Do behavioral measures of self-control assess construct-level variance?

A wide range of measures have been used to assess self-control including executive function tasks, delay of gratification tasks, and persistence and willpower tasks. The current study sought to examine the convergent and predictive validity of these measures, provide theoretical backing for why we might or might not expect high correlations between different indicators of the construct, and question whether such measures are assessing construct-level variance. The results largely replicated prior research, with the majority of correlations being small in magnitude and non-significant. Possible interpretations include indicators assessing distinct and unrelated subdomains of self-control, the inappropriate use of measures that maximize within person variance, indicators being plagued by large sources of error variance, or some combination of these.     

Perceived willpower self-efficacy fluctuates dynamically with affect and distress intolerance

People may experience subjective shifts in their self-efficacy for exerting willpower over time and based on context, and people who struggle with self-control may be particularly vulnerable to willpower self-efficacy fluctuations. Across four samples (college students without borderline features: n = 49; borderline features group: n = 50; current smokers: n = 61; chronic dieters: n = 92), participants completed one week of ecological momentary assessment (EMA) where momentary self-efficacy for willpower, positive affect, negative affect, tiredness, and distress intolerance were assessed randomly seven times per day. Results revealed that higher willpower self-efficacy was associated with lower negative affect and greater positive affect, and, lower willpower self-efficacy predicted subsequent distress intolerance via time-lagged analyses for the borderline features group.