Measurement of Negativity Bias in Personal Narratives Using Corpus-Based Emotion Dictionaries

• Published in: Journal of psycholinguistic research Vol. 40; no. 2; pp. 119 - 135
• Main Author: Cohen, Shuki J.
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01-04-2011; Springer; Springer Nature B.V
• Subjects: Adolescent; Affective Behavior; Behavioral Science and Psychology; Bias; Biological and medical sciences; Cognitive Psychology; Cognitive Style; College Students; Construction (Process); Content analysis; Depression (Psychology); Dictionaries; Emotional Response; Emotions; Epistemology; Female; Fundamental and applied biological sciences. Psychology; Grammar; Humans; Language; Male; Mental health; Mental Recall; Methodology. Experimentation; Modeling (Psychology); Narratives; Production and perception of spoken language; Psycholinguistics; Psychology; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Psychometrics; Psychometrics. Statistics. Methodology; Psychotherapy; Questionnaires; Research Assistants; Resistance (Psychology); Speech; Students; Transcripts (Written Records); Undergraduate Students; Verbal Stimuli; Vocabulary; Young Adult; Personal narratives; Text-analysis; Corpus linguistics; Negativity bias; Dysphoria; Human; Affect affectivity; Methodology; Cognition; Emotion emotionality; Dysphoria Negativity bias; Language; Narrative; Discourse analysis; Measurement method; Cognitive bias
• ISSN: 0090-6905; 1573-6555
• DOI: 10.1007/s10936-010-9158-7

This study presents a novel methodology for the measurement of negativity bias using positive and negative dictionaries of emotion words applied to autobiographical narratives. At odds with the cognitive theory of mood dysregulation, previous text-analytical studies have failed to find significant correlation between emotion dictionaries and negative affectivity or dysphoria. In the present study, an a priori list dictionary of emotion words was refined based on the actual use of these words in personal narratives collected from close to 500 college students. Half of the corpus was used to construct, via concordance analysis, the grammatical structures associated with the words in their emotional sense. The second half of the corpus served as a validation corpus. The resulting dictionary ignores words that are not used in their intended emotional sense, including negated emotions, homophones, frozen idioms etc. Correlations of the resulting corpus-based negative and positive emotion dictionaries with self-report measures of negative affectivity were in the expected direction, and were statistically significant, with medium effect size. The potential use of these dictionaries as implicit measures of negativity bias and in the analysis of psychotherapy transcripts is discussed.