Chapter 14 - Reasoning and hypothesis testing
Chapter Summary
Simulations of Key Experiments
Case Studies
Exploring dual-system theories of deductive reasoning
Dual-system, or dual-process, theories of reasoning hold that there is a distinction between two types of processes:
- System 1 is an intuitive, tacit or implicit process.
- System 2 is a rule-based, analytic or explicit process.
Since System 2 is by definition a slower system than System 1, it follows that, if System 2 can be inhibited, then we should be able to observe the effects of reasoning that are under the control of System 1.
In Experiment 1, Schroyens et al. (2003) attempted to manipulate such control by getting participants in one group to make quick decisions about the validity of a series of conclusions that followed pairs of premises. The problems were presented in the forms of affirmation of the consequent, modus ponens, modus tollens and denial of the antecedent (see Eysenck &Keane, 2015, p. 596).
One prediction would be that participants forced to make the quick decisions would have a bias to affirm most of the conclusions, but this did not occur. While participants tended to make more errors in the speeded condition, they did so more with conclusions that were invalid than with conclusions that were valid. This shows that, when System 2 is inhibited, participants are less able to reject invalid inferences. The smaller number of endorsements of the conclusions in the invalid problems compared to the valid problems shows that a good deal of reasoning was being carried out in the speeded group. The experiment confirms that System 1 is an important component in human reasoning.
In Experiment 2, the aim was to encourage participants to engage in System 2-type processing – that is, to attempt to employ explicit strategies of validation. This was achieved by informing one group of participants about how conclusions are evaluated in deductive reasoning, and that external factors that are irrelevant to the logic of the premises must be ignored. For example, in the literature it is known that participants can be influenced by the content of a premise even when it is irrelevant to the logic of the problem (so-called belief biases). A second group of participants were merely instructed to determine the truth or falsity of a conclusion given two premises.
The results showed an effect of instructions. The informed group (those informed about deductive reasoning) showed an increase in the accuracy of their ability to determine invalid inferences when compared with the uninformed group. This means that encouraging System 2 reasoning enables participants to reject invalid inferences.
The results provide support for the existence of two independent systems in deductive reasoning. Taken together, the results also give some indication of how the two systems operate. When there is a time constraint, participants are forced to use System 1 reasoning, a more implicit form than that of System 2, and hence are less well equipped to reject invalid inferences. When System 2 reasoning is encouraged, participants have more time to engage in a search for counterexamples, and hence are better at rejecting invalid conclusions. In sum, theories of deductive reasoning need to account for the existence of these two systems.
Reference
Schroyens, W., Schaeken, W. & Handley, S. (2003). In search of counter-examples: Deductive rationality in human reasoning. The Quarterly Journal of Experimental Psychology, 56A(7): 1129–45.