Parametric Order Constraints in Multinomial Processing Tree Models: An Extension of Knapp and Batchelder (2004)

TL;DR

This paper extends multinomial processing tree (MPT) models to incorporate order constraints on parameters, providing a method for order-restricted inference and demonstrating its application to confidence rating data.

Contribution

It introduces a way to construct order-constrained MPT models that are statistically equivalent to unconstrained models, extending previous mathematical analysis.

Findings

Order constraints can be incorporated into MPT models without loss of generality.

The method is illustrated with an analysis of confidence rating data.

The approach extends existing order-restricted inference techniques.

Abstract

Multinomial processing tree (MPT) models are tools for disentangling the contributions of latent cognitive processes in a given experimental paradigm. The present note analyzes MPT models subject to order constraints on subsets of its parameters. The constraints that we consider frequently arise in cases where the response categories are ordered in some sense such as in confidence-rating data, Likert scale data, where graded guessing tendencies or response biases are created via base-rate or payoff manipulations, in the analysis of contingency tables with order constraints, and in many other cases. We show how to construct an MPT model without order constraints that is statistically equivalent to the MPT model with order constraints. This new closure result extends the mathematical analysis of the MPT class, and it offers an approach to order-restricted inference that extends the approaches discussed by Knapp and Batchelder (2004). The usefulness of the method is illustrated by means of an analysis of an order-constrained version of the two-high-threshold model for confidence ratings.