One of the most prominent models for syllogistic reasoning is the Probability Heuristics Model (PHM) which has been formulated based on five heuristics. The contribution of this article is: (i) to provide an analysis of different formalizations of the PHM, (ii) to examine the impact of each heuristic, and (iii) to identify possible violations of underlying assumptions in present implementations. A systematic analysis of the model parameters shows a surprising variation of parameter values across experiments. A bayesian modeling approach is used to explain this variation. Implications for probabilistic approaches are discussed.

Return to previous page