Deductive and inductive conditional inferences: Two modes of reasoning

Abstract

A number of single- and dual-process theories provide competing explanations as to how reasoners evaluate conditional arguments. Some of these theories are typically linked to different instructions—namely deductive and inductive instructions. To assess whether responses under both instructions can be explained by a single process, or if they reflect two modes of conditional reasoning, we re-analysed four experiments that used both deductive and inductive instructions for conditional inference tasks. Our re-analysis provided evidence consistent with a single process. In two new experiments we established a double dissociation of deductive and inductive instructions when validity and plausibility of conditional problems were pitted against each other. This indicates that at least two processes contribute to conditional reasoning. We conclude that single-process theories of conditional reasoning cannot explain the observed results. Theories that postulate at least two processes are needed to account for our findings.

Keywords:

• Analytical reasoning
• Conditional reasoning
• Double dissociation
• Probabilistic reasoning
• State-trace analysis

View correction statement:

Corrigendum

Acknowledgments

We would like to thank Sieghard Beller, Jonathan St. B T. Evans, and two anonymous reviewers for comments on an earlier draft of this paper, and the research assistants at our lab for collecting the data. The research reported in this article was supported by Grant Kl 614/31-1 from the Deutsche Forschungsgemeinschaft to Karl Christoph Klauer.

Notes

¹ We use the terms deductive and inductive in the broadest possible sense. Deduction refers to all reasoning that entails the truth of the conclusion given the truth of the premises. Induction refers to all reasoning that does not entail the truth of the conclusion given the premises (this is also termed abductive reasoning).

² The two modes of reasoning are not synonyms for Type 1 and Type 2 processes. Rather, these distinctions are orthogonal to each other (see Rips, 2001). This aspect will be addressed in the General Discussion below.

³ Rips (2001; as well as Heit & Rotello, 2005, see below) used multiple forms of reasoning, including conditional reasoning (i.e., MP arguments). However, as the presented results are averaged over all different forms of reasoning, the specific effect on conditional arguments remains unclear.

⁴ When running the same analysis as in Experiment 1 (i.e., omitting the problems with the neutral conditionals) the 3-way interaction remained significant, F(1, 53) = 6.49, p = .01.

⁵ When running the same analysis as in experiment 1 (i.e. omitting the problems with neutral conditionals) the 3-way interaction was significant, F(1, 53) = 4.84, p = .03.