Capturing Human Cognitive Styles with Language: Towards an Experimental Evaluation Paradigm
This work addresses the validation gap in NLP for cognitive modeling by providing a more rigorous evaluation paradigm, though it is incremental as it builds on existing behavioral science methods.
The paper tackles the problem of evaluating language-based cognitive style models by introducing an experiment-based framework that compares linguistic features to human behavior in decision-making tasks, achieving moderate-to-high prediction accuracy (AUC ~ 0.8).
While NLP models often seek to capture cognitive states via language, the validity of predicted states is determined by comparing them to annotations created without access the cognitive states of the authors. In behavioral sciences, cognitive states are instead measured via experiments. Here, we introduce an experiment-based framework for evaluating language-based cognitive style models against human behavior. We explore the phenomenon of decision making, and its relationship to the linguistic style of an individual talking about a recent decision they made. The participants then follow a classical decision-making experiment that captures their cognitive style, determined by how preferences change during a decision exercise. We find that language features, intended to capture cognitive style, can predict participants' decision style with moderate-to-high accuracy (AUC ~ 0.8), demonstrating that cognitive style can be partly captured and revealed by discourse patterns.