Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

WorryWords: Norms of Anxiety Association for over 44k English Words

Saif M. Mohammad

Abstract

Anxiety, the anticipatory unease about a potential negative outcome, is a common and beneficial human emotion. However, there is still much that is not known, such as how anxiety relates to our body and how it manifests in language. This is especially pertinent given the increasing impact of anxiety-related disorders. In this work, we introduce WorryWords, the first large-scale repository of manually derived word--anxiety associations for over 44,450 English words. We show that the anxiety associations are highly reliable. We use WorryWords to study the relationship between anxiety and other emotion constructs, as well as the rate at which children acquire anxiety words with age. Finally, we show that using WorryWords alone, one can accurately track the change of anxiety in streams of text. The lexicon enables a wide variety of anxiety-related research in psychology, NLP, public health, and social sciences. WorryWords (and its translations to over 100 languages) is freely available. http://saifmohammad.com/worrywords.html

WorryWords: Norms of Anxiety Association for over 44k English Words

Abstract

Anxiety, the anticipatory unease about a potential negative outcome, is a common and beneficial human emotion. However, there is still much that is not known, such as how anxiety relates to our body and how it manifests in language. This is especially pertinent given the increasing impact of anxiety-related disorders. In this work, we introduce WorryWords, the first large-scale repository of manually derived word--anxiety associations for over 44,450 English words. We show that the anxiety associations are highly reliable. We use WorryWords to study the relationship between anxiety and other emotion constructs, as well as the rate at which children acquire anxiety words with age. Finally, we show that using WorryWords alone, one can accurately track the change of anxiety in streams of text. The lexicon enables a wide variety of anxiety-related research in psychology, NLP, public health, and social sciences. WorryWords (and its translations to over 100 languages) is freely available. http://saifmohammad.com/worrywords.html

Paper Structure

This paper contains 21 sections, 9 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Obtaining Human Ratings of Anxiety
  4. Reliability of the Annotations
  5. WorryWords: Research & Applications
  6. Using WorryWords to Answer Research Questions about Anxiety
  7. From a lexical standpoint, how is anxiety related to other emotions?
  8. At what rate do children acquire words associated with anxiety? How does this change with the Age of Acquisition?
  9. Capturing Anxiety Arcs in Text Streams using WorryWords
  10. Creating the Anxiety Arcs
  11. Results
  12. Conclusion
  13. Limitations
  14. Ethics Statement
  15. APPENDIX
  16. ...and 6 more sections

Figures (9)

  • Figure 1: Distribution of terms in WorryWords: percentage of terms associated with each class. (The total number is shown in parenthesis.)
  • Figure 2: Percent of WorryWords terms by AoA.
  • Figure 3: For all of the non-neutral WorryWords terms acquired at age x: the percentage corresponding to each of the non-neutral anxiety--calmness bins. (The six percentages for each age sum up to 100%.)
  • Figure 4: Predicted (green) anxiety arcs when applying four different types of association scores to one of the test data text streams: a. anxiety association from WorryWords, b. valence (from NRC VAD), c. arousal (from NRC VAD), and dominance (from NRC VAD). The gold (yellow) arc pertaining to the same text stream is repeated in all four figures for reference. Bin size used to create the arcs: 50 posts. On the right is the average correlation between 1000 pairs of gold and predicted arcs for each case.
  • Figure 5: Questionnaire: Detailed instructions.
  • ...and 4 more figures