Reply to Davis-Stober et al.: Violations of rationality in a psychophysical task are not aggregation artifacts.
Humans deviate from rational choice theory if their estimates of the attribute values for one alternative change as a function of the attribute values of competing alternatives (1). In our paper (2), we report such “context-dependent” (CD) deviations from rationality in the form of a frequent winner...
| Main Authors: | , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
National Academy of Sciences
2016
|
| Summary: | Humans deviate from rational choice theory if their estimates of the attribute values for one alternative change as a function of the attribute values of competing alternatives (1). In our paper (2), we report such “context-dependent” (CD) deviations from rationality in the form of a frequent winner (FW) effect and the corresponding weak stochastic transitivity (WST) violations. Davis-Stober et al. (3) claim that this observation is an artifact of aggregating over different trial types, and suggest that a context-independent (CI) decision model might explain our data. We thank Davis-Stober et al. (3) for drawing attention to a potentially important issue. However, we find their analysis to be misleading. Below, we comprehensively rebut their claims and present evidence that corroborates our original findings. |
|---|