ATTENTION, BODY, AND SELF
In Chapter 2, I gave an overview of what happens in the body and the brain as you meditate. With regard to the brain, three themes emerged: Changes in attention, changes in body awareness and the sense of embodiment, and changes in the sense of self. In this chapter, I offer a more detailed look at studies that flesh out these three themes in a more dynamic context, that is, as they unfold over the course of a meditation and over the course of a lifetime of meditating.
Tomasino et al.’s map is static; that is, it provides an insight as to what areas of the brain and what brain networks are active during meditation. Meditation, however, is a dynamic process, where periods of great clarity trade off with episodes of great dryness, and moments of sharp focus are interwoven with interludes of mind-wandering.
What happens when you sit down, close or half-close your eyes, and start meditating? Is there a neural switch, a brain circuit that turns on to propel you into a meditative state?
Klaus Bærentsen and colleagues1 used a simple but effective procedure to examine this switching process: They asked a group of 31 experienced meditators to do a form of on/off meditation inside the scanner—short alternations of 45 seconds of meditation and 45 seconds of rest. Such short bursts would tell us what happens as meditators sink into the early stages of meditation. They found three types of effects within those 45-second episodes. (Note that because this is just one study, the regions activated or deactivated do not always correspond to Tomasino et al.’s map.)
First, meditators activated the inferior parietal lobe, which, as we have seen, is a part of the executive control network that has a direct link to the default-mode network. Second, likely as a consequence of this activation, parts of the default-mode network (notably the precuneus and the posterior cingulate) were deactivated. As you may recall, both the precuneus and the posterior cingulate are also part of the narrative self. This suggests that, at the onset of meditation, the chattering narrative self gets actively shushed. A third effect concerned body awareness, in both guises. That is, Bærentsen et al. observed activation in the supplementary motor areas familiar from Tomasino et al.’s map, as well as activation of the primary somato-sensory cortex; both activations likely indicate a sharpening of specific bodily sensations. They also noted the equally familiar deactivation of the angular gyrus (more specifically the temporal parietal junction), indicating an increasing sense of disembodiment.
The nice surprise is that there are no surprises: By and large, these bursts of mini-meditations show the same pattern of results that appear on Tomasino et al.’s map. Attention switches on, the narrative self dims, sensations are played up, and whole-body awareness is turned down. This result suggests that meditation (or, more precisely, focused-attention meditation) really is a state of consciousness, just like sleep or wakefulness is a state of consciousness. By this I mean that meditation, like any other state of consciousness, presents a consistent pattern of coordinated interaction between specific brain systems—it forms what Tomasino et al. call a “meditation network.” Another way of saying this is that meditation is a whole package that by its nature implies the four flavors of experience I mentioned at the end of Chapter 2—increased control over attention, increased focus on body sensations, increased global disembodiment, and a quieting down of the storytelling mind. The finding that expert meditators can bring this entire package online within 45 seconds or so, that is, in about 10 or so breaths, suggests that the whole process unfolds rather quickly.
If you have ever tried to meditate, you know very well that even if you manage to get into the right state of mind within those 10 breaths, it still is far from guaranteed that the rest of the sit is going to be a blissful coasting on an ever-cresting wave of unbroken attention. On the contrary: For most (if not all) of us, attention needs to be reset or sharpened repeatedly over the course of a meditation session.
A study by Wendy Hasenkamp and colleagues2 illustrates very nicely how attention wavers throughout a session and how the meditating brain copes with those distractions. They put 14 meditators (with, on average, about 1,400 hours of lifetime practice) inside the scanner for 20 minutes and asked them to focus on their breath. Participants were given a button to press as soon as they realized that their mind had wandered.
The researchers took a slice of time of three seconds around the button press as an indicator of the brain state of becoming aware that the mind had drifted off. (The three-second time frame has to do with the way the brain was scanned in this study, that is, in time slices of 1.5 seconds.) They took the three-second slice of time right before the “becoming aware” slice as an indicator of the brain state of mind wandering. The three-second time slice right after the “becoming aware” slice was used as an indicator of the process of shifting back into the meditative state. Finally, they considered the three seconds after that as an indicator of the brain state of sustained attention. Thus the assumption is that there is a cycle: After a period of time of sustained attention, the mind wanders, gets caught in the act, and then attention shifts and you get back on track.
Hasenkamp et al. found that when the meditators’ minds wandered, many regions of the default-mode network were activated (the posterior cingulate cortex among them)—as you would expect. The phase of becoming aware mainly showed activation in the anterior insula and the anterior cingulate cortex, regions associated with the salience network. The state of shifting was associated with the executive attention network, here the lateral prefrontal cortex, and with the inferior parietal lobe. The researchers also found activation in the ventromedial cortex. This part of the brain specializes in self-related processing, particularly in emotion regulation (i.e., bringing down the level of emotion after a negative experience). Maybe this activation is related to the experience of letting go of the potentially emotionally grating experience of once again noticing that the mind has run off and then gently, with no hard feelings, bringing it back to where it needs to be. During the sustained attention phase, a part of the prefrontal executive network remained active, but activation in the parietal part of the network activation went back to baseline, maybe because activity in the default-mode network was now sufficiently dampened, and so the parietal meditation switch was no longer needed.
One intriguing finding was how often this cycle repeated, even within this well-trained group of yogis. On average, meditators pressed the button 15.5 times over the course of 20 minutes, that is, once every 80 seconds, or once every 15 breaths or so. (It remains to be seen if this estimate of what happens in the scanner is a reasonable estimate of what happens on the cushion. It is possible that sitting upright on your own well-worn meditation cushion in your usual meditation spot helps with concentration. On the other hand, it is also possible that being watched inside the scanner—brain and all—brings out the best in these veterans.) Equally intriguing is the finding that how often people pressed the button was not significantly related to meditation expertise.
It is clear, then, that during actual meditation there is some cycling back and forth between the attention systems and the default-mode network. I was able to find three other studies that looked at couplings between the default-mode networks and the attention networks during meditation, compared to the couplings when participants are waiting in the scanner, asked to just rest.3 Each of these studies found a tighter coupling between at least some regions of the default-mode network and at least one of the attention networks during meditation. That suggests that meditators stay on task while meditating: When the default-mode network is active—when the mind strays from the object of its focus—the attention system notices, clamps down, and corrects.4 As we’ve seen, while people wait and rest in the scanner, the mind likely goes off as well, but the relative uncoupling during rest suggests that distraction just happens and that people let it be. The tighter coupling during meditation reinforces the main point of Hasenkamp et al.’s study, namely that focused-attention meditation really is a dynamic process, a series of predictable cycles that occurs in a predictable manner.
The main conclusion here is that meditation is a very dynamic process—a cycle of setting a goal, drifting away from that goal, noticing the drift, and then returning to the goal, over and over again. Over the course of their 1,400 hours of lifetime experience, Hasenkamp et al.’s meditators must have gone through about 63,000 of such cycles: 63,000 times of noticing they were off the breath, 63,000 times of returning to the breath, and 63,000 times of getting lost again. That is a lot of drifting off focus and a lot of gentle correction.
I must admit that, as I write this, after my own accumulated 1,700 (and a few) hours, I still find this simple fact astounding: You set yourself an exceedingly simple goal (“focus,” “be quiet,” “be here and now”), and yet, almost immediately, the mind veers away. Hasenkamp et al.’s data suggest that the essential goal of this simple form of meditation is an unrealistic one: Every time you sit, you set yourself up for failure, in a sense—it’s highly unlikely that you will be focused, on task, here-and-now for very long. Yet this “failure” might actually be helpful. That is, perhaps it is exactly the repeated fumbles and their consequences, the gently tugging the mind by its sleeves and setting it back on its course, from which it will then invariably wander, and doing this over and over again that is one of the crucial aspects of this practice—it may be this cycling that builds up trait mindfulness.
All of this could lead us to conclude that quieting the mind is hard work. But that would be wrong, as very nice recent work by Kathleen Garrison and colleagues5 shows.
All of the studies we have looked at so far are still probing meditation from the outside—we are inferring what the mind is doing from what the brain is telling us, and this is risky business. Garrison and colleagues wanted to know how brain activation relates to life as seen from the inside. What does it actually feel like to sit down (or lay down, in the case of an fMRI study), focus your attention, and quiet down, and exactly how is brain activation related to that experience?
Garrison asked 10 seasoned meditators (with, on average, 10,000 hours of practice, collected over about 18 years) to meditate in the scanner for short bouts of a minute each.6 After each bout, they were asked to describe their meditation experience.
Here’s the crux: While the volunteers were meditating, the researchers monitored activation in the meditators’ posterior cingulate cortex (PCC). As we have seen, the PCC is a central part of the default-mode network, which is activated during self-related thinking—it is part of the circuitry of the narrative self, the self-as-story. This makes it a good indicator of whether the mind is drifting away from the actual focus of meditation.7 The researchers were able to capture activation in the PCC as it unfolded; they turned this activation into a real-time graph—showing activation up, in red, and deactivation down, in blue, with a new bar popping up every two seconds. Given that the BOLD response (the rushing of blood to the place where it is needed) is sluggish, this isn’t really real time: The graph lags behind by a few seconds, but it is close.
Then they had some fun with this.
In the first bout of meditation, participants were simply asked to concentrate on their breath. In the second bout, they were shown an example of a real-time PCC graph (not their own). They were asked to use this graph as the object of their meditation, paying attention to it as they would to any other object of concentration. In the third bout, they were shown the real-time feedback graph of their own PCC activation and deactivation and asked to use this graph as the object of meditation. They were also told that this graph reflected activation in “a particular region of their brain” and that there was a two- to four-second delay involved. Next, they were given three more bouts with the feedback graph and asked to “use their mind to make the graph go blue.” To conclude, they were given three final bouts with the feedback graph and asked to “use their mind to make the graph go red.”
The researchers wrote down the participants’ experiences. There were a total of 404 reported experiences. These were sorted into eight broader categories, forming four pairs of opposites: (a) concentration (experiences of focus, focus on the breath, and clarity) versus (b) distraction (muddled experiences); (c) observing sensory experiences (experiences of physical sensations, engagement with what they saw, heard or mentally experienced) versus (d) interpreting (self-related thinking, deliberating, engaging with memories); (e) not “efforting” (not trying: experiences of open awareness, calm, and acceptance) versus “efforting” (trying to change the experience); and (g) contentment (experiences of pleasure and equanimity) versus (h) discontentment (displeasure, restlessness, discomfort, negative emotions, and the like). These eight categories were then reduced to four: undistracted awareness (concentration and observing sensory experience) versus distracted awareness (distraction and interpreting) and effortless doing (not “efforting” and contentment) versus controlling (“efforting” and discontentment).
These opposites in experience also played out as opposites at the level of the brain. The first element of each pair of opposites (tending toward concentration and letting go) was associated with the graph turning blue—PCC deactivation. The second element of each pair of opposites (tending toward distraction and controlling) was associated with the graph turning red—PCC activation. Experiences that made the graph turn red include8: “I worried that I wasn’t using the graph as an object of meditation, so I tried, like, to look at it harder or somehow pay attention more to it’ (this is “efforting”), or “I began by thinking about a variety of things that need to be done, emails that need to be sent, things that I have not done in a timely fashion, that type of thing” (interpreting). Experiences that turned the graph blue include “Very smooth. It was very easy to concentrate” and “a concentrated meditation this time” (concentration) or “I maintained primary awareness on the full range of experience, including, just, awareness of the body and various touch points, the breath moving throughout the body, the sound being integrated into that sort of, sort of fuller awareness while watching the colors with relative ease … body awareness” (observing) and “The red bars correspond to times when I was trying to either force the experience or trying to think about, thinking about stuff in general, thinking about making [the graph] blue. And then when I could let it go, [the graph] turned blue” (effortless doing).
In many ways, these results confirm what we already know: Activation of the default-mode network is related to activating the narrative self and to spinning stories—to interpreting the experience, to deliberating, and to being distracted—and deactivation results in the opposite—a sense of concentration and a tendency to simply (and calmly) observe.
What is new here is that this deactivation of the PCC is not an effortful process. Although we know from Tomasino et al.’s map that meditation activates both the salience and the executive network, Garrison et al.’s results show that “efforting”—trying your best, working hard—isn’t what quiets the mind. On the contrary, trying hard really gets the PCC going, whereas the stripping away of effort, the effortless doing, the relaxing, the letting go is what settles the mind into the meditative groove. The act of mere observing might be especially important here—what Garrison et al. call the mindset “of not being pushed, pulled, or lost in mental content, feelings, or thoughts as they arise.” Meditators in that study described this as “letting things flow by” or “observing thinking.”
This is an interesting paradox: Trying too hard to meditate will likely lead to PCC activation and hence distraction and unease, whereas not trying and relaxing into the experience is likely to lead to PCC deactivation and hence contentment and concentration. One of Garrison et al.’s graphs that went quickest and most deeply in the blue deactivation territory was the graph of a meditator who described the episode as: “I noticed that the more I relaxed and stopped trying to do anything, the bluer it went.” This seems valuable advice for any meditator: The quality of attention in meditation should be gentle, not forceful. This is, in fact, advice that meditation teachers often give.
One aspect of the results that is particularly encouraging for meditators is that PCC deactivation is associated with feelings of contentment, equanimity, and pleasure. PCC activation, on the contrary, seems to be associated with displeasure. This is encouraging because it suggests that meditation has a direct emotional feedback mechanism built right in: If it brings you calm and peace, you’re doing it right; if it feels effortful and ill-fitting, you are probably doing it wrong. Thus one further piece of advice for the new meditator is to lean in the direction of delight in the experience. Again, this advice isn’t particularly new. (The caveat is that the meditators in this study were highly practiced—the mileage of novices might vary.)
Finally, I should note that the meditators in this study succeeded very well in the final two tasks, namely to meditate themselves into the red or the blue zone of PCC activation and deactivation. Thus meditators can use biofeedback to guide their experiences. Of course, few of us have an MRI scanner in our basement, so this knowledge is of little practical use. What is useful, though, is that the meditators learned to couple the biofeedback to their states of mind; after you’ve done that, it might suffice to be guided by your inner experiences. Specifically, some noted the relationship between turning the graph blue and the direct feeling of relaxation, as the quotes in the previous paragraphs show. Another technique was to connect directly to the sensory experience from an observer point of view: “I maintained primary awareness on the full range of experience, including, just, awareness of the body and various touch points, the breath moving throughout the body, the sound being integrated into that sort of, sort of fuller awareness while watching the colors with relative ease … body awareness.”
Although many meditation teachers sternly warn us against doing this, many of us do evaluate our meditation experiences: This was a good sit; this was a not-so-good-sit (and maybe once in a while we have an excellent sit). Is there anything to this feeling? Can the brain show us what is better and deeper about better and deeper sits?
Danny Wang and colleagues9 asked 10 long-term meditators (with, on average, about 20,000 hours of Kundalini yoga experience) to meditate for 24 minutes inside the scanner. The meditations were of the focused-attention type. In one practice, called Kirtan Kriya, they repeated a short four-syllable mantra while counting the syllables on their fingers; in the other, Shabad Kriya, they focused on the breath while repeating the mantra with each exhale. The researchers found that meditations that were rated as deeper and more intense were associated with lower activation in the medial prefrontal cortex and in the anterior cingulate cortex—areas that, as we have seen, deal with self-reference and the salience network. Thus stronger meditations tended to be associated with less self-centeredness and with a less active inner monitor. The study cannot tell us, however, what the direction of influence is. It might be the case that we consider meditations to be deep when we manage to set our selves and our fault-finding ways aside. Or it might be the case that our self and the little check-marker in our head is less active in deep meditation because everything runs along smoothly and there is less fault to find.
An additional finding in this study was that deeper meditations tended to be associated with activations in the left forebrain, in areas that are known to be associated with positive mood.10 Again, it isn’t clear from these results whether deeper sits make us happier or whether feeling happier during a sit makes us think the meditation is deeper. Kornfield11 also noted in his survey of retreatants that bliss is almost always associated with deep concentration of the mind.
The type of meditation that has been studied most in the scanner is focused-attention meditation; all of the studies I mentioned in this chapter so far are of that variety. We saw that this type of meditation involves an intricate dance between, on the one hand, the salience system and the executive system, which pull you toward the object of the focus of attention, and, on the other hand, the default-mode network, which seduces you into zoning out.
In an overview paper on how attention is regulated and monitored in the two main types of meditation—focused attention and open monitoring—Antoine Lutz and colleagues12 argue that open monitoring might be different on three counts. First, because open-monitoring meditation is, by definition, not focused on one particular object, there might be reason to assume that the executive system would be less involved. Instead, you might expect more activation in the salience system, which would be probing what is, right here, right now, relevant for your sit. Second, open-monitoring meditation involves the cultivation of awareness of the internal body states, and so you might expect more activation in regions that are concerned with the body, such as the somatosensory cortex. Finally, for some, and in some traditions, open monitoring might involve an attempt at emotion regulation, for instance, by labeling of feelings and emotions when they come up during a sit. This might be reflected in brain activation as well.
All three of these assumptions seem more than reasonable. Unfortunately, there are not yet enough studies that have probed open monitoring to warrant a meta-analysis. We can, however, look at the few existing studies and try to draw some tentative conclusions.
The most interesting studies would be those that directly contrast the focus-attention and the open-monitoring approach to meditation. I could find only two such studies. Sadly, they do not converge on the same conclusion.
The earliest of those two studies, by Antonietta Manna and colleagues,13 studied eight Buddhist monks from the Theravāda tradition, with 15,750 hours of accumulated meditation experience on average. During open monitoring, the left hemisphere, especially the left anterior insula and the left precuneus, was more active than during focused attention. Activation was also higher in midline structures and superior temporal areas—brain regions typically associated with self-awareness. The results suggest that, as Lutz et al. predicted, the salience network (here: the insula) might be more active, and awareness of internal states (here: precuneus, midline, and superior temporal areas) might be turned up as well.
One interesting result not predicted by Lutz et al. is that, unlike what Tomasino et al.’s map shows, the precuneus, typically related to the narrative self, was activated rather than deactivated in the open-monitoring portion of the meditation. Interestingly, activation in the language-related areas of the brain was not higher during open monitoring, suggesting that this heightened awareness of one’s internal state is not associated with the creation of an actual story involving words (or at least not more so than simply resting or focusing on the breath do).
The second study, by Judson Brewer and colleagues,14 contrasted focused-attention meditation, open-monitoring meditation, and loving-kindness meditation inside the scanner in a group of 12 practitioners with, on average, about 10,000 hours of practice. They found few differences between open-monitoring and focused-attention practices, but the one difference they did find—lower activation in the superior and medial temporal gyrus—stands in contradiction to the results from Manna and colleagues, who found higher activation in the superior temporal areas.
What to make of these results? The predictions made by Lutz et al. sound reasonable, but it seems that there is little support for them, at least in direct comparisons between focus-attention and open-monitoring meditation. One study provides support for two of the hypotheses; the other does not support any of them and contradicts one earlier finding. The sad conclusion, for now, is that we know all too little about the specifics of the differences between focused attention and open monitoring.
Two sections ago, I mentioned bliss.
People who don’t meditate sometimes think that meditators simply close their eyes and off they go into the sunset, riding magnificent waves of radiant bliss.
People who meditate know otherwise.
In some ways, of course, an absence of strong positive emotions is to be expected in attention-type meditations. In focused-attention meditation, there is a singular focus—typically on the breath—and when thoughts or emotions pop up, the meditator simply returns to the breath, without judgment and with acceptance. In open-monitoring meditation, you simply observe, without attachment, what comes up. Over time, this leads to equanimity vis-à-vis the original event that provoked the emotion.
There is one Buddhist practice, however, that has the effect of generating joy or bliss from within, without any external cues, at least during some of its stages—jhana meditation.15 Jhana meditation is a progression through eight sequential practices or stages, each further deepening the meditator’s concentration to the point of losing contact with the senses and hence the outside world. Stages 1 to 3 are associated with extreme bliss and the later stages more with a less ecstatic sense of deep peace and equanimity.
There is alas only a single brain-imaging study on jhana meditation, and it includes only a single meditator.16 This meditator, Leigh Brasington, is, however, a well-known teacher in this field, with 17 years of experience; the authors of the study claim he might be the one and only person in the United States with the requisite proficiency in jhana who was willing to participate in the study.
To test the joy hypothesis, the researchers probed the dopamine system. Dopamine is a neurotransmitter (i.e., a chemical messenger in the brain) that is released during pleasant events—your favorite food, sex, good music, money, and drugs are some things that get the dopamine response going. The researchers tapped into two crucial brain regions within the dopamine system, the nucleus accumbens and the medial orbitofrontal cortex. The nucleus accumbens is the basic brain structure from where the dopamine rush originates; the medial orbitofrontal cortex is activated a little later—it is where pleasure registers in awareness.
The results showed that both the nucleus accumbens and the medial orbitofrontal cortex were activated during Stage 2 jhana (Stage 1 was not recorded due to a technical difficulty). Interestingly, activation in the nucleus accumbens declined after Stage 2, but the orbitofrontal cortex stayed active until Stage 4. Thus, even as the dopamine reserves dwindled (as they are wont to do—no high lasts forever), the subjective experience of joy still lingered.
Brasington described his experience quite dramatically—Stage 2 was compared to “opening a birthday gift and getting exactly what you most wished for” and Stage 3 to postcoital bliss. As the researchers point out, however, the activation in the dopamine network wasn’t actually all that spectacular. What is most probably happening is that due to the intense concentration during jhana meditation, most cortical activity quiets down. In that quiet environment, even a modest reward signal from the nucleus accumbens will be detectable, and it will be felt as much more intense than it really is. Thus training the mind to be quiet might help to inflate simple feelings of contentment into something approaching rapture.
Finally, the astute reader has noticed that neither the nucleus accumbens nor the medial orbitofrontal cortex are part of Tomasino et al.’s map. This underscores again what I mentioned at the beginning of this section: Meditation in its most typical forms is not likely to catapult the meditator into euphoria on a day-by-day basis.
The focus of meditation manuals and teachers is almost exclusively on how to deploy attention and how to work with emotions. The day-to-day practice of meditation, however, comes with a wide variety of experiences, which often receive a lot less mention in meditation manuals. In this section, I focus on what we know about bodily and sensory experiences during meditation.
I start by mentioning that although most people assume humans have five senses (sight, sound, smell, taste, and touch), in reality we have quite a few more. The list also includes (but is not limited to) proprioception (knowing where your body parts are relative to other body parts), interoception (the sense of the physical condition of your body), thermoception (sense of temperature), equilibrioception (the sense of balance and acceleration), and nociception (awareness of pain), and even, some claim, a sense of time. Not all of these, obviously, have been investigated during meditation.
Earlier I mentioned Kornfield’s survey of about 160 meditators who had just finished either a two-week or three-month insight meditation retreat. Kornfield examined the data for “unusual” experiences, and found that most fell into three broad categories: somatic experiences (e.g., changes in proprioception and nociception), visual experiences, and mental experiences (in which Kornfield included experiences such as mood swings, rapture and bliss, changes in the perception of time, and out-of-body experiences). If you’re experiencing any of these in your sits and you would like to know whether what you’re experiencing is “normal,” that is, something many meditators experience, or just a few, you’re out of luck: Kornfield refused to provide exact frequency data for most of the experiences. What we do know: 55% of the questionnaires mentioned the somatic experience of “spontaneous movement”; spontaneous alterations of body perception (like a feeling of being switched to slow-motion, a feeling of floating, or a loss of body awareness) were reported “frequently” and so were visual hallucinations, either with the eyes open (this involves seeing color changes in the visual field, perceiving still objects as moving, having your perceptions sharpened or intensified, seeing vibrations in the air, or having LSD-like melting visions) or with the eyes closed (seeing flashes of light, or colors, or more full-flung visions, like beholding the Buddha). Mental experiences were also common: 47% of the reports mention dramatic mood swings (sudden heavy sadness or flatness, fear, anger, sexual fantasies, rapid switching between doubts, bliss, boredom, serenity, joy, etc.), perhaps offset by the rapture and bliss reported by 40% of the two-week and 95% of the three-month meditators.
In the remainder of this section, I concentrate on three items on Kornfield’s list that we do know a little about: body awareness experiences, visual hallucinations, and perceptual awareness during meditation (i.e., how much of the world gets through to you); I add what we know about how meditation alters pain perception.
Changes in body awareness during meditation are not unusual. I assume anyone who has ever sat recognizes that after a few minutes of concentration the body fades somewhat into the background: It becomes a little harder to tell where the bottom ends and the cushion begins; if your hands touch or hold each other, it might become less clear which hand is which, and so on. More extreme examples of changes in body awareness can easily be found—see Kornfield’s previously mentioned list.
It is, of course, impossible to link such experiences to whatever is happening inside the brain unless one has a precise account of what the meditator experiences. That is exactly what Aviva Berkovich-Ohana and colleagues17 set out to examine in a group of 16 Vipassanā meditators with, on average, 11,225 hours of meditation practice.18
The original intent of the researchers was to examine the perception of time and space during meditation (i.e., where you are in time and where you are in space). Their hypothesis was that this sense of where you are might be intimately linked to your body awareness—a sharper body image would lead to a sharper sense of time and space or vice versa.
The design of the study was quite complicated, but understanding it is worth the trouble, so bear with me. First, instead of waiting until meditators reported time and/or space experiences, the researchers provided the meditators with instructions. A first set of instructions was to “Try to be in the present moment” (let’s call this the “now” instruction) and “Try to be here” (“here” instruction)—routine meditation instructions. A second set of instructions was to “Try to be in the near past (in the same place—the lab)” (“then” instruction) or “Try to be elsewhere (at the moment, with the experimenters outside the shielded magnetoencephalography [MEG] room)” (“there” instruction). Finally, the meditators were instructed to “Try to be outside time” (“timelessness” instruction) or “Try not to be in the center of space” (“spacelessness” instruction). Participants cycled through these instructions, taking 90 seconds for each of the six meditations.
This design allowed the researchers to look at a few things these conditions have in common in terms of activation and what sets them apart. First, looking at what the pairs within each instruction have in common gives us an idea of what is overlapping in the brain’s representation of space and time. Second, looking at the contrast between the second and the first set of instructions (comparing “then” with “now” and “there” with “here”) gives us some idea of what brain regions are involved in memory, imagination, and the like. Third, the contrast between “timelessness” and “now” and between “spacelessness” and “here” tells us something about how the brain perceives timelessness and spacelessness. Here is where the second contrast comes in handy: We want to make sure that the third contrast isn’t simply the meditators’ imagination going wild, so it would be good to find that the third contrast involves different brain regions than the second contrast.
After the meditation session was over, meditators were interviewed about their experiences. Eight participants reported alterations in the sense of body boundaries. Four of those involved a more diffuse sense of self, that is, the feeling that the self spilled out of the body or that the body itself faded (e.g., “The experience of the body faded. There was a sense of body in the background, not in front of consciousness” and “a pleasant dissolution, something liquid-like”). For three participants, the body disappeared altogether (e.g., “The body as physical image was absent. There was a sense of open space without the bodily dimension” and “There was no awareness to bodily and self boundaries”). One participant reported an out-of-body experience (“I kept entering and leaving my body. Outside the body I felt short and small, like a little child, I shrank.”). Thus, even though the instructions did not emphasize the body or its boundaries in any way, the meditators’ attempts at meditating outside time or space had the effect of loosening the sense of body boundaries in half of them.
The contrast between “then” versus “now” and “there” versus “here” showed overlap in activation in the part of the default-mode network typically associated with mental time travel (i.e., reminiscing about the past or planning the future), as it should. Interestingly, these regions were not the ones that were activated when participants were meditating timelessly or spacelessly. Meditating timelessly involved activation not only in brain regions typically associated with time (right posterior parietal cortex) but also in regions typically associated with body awareness (right insula, right somatosensory, and medial posterior cingulate cortex). Likewise, spaceless meditation did not only involve regions associated with spatial processing (bilateral temporal gyrus, left thalamus, right postcentral gyrus, medial frontal gyrus, bilateral frontal cortices, and right parietal lobule) but also regions typically involved with interoception (bilateral posterior cingulated cortex and right insula). Thus meditators who try to meditate spacelessly and timelessly also deactivate some of the brain areas associated with situating the body in space and time, which might result in a feeling of disembodiment, as the reports noted in the previous paragraph suggest.
This result was confirmed in another analysis. Here, the researchers split up the participants in two groups: One group, about half of the participants, consisted of those meditators who reported a change in the sense of both space and time (e.g., “The center of space became endless, without a reference point. … Time was less relevant”); the other group consisted of the rest of the participants, that is, meditators who either reported a regular experience of both space and time (e.g., “The mind was in the present moment”) or a change in either the sense of time or space, but not both (e.g., “A sense of expansion, something open and wide”). The first group showed lower activation in the temporal parietal junction and the insula and increased activity in the cerebellum. As we discussed in the previous chapter, these regions are associated with embodiment.19 So what distinguishes people who report more timeless and spaceless awareness in meditation from those who do not is not any brain region that has to do with time perception or spatial awareness per se. Instead, the two groups differ in activation in brain areas that are associated with a sense of the body.
Taken together, the results from Berkovich-Ohana et al.’s study suggest that in meditation the senses of time, space, and body are intricately interwoven—when the experience of one of those changes, experience of one if not both of the other ones will change as well, for the simple reason that all three share activation in common brain regions. The final set of results cited here suggests that changes in the sense of embodiment might be the driving force. That is, changes in time and space are associated with changes in regions that situate the body in space and time, but when there are no changes in these body awareness regions, there are no changes in the perception of space and time. Thus meditators who are trying to achieve a timeless and/or spaceless state do so (knowingly or unknowingly) by altering the perception of the body—a more diffuse sense of body boundaries may well allow the meditator to feel free from the confines of time and space as well.
Another interesting side effect of meditation is the potential presence (in some people and some of the time) of visual experiences of the kind psychologists routinely call “hallucinations.” How common or uncommon these are is unknown—Kornfield simply mentions they happen “frequently.” My own anecdotal polling of fellow meditators suggests individual differences—some people encounter visual experiences while meditating and other people don’t, and both categories of people are surprised to hear of the other’s existence.20
I hasten to state that “hallucination” is a very strong word, often associated in people’s minds with mental illness. Here I simply intend the word’s neutral, descriptive meaning in psychologists’ lingo: Seeing something that isn’t physically there. The examples in Kornfield’s study include color changes in the visual field, light flashes, colored lights, fields of great brightness, “a luminous mind,” feeling as if someone is shining a spotlight on you or—for those who meditate with eyes open—increased clarity of vision, melting-like visions, or seeing air energies or vibrations. Kornfield also mentions more complex experiences, such as visions of the Buddha, a radiating cross, or—scarier—a hand-sized spider emerging out of the floor. Hallucinations can also occur during walking meditation—seeing sparkles of light while walking at night, for instance.
A more systematic study of these phenomena was undertaken by Jared Lindahl and colleagues.21 They looked in detail at the experiences of 28 people recruited from Buddhist meditation groups. This was not a random sample: The researchers included only yogis who had ever encountered—to quote their ad—“a meditation-related experience that was significant, unexpected, challenging, or was associated with physiological or psychological changes.” Out of these 28, nine participants, roughly one out of three, reported seeing lights or other forms of luminous experiences; these started occurring, on average, five years into their meditation practice.
The experiences fell into two types. The first type concerned distinct light forms. A typical example is “Sometimes there were, oftentimes just a white spot, sometimes multiple white spots. Sometimes the spots, or ‘little stars’ as I called them, would float together in a wave, like a group of birds migrating.” The second type of experience concerned patterns and fuzzy visual experiences, commonly described by the practitioners in this study as shimmering light, a pixilation of space, or a brightening of the visual field. A typical account is “Even with my eyes closed, there would be a lot of light in the visual field. Diffuse, but bright. … When I let go, I was totally enveloped inside this light. … I was seeing colors and lights and all kinds of things going on … Blue, purple, red” and “There was often a curtain, this internal curtain of light.”
Before I turn to Lindahl’s take on these phenomena, it might be good to note how meditation teachers react to such experiences. Lindahl gives a long overview of these reactions. In some traditions, visual experiences are viewed as positive, that is, as signals of progress on the meditative path. In some forms of Theravāda Buddhism,22 the first type of visual experience—spots of light—is often called a nimitta, or “sign.” Experiencing a nimitta is seen as an indication that your practice is progressing to a stage where deep concentration becomes possible. In the Tibetan tradition, both types of luminous experience—spots and fuzzy lights—are likewise sometimes interpreted as signs of progress—indications that the practitioner is making contact with her own clear and luminous mind.23 Other traditions, however, consider these experiences as side effects that can lead the meditator astray, making her believe she is further along on the path than she actually is. This idea is expressed clearest in contemporary forms of Theravāda Buddhism in Burma24 or in the Zen tradition.25 Still other traditions, like some Western forms of Theravāda26 or the Tibetan Dzogchen tradition,27 encourage the meditator to read and sometimes even manipulate these experiences in particular ways to advance her practice. For instance, Ajahn Brahm suggests that meditators first learn to recognize the nimitta and then try to “shine it up” or make it more radiant by focusing on its center, to finally stabilize it. If the meditator is able to stay with the nimitta with a one-pointed mind, she will eventually enter the jhana state.
What can science tell us about those light experiences?
To answer this question, we must take two characteristics of these visions into account. The first is that these hallucinations don’t occur early on in practice—they only appear after some expertise in meditation has been developed. Half of the visions reported in the Lindahl et al. study also debuted during retreats, that is, during periods of extensive, concentrated practice. Second, the visions described by the meditators in the Lindahl et al. study are all what psychologists call “simple” hallucinations. This is important because both characteristics rule out one trivial explanation, namely that the meditators are having hypnagogic hallucinations—that is, the kind of visions you have when you are falling asleep. If the yogis were falling asleep, the visions would to be more akin to dreams—multisensory (typically, hearing things as well as seeing things), widescreen, without insight in their nature, and uncontrollable.
Lindahl et al. note that in nonmeditators such simple hallucinations occur after sensory loss, for instance when people are submitted to extended periods of isolation, silence, darkness, and immobility, or when they are living in a very monotonous environment—like an isolation cell in a prison. Visual impairment can be a cause as well (e.g., older adults with very poor vision can develop hallucinations; the so-called Charles Bonnet Syndrome). In fact, it doesn’t take long for sensory loss–driven hallucinations to occur—15 minutes in a fully darkened room that completely dampens sound typically does the trick.28 You likely don’t have access to an anechoic chamber (aka, “dead room”), but you can mimic the effect easily by blindfolding yourself and blasting a white noise app through your headphones loud enough to drown out external sounds; half an hour of this often leads to (sometimes remarkably detailed) visual hallucinations. The bonus is that you are likely to also experience auditory hallucinations.29
Sensory-loss-driven hallucinations are associated with activation in the occipital cortex, the seat of the early visual system located in the back of the brain.30 It is this activity that appears to cause the hallucinations, rather than the other way around: In an fMRI study with Charles-Bonnet patients, activation started ramping up in the visual system before the patient reported the hallucination.31 In the same study, the researchers also found a logical connection between the content of the hallucination and the brain region that was activated: A participant who reported seeing faces showed activation in the left middle fusiform gyrus, an area associated with face perception; patients seeing featureless colors showed activation in V4, a region crucially involved in processing color, and so on.
What is then likely happening is that lack of input into the sensory regions of the brain leads to spontaneous firing of neurons within those regions through a mechanism called homeostatic plasticity—adjustments to keep the activity levels within neuronal circuits stable. This can lead to two things. One is lowering of the firing threshold of the particular brain region, which is a fancy way of saying that the region will be easier to stimulate (even a little light that falls on your closed eyelids might look very bright and colorful); the other is that the neuronal circuit starts firing above threshold even in the absence of an external stimulus (i.e., you hallucinate).
Meditation is a close cousin to sensory deprivation. You sit as immobile as you can in a quiet room, alone or with no social interaction, often with the lights dimmed, and you either fix your gaze on a point on the wall or floor or close your eyes altogether. Although this isn’t an actual situation of sensory deprivation (sounds and smells gets through, and, for those who do not close their eyes, vision remains engaged), it is clearly one of diminished input.
Lindahl et al. speculate that a keen attentional focus, engaged with something very monotonous and undifferentiated like the breath at the exclusion of everything else, might be the additional crucial element to bring the meditator into a zone of deprivation. This might explain why meditation-based hallucinations typically appear after only a few years of practice and why they are more likely to surface during retreats—the attentional focus, with its active clamping down of all that is not the immediate object of attention, needs to be strong enough to create the effect. In line with this conjecture, seven out of nine practitioners in the Lindahl et al. study who mentioned light experiences connected the arising of these experiences with a period of enhanced concentration.
All of this, for now, remains largely speculative. There are no fMRI studies of hallucinations during meditation. There is, however, one EEG study,32 which found that EEG for two meditators who reported light experiences during a testing session showed strong alpha blocking, likely a sign of the brain clamping down on external and internal input—as Lindahl et al. speculated.
So which of the three Buddhist explanations for these phenomena is correct? Intriguingly, Lindahl et al.’s study suggests that all three have their merit.
First, the finding that enhanced attention might be necessary for the hallucinations to occur fits well with traditions that claim that such hallucinations are a sign of progress: Nimitta or the luminous mind appear after a long build-up of attentional muscle; they can thus herald an advance in singularity of focus. Note here that although the results suggest that these visions are signposts for progress in attention, the inverse is not necessarily true. That is, individuals who never experience such visions are not necessarily not making progress—there is no reason why attentional focus should always and automatically lead to visual experiences.
Second, the traditions that claim that the nimitta are nothing more than side effects are correct too: The hallucinations don’t seem to play any direct crucial role in progress in any other aspect of meditation.
Third, the traditions that suggest that the nimitta can be used as guidance for increasing concentration may also be correct. The EEG study suggests that the presence of nimitta may be indicative of the strength of attention in the moment. Developing the ability to maintain or stabilize the nimitta may then be a very good feedback mechanism for further concentration training. One study on sensory deprivation33 found that participants often tried to play with their hallucinations and that those research subjects who were able to shift their attention to different aspects of the experience (e.g., fluctuate between what they thought they were hearing in a sea of white noise and their internal body states) or who were able to zoom in on the hallucinations were also the ones who reported such perceptions more frequently. Thus the ability to shift attention or zoom in on the visions might be a hallmark of increased concentration and may be exactly what is needed to bring you to the threshold of absorption, and maybe beyond, as advocates of the third position claim.
Meditation is often done with the eyes closed or half-closed. It is, of course, impossible to close your ears, and sounds will, almost by definition, intrude. (A possible exception concerns the deeper stages of meditative absorption, where the claim is that all senses, except the sense of mind, fall away.)34
From what we have seen so far, we might formulate two expectations for focused-attention meditation. One is the general expectation that focusing attention simply works, and so sounds and noise might become less noticeable. The other expectation is that the occasional paradoxical episode will occur as well, where deep concentration leads to hypersensitivity to sounds, just like visual isolation leads to lower visual thresholds. Fainter noises will then be perceived more clearly, and ordinary noises may sound louder. When my wife and I sit at home, sometimes one or the other of our cats likes to sit with us. Being a cat, her initial curling-up session might abruptly erupt into a fur-licking fest. When this happens, I often misjudge the cat as being much closer by than she actually is. In meditation halls, it can sometimes seem as if your neighbor is breathing right into your ear. And I once witnessed an otherwise unflappable long-term practitioner of Vipassanā meditation get up quite resolutely 10 minutes into a sitting, tear our newly acquired clock off the wall, and throw it into the hallway, after which he returned, serenity re-embodied, back to his cushion.
During open-monitoring meditation, in contrast, the practitioner is expected to observe his internal and external environment, and so we might expect that sounds do get through and may even be noticed more quickly or hit with higher intensity.
The literature on auditory perception and meditation is large; much of that literature investigates the effects of Transcendental Meditation®. Most studies ask the participants to meditate and then present them with sounds during or right after the meditation period—clicks at regular or irregular intervals, most often—and measure how the brain processes these sounds. Most of the work has been done using EEG.
When we look at the literature, however, the picture isn’t clear at all.35 For the six studies that examined open-monitoring meditation, the evidence is mixed: Three studies find enhancement (i.e., stimuli are processed better or faster), one finds no difference, and two find suppression (i.e., stimuli are processed less well or slower). For the nine studies that examined focused-attention meditation, the results suggest maybe a suppression effect: One study finds enhancement, three find no difference, and five find evidence for suppression. It is precarious to draw general conclusions from such diverse findings.
Here I highlight two results from this group of studies that might give you a sense of the complexity of the literature. One comes from the McEvoy et al. paper. In this study, the researchers tested five expert practitioners of Transcendental Meditation®. They examined how the brain reacted to very short clicks (lasting 1 millisecond, that is, 1/1000th of a second), presented at 20 clicks/second immediately before and after meditation. The researchers recorded EEG in the brainstem; this EEG measured very early processing (i.e., the brainwaves started between 5.5 and 9 milliseconds after the stimulus occurred, depending on how loud the click was)—way before the sound signal reached the cortex and stood even a chance of being represented in awareness.
It turned out that the brainstem reacted differently based on the intensity of the clicks. When clicks were presented quietly (at 5 to 40 decibels, the sound level of, at most, a whisper), there were no differences in processing before and after meditation. Between 40 and 50 decibels (about the level of ambient urban noise or a very quiet conversation), the brainstem response was slightly delayed after meditation, suggesting that sounds at these levels are processed less well during meditation. At 60 to 70 decibels, however (the sound level of a normal conversation or of background music), there was a speed-up after meditation, suggesting that meditators are more sensitive to these sound levels during meditation than before meditation.
So, in this study, meditators were able to shut out (at least to some extent) sounds that occurred at the levels typical for a quiet meditation hall (from a whisper to birdsong and ambient traffic noise), but they became hypersensitive to sound levels just a little louder than that—the level of casual conversation. (Maybe this explains why nothing grates meditators more than someone having a conversation right outside the meditation room.)
In the second study, Cahn and colleagues tested 16 expert Vipassanā meditators (with, on average, 20 years of experience). They compared EEG recorded during a meditation period with EEG recorded during a period of mind-wandering. During the last four minutes of either meditation or mind-wandering, the researchers played a series of 250 sound stimuli to their volunteers, one per second; the subjects were asked simply to continue what they were doing and ignore the sounds. There were three types of sounds: a low sound, a high sound, and a burst of white noise; the high sound was played 80% of the time and the high sound and the white noise each 10% of the time.
Why the 80% versus 10%? When you repeat the same stimulus over and over, the brain tends to tune it out—a process called habituation. A good example is the ticking of a clock: After a while, the sound of the ticking fades away in your mind, and finally it just slips out of awareness altogether. (This example also shows that habituation is fragile. When attention turns to the clock again, the opposite occurs—the clock now seems louder and more obnoxious than ever before: sensitization.) In the Cahn et al. study, the high tones should have led to habituation, because they were presented very frequently. That did indeed happen for the mind-wandering condition, but it did not for the meditation condition. (This is likely what happened to my clock-throwing friend: The ticking unfortunately failed to habituate.) In the mind-wandering condition, the brain reacted differently to the infrequent sounds (the low tones and the white noise) than to the frequent sound (the high tone), as it should. In the meditation condition, however, the brain did not react differently to the two types of sound.
All of this suggests that, while you meditate, you take in every sound as it presents itself, every moment anew, as if the moments that came before never were—truly engaging with each moment and each event as it arises and passes.
Finally, I would be amiss not to mention one rather spectacular study that demonstrates quite dramatically how far a truly exceptional meditator can go in locking out distractions. In this study, by Bob Levenson, Paul Ekman, and Matthieu Ricard,36 the two first authors tested the third author, a French monk who, at the time of the study, had been practicing in the Tibetan tradition for more than 30 years. They subjected him to what the paper describes as “a 115 decibels, 100 milliseconds acoustic stimulus.” This short burst of noise was meant to sound like a gunshot; 115 decibels is about the noise level of a rock concert, of a bass drum being hit, or of sandblasting (it falls short of the sound level of a real gunshot, which is around 150 to 165 decibels). The idea was to provoke what is called, unsurprisingly, the startle reflex: You jump up, and your parasympathetic nervous system goes wild—your heart rate and blood pressure go up, you breathe more rapidly, and you start to sweat.
There were four conditions in this experiment. The first was an open-presence meditation, in which Ricard went into a state of open monitoring that the paper describes as “very vast, clear, vivid, lucid and fully resting in the moment.” After Ricard indicated that he had reached this state, there was a 60-second waiting period, and then a 20-second countdown, at the end of which the fake gunshot sound was blasted from loudspeakers located right behind the monk’s head. In the second condition, Ricard first went into a state of focused attention, with his “mind gathered into a point.” After he indicated that he had reached this state, the same waiting period-then-countdown-then-blast scenario followed. In the third condition, the same scenario was followed after Ricard entered a state of distraction—thinking about a particular incident from the past. Finally, there was an “unanticipated” startle condition: Ricard was simply blasted with the noise, unannounced.
The researchers measured the strength of Ricard’s startle reflex by polygraph (which records automatic physiological responses such as heart rate and skin conductance) and by examining his facial expression. During the study, Ricard was seated on a chair on casters, which had a motion detector attached—yet another way of measuring startle.
Before anything else, the researchers compared Ricard’s unanticipated startle response to that of other people of his age and found that Ricard had a perfectly normal startle response.
The meditation results showed that both meditation techniques generally led to much less of a startle response than distraction did, with open monitoring generally yielding less of a response than focused attention. This is remarkable because typically when research volunteers are asked to suppress their startle reflex, this actually leads to a larger response—bracing yourself is counterproductive. Meditation works differently: It is neither a pushing away, as when you are asked to suppress, nor a looking (or hearing) away, as in the distraction condition. Rather, it is a way of being with what happens when it happens. Ricard’s own description of his experience during open monitoring was that he was fully in the present moment and the fake gunshot was just one more of those present moments—in a way, there was nothing to be startled about. In the focused-attention condition, Ricard reported being fixed on the incoming event and thus a little more outside the present moment, not resting in a state of no expectation but more in a state in between open acceptance and distraction. This might explain why the startle response was a little more outspoken in focused attention than in open monitoring.
Pain is one of the certainties in life. It is also one of the certainties in meditation: Almost every meditator who goes through longer retreats will have to deal with a significant amount of pain in her joints or back sooner or later. Interestingly, meditation itself provides some remedy for the pain it provokes.
I was able to find seven relevant papers.37 In all of these studies, meditators were presented with painful stimuli that were carefully controlled by the researchers. This involved things like hot patches applied to the skin (a typical value is 48˚C, or 118˚F, for a few seconds), low-level electric shocks (below 400 mA, typically lasting for a few seconds), or laser stimulation (supposed to imitate a needle prick). In most of the studies, the painful stimulus was presented during meditation; in a few, the stimulus occurred right after a meditation session. Some studies compared meditators with nonmeditators; some compared meditation states with various nonmeditative conditions.
The main conclusion is that meditation makes painful experiences less unpleasant—this was found in all five studies that included this measure. Interestingly, that does not mean that the pain is necessarily felt less sharply: In three of the seven studies, meditators rated the pain just as intense during meditation as outside of meditation; the other four found a decrease in intensity ratings. This suggests that the main effect of the meditative experience on pain is not that it removes or dulls the ache. Rather, meditation makes the experience a tad more bearable.
Why is that?
One reason meditation could make pain more tolerable is the parasympathetic response associated with meditation—the calming of the body.38 The evidence here is mixed: One study39 found that all of the changes in the meditators’ reports of pain intensity and unpleasantness could be explained by the slowing down of breathing rate; another40 found that meditation decreased pain intensity ratings but simple relaxation did not.
A second potential mechanism is distraction. You are now focusing your attention elsewhere—like on the breath. One finding that suggests that this might be the case is a study by Fadel Zeidan and colleagues—they found that occupying participants’ minds by having them work on a simple math problem (counting backward by sevens, starting from 1,000) lowered pain ratings as well, although not as much as meditation does.
Another possible piece of evidence for the meditation-as-distraction view comes from a study on a yoga master who claims he is unable to feel pain during meditation by “concentrating on not feeling pain.”41 The man is clearly not all talk—he likes to demonstrate his insensitivity to pain by sticking needles into his tongue and cheek. The researchers applied laser stimulation to one of the yogi’s hands or feet at unpredictable points in time. The yogi reported no pain during meditation but did indicate pain while not meditating. Brain imaging confirmed this report: During meditation, there was little or no increase in activation in regions of the brain that are typically associated with the intensity of pain perception (the thalamus, the somatosensory cortex).
A third potential mechanism is the nonjudgmental, accepting nature of meditation—remember Kabat-Zinn’s definition of mindfulness?—especially in its open-monitoring form. This mindful attention can come in a number of flavors.
One is to direct your attention carefully to the actual sensations involved in the experience. In support of this hypothesis, Tim Gard and colleagues found that painful stimulation during meditation activated the posterior insula and the secondary somatosensory cortex. These are brain regions typically associated with pain intensity—the higher the activation, the sharper the pain. But Gard et al. found the actual opposite: Meditators with higher activation in these pain regions rated their pain as less intense. The best explanation is that pain feels less painful when you carefully zoom in on the exact nature of the experience—what it actually feels like moment to moment—rather than labeling it as “pain” and sticking it into the “unpleasant” category.
A second form of nonjudgmental attention is to let go, that is, to disengage the thinking mind and to give up control—to just be with the pain rather than try to influence it. In support of this hypothesis, Gard et al. found that meditators deactivated part of the prefrontal control system of the brain during painful stimulation.
A third way to be mindful with pain is to dampen (or maybe even completely abandon) your usual emotional response to pain. You could reassess the situation and realize “that all components of the experience of pain are merely mental events, and thus do not necessarily need to be acted upon,”42 This type of reexamination is typically the task of part of the salience network, notably the dorsal anterior cingulate cortex and the anterior insula. Indeed, two studies43 reported an increase of activation in these brain regions during painful stimulation.
These three mechanisms, of course, aren’t incompatible with each other—some may be more at play in some studies, some may be more active in some people than in others, and you can even apply more than one of them at the same time.
On a practical note, this leaves meditators with a nice bag of relatively simple tricks to deal with unavoidable pain: You can focus your attention elsewhere, for instance on the breath; you can get attuned to and relish in the relaxing effects of meditation; or you can take an attitude of openness and acceptance toward the true reality of the experience of the pain as it unfolds, moment by moment.
Earlier I discussed changes in attention and in bodily awareness. This section deals with the third aspect of Tomasino et al.’s map, namely the self. Before I review the studies on how meditation changes the experience of self, I briefly discuss how psychologists see the self (or, rather, the sense or experience of self) and how the brain implements it.
What constitutes the self is still a matter of great psychological and philosophical debate.44 As I briefly discussed in Chapter 2, most psychologists and neuroscientists make a distinction between, on the one hand, the narrative self, that is, the self-as-story, and, on the other, the core self or minimal self, that is, the self-of-momentary-awareness. One difference is that the narrative self extends over time—it has a past, it lives a present, and it projects itself into a future. The core self does not—it just is, right here, right now.
The narrative self is the aspect of the self that is being interrogated at parties (“So, tell me about yourself,” and we know what is expected: You talk about your job, your kids, your spouse, your hobbies, and your recent vacation); it is also the self you consult when you make life decisions (“Should I take that job offer?”—and off you go, dreaming up scenarios, referring to your past experiences, and imagining yourself in new surroundings). The narrative self is crucial in your sense of who you are: It anchors you in your relationships with the world and the people around you, and it connects you with your past and future selves. We can see this, for instance, in people where this sense of self starts to dissolve, as in Alzheimer’s disease: From the inside, there can be a real sense of being lost; from the outside, you likewise can get the sense that you are losing a loved one who no longer understands her connection to you.
The narrative self is also a story in another sense: It is a fiction—clearly, a useful one—that is no more than a bundle of momentary impressions that are strung together by memory and the imagination (to paraphrase Hume) but that gives us a helpful sense of continuity. The classical example to illustrate this is to look back at a picture of yourself as a child. Are you the same person as that seven-year-old in that little turquoise jumpsuit? The only reason you can affirm that that child is you is by pointing at the continuity that links you to him, but if you were to meet that child, here and now, he would likely not feel like he was “you” at all.
The minimal, living-in-the-moment, core self is more basic. After all else is stripped away, after all my memories have faded and all my plans are forsaken, I likely would still have the sense that there is an “I” here that “I” experience, an entity that is distinct from the rest of the world, that has a vantage point anchored in this particular body that “I” recognize as “my” own and is capable of doing things of its own accord— “I” am typing these words with “my” hands, because “I” think them, and these words are “mine.” This self is short-lived; it’s a series of transient selves, a process, born as each of its experiences arises and thus born again with each new experience or, rather, with each shift of attention. The narrative self is likely uniquely human, if only because it seems very much tied to language,45 but the minimal self most probably is not—our dog,46 who is right now nuzzling my hand, likely has as much a sense of ownership of his body and of agency as I have and of me-not-being-him and him-not-being-me as I do, but it’s unlikely he has a complicated story to tell about himself to his dog park buddies.
These two selves are represented in different parts of the brain, as we have seen. The self-of-momentary experience is often seen as located in the thalamus and the brainstem, the somatosensory cortices, and the insula—those parts of the brain that are sensitive to the current state of the body and its interplay with the environment. The self-as-story is associated mostly with activation in the precuneus and the posterior cingulate, regions that are part of the default-mode network. And, as explained already, at a less self-absorbed, maybe more emotional level, this self is associated with activation in the medial prefrontal cortex as well—also part of the default-mode network.
In a very elegant fMRI study, Norman Farb and colleagues47 demonstrated that even beginning meditators are capable of turning off the narrative self. They showed participants (half of whom were trained in mindfulness through an eight-week MBSR program; the other half were complete novices) lists of personality traits, one word every second. Half of the words were positive (like “confident”) and the other half negative (like “melancholy”).
There were two conditions. One was a narrative-focus condition, designed (as the name suggests) to tap the narrative self. In this condition the participants were asked to judge what was occurring in their minds as they were trying to figure out what the words meant to them; they were also explicitly allowed to get caught up in their train of thought. The other condition was an experience-focus condition, designed to tap into the self-of-momentary-awareness. For that condition participants were asked to just sense, without judging, what was occurring in their mind, bodies, and feelings as they read the words, without purpose or goal. If they got distracted by a particular thought or memory, they were asked to gently return to their current experiences.
The narrative-focus condition did indeed yield, as you would expect, activation in the posterior cingulate and the medial prefrontal cortex, as well as in language areas and the hippocampus (a memory structure). In the experience-focus condition, the mindfulness-trained participants showed suppression of the medial prefrontal cortex—in other words, they were deactivating part of the narrative-self network—and an increase in activation in the insula and the secondary somatosensory cortex—areas associated with the self-as-momentary experience. These results confirm that meditators (even with relatively little prior practice) can tune down the narrative self and tune up the experiencing self simply when you ask them to do so.
One very interesting additional and unexpected result was that, in novice meditators, the two selves were correlated—when the self-as-momentary experience was activated, so was the narrative self, and vice versa. This was not the case in the mindfulness-trained meditators: For them, the two systems were decoupled. So, in people who are new to meditating, engaging the narrative self may be an automatic response, a habit. Even a little meditation experience, however, allows people to step out of this habit and free themselves from getting caught up in stories about I-me-mine, at least for the duration of a sit.48
One mechanism that makes the decoupling of the narrative and the core self possible could be the time difference in activation of the two types of self. Two studies49 have shown that when people judge whether a word applies to themselves or someone else, the brain structures that build the self-of-momentary-awareness come online very quickly—after about 150 milliseconds; the narrative self takes about 500 to 800 milliseconds to become operational. Meditators apparently can learn to exploit this gap and stop the activation emanating from the immediate self before it spreads to the narrative self.
This dampening of the narrative self seems to lead to greater happiness in the moment. One study that looked at shifting from a focus on the narrative self to a focus on the self-of-momentary-awareness during meditation showed that this shift was accompanied by a marked decrease of negative and mixed negative/positive emotions.50 This may be directly related to the deactivation of the medial prefrontal cortex and the posterior cingulate.51 Thus narrative-self-less meditating may be one way to (at least momentarily) lift the dark clouds of the mind.
Astonishingly, it is possible to go further still in meditation and silence even the self-of-momentary-awareness, as demonstrated by a clever study by Yair Dor-Ziderman and colleagues.52 They tested 12 long-term practitioners of Vipassanā in an MEG machine.53 (This study is part of a larger study, of which the Berkovich-Ohana and colleagues study mentioned earlier—on timelessness and spacelessness—was another subset.)
As in the Berkovich-Ohana et al. study, participants were asked to cycle through different meditation conditions. The first condition was designed to activate the narrative self (“Try to think what characterizes you”). The second was a minimal-self condition, designed to tap into the self-of-momentary-awareness (“Try to experience what is happening to you at the present moment”). Finally, the third condition was a selfless condition (“Try to experience what is happening at the present moment, when you are not in the center”). The hope here was that this last instruction would deactivate both the narrative self and the self-of-momentary-awareness. Immediately after each bout of meditation, the meditators also told the researchers about their experiences and rated the quality and stability of their meditation.
The contrast between the narrative-self and the minimal-self condition showed a reduction in activation in the medial prefrontal cortex, as well as a more global left frontal deactivation. The contrast between the minimal-self and the selfless condition showed an even further decrease in medial prefrontal activation, as well as a reduction in activation in the precuneus and the inferior parietal lobe. We have met both of these regions before—the precuneus as part of the self-referential network and the inferior parietal lobe as a switch to the default-mode network. Additionally, both regions are known to be involved in the feeling of agency (i.e., the feeling that “I” am the one doing this), in discriminating between self and others, and in differentiation between a third-person and a first-person perspective—all functions of the minimal self. The results here then strongly suggest that meditators can indeed dim down the activation in the core self.
When the core self is turned all the way down, what remains should by definition be selfless. What does this feel like?
Three meditators described their experiences as a general relaxation or quieting of body, thought, or experience (examples include: “less judgmental element; less naming of the experience, less verbally” and “very pleasant and relaxed and quiet. … devoid of effort”). These were also the meditators with the least amount of meditation experience (note that the smallest number of accumulated hours was 1,290, which is still quite respectable).
Three participants with more accumulated practice described altered experiences of their body, their senses, or a disorientation in space (“Like in a dream. Sensations of all kinds of things flickering” and “as if I took a step back and am looking at myself from the back. I see myself but I am also aware of what is happening around”).
Finally, the four most seasoned meditators described experiences without any sense of ownership or agency (“I understood that it was just a sensation, it was not the hand itself, and the sensation was liberated, and so on in other areas. There were jumps of liberation; there was a deep thought that all this was not mine” and “It was emptiness, as if the self fell out of the picture. There was an experience but it had no address, it was not attached to a center or subject. … There was no sense of an object there running the show”). This lack-of-ownership experience seems to be closest to what you would assume selflessness would feel like, and it was the only type of experience associated with a decrease in activation in the inferior parietal lobe, as well as in the thalamus. The thalamus is an ancient brain structure that acts as a hub—it receives input from the eyes, ears, and spinal cord and relays that information to the cerebral cortex; it is also implicated in regulating alertness and sleep. Thus, for this very accomplished group of meditators, this specific experience of selflessness was indeed associated with the dimming of brain regions that are associated with a core, minimal self—as self-less as it could possibly get.
In some forms of Buddhism, notably Zen, the self-less experience, labeled kenshō or satori, is considered the peak experience of the meditative mind, where “the sense of selfhood is dissolved and an ‘unattached, self-less, impersonal’ awareness remains.”54 The Dor-Ziderman et al. study suggests that the self-less meditators (or should we say meditations?) might have been self-less in almost a literal sense. That is, this meditative state was associated with a stripping away of activation in regions associated with the self, not with increased activation in some other area of the brain. I point this out because previous research on spiritual experiences has suggested that there is an on switch for mystical experiences, sometimes nicknamed the “God spot” or the “God module.”55 Here we find the opposite: There is no selflessness, emptiness, or kenshō module—the self-less experience is exactly what it claims to be: the peeling or falling away of the self, the loss first of your own identity, then of identity altogether. It is not the attainment of something new but rather the letting go, the unplugging of your habitual way of viewing the world from the lens of an “I.”
On an interesting side note, all participants in the study felt very successful in their endeavor to meditate selflessly, probably because all of them experienced reduced activation in the precuneus and the inferior parietal lobe, both of which may be associated with some change in inner experience. Clearly, however, only the meditators with a tremendous amount of meditation experience succeeded in actually meditating with only minimal stirrings of a sense of self, although the less accomplished meditators were clearly convinced that they succeeded in doing this as well.
All of the studies described here pertain to the attention-based practices—focused attention (including jhana) and open monitoring.
What about the third type of Buddhist practice—heart practices such as metta practice or compassion meditation?
The sad answer is that we don’t know much about brain activation during the heart practices. I could only find six papers; between them, they reported results from four different studies. Only one of those studies—the study by Brewer and colleagues we have already encountered—provided an explicit comparison with other types of meditation.
In the first fMRI study on heart practices, Antoine Lutz and colleagues56 compared novices and expert meditators. The experts (16 in the 2008 study; data from 10 of them were reanalyzed in the 2009 study) had accumulated between 10,000 and 50,000 hours of meditation experience. All had extensive experience with compassion meditation. They meditated inside the scanner, cycling a total of eight times between 3 minutes of compassion meditation57 and 1.6 minutes of rest. Every 6 to 10 seconds, a sound was played over earphones. The sound could either be neutral (e.g., background noise at a restaurant), positive (e.g., a baby laughing), or negative (e.g., a woman screaming). Participants were simply asked to continue meditating or resting and to ignore the sounds. The researchers, of course, were interested in how the brain reacted to these sounds.
In general, they found that expert meditators activated parts of the default-mode network more than novices did. More specifically, they activated parts of the network that are associated with present-state awareness (temporal parietal junction) as well as the core of the network (precuneus and posterior cingulate cortex). The default-mode network is also often active when people are trying to figure out other people’s intentions and feeling, and Lutz et al. speculate that this is what might be going on here: mentalizing, putting oneself in the place of all the beings one feels compassion toward.
The anterior insula and the anterior cingulate cortex—that is, the salience network—were also more active during compassion meditation than during rest, as was the amygdala—the emotion center of the brain. Interestingly, when emotional sounds were played, the right insular cortex was the only brain area that reacted differently in novices and experts: Only in experts did it react more strongly to negative sounds. This activity was also related with the degree to which meditators indicated they had successfully entered the meditative state: The deeper the meditation, the stronger the activation in the insula. These results suggest that the meditation set up the participants to check out emotional stimuli—perhaps a readiness to empathize with others—and experts in compassion meditation perceived negative stimuli—the sound of a suffering fellow human being appealing to their compassion—as acutely pertinent.
Compassion meditation also led to an increase in heart rate but not breathing frequency. The change in heart rate correlated with the activation in the insula, and especially so in experts. This suggests that compassion practice effectively counteracts some of the usual parasympathetic, calming effect of meditation. Teachers often prescribe heart practices when meditators feel drowsy or sleepy,58 and this may be the reason: It is, indeed, regardless of its emotional impact, an invigorating exercise.
In the second study on heart practices, Maria Engström and Birgitta Söderfeldt59 tested a single highly experienced meditator (who, besides her daily practice, had also participated in two three-year-long traditional Tibetan Buddhist retreats). In the scanner, she cycled three times through 30 seconds of compassion meditation, accompanied by a mantra, and two periods of repeating sentences. Like Lutz and colleagues, the researchers found activation in the anterior cingulate and the right insula, as well as in the right caudate. The caudate is implicated in, among other things, processing of emotions.
In the third study, Tatia Lee and colleagues60 examined 12 meditators with, on average, about 7,500 hours of compassion-meditation practice. During meditation, the participants viewed a set of 20 neutral, 20 happy, and 20 sad pictures. They rated how emotional they thought each picture was. The main finding was that experts who were viewing sad pictures activated the left medial frontal gyrus and the left caudate more strongly than novices did; while they were viewing happy pictures, experts showed higher activation in the left anterior cingulate, the right medial frontal gyrus, and the right precuneus. This suggests that compassion meditation experts react more strongly to a display of sad emotions (anterior cingulate) and identify more quickly with happy sentiments (left caudate). In both cases, they were also more efficient in regulating these emotions (the middle or inferior frontal gyrus).
A different picture, however, emerges from the fourth study, by Judson Brewer and colleagues61 (12 meditators with about 10,000 hours of practice). In this study, loving-kindness meditation specifically deactivated the amygdala and the hippocampus. The hippocampus tends to be associated with, among other things, the retrieval of memories and planning for the future. Its deactivation suggests that meditators were on focus, that is, not mind-wandering, and likely less self-centered. The amygdala is intimately linked with the processing of emotions. Its deactivation might mean that emotional stimuli in general are temporarily shut out of awareness—a conclusion opposite to that of Lutz and colleagues. It could, however, also mean that the brain’s alarm system calms down as the meditator infuses himself with the emotions and sensations of loving-kindness (the amygdalae are notable for reacting very quickly to fearsome stimuli62).
Finally, Kathleen Garrison and colleagues63 replicated and extended the results from the Brewer et al. study in a group of 20 expert meditators (with about 10,000 hours of accumulated practice, on average). Specifically, they found that experts showed lower connectivity between the inferior frontal gyrus and the posterior insula and the rest of the brain. The inferior frontal gyrus is often implicated in emotional processing and empathy,64 and so Garrison et al.’s result suggests that experts engage in less emotional processing during loving-kindness meditation than novices do.
How to summarize these diverse results?
For one, it is clear that, unlike the two attention-based practices, loving-kindness and compassion meditation have a clear impact on emotion-related structures in the brain. There is an interesting discrepancy here, however: Three studies show a heightened sensitivity or receptivity to emotions; two show a decrease in emotional processing. I think the key difference is that the second set of studies looked at loving-kindness/compassion meditation as it occurred, undisturbed. The first set of studies, in contrast, interrupted the meditation with emotional stimuli—sounds of distress or joy, or pictures of happy or sad events. The differences in emotional processes between the two sets of studies might be telling here: In the first set of studies, the salience system was engaged, as was the system related to processing of the self, but this was not the case in the second set of studies.
What might be happening, then, perhaps, is that during uninterrupted loving-kindness meditation the self-related network and the salience system are allowed to go silent because the object of meditation is clear and predictable; that is, there is little room for mind-wandering. When emotional sounds or pictures puncture your loving-kindness meditation, however, the salience system reacts swiftly, and the connection between the perceived suffering and yourself is quickly established. Thus loving-kindness meditation appears to be both a practice that is cool and composed and a practice that primes you for compassion and loving-kindness, if the need to exercise those would arise.
No one is born a meditator—it takes time, effort, and dedication to develop the skill. Most (but not all) of the studies I describe in this chapter were done on highly accomplished meditators, with often tens of thousands of hours of practice under their belts (or robes, in many cases). Given that there are few studies of less accomplished meditators (people like me, and maybe you), it is hard to trace the development of meditation expertise from monkey mind to monk’s mind.
Here is what we do know.
First, we know that, over time, meditators develop the skills associated with the type of meditation they engage in.
In her meta-analysis on attention practices, Tomasino et al. also looked at differences between long-term and shorter term meditators (the dividing line was at 5,000 hours of total accumulated practice—quite a lot). They found that long-term meditators activated the attention brain circuits to a lesser extent than shorter term practitioners did, suggesting that as the meditation practice matures, less effort is needed to sustain a sitting.
Tomasino et al.’s contrast was between studies; that is, they contrasted studies in which the average level of expertise was high with those where it was lower. A potential problem here is that studies differ in many aspects—maybe the studies with long-term practitioners used simpler meditations and thus placed lower demands on the yogi’s attentional systems.
When we look at studies that include individuals of different levels of expertise, we find a more mixed pattern. Antonietta Manna and colleagues65 found the same result as Tomasino et al. when comparing monks with about 16,000 hours of practice with novices—monks showed decreased activation in attention structures. Baron Short and colleagues66 found the opposite result when looking at lay meditators (splitting the sample in meditators who had an accumulated record of less than or more than about 2,000 hours)—increased expertise leads to increased activation in attention structures. Julie Brefczynski-Lewis and colleagues67 found the solution to the riddle: They obtained both results when they split their group of meditators into three—experts (19,000 hours of accumulated practice) showed stronger activation than novices, but experts with 44,000 hours of accumulated practice showed less activation than novices. What seems to be happening, then, is that meditators initially learn to build up attentional effort (and this takes many years); when they have done so, they slowly become more efficient. Part of the growing efficiency is that expertise builds stronger connections within the attention network, as we will see in a bit more detail in the next chapter. Likewise, experts show stronger couplings between the attention network and the default-mode network, which makes it easier to suppress mind-wandering, as Hasenkamp and Barsalou found.68
Part of this fine-tuning of attention might be developing a better sense of the sometimes subtle signals that tell you that the meditation is going where it is supposed to go. Earlier I discussed the findings from Kathleen Garrison et al.’s study where research participants were shown a real-time graph of their PCC activity and were asked to meditate themselves into either the red or the blue zone of PCC activation or deactivation. An important additional finding from this study was that only long-term meditators were able to do this; novices were not. Expert meditators may thus be more sensitive to nuances in their inner states, connect those with the graphs, and use this knowledge to guide them toward deeper or shallower states of concentration, as the task required. This result also casts doubt on some of the claims made by commercial enterprises (I won’t name any here) that meditation can be taught through biofeedback. The reason for the doubt is twofold: One is that collecting reliable default-mode network signals is hard to do outside the scanner69; the other is that you might need a lot of experience with meditation before you can use such signals to your advantage. We do know that biofeedback can teach you to relax,70 but meditation is much more than chilling out.
A second finding is that, over time, there seems to be a trend toward disembodiment and selflessness. Tomasino et al.’s meta-analysis found that long-term meditators show more activation in the supplemental motor areas and in the superior medial gyrus than less experienced meditators—evidence for a growing disembodiment with longer practice. Previously I cited the results from the study on selflessness by Dor-Ziderman and colleagues, which suggests that true selflessness, as characterized by a lack of ownership over experiences, can be reached only after an extremely large amount of practice. Both of these outcomes of meditation are side effects—they are not the actual goal of practice (at least not in these traditions). This perhaps implies that they are a natural outflow of the amount of time spent in meditation. Tomasino et al. describe this as a strategy shift: Seasoned meditators might focus less on controlling their attention and instead concentrate more on disembodiment. This might fit with the (non-)strategy of “no-efforting” that Garrison et al. found in their expert meditators.
Third, there is also some evidence that more experienced meditators take less of a judgmental, evaluative, or emotional stance in meditation. Two studies support this idea. First, Manna et al. found that novices simultaneously activate the anterior cingulate (part of the salience network) and the lateral orbitofrontal cortex during open monitoring. The lateral orbitofrontal cortex is involved in affective and cognitive evaluations, and so it appears that either whatever you are monitoring is being evaluated, or you check in on your evaluations, or both. This coupling was absent in expert meditators, suggesting that experts let go of their evaluative mindset. Second, Brefczynski-Lewis et al. presented emotional sounds toward the end of short meditation periods. They found a strong negative relation between accumulated number of hours of practice and activation in the amygdala (a region associated with gut emotionality) and the posterior cingulate cortex (part of the default-mode network), suggesting less emotional reactivity and less distraction in highly accomplished meditators.
Fourth, over time, meditators become more meditative in daily life—or at least they might use idle moments as meditative opportunities. This isn’t really something researchers were looking for, so it’s a bonus finding. They happened to stumble across this result when comparing the brains of meditators with those of nonmeditators when both were lying idle inside the scanner or the testing room and asked to do nothing in particular. This is illustrated in three examples.
First, in an EEG study, Rael Cahn et al.71 found that long-term Vipassanā meditators (with, on average, 19 years of practice) showed just as much alpha power in their brainwaves during rest as during meditation, and this level was higher than that of meditators with fewer years of practice (2.5 years, on average). As we have seen, meditators likely generate these alpha waves during meditation as they turn off distracting thoughts. The new finding is that experts also generate those waves when they are just sitting around, not doing anything in particular. Cahn et al. interpret this as a trait difference: The meditative habit starts seeping into every aspect of daily life; it becomes part of your personality. Another interpretation is that meditators simply slip into meditative states whenever there isn’t anything in particular that needs their attention—like when they are waiting for the researchers to start up their experiment.
Second, in their fMRI study, Judson Brewer and colleagues found a strong coupling between the default-mode network and the salience network in long-term meditators, regardless of what type of meditation they were engaged in. But this didn’t happen only during meditation: The same coupling was found during a rest period. As I mentioned, Hasenkamp and Barsalou obtained a similar result: Longer meditation practice was associated with a stronger coupling between the default-mode network and the attentional network at rest. These two studies suggest that the meditative state carries over, with meditators bringing more attention to their inner states, at least when resting inside the scanner.
Third, Antonietta Manna and colleagues found that the brain patterns of Buddhist monks during rest resembled those of open-monitoring meditation but not focused-attention meditation; novices’ brains did not look meditative at all. The conclusion here would be that monks practice open monitoring also in nonmeditative conditions or, again, at least when waiting inside the scanner.
The three key findings concerning the long-term development of meditation—meditation becomes less effortful over time, during meditation meditators become more self-less as well as less judgmental, and the meditative state increasingly spills over into nonmeditative moments—suggest that long-term meditation (or maybe even short-term meditation) might leave lasting imprints on the brain and likely leaves traces on behavior off the cushion as well. We explore those possibilities in the remainder of this book.
A final note on this: None of the studies cited here suggest that there are cut-offs or stages in meditation practice. Rather, when researchers plot activation (or some other measure of how meditation is implemented in the brain) as a function of accumulated practice, the plots all look wonderfully continuous even if—in the attention case—curved. There are, as far as we know, no steps, no plateaus, no sudden jumps. Thus meditation expertise (like just about any other form of expertise) is built gradually, and also—another fascinating point to consider—it never ends. The Buddha himself, after his peak experience at age 35, kept practicing for another 45 years—tradition has it that he died while meditating, at age 80.
There is one exception to this pattern, as we have seen, and that is the experience of selflessness,72 which, in its form of loss of ownership, is something that does look like a qualitative jump and one that can only be made after a tremendous amount of practice. This may be one reason why this experience, in some Buddhist traditions, is considered the hallmark of awakening.