HughRistik comments on Willpower: not a limited resource? - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (28)
Right. I'm not convinced that priming people so close to a task tells us much about their actual beliefs in general, and how they will behave outside a lab: it just tells us what people believe they believe. It's like the quick fix people get after motivational seminars that fades away.
The manipulation didn't measure the effect of beliefs; it measured the effect of the cognitive affirmation of a belief. That's not really a measure of truly "implicit"
The results of this study could still be explained by some third variable underlying both self-control and implicit beliefs about self-control (e.g. Conscientiousness, sleep deprivation, akrasia, perception of a task as difficult...).
Study 4 does shed some light on this problem:
Basically, your self-control doesn't predict your beliefs a month later. This potentially rules out some stable third variable, like Conscientiousness. But it doesn't rule out other fluctuating third variables. For sleep deprivation, for instance, we would not primarily expect your sleep last month to influence both your self-control and your beliefs about your self-control now; it's your sleep this month that matters most. Same thing with stress, workload, etc...
The study concludes:
This study does suggest that beliefs about willpower may have some sort of effect (at least, if you are motivated/demotivated or engaging in affirmations), but it's very weak evidence. The lab manipulation they attempted of people's beliefs about their self-control is consistent with the author's hypotheses, but not terribly convincing. The longitudinal study could still be explained by third underlying causal variables. I still find the willpower depletion hypothesis plausible, at least for some people, on some types of tasks.
UPDATE: I found this observation in the notes:
This does seem to rule out self-control as an underlying third variable. Though it's also a bit strange in light of the results of the current study: if resource theory isn't related to self-control, why is it predicting performance on tasks like the Stroop test that supposedly measure self-control? Maybe it's Jonathan Graehl's hypothesis of making excuses to exercise less self-control than one is capable of. Or maybe "self-control" is being operationalized in different ways.
Unfortunately, I can't find the German study they are referring to through they link in the references; all I can find is the scale it used.