Why Are Parsing Actions for Understanding Message Hierarchies Not Random?

2025-06-27

AI Summary

Human syntactic parsing is non-random, yet random action selection would require a universal symbolic system—contrary to empirical evidence of systematic biases in human parsing strategies. Method: The authors introduce (1) deeply nested synthetic inputs that markedly increase semantic failure rates under random parsing, and (2) a surprisal-weighted term in the objective function to capture human sensitivity to prediction uncertainty. Leveraging an emerging communication framework, they construct a hierarchical biased parser and conduct iterative interactive training and evaluation. Contribution/Results: Experiments show that under distributions approximating natural language, random-parsing agents suffer sharp declines in communication accuracy, whereas non-random, surprisal-guided hierarchical parsers maintain high performance. This provides the first computational modeling evidence that non-random, surprisal-modulated hierarchical parsing is a necessary mechanism for efficient language understanding.

Technology Category

Natural Language Processing: Syntax — Tagging, Chunking & ParsingCognitive Modeling & Cognitive Systems: Simulating Human BehaviorHumans and AI: Human-Aware Planning and Behavior Prediction

Application Category

Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactionsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingUser Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

If humans understood language by randomly selecting parsing actions, it might have been necessary to construct a robust symbolic system capable of being interpreted under any hierarchical structure. However, human parsing strategies do not seem to follow such a random pattern. Why is that the case? In fact, a previous study on emergent communication using models with hierarchical biases have reported that agents adopting random parsing strategies$unicode{x2013}$ones that deviate significantly from human language comprehension$unicode{x2013}$can achieve high communication accuracy. In this study, we investigate this issue by making two simple and natural modifications to the experimental setup: (I) we use more complex inputs that have hierarchical structures, such that random parsing makes semantic interpretation more difficult, and (II) we incorporate a surprisal-related term, which is known to influence the order of words and characters in natural language, into the objective function. With these changes, we evaluate whether agents employing random parsing strategies still maintain high communication accuracy.

Problem

• Investigates why human parsing strategies are not random
• Tests random parsing on complex hierarchical inputs
• Evaluates impact of surprisal on communication accuracy

Innovation

• Complex hierarchical inputs challenge random parsing
• Surprisal term in objective guides parsing order
• Evaluate random parsing accuracy under new conditions