Objective: Research indicates that the language used in addiction matters; therefore, the purpose of this study was to investigate whether differences in discursive and psychological processes exist in treatment manuals used for addiction treatment. Method: Using a synchronic corpus linguistic design, twelve-step facilitation, cognitive behavioral therapy, and motivation enhancement therapy manuals from Project MATCH were analyzed using LIWC2015. The log-likelihood ratio test was used to examine if difference in variables exist between the three manuals, with post hoc analysis to further examine differences. Bayesian information criterion was used to measure effect size, which ranged from weak to very strong. Results: Statistically significant differences exist in analytical thinking, authentic, emotional tone, first-person singular pronouns, second-person pronouns, third-person plural pronouns, negative emotion, male, and biological process words. Conclusions: Differences exist in psycholinguistic, linguistic, psychological, and physical processes between TSF, MET, and CBT. The greatest differences across all variables were between TSF and MET. Physical and mental aspects are addressed in TSF, whereas MET focuses on mental aspects. Knowledge of these and other differences support treatment matching for improved client outcomes. Findings of this study may have relevance to clinicians who provide addiction treatment, manual writers, addiction researchers, and clients with addiction issues. Further study of I words, which correlate with truth-telling, versus we words which support community, is suggested.

Research indicates that the language used in addiction matters; therefore, the purpose of this study was to examine whether differences in discursive and psychological processes exist in 12-step texts used for substance and behavioral addictions. Using a synchronic corpus linguistic design, writings on the first step of Alcoholics Anonymous, Debtors Anonymous, Gamblers Anonymous, Narcotics Anonymous, Overeaters Anonymous, and Sex and Love Addicts Anonymous were analyzed using LIWC2015. The log-likelihood ratio test was used to examine if differences exist between the six texts, with post hoc analysis to further examine differences. Bayesian information criterion was used to measure effect size, which ranged from weak to very strong. Statistically significant differences exist in analytic, authentic, emotion tone, first-person singular pronouns, first-person plural pronouns, third-person singular pronouns, third-person plural pronouns, male, and biological process words. Results demonstrate that differences exist in the broad psycholinguistic, specific linguistic, and psychological and physical process words used in the texts. This research is the first study to compare linguistic components of different 12-step programs, which often fill a gap created by lags in research. Findings may have relevance to writers of 12-step literature, clinicians who provide treatment, and researchers who study substance use and behavioral addictions, and individuals with these issues.