In healthy subjects, success in motor inhibitory control and in affect regulation were positively correlated, and success on both tasks was positively correlated with gray matter intensity in overlapping areas of right IFGpo. Furthermore, motor inhibitory control and gray matter intensity in right IFGpo correlated with methamphetamine craving in METH subjects, who exhibited deficits in inhibitory control and in gray matter integrity in this region.
These results support previous reports suggesting a role of the right IFG in inhibitory control. Successfully withholding a prepotent motor response depends on structural integrity of right IFGpo and activates this region (Aron, Robbins, and Poldrack, 2004
; Leung and Cai, 2007
; Forstmann et al., 2008
). Conjunction and meta-analyses have identified activation in right IFGpo during both motor (e.g., SST, go/no-go) and cognitive (e.g., Wisconsin Card Sorting Task, flanker task) inhibitory control, with some overlap between the two (Bunge et al., 2002
; Buchsbaum et al., 2005
; Nee, Wager, and Jonides, 2007
). Importantly, downregulation of negative emotions has also been associated with increased activity in right IFG (Lieberman et al., 2007
), including right IFGpo (Hariri, Bookheimer, and Mazziotta, 2000
; Ochsner et al., 2004
; Phan et al., 2005
). Moreover, consistent with the purported role of IFG in inhibitory control across domains, activating right IFG during motor inhibitory control has an unintended inhibitory effect on emotion-related amygdala activation (Berkman et al., 2009
). Our findings complement these previous reports and suggest that gray matter in an overlapping region of right IFG may support both affective and non-affective inhibitory control.
A recent study suggests that right IFGpo may participate in control of craving (Volkow et al., 2010
). Although we did not measure craving control explicitly, the correlation of gray matter intensity in right IFGpo with self-reported spontaneous craving is consistent with a role of this region in downregulation of craving. Greater gray matter intensity in this region was associated with lower methamphetamine craving in hospitalized subjects to whom the drug was unavailable and who would optimally downregulate craving in order to cope. However, the additional correlation of craving with left IFGpo and right pars triangularis
suggests an association of craving with overall gray matter degradation or with addiction severity in general. The absence of correlation between craving and right MFG, however, weakens that possibility.
Impaired motor inhibitory control has been related to deficits in complex and affective self-control across several psychiatric disorders. Boys with ADHD have slower SSRT along with emotion dysregulation (Nigg, 2001
). Cocaine abusers have slower SSRT (Fillmore and Rush, 2002
), and neural activity during stopping in this group correlates with self-reported ability to regulate emotions (Li et al., 2007). Poor stopping has also been associated with some eating and anxiety disorders (Nederkoorn, Van Eijs, and Jansen, 2004
; Chamberlain and Sahakian, 2007
). Investigators of the neurocognitive underpinnings of these disorders often use simple motor inhibitory control tasks, with the assumption that these tasks share psychological and neural components with the more complex processes of impulsivity and affect regulation that are compromised in the disorder. Our results support this assumption and the notion that a deficit in motor inhibitory control may be a risk factor for addiction and other psychiatric disorders of self-control.
Despite differences between motor inhibitory control and affect regulation, it is plausible that they share a common substrate. The IFG directly projects to the subthalamic nucleus, which can inhibit motor and limbic responses via the basal ganglia (see Aron et al., 2007
). Alternatively, the IFG may support a broader attention-orienting or information-selection function in motor (Chambers et al., 2009
) and affective (Wager et al., 2008
) self-control. Whatever the specific computation of the right IFG, our results suggest that it supports motor inhibitory control in the stop-signal task and affect inhibitory control in the reappraisal task.
Undoubtedly, the right IFG is not the only region important for inhibitory control, as other regions within the prefrontal cortex have also been implicated (Cohen and Lieberman, 2010
). However, right IFGpo seems to be a particularly likely region to support inhibitory control across psychological domains. Among all ROIs used in this study, right IFGpo was the only one in which gray matter intensity correlated with all three behavioral measures (motor and affective inhibitory control, and craving). In fact, SSRT and reappraisal success were not correlated with gray matter intensity in any other ROI. These results do not conflict with reports of neural activation in other prefrontal regions during inhibitory control, as greater gray matter intensity in a region is not necessarily coupled with greater or lower activation in that region (Jacobson et al., 2010
). More notably, our results are consistent with lesion studies showing that the right IFGpo is one of the only regions critical for motor inhibitory control (Aron et al., 2004
Although some have suggested an exclusively attention-orienting role of right IFG in motor inhibitory control (Hampshire et al., 2010
), there are several lines of recent evidence arguing for right IFG's role as not merely one of monitoring or orienting attention, but as implementing inhibitory control itself. A recent study used TMS to compare disruption of the right IFG (pars opercularis region) with disruption of the more dorsal right inferior frontal juncture region (IFJ) (Verbruggen, Aron, Stevens, and Chambers, 2010
). Disruption of both regions affected response inhibition speed, but apparently by affecting different processes. It was argued that the right IFJ implements attentional orienting, while the more ventral sector of rIFG (pars opercularis region) implements inhibitory control. This is highly consistent with a recent fMRI study which found that IFJ was activated in relation to the ‘oddball effect’ of infrequent NoGo stimuli, while the more ventral IFG region was activated by the putative response inhibition requirement when controlling for the oddball effect (Chikazoe et al., 2009
Strong evidence for an inhibitory control function for IFG was recently provided by a paired-pulse TMS study, which showed that when action countermanding was required, the connection between right IFG and M1 became strongly inhibitory (Buch et al., 2010
). Further, a recent study with the stop signal paradigm, which recorded from the right IFG in patients being evaluated for epilepsy, reported a significantly greater response for successful than unsuccessful stop trials (Swann et al., 2009
). This comparison controls for the ‘oddball effect’. Importantly in that study the increased neural response on successful stop trials was within the timescale of SSRT, and it occurred in the beta frequency band. Given that increases in the beta frequency band are strongly associated with stopping movement, including within the primary motor cortex (Swann et al., 2009
) and the subthalamic nucleus (Kuhn et al., 2004
) (which may participate in a structurally-connected functional circuit with the right IFG and dorsomedial frontal cortex (Aron and Poldrack, 2006
; Aron et al., 2007
), these findings are much better reconcilable with an inhibitory control function for right IFG rather than merely a monitoring/attentional-orienting role.
All current Axis I diagnoses, except amphetamine (methamphetamine) dependence and nicotine dependence, were exclusionary for both groups, but a greater proportion of METH than Control participants were smokers. It is therefore plausible that smoking status may have affected the behavioral and gray matter assessments, especially in the METH group. Our results and the literature, however, do not support this possibility. As in a prior report (Monterosso et al., 2005
), SSRT did not differ between smoking (n=8) and non-smoking (n=17) Control participants (p
=.71) in the current study. We obtained similar results (p
=.32) in a larger sample of Control participants (26 smokers, 25 non-smokers) who completed the SST for another study in our laboratory. In this larger sample, comparing smoker METH participants (n=31) to smoker Control participants (n=26) still yielded a reliable group difference in SSRT (p
=.005). The effect of smoking on reappraisal success is more difficult to assess, due to the small sample size. Reappraisal success did not differ between smoking (n=7) and non-smoking (n=16) Control participants (p
=.88). Comparing smoker METH participants (n=22) to smoker Control participants (n=7) yielded a trend toward a significant difference in reappraisal success (p
=.14), suggesting that this question may warrant further study.
The results presented here demonstrate that “cold” motor and “hot” affective self-control involve a common substrate in the IFG, which is compromised in stimulant dependence, a disorder that features self-control deficits and craving. Our findings suggest that a shared neurocognitive system subserves different kinds of self-control, although the specific computation carried out by this system is still unclear. An important implication of these findings is that defects in an IFG self-control system may underlie different disorders of self-control (Chamberlain and Sahakian, 2007
). Conversely, strengthening this system in one psychological domain may enhance self-control in another domain (Muraven and Baumeister, 2000
). Further research is needed to support these hypotheses.