Breaking Bad: Norms for Valence, Arousal, and Dominance for over 10k English Multiword Expressions

2025-11-24

AI Summary

Existing sentiment lexicons (e.g., NRC VAD) provide only word-level Valence, Arousal, and Dominance (VAD) scores, lacking systematic affective annotations for English multi-word expressions (MWEs) and their constituent words—limiting phrase-level sentiment analysis and cross-disciplinary research. To address this, we construct the first large-scale, human-annotated MWE VAD lexicon, comprising 10,000 MWEs and 25,000 words, with high-reliability ratings collected via psycholinguistic protocols. Structural validity is confirmed through factor analysis. NRC VAD v2—the resulting resource—introduces the first fine-grained, three-dimensional affective quantification of MWEs, uncovering patterns in emotional intensity and compositionality. It substantially extends the granularity of sentiment lexicons beyond the lexical level, enabling precise phrase-level affective modeling. As a publicly available, high-fidelity, phrase-level affective resource, NRC VAD v2 advances research in NLP, psychology, and cognitive science.

Technology Category

• Natural Language Processing: Lexical Semantics and Morphology
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Application Domains: Humanities & Computational Social Science

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web

Abstract

Factor analysis studies have shown that the primary dimensions of word meaning are Valence (V), Arousal (A), and Dominance (D). Existing lexicons such as the NRC VAD Lexicon, published in 2018, include VAD association ratings for words. Here, we present a complement to it, which has human ratings of valence, arousal, and dominance for 10k English Multiword Expressions (MWEs) and their constituent words. We also increase the coverage of unigrams, especially words that have become more common since 2018. In all, the new NRC VAD Lexicon v2 now has entries for 10k MWEs and 25k words, in addition to the entries in v1. We show that the associations are highly reliable. We use the lexicon to examine emotional characteristics of MWEs, including: 1. The degree to which MWEs (idioms, noun compounds, and verb particle constructions) exhibit strong emotionality; 2. The degree of emotional compositionality in MWEs. The lexicon enables a wide variety of research in NLP, Psychology, Public Health, Digital Humanities, and Social Sciences. The NRC VAD Lexicon v2 is freely available through the project webpage: http://saifmohammad.com/WebPages/nrc-vad.html

Problem

• Creating emotional ratings for 10k English multiword expressions
• Expanding lexicon coverage with 25k additional unigram entries
• Analyzing emotional characteristics and compositionality of MWEs

Innovation

• Expanded lexicon with 10k multiword expression ratings
• Added 25k new unigram entries for broader coverage
• Provided reliable valence arousal dominance human annotations