We have argued that the evolved architecture of the human mind includes a number of functionally distinct, domain-specific reasoning mechanisms. Each is specialized for reasoning about a different, adaptively important domain of human life, such as social exchange (Cosmides, 1985, 1989; Cosmides & Tooby, 1989, 1992) and hazardous situations (Cosmides & Tooby, 1997; Fiddick, 1998).

Some, though not all, of the data on which these claims were based was derived using a test of conditional reasoning called the Wason selection task.

To show that a cognitive mechanism is an adaptation, one needs to show that it has design features that solve an adaptive problem with special efficiency, AND that it is not more parsimoniously explained as a byproduct of some other, more general mechanism. Recently, Sperber, Girotto, & Cara (1995) claimed that all reasoning results derived from the Wason selection task - including ours - could be more parsimoniously explained using relevance theory. Indeed, their article was entitled, "Relevance theory explains the selection task".

Such claims deserve to be taken seriously. After examining relevance theory and conducting some experiments, we found that we did not agree that relevance theory explains the selection task, so we wrote "No Interpretation without representation: The role of domain-specific representations and inferences in the Wason selection task" (Fiddick, Cosmides, & Tooby, in press).

In addition to evaluating relevance theory, this paper also summarizes the data that speak against other alternative theories that have been put forth during the last decade.

Below is a Table of Contents for this paper. To read a preprint of it, CLICK HERE. For reasons of copyright, we will have to take this preprint off of the web once the paper comes out in Cognition.

Laurence Fiddick, Leda Cosmides, & John Tooby

INTRODUCTION

The Wason Selection Task
Relevance Theory and the Selection Task
Choosing the not-Q card
Ecological rationality and two applications: Social contract theory and hazard management theory
   Social Contract Theory
   Hazard Management (Precaution) Theory
Strategies for evaluating competing theories
Theories differ in their scope of application: The case of the missing content effects
   Dissociations and selective deficits
   Manipulations that regulate content effects
   Predictions about conflicting mechanisms: The principle of pre-emptive specificity
Relevant with respect to which representations? Areas of controversy, agreement, and complementarity

EXPERIMENT 1: ARE LOGICAL CONNECTIVES NECESSARY?

Method
Predictions
   Social Contract Theory
   Relevance Theory
Results and Discussion
   Relevance theory does not explain perspective shifts
   Relevance theory makes different predictions for social laws and for trades.
Conclusions from Experiment 1

EXPERIMENT 2: ARE DENIALS NECESSARY?
   Using pragmatics to block inference (c).
   Varying the testing strategy
Method
Predictions
   Relevance Theory
   Hazard Management Theory
Results and Discussion
   Implications
   Do the instructions for the precaution version highlight violations?
   A parallel case with social contracts

EXPERIMENT 3: ONE NOTION OF VIOLATION, OR MANY?
   The problem with logic
   Different kinds of violations
Method
Predictions
   Social Contract Theory and Hazard Management Theory
   Relevance Theory
Results and Discussion
   Implications
   Can Relevance Theory explain the choice of "not-P & Q" in the Privilege condition?

GENERAL DISCUSSION

Social contract and hazard management algorithms: some new design features
Common ground between ecological rationality and relevance theory: Implications for communication and culture
   What's mutually manifest? Pre-emption and the origin of background assumptions.
The uses of the Wason selection task
   A challenge for RT
   The benefits of using the selection task

CONCLUSION

APPENDIX 1. About two misreadings of social contract theory

APPENDIX 2. Instructions and RT_Wason