Chater, N., & Oaksford, M. (1999). Ten years of the rational analysis of cognition. Trends In Cognitive Science, 3(2), 57–65. doi:10.1016/S1364-6613(98)01273-X

Summary

Chater and Oaksford re-summarize the definition of rational analysis, and describe a two areas where rational analysis has proved to be fruitful:

• Memory
  • which items are remembered better is a function of the cost of retrieving a memory, the utility of the goal, and the probability that the memory will be relevant.
• Reasoning (specifically, Wason selection task)
  • to verify “if p then q”, participants should choose p and not-q, but they choose p and q
  • assume “most properties and events are rare”
  • then, the pattern of expected informativeness EI(p) > EI(q) > EI(not-q) > EI(not-p)

Methods

n/a

Algorithm

n/a

Takeaways

Interestingly, they claim to analyze a whole decade of research, but it doesn’t seem like the research they discuss is that extensive (especially since the memory example is also discussed by Anderson in 1991). If anything, based on their article it seems as if rational analysis was largely ignored in the 90s.

However, the examples are good ones, and do illustrate the way in which rational analysis can be very helpful at explaining otherwise seemingly illogical behaviors (e.g. in the Wason selection task).