Yoga has a positive effect on the structure and function of key brain areas associated with memory


Scientists have known for decades that aerobic exercise strengthens the brain and contributes to the growth of new neurons, but few studies have examined how yoga affects the brain.

A review of the science finds evidence that yoga enhances many of the same brain structures and functions that benefit from aerobic exercise.

The review, published in the journal Brain Plasticity, focused on 11 studies of the relationship between yoga practice and brain health.

Five of the studies engaged individuals with no background in yoga practice in one or more yoga sessions per week over a period of 10-24 weeks, comparing brain health at the beginning and end of the intervention.

The other studies measured brain differences between individuals who regularly practice yoga and those who don’t.

Each of the studies used brain-imaging techniques such as MRI, functional MRI or single-photon emission computerized tomography.

All involved Hatha yoga, which includes body movements, meditation and breathing exercises.

“From these 11 studies, we identified some brain regions that consistently come up, and they are surprisingly not very different from what we see with exercise research,” said University of Illinois kinesiology and community health professor Neha Gothe, who led the research with Wayne State University psychology professor Jessica Damoiseaux.

“For example, we see increases in the volume of the hippocampus with yoga practice,” Gothe said.

Many studies looking at the brain effects of aerobic exercise have shown a similar increase in hippocampus size over time, she said.

The hippocampus is involved in memory processing and is known to shrink with age, Gothe said.

“It is also the structure that is first affected in dementia and Alzheimer’s disease.”

Though many of the studies are exploratory and not conclusive, the research points to other important brain changes associated with regular yoga practice, Damoiseaux said.

The amygdala, a brain structure that contributes to emotional regulation, tends to be larger in yoga practitioners than in their peers who do not practice yoga.

The prefrontal cortex, cingulate cortex and brain networks such as the default mode network also tend to be larger or more efficient in those who regularly practice yoga.

“The prefrontal cortex, a brain region just behind the forehead, is essential to planning, decision-making, multitasking, thinking about your options and picking the right option,” Damoiseaux said.

“The default mode network is a set of brain regions involved in thinking about the self, planning and memory.”

Like the amygdala, the cingulate cortex is part of the limbic system, a circuit of structures that plays a key role in emotional regulation, learning and memory, she said.

The studies also find that the brain changes seen in individuals practicing yoga are associated with better performance on cognitive tests or measures of emotional regulation.

The discovery that yoga may have similar effects on the brain to aerobic exercise is intriguing and warrants more study, Gothe said.

“Yoga is not aerobic in nature, so there must be other mechanisms leading to these brain changes,” she said.

“So far, we don’t have the evidence to identify what those mechanisms are.”

She suspects that enhancing emotional regulation is a key to yoga’s positive effects on the brain.

Studies link stress in humans and animals to shrinkage of the hippocampus and poorer performance on tests of memory, for example, she said.

A review of the science finds evidence that yoga enhances many of the same brain structures and functions that benefit from aerobic exercise.

“In one of my previous studies, we were looking at how yoga changes the cortisol stress response,” Gothe said.

“We found that those who had done yoga for eight weeks had an attenuated cortisol response to stress that was associated with better performance on tests of decision-making, task-switching and attention.”

Yoga helps people with or without anxiety disorders manage their stress, Gothe said.

“The practice of yoga helps improve emotional regulation to reduce stress, anxiety and depression,” she said. “And that seems to improve brain functioning.”

The researchers say there is a need for more – and more rigorous – research into yoga’s effects on the brain.

They recommend large intervention studies that engage participants in yoga for months, match yoga groups with active control groups, and measure changes in the brain and performance on cognitive tests using standard approaches that allow for easy comparisons with other types of exercise.

“The science is pointing to yoga being beneficial for healthy brain function, but we need more rigorous and well-controlled intervention studies to confirm these initial findings,” Damoiseaux said.

Gothe is an affiliate of the Beckman Institute for Advanced Science and Technology at the U. of I. Damoiseaux is an affiliate of the Institute of Gerontology at WSU.

Yoga, a mind-body activity that has components centering on meditation, breathing and postures has become increasingly popular in recent years. It’s health benefits are being systematically investigated and it is acknowledged today as an effective therapy for a variety of physical conditions such as pain and associated disability (Büssing et al., 2012; Cramer et al., 2013a), arthritis (Haaz and Bartlett, 2011), rheumatic diseases (Cramer et al., 2013b), cardiopulmonary and musculoskeletal function (Raub, 2002) as well as some psychological conditions including depression (Uebelacker et al., 2010) and anxiety (Kirkwood et al., 2005).

In addition to the mounting evidence for the health benefits of yoga, its effects on cognition have become a particularly important area of inquiry in recent years. A systematic review and meta-analysis examined the acute (short term, single bout) and intervention effects of yoga on cognitive function across 15 randomized trials and seven acute studies (Gothe and McAuley, 2015). The effect sizes observed for the different cognitive functions including attention, processing speed, executive functions and memory ranged from g = 0.18 to g = 0.29 for the randomized trials and were even greater in magnitude for acute studies of yoga, ranging from g = 0.39 to g = 0.78.

The studies reviewed in this meta-analysis used behavioral measures to assess cognitive function, such as computer- and paper pencil-based tests of executive function, attention, processing speed and memory.

Fewer studies have examined the neurobiological correlates of yoga practice using advanced imaging techniques. Froeliger B. E. et al. (2012) and Froeliger B. et al. (2012) examined structural and functional brain differences in Hatha yoga meditation practitioners and meditation naïve controls. Seven Hatha yoga meditation practitioners were compared with seven matched controls and were found to have significantly larger gray matter volume in the prefrontal cortical regions, including the middle and orbital frontal gyri for the yoga group than the controls.

In subcortical regions, yogis were found to have a significantly larger left parahippocampal gyri and hippocampus than controls. A similar influence of yoga was observed in an intervention study, which revealed a bilateral increase in hippocampal size following a 6 month yoga intervention in older adults (Hariprasad et al., 2013).

In a second study (Froeliger B. E. et al., 2012; Froeliger B. et al., 2012) the effects of yoga meditation on emotion-cognition interactions were investigated as the same subjects performed an affective Stroop task in the MRI and found that while viewing negative emotional images yoga meditation practitioners displayed less activation in the dorsolateral prefrontal cortex than controls.

Executive function refers to a subset of goal-directed processes such as planning, decision making, working memory, cognitive flexibility, abstract thinking, and has been repeatedly shown to improve with regular erobic exercise (Smith et al., 2010) as well as yoga (Gothe et al., 2014; Gothe and McAuley, 2015). Working memory includes a subset of processes involved in the active encoding, maintenance and manipulation of information, and its retrieval typically following a short period of time (Miyake et al., 2000; Kane et al., 2007).

Encoding occurs when information is first perceived while maintenance refers to the retention of that information over a short delay. Retrieval is the process of later recalling that information. In a study by Gothe et al. (2014), an 8 week yoga intervention had the most significant impact on working memory performance among a sample of middle aged and older adults compared to a stretching and strengthening control group.

Acute effects of yoga on executive function have also shown to improve performance on working memory and inhibitory control measures (Gothe et al., 2013) following a brief 30 min Hatha yoga sequence. It is unknown, however, if these differences in working memory stem from benefits to a single subcomponent process in working memory (i.e., encoding, maintenance, or retrieval) or from benefits to multiple subcomponent processes.

The purpose of this study was to examine structural and functional differences between experienced yoga practitioners and age- and sex-matched controls.

Based on recent evidence from imaging studies showing greater hippocampal volume after yoga practice, we hypothesized finding greater hippocampal volume in yoga practitioners compared to controls.

Research from human cognitive studies demonstrates some specificity, such that exercise influences some brain regions like the hippocampus selectively, and has minimal or no influence on others (Hillman et al., 2008; Erickson et al., 2011). We therefore included two brain regions: thalamus and caudate nucleus as controls, to test whether the regional specificity is also observed among long term yoga practitioners.

Given that working memory seems to be the domain of cognition that is most significantly impacted by yoga training, the present study also aimed to identify the neural correlates of working memory performance among experienced yoga practitioners using functional imaging techniques.

We used the Sternberg task (Sternberg, 1966), which requires participants to encode a series of stimuli into their working memory to decide whether a probe stimulus that is presented at a later time point was present in the encoded series. The task captures the three processes associated with working memory: encoding of information, storage in short term memory, and the retrieval of information in response to the probe.

Distinguishing between these three sub-component processes allowed us to pinpoint which specific component(s) showed differences in brain activation between experienced yoga practitioners and controls.

Given the known cognitive benefits of yoga, but limited knowledge of its neural correlates, we expected to find different brain activation patterns during the Sternberg task in experienced yoga practitioners compared to controls without a predilection for specific sub-component processes. We also predicted that the experienced yoga practitioners would perform the Sternberg working memory task with greater accuracy than the controls.


This study examined the similarities and differences in brain structure and function between 13 experienced yoga practitioners and age- and sex-matched controls. We found a significant difference in the left hippocampal volume, where experienced yoga practitioners exhibited larger gray matter volume than control participants.

Additionally, during the performance of the Sternberg working memory task, experienced yoga practitioners exhibited less activation in the left dorsolateral prefrontal cortex region than controls. Together, these findings contribute to our limited understanding of the neurological correlates of yoga practice.

Differences in gray matter volume have been reported in previous yoga studies. Froeliger B. E. et al. (2012) and Froeliger B. et al. (2012) found yoga meditation practitioners (N = 7, females = 6, mean age = 36.4 years) to have significantly higher gray matter volume in a number of regions including the left para-hippocampal gyrus, hippocampus and insula. Our findings also corroborate with the evidence from a yoga-based intervention study that examined the effects of a 6-month yoga intervention on cortical structures among seven healthy older adults (Hariprasad et al., 2013) (age range 69–81 years., males = 4).

Their results also showed a significant increase in bilateral hippocampus gray matter volume. The hippocampus is known to be critically involved in learning and memory processes (Squire, 1992).

Yoga effects on hippocampal volume are also aligned with findings from the erobic exercise (Erickson et al., 2011) and mindfulness literature (Hölzel et al., 2011). Future research needs to examine the underlying mechanisms, other cortical and subcortical regions, and the similarities and differences within the different forms of exercise (such as yoga vs. erobic) that lead to similar neurobiological effects.

In addition to examining the subcortical volumes, we compared activation during the performance of the Sternberg working memory task between the two groups. During the encoding phase of the task we observed activation in the middle frontal, precentral and paracingulate gyri for all of the participants, which is in line with previous studies (Bedwell et al., 2005). However, there was a significant difference in the activation of the left dorsolateral prefrontal cortex.

Experienced yoga practitioners exhibited less activation in this region than the controls. The dorsolateral prefrontal cortex is a region that is typically activated during the encoding phase of verbal working memory tasks (Bedwell et al., 2005) and may be sensitive to increasing load during encoding, such that this area is engaged more as task load is increased (Rypma and D’Esposito, 1999). Less dependency on recruiting the dorsolateral prefrontal cortex to perform the task may be reflective of increased efficiency by experienced yoga practitioners.

This is in line with behavioral studies that suggest yoga practice has a beneficial influence on working memory performance (Gothe et al., 2014; Gothe and McAuley, 2015). However, because the task load was relatively low during all trials, it is not surprising that we did not observe any objective differences in task performance as determined by the accuracy and reaction time of participants.

Future studies should strive to use tasks with a higher cognitive load so that the association between task performance and brain measures can be elucidated. It is also important to note that the behavioral performance of the two groups was the same, in spite of significant differences observed in the left hippocampal volume.

Several functions of the hippocampus and dorsolateral prefrontal cortex can be implicated in the practice of yoga. Both of these regions are involved in the modulation of cortical arousal and emotional regulation (Milad et al., 2007). Yoga practice is holistic and involves a combination of physical exercises, breathing and meditation that include relaxation and yoga has been shown to have psychological effects including decreased anxiety (Kirkwood et al., 2005) and stress (Chong et al., 2011).

In addition to improvements in mood and anxiety, imaging studies have also shown yoga interventions to increase thalamic gamma aminobutyric acid levels in healthy young adults (Streeter et al., 2010). Preliminary evidence also suggests that yoga has a down regulating effect on both the sympathetic nervous system and the hypothalamic-pituitary adrenal axis in response to stress (Ross and Thomas, 2010).

More recently, salivary cortisol and self-reported affect were found to mediate the relationship between yoga practice and improvement in behavioral measures of cognitive performance, specifically working memory and mental flexibility (Gothe et al., 2016). It appears that regular yoga practice may result in optimal regulation of affect and emotion for the practitioner, which may result in the effective activation of the dorsolateral prefrontal cortex as observed in the present study.

Because of the cross-sectional nature of the study our results should be interpreted as tentative. Although this small sample size may have impacted our power and undermined our ability to detect some of the effects of yoga, we were able to recruit and test twice as many participants compared to previous studies. Given the pilot nature of this study, we were limited in our measurements and chose to examine the subcomponents of working memory based on the preliminary studies and literature.

Future studies should examine whether the effects of yoga practice are selective in impacting working memory, or also influence other executive functions and corresponding brain regions. While our groups were well matched on age, sex and education levels, future studies could also include a measure of intelligence and account for menopausal status for female participants as it has been recently shown to affect functional connectivity and hippocampal volume (Lisofsky et al., 2015).

It is possible that differences in cerebral blood flow could underlie the differences we observed in the dorsolateral prefrontal cortex, as such an effect had been shown in the aging (Moses et al., 2014) and cognitive training literature(Chapman et al., 2013). Nevertheless, there are a number of strengths to this study worth noting. Contrary to previous studies (Froeliger B. E. et al., 2012; Froeliger B. et al., 2012; Hariprasad et al., 2013), we also accounted for differences in physical activity levels and cardiorespiratory fitness between the study groups, using an established measure of self-reported physical activity as well as an objective assessment of cardiorespiratory fitness.

Importantly, these characteristics did not differ between the yoga and control groups, indicating that the observed differences in brain structure and function can be attributed to differences in yoga practice specifically rather than to exercise or fitness levels of the study participants.

Another strength of the present study is its specific focus on working memory, as opposed to a measuring cognition as a congregate of memory, attention, processing speed and other executive functions. It is important to make a distinction between the subcomponent processes of cognitive functions, as they are likely to recruit differential brain resources and regions.

The differences we observed in encoding but not in the maintenance or retrieval phases of the working memory task demonstrate the value of investigating working memory as well as other executive functions at their subcomponent levels. While yoga practice may not exhibit differences on overall task performance, specific brain functions (such as encoding alone) may show differential brain activation or patterns as observed in our study.

Finally, our sample characteristics were different from previous studies, where researchers have recruited yoga meditation practitioners, i.e., participants with a significant meditation and/or mindfulness practice (Froeliger B. E. et al., 2012; Froeliger B. et al., 2012).

Although the practice of yoga involves meditation, the postures and breathing exercise are just as important in Hatha yoga styles of practice.

Our sample of experienced yoga practitioners primarily reported the practice of yoga postures (66% of their yoga practice time) as compared to previous studies where the samples were categorized as yoga meditation practitioners.

In conclusion, the present study contributes to the topical field of yoga and cognition and to our understanding of the neurobiological correlates of yoga practice. The regions affected by yoga practice, i.e., the hippocampus and the prefrontal cortex, also show significant age related changes (Grady, 2012; Toga, 2015).

Therefore, behavioral interventions like yoga may hold promise to mitigate age-related and neurodegenerative declines. Systematic randomized trials of yoga based exercise, as well as long term longitudinal studies on yoga practitioners, are needed to identify the extent and scope of cognitive and neurobiological changes and their underlying mechanisms that occur as a function of yoga practice.

University of Illinois
Media Contacts:
Neha Gothe – University of Illinois
Image Source:
The image is credited to Neha Gothe.

Original Research: Open access
“Yoga Effects on Brain Health: A Systematic Review of the Current Literature”. Neha Gothe et al.
Brain Plasticity doi:10.3233/BPL-190084.


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.