Not all logic is loud; some choices whisper, and I listen.
Postdoctoral Researcher, Aix-Marseille School of Economics (AMSE)
Email: pegah.RAHMANI@univ-amu.fr | Download CV
I am a postdoctoral researcher at the Aix-Marseille School of Economics (AMSE). I received my PhD in Economics from the University of Toronto in 2026, under the supervision of Professor Yoram Halevy.
My research focuses on decision theory and behavioral economics, particularly on decision-making under uncertainty and context-dependent preferences. I study correlation-sensitive and incomplete preferences, as well as context-dependent choice, using both theoretical and experimental approaches.
Interests: Decision Theory, Microeconomic Theory, Behavioral Economics, Experimental Economics
This paper provides an axiomatic characterization of incomplete correlation-sensitive preferences, allowing evaluations to depend on the joint distribution of outcomes across alternatives without requiring every comparison to be resolved. By relaxing Completeness while retaining Strong Independence and introducing conditions accommodating incomparability, the framework extends complete correlation-sensitive representations. It also offers an alternative interpretation of observed violations of the Sure-Thing Principle: although the underlying preference relation is invariant to changes in common consequences, such changes may influence the context-dependent resolution of incomparable alternatives, potentially generating observed choice reversals. Finally, the paper shows that even when Transitivity is restricted to jointly compatible comparisons, it forces comparison functions to be additively separable, eliminating correlation sensitivity and reducing the representation to expected multi-utility.
A well-known phenomenon in the decision science literature (Loomes and Sugden 1982, Bell 1982, and Fishburn 1982) is that anticipated regret affects choices and valuations. We analyze Kahneman and Tversky’s (1979) famous decision problem of the certainty effect – a special case of the common ratio effect ´a la Allais (1953) as well as extensively documented probability insensitivity in mid-ranges. We propose that these phenomena are, in fact, manifestations of anticipated regret; offer a behavioral definition of anticipated regret without committing to a specific functional representation; and document evidence of anticipated regret in a controlled lab setting. We find that more than half of our participants exhibit strict Certainty Effect, and about two-fifths of them exhibit aversion to anticipated regret.
Rank-based salience has been proposed as a general mechanism for context-dependent choice, distorting decision weights according to outcome comparisons across states. Because the model allows a wide range of functional forms for salience, it is flexible enough to fit many observed behaviors. We study its predictions in risky choice by introducing dominated distractors into binary lotteries. Despite its generality, the model yields a sharp restriction: if any menu-induced reversal occurs, it must be from the safer lottery to the riskier one. Our experimental findings contradict this prediction. In several cases, subjects instead shift toward the safer option when a dominated distractor is added. This shows that even in its most general formulation, rank-based salience cannot capture the full range of attraction effects in risky choice, highlighting a structural limitation of the model.
Course Instructor:
Course materials are available upon request.
TA:
Phone: +33 6 52 67 52 43
Email: pegah.RAHMANI@univ-amu.fr
Office:
2.53, Aix-Marseille School of Economics (AMSE)
Aix-Marseille Université · Marseille, France
Links: TEEL @ UofT | LinkedIn