Date of Award
2023
Document Type
Doctoral Thesis
Degree Name
Doctor of Philosophy
Department
Computer Science
First Advisor
Dr Ruairi O'Reilly
Second Advisor
Dr Aoife Johnson
Third Advisor
Dr Áine de Roiste
Abstract
In the field of Psychology, it has long been assumed that one’s personality traits are linked to one’s emotional states. Yet there is a scarce amount of research that has directly quantified the relationships between basic emotional states and Big Five personality traits. Most empirical research has investigated how attributes of emotional states map to a narrow selection of personality traits (Extraversion and Neuroticism) via a single modality (questionnaires). This narrow focus restricts the field’s understanding of the relationship between emotional states and personality traits.
In this research, the Personality Emotion Mapping (PEM) model was developed to map the relationships between basic emotional states and Big Five personality traits and their sub-traits across multiple modalities (questionnaire, visual, audio, and text). PEM was employed in two empirical studies (N1 = 38, N2 = 203), where participants’ basic emotional states were measured using both questionnaire and automatic emotion classification (emotive facial expression analysis, semantic analysis, affective prosody) methodologies. These emotional states were then mapped to Big Five personality traits and their sub-traits using both linear and non-linear approaches.
The results demonstrate that i) Conscientiousness (Anger, Disgust, Fear, Joy, and Sadness) and Agreeableness (Disgust, Fear, Sadness) map to several basic emotional states, ii) Openness to Experience is not linked with any basic emotional state, and iii) that several state-relationships can only be mapped at the sub-trait level. Additionally, the results corroborate previous research findings mapping Extraversion to Joy and Neuroticism to Sadness and Fear. Contrary to expectations, these results do not replicate across visual, audio, and text modalities, due to the poor performance of their respective automatic emotion classification methodologies. This research indicates that basic emotional states are important to understanding and predicting one’s personality traits.
Recommended Citation
Ryan Donovan, Brendan, "Modelling the Relationship Between Personality Traits and Basic Emotions: A Multi-Modal and Affective Computing Approach" (2023). Theses [online].
Available at: https://sword.cit.ie/allthe/822
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Access Level
info:eu-repo/semantics/openAccess
