.png){kind=icon}
.png){kind=icon}
Return
Measuring utility with diffusion models
.png){kind=icon}
OA
AI
DOI：10.1126/sciadv.adf1665.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The drift diffusion model (DDM) is a prominent account of how people make decisions. Many of these decisions involve comparing two alternatives based on differences of perceived stimulus magnitudes, such as economic values. Here, we propose a consistent estimator for the parameters of a DDM in such cases. This estimator allows us to derive decision thresholds, drift rates, and subjective percepts (i.e., utilities in economic choice) directly from the experimental data. This eliminates the need to measure these values separately or to assume specific functional forms for them. Our method also allows one to predict drift rates for comparisons that did not occur in the dataset. We apply the method to two datasets, one comparing probabilities of earning a fixed reward and one comparing objects of variable reward value. Our analysis indicates that both datasets conform well to the DDM. We find that utilities are linear in probability and slightly convex in reward.
Keywords:
SEQUENTIAL SAMPLING MODELS
DECISION-MAKING
RESPONSE-TIME
PROBABILITY
ACCURACY
CHOICE
SPEED
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
12.5
Papers:
2.0W
Citations:
18.1W
Organization
View more organizations