Reading Cognition as Decisions Unfold in Words - A Factorized Inverse Decision Model

Sep 15, 2026

Uploaded by Finn

Inverse decision modeling infers latent properties of decision processes from observed behavior, but action-trajectory-based formulations omit response dynamics present in verbalized cognitive tasks, such as verbal production, interaction, and hesitation. A factorized inverse decision model (FIDM) decomposes each participant’s task-execution likelihood into an action factor and an effort factor with separate individual-specific parameters. From raw verbal transcripts, a language model generates structured execution traces for factorized inference. Experiments on 400 older adults show selective recovery, preserved distinctions under matched summaries, and complementary gains in cognitive-status classification.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Abstract

Introduction

Related Work

Inverse Decision Modeling

arXiv : 2608 .09222v2 [cs .CL] 2 1 Aug 2026 Reading Cognition as Decisions Unfold in Words: A Factorized Inverse Decision Model Jiawen Kang1 Dongrui Han1 Xixin Wu1 Helen Meng1 1The Chinese University of Hong Kong {[jwkang}@se.cuhk.edu.hk](jwkang}@se.cuhk.edu.hk) Abstract Inverse decision modeling infers latent properties of decision processes from observed behavior, but existing formulations rely primarily on action trajectories. In verbalized cognitive tasks, task execution also produces response dynamics that action-only formulations leave unmodeled, such as verbal production, interaction, and hesitation. We propose a factorized inverse decision model (FIDM) that decomposes each individual’s task-execution likelihood into an action factor and an effort factor, governed by separate individual-specific parameters. From raw verbal transcripts, a language model produces structured task-execution traces for factorized inference. On data from 400 older adults performing a groceryshopping dialog task for cognitive screening, controlled recovery shows selective estimation of the intended factors, while matched semi-synthetic conditions show that FIDM preserves action-execution distinctions even when aggregate behavioral summaries are matched. Action evidence further localizes task-defined deviations across participants. In cognitive-status classification, FIDM provides information complementary to clinical scores, trajectory summaries, and frozen language representations, with consistent gains across all evaluated baselines in the binary setting. 1 Introduction Human decision behavior exhibits rich variability: when driving home, one may cut quickly through busy intersections or cruise along a quiet waterfront; a football attack may drive straight at goal or unfold through patient passes and detours. Such variations reflect the underlying cognitive processes of decision-makers, including interpreting goals, forming plans, selecting actions, and organizing their execution. Inferring properties of the underlying decision process from observed behavior is a central problem in fields like cognitive science and behavioral economics [1–3] . In this work, we model decision trajectories from a verbalized cognitive task to infer interpretable participant descriptors and characterize cognitive status. Inverse decision modeling (IDM) offers a unifying, model-based perspective for interpreting observed decision behavior [4, 5] . IDM specifies how a decision process generates behavior and reasons backward from observations to the factors that shaped it. The forward process is typically formulated as a Markov decision process that structures states, actions, and transitions under reinforcement learning or planning models. Different instantiations recover different latent quantities: inverse reinforcement learning recovers reward functions from demonstrated behavior [6, 7], Bayesian inverse planning infers goals or beliefs underlying observed actions [8, 9], and inverse constraint learning identifies constraints that shape feasible behavior [10] . Existing IDM formulations base inference primarily on action trajectories. In cognitive assessment, however, task execution performance reflects both the actions participants select and the response dynamics that accompany each choice. This distinction is especially pronounced when a task requires Preprint. Figure 1: Left: spatial trajectories show that participant P0001 (Health Control, HC) follows a near-shortest route, while participant P0107 (Mild Cognitive Impairment, MCI) backtracks and exits without checkout. Right: transition-aligned verbal output, derived from character counts of route descriptions, shows that P0001 verbalizes concisely, while participant P0161 (MCI) produces elevated verbal effort with self-directed verbalization. coordinating multiple steps toward a goal: two participants may select similar actions yet differ markedly in their fluency, hesitation, a

FAQs

What limitation do existing inverse decision modeling methods have for verbalized tasks?

They primarily rely on action trajectories and do not model response dynamics created during task execution, such as verbal production, interaction, and hesitation.

How does the factorized inverse decision model (FIDM) represent task execution?

FIDM factorizes each participant’s task-execution likelihood into an action factor and an effort factor, each governed by separate individual-specific parameters.

What evidence does FIDM provide for cognitive-status classification?

On the grocery-shopping dialog task data, FIDM yields complementary information to clinical scores, trajectory summaries, and frozen language representations, with consistent gains in the binary setting.

Sep 15, 2026

Uploaded by Finn