Journal article
Distinctive Features of Emotion Concepts
01 Jan 2021
Abstract
Some theories of abstract word representation posit a role for affective information; however, little has been done to delineate how this holds for the class of emotion concepts relative to other types of abstract concepts. Work using distributional semantic models has shown that emotion words tend to co-occur with other highly affective words, complicating the picture of how affective information contributes to the grounding of abstract concepts generally and emotion concepts specifically. Here, a novel set of data from 180 participants collected using a feature generation paradigm with emotion, non-emotion-related abstract, and concrete stimuli is leveraged to test how the distributional properties of emotion concepts and their associated features differs from those of other abstract concepts. Using co-occurrence statistics and normative ratings for affect, results show the importance of affect differs between emotion and non-emotion-related abstract words.
Metrics
6 Record Views
Details
- Title
- Distinctive Features of Emotion Concepts
- Creators
- Alexandra E KellyEvangelia G Chrysikou
- Publisher
- eScholarship, University of California
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- College of Arts and Sciences
- Identifiers
- 991020531855004721