Categorical Judgment with a Variable Decision Rule

TL;DR

This paper introduces a new Thurstonian rating scale model with a variable decision rule that combines multiple decision strategies, improving data fit over classical models and requiring extensive data for accurate parameter recovery.

Contribution

The paper presents a novel VDR model that integrates three decision rules, validated through simulations and reanalyses, enhancing the understanding of categorical judgment processes.

Findings

VDR model fits simulated data tightly.

Larger datasets improve parameter recovery.

VDR outperforms classical signal detection models.

Abstract

A new Thurstonian rating scale model uses a variable decision rule (VDR) that incorporates three previously formulated, distinct decision rules. The model includes probabilities for choosing each rule, along with Gaussian representation and criterion densities. Numerical optimisation techniques were validated through demonstrating that the model fits simulated data tightly. For simulations with 400 trials per stimulus (tps), useful information emerged about the generating parameters. However, larger experiments (e.g. 4000 tps) proved desirable for better recovery of generating parameters and to support trustworthy choices between competing models by the Akaike Information Criterion. In reanalyses of experiments by others, the VDR model explained most of the data better than did classical signal detection theory models.