École d'été | 27 mai 2026 : Probabilistic theories, dilemma inferences par David Over

Value of the Information & Strength of the Argument

The talk provides a valuable overview of current debates in the psychology of reasoning, clearly articulating the Bayesian position and its advantages. The argument is well-structured, using intuitive examples to illustrate theoretical points. Over effectively critiques inferentialism by highlighting its problems with independence conditionals and walrus conditionals, and he supports his claims with references to experimental studies. However, the argument is primarily theoretical and relies on the speaker’s interpretation of the literature; it does not present new experimental data. The discussion of dilemma inferences is insightful, showing how the Bayesian framework handles them naturally.

Scientific Rigor, Source Quality, Title Accuracy

The presentation demonstrates scientific rigor by referencing key works in the field (e.g., Ramsey, de Finetti, Edgington, Johnson-Laird, Skovgaard-Olsen, Lassiter). The sources are relevant and up-to-date, including a 2022 paper by Johnson-Laird. The title accurately reflects the content, which focuses on probabilistic theories and dilemma inferences. The talk is well-organized and the speaker is transparent about his own theoretical commitments. No comments were provided for analysis.

The title accurately reflects the content, which focuses on probabilistic theories, dilemma inferences, and decision-making.

8/10

Presentation by a recognized expert in the psychology of reasoning, with references to established theories and recent experiments. The argument is coherent and acknowledges limitations, but relies on personal interpretation and ongoing debates.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 02:00Introduction and overview of the talk: new paradigm in psychology of reasoning, probability of conditionals, and the plan to discuss dilemma inferences.
• 05:00Introduction of Humpty Dumpty examples: dependence and independence conditionals, and the dilemma inference.
• 10:00Discussion of the material conditional hypothesis and its problems, leading to the suppositional conditional.
• 15:00Explanation of the Bayesian approach: reasoning from degrees of belief, not assumptions, with the pregnancy test example.
• 20:00Introduction of inferentialism and its requirement that A raises the probability of C; critique of this view.
• 25:00Discussion of independence conditionals and their importance, including concessive conditionals like 'even if'.
• 30:00Introduction of 'walrus conditionals' (pragmatically odd conditionals) and Edgington's prediction about them.
• 35:00Review of experiments on walrus conditionals, highlighting confounds and the need for unconfounded designs.
• 40:00Discussion of jabberwocky conditionals and the distinction between semantic and pragmatic nonsense.
• 45:00Conclusion: Bayesian approach provides the best account of reasoning and decision-making, especially for dilemma inferences.

• Johnson-Laird, P. N., & Ruth, M. (1991). Deduction. — Mentioned as proposing the material conditional hypothesis in traditional psychology of reasoning.
• Ramsey, F. P. (1926). Truth and probability. — Foundational work for the suppositional conditional and conditional probability.
• de Finetti, B. (1937). La prévision: ses lois logiques, ses sources subjectives. — Another foundational source for the suppositional conditional.
• Douven, I., & Verbrugge, S. (2010). The Adams family of conditionals. — Referenced as a summary of inferentialist positions.
• Skovgaard-Olsen, N., Singmann, H., & Klauer, K. C. (2016). The relevance effect and conditionals. — Experiments on conditionals with irrelevant antecedents, supporting the conditional probability hypothesis but also showing low acceptability for walrus conditionals.
• Lassiter, D. (2020). Pragmatic discourse coherence and the interpretation of conditionals. — Linguistic analysis explaining why walrus conditionals are pragmatically odd, using discourse coherence.
• Johnson-Laird, P. N., et al. (2022). The probability of conditionals: A review. — Recent paper where Johnson-Laird revises his view and accepts the conditional probability hypothesis.

• Edgington, D. (1995). On conditionals. — Classic paper supporting the suppositional conditional and predicting problems with walrus conditionals.
• Evans, J. St. B. T., & Over, D. E. (2004). If. — Book by the speaker and colleague, presenting the suppositional conditional and Bayesian approach.

• Douven, I. (2016). The epistemology of indicative conditionals.