Evaluating conditional arguments with uncertain premises

ABSTRACT

Treating conditionals as probabilistic statements has been referred to as a defining feature of the “new paradigm” in cognitive psychology. Doing so is attractive for several reasons, but it complicates the problem of assessing the merits of conditional arguments. We consider several variables that relate to judging the persuasiveness of conditional arguments with uncertain (probabilistic) premises. We also explore ways of judging the consistency of people's beliefs as represented by components of conditional arguments. Experimental results provide evidence that inconsistencies in beliefs are more prevalent if the arguments’ components are spatially and temporally dispersed than if they are contiguous in space and time. This supports the idea that assuring the consistency among even a small number of beliefs is difficult to do, especially if the beliefs in question are not brought into consciousness at the same time; but consistency can be enhanced when beliefs are considered simultaneously or nearly so.

KEYWORDS:

• Conditional reasoning
• consistency
• coherence

Acknowledgments

This study was supported by the Office of Naval Research (Grant N00014-14-1-0529). The authors gratefully acknowledge very helpful comments on this MS from TAR action editor Henry Markovits, reviewer David Over and an anonymous reviewer. We also thank Nikita Shukla for help in running participants and tabulating data. Correspondence concerning this article should be addressed to Raymond Nickerson, 5 Gleason Rd., Bedford, MA 01730. E-mail: [email protected].

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Where Over and Cruz use c and f, we use A and C as reminders that they represent, respectively, the Antecedent and Consequent of the conditional. Also, we use ¬ to represent not.

² Originally, we prepared 80 statements as a good round number. The last four statements referred to Patriots football quarterback “Steve Brady.” After running a few participants it dawned on the first author that he had conflated two of his favourite Patriots quarterbacks, Steve Grogan (of times past) and Tom Brady. Given that the participants were college students in the Boston area, many of whom would perhaps notice the gaffe and wonder what was going on (though no one mentioned it), it seemed best to delete the last four statements, so we did so and proceeded with the remaining 76.

Additional information

Funding

U.S. Office of Naval Research [grant number Grant N00014-14-1-0529].