A systematic investigation into the reliability of inter-temporal choice model parameters

Area of Science:

• Behavioral Economics
• Cognitive Psychology
• Decision Science

Background:

• Inter-temporal choice models are widely used to understand decision-making.
• Parameter estimates from these models are often treated as reliable measures of cognitive processes.
• The reliability of these parameter estimates has not been rigorously examined.

Purpose of the Study:

• To assess the reliability of parameter estimates from prominent inter-temporal choice models.
• To identify factors influencing the reliability of these estimates, including model type and experimental design.
• To provide recommendations for improving the reliability of inter-temporal choice models for measurement.

Main Methods:

• Fitting 11 prominent inter-temporal choice models to data from three representative experiments.
• Analyzing the consistency of parameter estimates for individuals across different choice sets.
• Conducting parameter recovery analyses to evaluate model performance.

Main Results:

• Generally low correlations were observed between parameters estimated for the same individual using different choice sets.
• Parameter recovery varied significantly across different models and experimental designs.
• Evidence suggests that many previously reported parameter estimates may be unreliable.

Conclusions:

• Parameter estimates from many inter-temporal choice models lack sufficient reliability for accurate measurement.
• The choice of model and experimental design significantly impacts the reliability of parameter estimates.
• Future research should focus on enhancing the reliability of inter-temporal choice models to ensure valid interpretations.