|
Defining and Testing Context Effects| title | Defining and Testing Context Effects |
|---|
| start_date | 2025/06/04 |
|---|
| schedule | 11h-12h |
|---|
| online | no |
|---|
| location_info | salle R1-R15 |
|---|
| summary | What is “context« and how can we best test for its influence (or the lack thereof)? A principled research program on context effects requires a unity between verbal definitions, their mathematical representations, and the observational opportunities created by the experimental design. The diversity in settings covered by the literature on contextual influences, spanning from low-level foraging tasks in amoeboid organisms to real-world market behavior, goes hand-in-hand with an almost equal diversity of testing approaches designed to assess the presence of context effects. The present work reviews and critically evaluates existing approaches for testing context effects and identifies a mismatch between a naïve verbal definition commonly put forth and the formalisms purported to represent it. We propose a new approach for testing context effects through the lens of random-preference models. We show that the ability of typical experimental designs to detect context effects is very limited primarily due to their reliance on implausible testing standards. We demonstrate how testing power can be bolstered by linking individual choices across different contexts, providing highly diagnostic omnibus tests for context effects. A reanalysis of existing benchmark datasets confirms that context effects occur more frequently than traditional approaches suggest. We also observe a large degree of heterogeneity in the manifestation of context effects, a finding that speaks against a strictly regimented taxonomy comprised of “attraction », “compromise« , and “similarity » effects. Beyond conceptual clarification, the present work establishes a modeling framework that is able to accommodate existing standards for context effects and also enables the development of richer interpretations that include mixtures of cognitive processes. |
|---|
| responsibles | Chassagnon, Apouey |
|---|
Workflow history| from state (1) | to state | comment | date |
|---|
| submitted | published | | 2025/05/20 09:52 UTC |
| |
|