Skip to main content

Berkout. 2023

APA Citation

Berkout, O.V. (2023). Assessing psychological inflexibility in text: An examination of the inflexitext program. Journal of Contextual Behavioral Science, 28, 10-17. https://doi.org/10.1016/j.jcbs.2023.03.001

 

Publication Topic
ACT: Empirical
Publication Type
Article
Language
English
Keyword(s)
Natural language processing, Psychological inflexibility, Digital phenotyping, Text data
Abstract

We generate a large amount of text through interactions with technology. Identifying digital markers of psychological constructs in this data may provide novel opportunities for research, assessment, and intervention. Psychological inflexibility has been broadly related to distress and poorer well-being and represents a promising target for such efforts. The present study examined the performance of Inflexitext, a program scoring psychological inflexibility in text, within essays about an important emotional issue by 630 undergraduates. Relationships between Inflexitext and self-report measures of psychological inflexibility processes, self-compassion, distress, and state positive and negative affect, use of words from positive and negative emotion categories, and human coder ratings were examined. Inflexitext scores demonstrated small theoretically consistent relationships with measures of cognitive fusion, challenges in pursuing values, self-compassion, distress, state positive and negative affect, and use of positive and negative emotion terms. Inflexitext also evidenced a moderate relationship with human coder ratings of inflexibility and was moderately predictive of coder classification of essays as containing any inflexibility in a Receiver Operating Characteristic (ROC) analysis. However, a comparison of areas under the curve (AUCs) associated with Inflexitext and use of negative emotion terms did not demonstrate a statistically significant difference in ability to classify essays as containing inflexibility. Inflexitext scores also did not evidence expected associations with self-reported progress towards values, experiential avoidance, and state experiential avoidance while writing the essay. Limitations of current findings and implications for future research are discussed.

To find the full text version of this article and others (as well as download a full text pdf.), ACBS members can visit the ScienceDirect homepage here