Corrugator activity reflects integrated simulation and multifaceted evaluation when processing affective language: Frowning responses to reading about emotions of morally good and bad characters
Open Access
- 7 August 2026
- journal article
- Published by Springer Nature in Cognitive, Affective, & Behavioral Neuroscience
- Vol. 26 (5) , 2118-2138
Abstract
Corrugator supercilii (“frowning muscle”) activity tracks affective valence of linguistic stimuli. According to embodied cognition accounts, this reflects simulation of the emotion at lexical and/or situational levels. Additionally, it may indicate evaluation, i.e., how we feel about an affective event within a context. This study used facial electromyography (fEMG) to examine corrugator responses as participants read short stories. Each story included segments, describing some character’s morally good or bad behavior, a behavior rating, and sentence-initial phrases describing the character’s positive or negative affective states (e.g., “Mark is frustrated”), followed by the reason for this. Our prior studies showed substantially more frowning when good characters were “frustrated” compared with “happy” but minimal differences for bad characters, suggesting a multiple-drivers account where both simulation and evaluation influence responses. Crucially, they can counteract in the context of bad characters, e.g., “happy” has positive valence but is unfair in context and the opposite applies to “frustrated.” To emphasize characters’ moral status and amplify fairness-based evaluations, an embedded morality judgment rating task was introduced. We hypothesized stronger frowning for “happy” than “frustrated” bad characters due to fairness considerations now outweighing simulation. However, results revealed increased frowning for “frustrated” bad characters. The rating task might have unexpectedly reduced, rather than boosted, evaluative processing through cognitive interference. Nevertheless, previously observed frowning patterns for moral characters were replicated. These findings support a multiple-drivers model but suggest that evaluative processes are more complex and multifaceted than current models capture, calling for revisions to incorporate these complexities into affective processing theories.Keywords
All Related Versions
Funding Information
- European Research Council (101088098 - MEMLANG)
This publication has 49 references indexed in Scilit:
- Using Facial EMG to Track Emotion During Language Comprehension: Past, Present, and Future2023
- Electromyographic (EMG) Responses of Facial Muscles During Language Processing2023
- Enjoying vs. smiling: Facial muscular activation in response to emotional languageBiological Psychology, 2016
- Examining subjective and physiological responses to norm violation using text‐based vignettesInternational Journal of Psychology, 2016
- Action-based language: A theory of language acquisition, comprehension, and productionCortex, 2012
- Repeatability of facial electromyography (EMG) activity over corrugator supercilii and zygomaticus major on differentiating various emotionsJournal of Ambient Intelligence and Humanized Computing, 2011
- Grounding emotion in situated conceptualizationNeuropsychologia, 2011
- Examples of mixed-effects modeling with crossed random effects and with binomial dataJournal of Memory and Language, 2008
- On multi-level modeling of data from repeated measures designs: a tutorialSpeech Communication, 2004
- The parsimony principle applied to multivariate calibrationAnalytica Chimica Acta, 1993