TaskSense: Cognitive Chain Modeling and Difficulty Estimation for GUI Tasks

TL;DR

This paper introduces Cognitive Chain, a framework that models GUI task difficulty based on cognitive steps, providing a more accurate measure than motor actions and revealing agent capability gaps.

Contribution

The paper presents a novel cognitive chain model and an LLM-based method to automatically extract cognitive processes from GUI tasks, improving difficulty estimation.

Findings

Cognitive difficulty correlates with user completion time (R^2=0.46).

State-of-the-art GUI agents struggle with cognitively demanding tasks.

The framework enables better assessment of agent capabilities and human-AI interaction.

Abstract

Measuring GUI task difficulty is crucial for user behavior analysis and agent capability evaluation. Yet, existing benchmarks typically quantify difficulty based on motor actions (e.g., step counts), overlooking the cognitive demands underlying task completion. In this work, we propose Cognitive Chain, a novel framework that models task difficulty from a cognitive perspective. A cognitive chain decomposes the cognitive processes preceding a motor action into a sequence of cognitive steps (e.g., finding, deciding, computing), each with a difficulty index grounded in information theories. We develop an LLM-based method to automatically extract cognitive chains from task execution traces. Validation with linear regression shows that our estimated cognitive difficulty correlates well with user completion time (step-level R-square=0.46 after annotation). Assessment of state-of-the-art GUI agents shows reduced success on cognitively demanding tasks, revealing capability gaps and Human-AI consistency patterns. We conclude by discussing potential applications in agent training, capability assessment, and human-agent delegation optimization.