Examining causal components and a mediating process underlying self-generated health arguments for exercise and smoking cessation

Austin S. Baldwin, Alexander J. Rothman, Mark W.Vander Weg, Alan J. Christensen

Research output: Contribution to journalArticlepeer-review

22 Scopus citations


Objective: Self-persuasion- generating one's own arguments for engaging in a specific behavior-can be an effective strategy to promote health behavior change, yet the causal processes that explain why it is effective are not well-specified. We sought to elucidate specific causal components and a mediating process of self-persuasion in two health behavior domains: physical activity and smoking. Methods: In two experiments, participants were randomized to write or read arguments about regular exercise (Study 1: N = 76; college students) or smoking cessation (Study 2: N = 107; daily smokers). In Study 2, we also manipulated the argument content (matched vs. mismatched participants' own concerns about smoking) to isolate its effect from the effect of argument source (self vs. other). Study outcomes included participants' reports of argument ratings, attitudes, behavioral intentions (Studies 1 & 2), and cessation attempts at 1 month (Study 2). Results: In Study 1, self-generated arguments about exercise were evaluated more positively than other arguments (p = .01, d = .63), and this biased processing mediated the self-generated argument effect on attitudes toward exercise (β = .08, 95% CI = .01,.18). In Study 2, the findings suggested that biased processing occurs because self-generated argument content matches people's own health concerns and not because of the argument source (self vs. other). In addition, self-generated arguments indirectly led to greater behavior change intentions (Studies 1 & 2) and a greater likelihood of a smoking cessation attempt (Study 2). Conclusions: The findings elucidate a causal component and a mediating process that explain why self-persuasion and related behavior change interventions, such as motivational interviewing, are effective. Findings also suggest that self-generated arguments may be an efficient way to deliver message interventions aimed at changing health behaviors.

Original languageEnglish (US)
Pages (from-to)1209-1217
Number of pages9
JournalHealth Psychology
Issue number12
StatePublished - Dec 2013


  • Exercise
  • Health messages
  • Mediators
  • Self-persuasion
  • Smoking


Dive into the research topics of 'Examining causal components and a mediating process underlying self-generated health arguments for exercise and smoking cessation'. Together they form a unique fingerprint.

Cite this