AI-enhanced semantic feature norms for 786 concepts

TL;DR

This paper presents NOVA, an AI-augmented semantic feature norm dataset that combines human and large language model responses, significantly improving coverage and predictive power for human semantic similarity judgments.

Contribution

The study introduces NOVA, a novel dataset that enhances traditional human semantic feature norms with AI-generated responses validated against human judgments, advancing cognitive science research tools.

Findings

NOVA has higher feature density and concept overlap.

NOVA outperforms human-only norms and word-embedding models in predicting semantic similarity.

Human conceptual knowledge is richer than previous norm datasets.

Abstract

Semantic feature norms have been foundational in the study of human conceptual knowledge, yet traditional methods face trade-offs between concept/feature coverage and verifiability of quality due to the labor-intensive nature of norming studies. Here, we introduce a novel approach that augments a dataset of human-generated feature norms with responses from large language models (LLMs) while verifying the quality of norms against reliable human judgments. We find that our AI-enhanced feature norm dataset, NOVA: Norms Optimized Via AI, shows much higher feature density and overlap among concepts while outperforming a comparable human-only norm dataset and word-embedding models in predicting people's semantic similarity judgments. Taken together, we demonstrate that human conceptual knowledge is richer than captured in previous norm datasets and show that, with proper validation, LLMs can serve as powerful tools for cognitive science research.