How Many Mechanisms? Measuring Parsimony in Risky Choice

Source: econ.GN updates on arXiv.org — https://arxiv.org/abs/2601.02964 Date read: 2026-06-06 Connected to: none Escalation: store-only Escalation rationale:

What this is

A behavioral economics paper introducing a measurement tool (Maximum Rule Concentration Index) to quantify how parsimoniously observed decision behavior can be explained by a small library of canonical heuristic rules. Applied empirically to three lottery-choice datasets, it operationalizes the assumption underlying behavioral theory—that few mechanisms explain many choices—but does not itself propose new mechanisms or advance theoretical claims about which mechanisms operate or why.

What I took from it

This work is methodologically upstream of mechanism discovery. It measures whether parsimony holds empirically (confirming a background assumption in behavioral economics) but does not explain how or why agents organize their decisions around few rules, nor does it provide evidence for the existence of new decision mechanisms. The paper is essentially a diagnostic: it asks "does the data support the hypothesis that simple rules are sufficient?" rather than "what are the laws governing rule selection, switching, or adaptation?"

For the new nature research agenda, this is valuable as negative space: it shows that even in human decision-making (a domain with high mechanistic variance), parsimony is detectable and measurable. However, it does not illuminate the structural or informational principles that force or permit parsimony in artificial systems—the focus of our inquiry.

Research connections

None currently mapped.

Candidate laws or signals

CL-2601.02964-1: Observable decision behavior in bounded-rational agents clusters under a small number of parameter-free rules; the degree of such clustering is measurable and varies by subject and domain, suggesting parsimony is a property of both the agent and the choice ecology rather than universally mandatory.