I’m pleased to share that my paper has now been published in the Journal of Mathematical Psychology:

Structural identifiability of Order-of-Processing models: When do different cognitive architectures generate identical response time distributions

Order-of-processing work figure

This paper asks a simple but important question: when can different Order-of-Processing (OP) architectures produce the same response-time distribution, and when can we recover the underlying parameters uniquely from the observed data?

The study develops a theory of structural identifiability for a restricted path-based class of OP models. Using moment-generating functions, Laplace-transform arguments, and diagram symmetries, the paper shows that the multiset of rate parameters is generically recoverable from the reduced rational transform under stated regularity and no-cancellation conditions. However, the assignment of those rates to specific process labels can fail when the OP diagram has nontrivial permutation symmetries.

The results clarify the inferential scope of OP-model analyses and show when additional constraints or experimentally induced asymmetries are needed before fitted parameters can be interpreted psychologically.

Read the paper here: