Quantifying Simplicity: How to Measure Sub-Processes and Bottlenecks of Decision Strategies Using a Cognitive Architecture


Introduction What makes a decision strategy simple or complex? In this project, we investigated the time costs for cognitive sub-processes and bottlenecks of decision strategies. In order to gauge these time costs, we formally implemented two prominent decision strategies as well as a working memory (WM) load manipulation within the cognitive architecture ACT-R (Anderson, 2007) and compared the performance of the strategies under varying degrees of WM load. We tested the simulation results from this analysis in an empirical study. The first tested strategy is tallying (TALLY), which integrates across several attributes to make a decision (Gigerenzer & Goldstein, 1996). The second strategy is the take-the-best heuristic (TTB), which relies on one best attribute (or reason) and considers further attributes only if the decision alternatives do not differ on that reason. As TTB does not integrate across multiple attributes and often searches only part of the information, it is considered a relatively “simple” strategy (Gigerenzer, Todd, & the ABC Research Group, 1999). Using TALLY and TTB as paradigmatic examples of integrative and one-reason decision making strategies, respectively, we evaluated the hypothesis that integrative strategies induce higher cognitive costs than one-reason strategies—as indicated, for instance, by longer response times and lower execution accuracy when set under WM load (cf. Payne, Bettman, & Johnson, 1993).