Emotion Recognition based on Psychological Components in Guided Narratives for Emotion Regulation

Patrick Paroubek

Abstract:

Emotion regulation is a crucial element in deal-
ing with emotional events and has positive ef-
fects on mental health. This paper aims to pro-
vide a more comprehensive understanding of
emotional events by introducing a new French
corpus of emotional narratives collected using
a questionnaire for emotion regulation. We
follow the theoretical framework of the Com-
ponent Process Model which considers emo-
tions as dynamic processes composed of four
interrelated components (BEHAVIOR, FEELING,
THINKING and TERRITORY). Each narrative is
related to a discrete emotion and is structured
based on all emotion components by the writ-
ers. We study the interaction of components
and their impact on emotion classification with
machine learning methods and pre-trained lan-
guage models. Our results show that each com-
ponent improves prediction performance, and
that the best results are achieved by jointly con-
sidering all components. Our results also show
the effectiveness of pre-trained language mod-
els in predicting discrete emotion from certain
components, which reveal differences in how
emotion components are expressed.