guest ::
| Home > Workflow collections > In process > Embodied Emotions in Natural Language |
| Home > Workflow collections > In process > Embodied Emotions in Natural Language |
| Preprint | FZJ-2026-05162 |
; ; ; ; ; ; ;
2026
This record in other databases:
Please use a persistent id in citations: doi:10.31234/osf.io/58f3t_v1
Abstract: Emotions are often felt in the body, and embodiment is a prominent mechanism supporting language processing. Here we utilized computational linguistics to map embodied representations of emotions in an English text corpus. The data revealed consistent embodied patterns for emotion categories. Comparison between semantics-derived and self-reported bodily maps of emotions revealed similar, discrete bodily representations of different emotions in texts and self-reports. Hierarchical clustering revealed three clusters: positive emotions within the head and chest, negative emotions within the torso and legs, and negative social emotions mostly within the head. Replication in a Finnish corpus yielded consistent results. These findings show that the bodily organization of emotional experience is reflected in the statistical structure of language, linking linguistic meaning to the subjective bodily experience of emotion.
|
The record appears in these collections: |
PDF