The present study identified significant differences in gray matter volume and self-reported cognitive failures between hatha yoga meditation practitioners (YMP) and a sample of well-matched controls (CG), such that YMP exhibited volumetrically larger brain structures and fewer lapses in executive function in daily life. Structural differences were particularly evident in brain regions subserving higher-order control of cognitive and motor responses. Concomitantly, the extent to which YMP and CG differed with regard to gray matter volume in these regions was significantly associated with the occurrence of self-reported cognitive failures. Moreover, yoga meditation experience was significantly predictive of gray matter volume in many of these same neuroanatomical regions. Taken together, study findings suggest that the practice of hatha yoga (a multimodal discipline involving physical postures, breathing exercises, and meditation) is associated with enhanced cognitive function coupled with enlargement of brain structures held to instantiate executive control.
VBM analysis indicated that, on the whole, YMP exhibited significantly larger prefrontal cortical regions (including the middle and orbital frontal gyri) than the CG. Experimental and legion studies indicate these brain structures are recruited during tasks that involve cognitive control [43
], inhibition of automatized or prepotent responses [44
], the contextually appropriate selection and coordination of actions [45
], and reward evaluation and decision making [46
]. Self-report data from the Cognitive Failures Questionnaire [33
] indicate that greater gray matter volume in these regions was associated with making fewer errors in attention, memory, and motor function in everyday tasks. Relative to the CG, YMP also exhibited significantly greater gray matter volume in the cerebellum, a brain structure known for decades as integral to the precise coordination and timing of body movements [48
], but more recently has been acknowledged to be involved in executive function [49
]. Common to both of these domains, the cerebellum may predict the consequences of planned actions, be they motor behaviors or mental operations, and use these predictions to update action plans [50
]. Putatively, the integration of cognitive and motor control is mediated by anatomical connections between units in the cerebellum and regions of prefrontal cortex [51
Hypothetically, prolonged practice of hatha yoga might stimulate frontocerebellar connectivity and neuroplasticity by virtue of the intense, multimodal, cognitive, and motor skill learning that such practice involves. The word yoga, stemming from the Sanskrit word yuj
, “to yoke” or “bind together,” refers to the primary aim of the practice: to unify mind and body by cultivating heightened mindfulness and self-discipline, ultimately leading to equanimity and insight [52
]. Hatha yoga involves the complex training context, high task variability, increasing task difficulty, motivated states of arousal, and long duration of training believed to be requisite to process-specific learning undergirded by brain plasticity [13
]. Indeed, the practice of hatha yoga demands exquisite executive control to coordinate body posture and breathing while maintaining attentional focus on proprioceptive and interoceptive feedback in the face of distracting thoughts and bodily discomfort. In addition, ardent motivation is needed to endure the rigors of yoga, which increase in difficulty as the practice deepens. Great precision is required to move into progressively more challenging physical postures while timing the positioning of limbs with respiration. In light of these characteristics, disciplined pursuit of yoga meditation may foster cognitive plasticity through the intensive mental training this practice entails.
Although study findings are preliminary, they suggest that yoga and/or meditation practice may serve as an effective treatment intervention for disorders with concomitant GM volume atrophy and cognitive difficulties. For example, results from the current study may be meaningfully contrasted with extant literature demonstrating that GM atrophy is associated with a broad array of psychiatric conditions including depression [53
], age-related mild cognitive impairment and depression [54
], posttraumatic stress disorder [55
], and chronic pain [57
]. In addition, substance use disorders are associated with decreased GM volume; including addictive use of alcohol [58
], cigarettes [59
], and psychostimulants [61
]. Importantly, GM volume reductions in frontal and limbic regions are found to be associated with deficits in cognition function [54
]. Consistent with the hypothesis that yoga meditation practice may remediate psychiatric conditions, a recent review paper of over 90 studies found that mind-body therapies improved depressive symptoms in patients suffering from a wide range of ailments [62
]. With regard to addictive disorders, Yoga has been reported to improve recovery from substance abuse disorders [63
] and improve smoking cessation outcomes among nicotine dependent individuals [64
]. Whether such therapeutic benefits derive from increased mindfulness and neuroplasticity stemming from state-by-trait interactions remains to be determined by future research.
To be clear, due to the cross-sectional nature of this study, no causal inferences can be drawn between the practice of hatha yoga, increased gray matter volume, and cognitive function. Indeed, it is possible that the observed neuroanatomical and cognitive differences between YMP and CG were extant prior to the initiation of hatha yoga and may reflect a preexisting propensity to engage in contemplative practice. However, the fact that number of years of yoga experience was significantly associated with gray matter volume suggests that duration of yoga practice may contribute in part to the observed volumetric differences in brain structure, possibly by stimulating neuroplasticity. In addition, the current study design did not allow us to differentiate effects that may be due to yoga versus those that may be due to meditation outside of the context of any yoga practice. Future research may address this issue by comparing yoga practitioners with yoga-naïve meditation practitioners. Furthermore, the modest sample size may have limited the statistical power of our analyses. In this regard, it is notable that robust between-groups differences were observed for gray matter volume and self-reported cognitive failures. Future research should readress these limitations by employing randomized, controlled, longitudinal designs, where yoga-naïve subjects are scanned at baseline, randomly allocated to receive either yoga training or a comparable control group and then followed for a prolonged period of time before receiving additional MRI. In addition, behavioral measures of executive function (e.g., GoNoGo task, Stop Signal Task) should be used to assess whether volumetric differences between YMP and CG correspond with objective indices of cognitive performance enhancement.