Mindfulness doesn't reduce impulsive behavior.

Korponay, Davidson and colleagues present results of a study that were contrary to their expectation that the practice of mindfulness meditation would correlate with a reduction in impulsive behaviors (like having that second dish of ice cream). What they found is that neither short-term nor long-term meditation appears to be effective for reducing impulsivity that is not related to attentional difficulties, but rather is a function of motor control and planning capacities. Here is their detailed abstract:

Interest has grown in using mindfulness meditation to treat conditions featuring excessive impulsivity. However, while prior studies find that mindfulness practice can improve attention, it remains unclear whether it improves other cognitive faculties whose deficiency can contribute to impulsivity. Here, an eight-week mindfulness intervention did not reduce impulsivity on the go/no-go task or Barratt Impulsiveness Scale (BIS-11), nor produce changes in neural correlates of impulsivity (i.e. frontostriatal gray matter, functional connectivity, and dopamine levels) compared to active or wait-list control groups. Separately, long-term meditators (LTMs) did not perform differently than meditation-naïve participants (MNPs) on the go/no-go task. However, LTMs self-reported lower attentional impulsivity, but higher motor and non-planning impulsivity on the BIS-11 than MNPs. LTMs had less striatal gray matter, greater cortico-striatal-thalamic functional connectivity, and lower spontaneous eye-blink rate (a physiological dopamine indicator) than MNPs. LTM total lifetime practice hours (TLPH) did not signifcantly relate to impulsivity or neurobiological metrics. Findings suggest that neither short nor long-term mindfulness practice may be efective for redressing impulsive behavior derived from inhibitory motor control or planning capacity defcits in healthy adults. Given the absence of TLPH relationships to impulsivity or neurobiological metrics, diferences between LTMs and MNPs may be attributable to pre-existing diferences.