IISER MOHALI

Abstract: Probability distortions are a cornerstone of behavioral economics' prospect theory, describing the non-linear weighting of probabilities in human decision-making. However, the normative (why?) and mechanistic (how?) underpinnings of these distortions remain largely unexplored.

Our research bridges this gap by linking probability distortions to perceptual illusions in neuroscience. These illusions, where humans misjudge angles due to environmental statistics and neural noise, inspire a computational neuroscience approach. Leveraging principles like the Efficient Coding Principle and the Bayesian Brain Hypothesis, we propose that probability distortions stem from noisy neural computations and the bounded nature of probabilities (0 to 1).

Experimental manipulations of noise and boundaries confirmed our theory, showing that these distortions are predictable and rooted in fundamental neural processes. This provides a novel, normatively grounded explanation for these biases. Additionally, for physics students, speculative connections to statistical physics, including parallels with magnetization, will also be discussed.

We hope to see you all there for what promises to be an engaging and thought-provoking session!