Accepted Manuscript

Event-related potential correlates of mindfulness meditation competence

Rachel Atchley, Dan Klee, Tabatha Memmott, Elena Goodrich, Helané

Wahbeh, Barry Oken

PII: S0306-4522(16)00079-8
DOI: http://dx.doi.org/10.1016/j.neuroscience.2016.01.051
Reference: NSC 16869

To appear in: Neuroscience

Accepted Date: 23 January 2016

Please cite this article as: R. Atchley, D. Klee, T. Memmott, E. Goodrich, H. Wahbeh, B. Oken, Event-related
potential correlates of mindfulness meditation competence, Neuroscience (2016), doi: http://dx.doi.org/10.1016/
j.neuroscience.2016.01.051

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 1

Title: Event-related potential correlates of mindfulness meditation competence

Authors: Rachel Atchley, PhD; Dan Klee, BS1; Tabatha Memmott, BA, BS1; Elena

Goodrich, MDiv1; Helané Wahbeh, ND, MCR1, and Barry Oken, MD, MS1.

Author Affiliations: 1Oregon Health & Science University, Portland, OR USA

Corresponding Author Information:

Name: Rachel Atchley, PhD

Address: Oregon Health & Science University

3251 SW Sam Jackson Park Rd

Hatfield Research Center, CR-120

Portland, OR 97239

Email: [email protected]

Phone: 503-494-5650
 2

Abstract

Objective: This cross-sectional study evaluated event-related potentials (ERPs) across

three groups: naïve, novice, and experienced meditators as potential physiological

markers of mindfulness meditation competence.

Methods: Electroencephalographic (EEG) data was collected during a target tone

detection task and a Breath Counting task. The Breath Counting task served as the

mindfulness meditation condition for the novice and experienced meditator groups.

Participants were instructed to respond to target tones with a button press in the first

task (Tones), and then ignore the primed tones while breath counting. The primary

outcomes were ERP responses to target tones, namely the N2 and P3, as markers of

stimulus discrimination and attention, respectively.

Results: As expected, P3 amplitudes elicited by target tones were attenuated within

groups during the Breath Counting task in comparison to the Tones task (p < .001).

There was a task by group interaction for P3 (p = .039). Both meditator groups

displayed greater change in peak-to-trough P3 amplitudes, with higher amplitudes

during the Tones condition and more pronounced reductions in P3 amplitudes during

the Breath Counting meditation task in comparison to the naïve group.

Conclusions: Meditators had stronger P3 amplitude responses to target tones when

instructed to attend to the tones, and a greater attenuation of P3 amplitudes when

instructed to ignore the same tones during the Breath Counting task. This study

introduces the idea of identifying ERP markers as a means of measuring mindfulness

meditation competence, and results suggest this may be a valid approach. This
 3

information has the potential to improve mindfulness meditation interventions by

allowing objective assessment of mindfulness meditation quality.

Keywords: Meditation, mindfulness, event-related potentials, attention

Mindfulness meditation is an effective complementary medicine technique in the

treatment of anxiety, depression, and chronic pain (Kabat-Zinn et al., 1992; Miller et al.,

1995; Teasdale et al., 2000; Zeidan et al., 2010; Chiesa & Serretti, 2010; Grossman et

al., 2004). However, much remains to be learned about the neurophysiological

mechanisms underlying the benefits of mindfulness meditation (Ospina et al., 2007;

Wahbeh et al., 2008), despite reported changes in electroencephalography (EEG) and

magnetic resonance imaging (MRI) (Chiesa & Serretti, 2010; Fox et al., 2014;

Dickenson et al., 2013; Kilpatrick et al., 2011; Cahn & Polich, 2006). A more

straightforward EEG marker could enable researchers to more accurately assess the

efficacy of mindfulness meditation trainings, as more and more meditation interventions

are emerging. More accurate assessment would further elucidate the brain mechanisms

in mindfulness meditation, assist in the development of effective dose in clinical trials,

and facilitate mindfulness meditation training in the future.

Mindfulness meditation competence, for the purposes of this experiment, is

delineated as the ability to attend to breathing while ignoring other stimuli. EEG is

appealing way to assess brain changes in meditation because it allows for brain activity

to be inexpensively, noninvasively, and unobtrusively recorded. Considering that

mindfulness meditation can improve aspects of cognition (Brefczynski-Lewis et al.,

 4

2007; Lutz et al., 2009), EEG may be of additional benefit since it can evaluate these

changes.

The aim of this study was to identify an electrophysiological correlate of

mindfulness meditation competence using event-related potentials (ERPs). To achieve

this aim, we investigated brain changes with EEG during mindfulness meditation

practice among participants with differing levels of experience: naïve, novice, and

experienced.

Jon Kabat-Zinn’s (1994) definition of mindfulness is “paying attention in a

particular way: on purpose, in the present moment, and non-judgmentally.” We

examined the ERP correlates of “paying attention”, not the motivational or emotional

processes that accompany paying attention. We acknowledge that there are many

definitions and components of mindfulness. In this particular experiment we are

delineating the ability to keep awareness in the present moment because we were

interested in developing a biomarker for the focused attention aspect of mindfulness.

During mindfulness meditation practice and interventions such as mindfulness-based

stress reduction and mindfulness-based cognitive therapy, participants both practice

mindful attention to the present moment during daily living and cultivate concentrative

meditational techniques. The Body Scan and Sitting Meditation, which are the two

primary practices in the aforementioned secular interventions, are both concentrative

techniques. Mindfulness meditation competence must be associated with a cognitive

surrogate in order to be evaluated, and attention is an ideal marker given the results of

past studies (Brefcyznski-Lewis et al. 2007; Chan & Woollacott 2007; Chiesa et al.,

2011; Lutz et al., 2008). We used a breath counting task because non-meditators can
 5

do it and also because it has shown potential as an index of mindfulness (Levinson et

al., 2014; Milz et al., 2014).

ERP components are predictable patterns of brain activity that occur in response

to specific stimuli, and we expected two components in particular to display meditation

experience-related differences during a quiet, potentially meditative state. The first was

the P3, which generally indicates stronger attentional focus on a particular stimulus. We

predicted that meditators would have greater attentional expertise, as observed in

earlier work (Brefcyznski-Lewis et al. 2007; Chan & Woollacott 2007; Chiesa et al.,

2011; Lutz et al., 2008). Thus we hypothesized that meditators would also have greater

P3 amplitudes than non-meditators, as previously shown (Lutz et al., 2009). Differences

in attentional focus and engagement have been previously observed between

meditators and non-meditators (Cahn & Polich 2009; Chiesa et al. 2011; Delgado-

Pastor et al. 2013; Tang et al., 2009), although the relationship between meditation

training and attentional priming effects remains unknown. We also predicted that P3

amplitudes would differ between experienced and novice meditators, expecting that

participants with the most meditation experience would also have the most attentional

control.

The second component of interest in this study was the N2, which is elicited by

stimuli similar to those that elicit the P3, specifically those that involve greater task-

related attention or novelty (Luck, 2014). The N2 is elicited in response to target tones

in an oddball task. We hypothesized that N2 amplitudes in the Tones task would be

greater in meditators due to meditation-related attentional training. Specifically,

experienced meditators were expected to have the most pronounced difference from
 6

naïve participants while novices’ performance would fall in the middle. To test these

hypotheses, we compared the N2 and P3 components of experienced, novice, and

naïve meditators during an odd-ball paradigm, initially while participants were instructed

to attend to the target tones and then while they were told to ignore the tones and focus

on their breathing. We were particularly interested in the effects of primed ERP

responses to target tones during the Breath Counting task and how these responses

differed from those in the Tones task.

Methods

Participants. Forty-two participants were recruited from the Portland metropolitan

area utilizing internet-based, flyer, and word-of-mouth strategies. All underwent a 30-

minute telephone screening. Inclusion criteria for participants in all groups were as

follows: 1) Age 25-75 years; 2) Good general past and present medical health; 3) Stable

on all medications for at least two months; and 4) Cognitively intact, as determined by a

score of ≥31 on the Telephone Interview for Cognitive Status (TICS) (Welsh et al.,

1993). Exclusion criteria were as follows: 1) Significant medical or neurological

disorder/disease; 2) Significant visual or hearing impairment; 3) Medications that would

affect outcome measures (e.g. benzodiazepines or neuroleptics); and 4) Significant

untreated depression, as determined by a score of ≥10 on the 10-item Center for

Epidemiologic Studies Depression Scale (CESD-10) (Andresen et al., 1994).

The naïve, novice, and experienced groups were carefully defined, and each

subject’s telephone screening was thoroughly reviewed by the principal investigator.

Information about lifetime hours of practice was collected via verbal self-report.

Meditation was defined as any practice of attention training, in which one is intentionally
 7

working to improve his or her ability to pay attention to the present moment. Further

details about the types of practice are described by group below. Mind-body movement

practices were defined as any activity in which one is practicing attention training. Mind-

body movement-based practices included yoga, tai chi, martial arts, qi gong, mindful

eating, walking, running, and swimming. The practices included foci of attention were

reported to be breath, physical sensations, and sound (Table 1).

The three groups were evaluated by specific criteria. Potential naïve subjects

were excluded if they had taken a meditation or other mind-body movement class within

the past two years or if they had a daily meditation practice of five minutes or longer in

the past 30 days. Novice subjects were included if they had some formal training,

meaning they attended classes that were described and approved by the principal

investigator, practiced at least three days a week, and had ≤1,000 hours’ lifetime

experience of combined meditation and mind-body practices. All novice meditators took

a meditation class and continued practicing meditation on the breath, as well as a

myriad of other methods, including the body scan, compassion practices, and

mindfulness of sounds. To qualify as an experienced meditator, a subject needed

≥5,000 lifetime meditation and mind-body movement practice hours. All had ≥4,000

hours of meditation experience, and all self-identified as either Tibetan or Zen

Buddhists. Experienced meditators reported many different types of meditation practice

in addition to those practiced by the novices, including but not limited to the following:

mantra and prayer recitation, visualization, analytical meditation on philosophical topics,

non-conceptual awareness, and open awareness. The protocol for this study was

approved by the OHSU Institutional Review Board. Participants attended a 3.5-hour

 8

laboratory session where they were connected to physiological data collection

equipment and instructed in the tasks described below.

Our experiment was designed to measure changes in brain activity during Breath

Counting meditation and assess priming as a means of attentional focus; therefore, our

primary measures were the attentional N2/P3 ERP components and P3b specifically in

response to infrequent target tones across tasks. Given the age of our sample (M age =

49 years), it is worth noting that P3a and P3b tend to overlap as age increases (Alperin

et al., 2014) and that general P3 anteriorization occurs as age increases (Kopp et al.,

2014; Oken & Kaye 1992).

Tones Task. For the first cognitive task, participants were seated with eyes

closed for 15 minutes and asked to press a button when they identified an infrequent

high-pitched target tone (2000Hz). Participants also heard infrequent, non-target low

tones (low-500Hz) and frequent standard tones (1000Hz). Tones were presented at

1500-2500ms second intervals. In each block of 10 tones, there was 1 high frequency

target tone, 1 low frequency distractor tone, and 8 standard tones. The minimum

number of standard tones between each target tone was 2. On average, there were 45

target tones presented in 15 minutes. Participants practiced this task before beginning

the experiment.

Breath Counting Task. To enable some experimental matching of state between

meditators and non-meditators, there was a Breath Counting session with the eyes

closed. This task was described simply as breath counting to naïve participants and as

a breath counting meditation for meditators. The rationale for using Breath Counting as

a mindfulness meditation task is that it is easy to comprehend and attempt by

 9

participants in all three groups, and all novice and experienced meditators had

experience with meditation focusing on the breath. The same tones were presented in

the background during the Breath Counting task as during the Tones task, but

participants were instructed to ignore the tones and keep track of their breath count.

The prior exposure to the tones served as a prime.

Importantly, the purpose of the Breath Counting task is to allow for between-

subjects comparisons of priming effects during a task that could be performed by both

meditators and non-meditators and also be considered a meditative state. Participants

were instructed to press a button if they lost count of their breaths and continue

counting using their best estimate of their breath count to that point. The Meditation

Depth Index (MEDI) (Piron, 2001) and Stanford Sleepiness Scale (SSS) (Herscovitch,

1981) were administered after the Breath Counting session. The MEDI assessed the

subjective quality of attentional focus to breath counting. The Stanford Sleepiness Scale

was administered after the Breath Counting sessions to assess potential differences

that might be attributed to drowsiness from prolonged eye closure, although none were

found. Additionally, we calculated theta and alpha power for one-second epochs prior to

all stimulus tones to evaluate drowsiness.

Physiological Recordings. EEG was collected using a BioSemi 32- channel

ActiveTwo system using passive-sensor Ag-AgCl electrodes (BioSemi, Amsterdam, the

Netherlands). Electrodes were placed according to a Modified Combinatorial

Nomenclature (MCN) at AF3/4, F7/8, F3/4, FC1/2, FC5/6, T7/8, C3/4, CP1/2, CP5/6,

P7/8, P3/4, P03/4, O1/2, Oz, Fz, Cz, and Pz. All EEG data were acquired at a sampling

rate of 1024Hz with a resolution of 24 bits. Amplification occurred with a bandwidth of

 10

0.1-70.0Hz. Online recordings referenced a common mode sense (CMS) and driven

right leg (DRL) electrodes, halfway between Cz and C3/4.

Respiration has been shown to be important for meditation in general (Ahani et

al., 2014) and also important for the specific nature of the Breath Counting meditative

task. It was measured with a light elastic piezoelectric belt (Ambu-Sleepmate, Maryland)

around the participant’s chest near the diaphragm. Respiration rate was analyzed in

BrainVision Analyzer (Version 2.0.1.3931, Professional Edition). Breaths were labeled

semi-automatically using a voltage trigger to label peak values. Specifically, a new

breath was counted whenever the respiratory voltage reading crossed zero in a positive

direction.

Electrophysiological Analyses. Data were analyzed from the Fz, Cz, and Pz

midline electrode sites. During offline processing data were re-referenced to linked

mastoids. Data were filtered offline using a 70Hz low pass filter, a 0.1Hz high pass filter,

and a 60Hz notch filter and analyzed in BrainVision Analyzer. Visual, semi-automatic

artifact rejection was employed to remove data segments contaminated by non-

neuronal activity such as muscle activity, electrode malfunction, or abrupt head

movements. Criteria for semi-automatic artifact rejections were as follows: maximal

allowed voltage step of 50µV/ms; maximal allowed absolute difference 125µV within

interval length of 50ms; maximum EEG amplitudes allowed was +/- 75 µV and the

lowest allowed EEG activity was .5µV within interval length of 100ms. The EEG activity

200ms before and 200ms after artifact events were eliminated.

To generate the ERPs for processing of the N2 and P3 peaks, data were

segmented -200ms before to 1000ms after stimulus presentations. The tone types were
 11

then separated using distinct markers, averaged, and baseline corrected using 200ms

before stimulus presentation. A semi-automated peak detection tool was implemented

using these intervals N2 (200-300ms) and P3 (275-450ms). All peak estimates were

subject to visual inspection, and temporal windows were adjusted and reapplied if

necessary. These criteria were decided on a case-by-case basis for target tones. Inter-

rater reliability was achieved by having two researchers separately view the data and

agree upon window settings/peak identifications. Automated peak analyses revealed

one naïve participant with no discernible N1, P2, or N2 during Breath Counting,

although there was no issue with this person’s P3. Two experienced meditators had no

discernible N2 or P3 during Breath Counting. In these three cases, automated

approximate values were kept in an effort to represent a complete, objective data set.

Following detection, discrete peak voltage values were exported from Brain Vision

Analyzer for further processing. Final N2 and P3 values were computed as the change

in voltage from the previous peak estimate (i.e., N2final = P1raw – N2raw, and P3final =

P3raw – N2raw). This computation was intended to control for deceptive individual

differences in absolute peak amplitudes resulting from general positive or negative drift

across the multiple ERP components.

EEG frequency analysis was utilized to assess drowsiness and thus ensure ERP

differences were not simply related to differences in levels of alertness. The analysis

was performed on one-second epochs prior to all stimulus tones using the same pre-

processing and filtering techniques as the ERPs. A periodic Hanning window was

applied with a tapering of 20%. The frequency spectrum was generated using the Fast

Fourier Transformation (FFT). The resulting epoch FFTs were averaged and exported,
 12

then the square root of power in frontal-midline theta (4-8 Hz) and posterior alpha (8-13

Hz) were calculated. The square root transform was used to improve normality of FFT

data (Oken & Chiappa, 1988).

Statistical Analyses. Results were assessed using mixed design ANOVAs for N2 and

P3 amplitudes and total task respiration counts (SPSS v.21). Mixed 2 (Tones or Breath

Counting, within subjects) by 3 (naïve, novice, or experienced meditator, between

subjects) ANOVAs were conducted to assess changes in event-related potentials by

task and by group. Univariate ANOVAs were used to assess N2 and P3 difference

values by group. Independent t-tests were conducted to follow up on potential group

differences in these measures as well as the MEDI and SSS. A separate 2x2 mixed

ANOVA that pooled novice and experienced meditator groups was also calculated.

Results

The overall sample had an underrepresented racial and ethnic representation of

7% African American, 7% Asian American, 2% Native American, 2% Hispanic, and 5%

of more than one race (Table 1). A one-way ANOVA determined there were no

significant demographic differences among the groups. Novices had been practicing

meditation and/or mind-body activities for an average of 2.4 (SD = 2.5) years and

currently practiced 12.6 (SD = 7.7) minutes per day. Experienced meditators had been

practicing meditation and/or mind-body activities for an average of 22.6 (SD = 13.1)

years and presently practiced an average of 122.1 (SD = 180.9) minutes per day. High-

pitched target tones were the focus of ERP analyses from the Fz, Cz, and Pz sites

because we were looking for N2/P3 priming effects from the Tones condition to the

Breath Counting condition (see Table 2 for task summary). The non-target infrequent

and standard tone results as well as the N1/P2 components are discussed separately.
 13

Please note that one novice participant was excluded from Pz analyses due to electrode

failure at that particular site.

P3 ERP. IN all three groups there was a main within-subjects effect of task on P3

amplitudes elicited by infrequent target tones, F(1,39) = 58.02, p < .001, with primed P3

amplitudes attenuated during the Breath Counting task (M = 5.12µV) compared to the

Tones task (M = 5.96µV) across the Fz, Cz, and Pz midline electrode sites. There were

no significant differences between the novice and experienced groups, so in order to

further assess the effects of meditation experience, we pooled novice and experienced

meditators into one meditation group and repeated the above analyses. For the pooled

meditator group, there was a priming effect on P3 amplitude following infrequent target

tones at the Fz site: F(1,40) = 42.34, p < .001; Cz site: F(1,40) = 34.31; p < .001; and Pz

site: F(1,40) = 42.23, p < .001. There was an interaction for target P3s at the Cz site as

well, F(1,40) = 4.56, p = .039 (Figure 1), with meditators showing a greater change in

peak-to-trough P3 amplitudes from the Tones task to the Breath Counting task. See

Figure 2 for average ERP waveforms elicited by target tones at the Cz site. Figures 3

and 4 contain summaries of topographic distributions and general ERP amplitudes

across Tone conditions.

N2 ERP. With three groups there were interactions for N2 responses to target

tones at the Cz site, F(2,39) = 3.96, p = .027, and Pz site, F(2,38) = 3.25, p = .05. With

pooled meditators, these group interactions were still present at the Cz site, F(1,40) =

6.25, p = .017, and Pz site, F(1,39) = 4.32, p = .044, for high/infrequent target tones

(Figures 5 & 6). At the Cz electrode site the effect was more pronounced, with

meditators again showing a greater difference in N2 peak-to-trough amplitudes from the

 14

Tones task to the Breath Counting task. In both cases meditators on average had

greater N2 amplitudes during the Tones task and more attenuated N2 amplitudes during

the Breath Counting task in comparison to naïve participants.

Exploratory ERPs. This section will cover results for N1 and P2 ERPs as well as

responses to low infrequent non-target tones and frequent standard tones. These

analyses are exploratory, as N2/P3 responses to target tones were our primary

analyses. Thus, only group effects are of interest with these variables. We will report

pooled group results (meditators versus naïve participants) as they were the strongest.

Additional analyses on standard and low tones for P3 revealed no group effects

or interactions regarding low or standard tones. For N2 at the Pz site, there was a

significant group effect in which meditators (M = -5.98µV) had higher amplitude

responses to low tones than naive participants (M = -3.46µV) during the Tones task.

Meditators (M = -6.43µV) also had higher amplitude N2s than naïve participants (M = -

4.27µV) during Breath Counting. There were no significant results for N2 responses to

standard tones.

For N1, there was a general reduction in amplitude from the Tones task to the

Breath Counting task for all tone types (standard, high, low) at all electrode sites of

interest (Fz, Cz, Pz), all p values < .005, with the exception of Pz low tones, for which

there was no task effect. A group difference in N1 amplitudes between naïve

participants and meditators was observed for low tones at the Cz midline site, F(1,40) =

5.94, p = .019. During Tones, meditators (M = -11.67µV) exhibited higher amplitude N1s

than naïve participants (M = -6.83µV).

 15

P2 results were more scattered. P2 amplitudes went down from Tones to Breath

Counting but only for standard tones across Fz, Cz, and Pz, with all p values < .008.

Like N1, there was a group effect for low tones at the Cz site, F(1,40) = 4.53, p = .040,

but the effect was for the Breath Counting task as meditators (M = 15.72µV) exhibited

higher amplitude P2s in comparison to naïve participants (M = 11.39µV). ERP

waveform graphs are included for standard and low tones (Figures 7 & 8). These results

will not be discussed further since they were exploratory and separate from our original

research question.

Respiration. Respiration data were assessed to supplement ERP results. Two

(Tones or Breath Counting, within subjects) by 3 (experienced or novice or naïve

meditator, between subjects) mixed ANOVAs revealed a main effect of task on total

automated respiration count for the Breath Counting session after the Tones task,

F(1,41) = 51.78, p < 001. Total respiration count decreased during the Breath Counting

task (M = 128.67) in comparison to the Tones task (M = 203.70). These tasks were

equivalent in length at 15 minutes. When the tasks were further divided into three

equivalent segments, there were significant respiration differences between the tasks at

each time point for the first segment, F(1,41) = 49.10, p < 001, second segment, F(1,41)

= 34.75, p < 001, and final segment, F(1,39) = 40.79, p < 001. Overall, there were no

interactions or group effects.

Drowsiness. There were no differences between groups after the Tones or

Breath Counting tasks in the frequency content analyzed. We also assessed theta/alpha

ratios to assess drowsiness in an objective way using a 2 (task, Tones or Breath

Counting) by 3 (group, naïve, novice, or experienced meditator) mixed ANOVA. Again,

 16

there were no differences across groups in drowsiness measures. When theta and

alpha levels were measured separately however, there was a significant task by group

theta interaction, F(2,36) = 5.12, p = 01. When broken down by t-tests, we see that

naïve participants had greater theta (Theta M = .94) than experienced meditators (Theta

M = .54) during the Breath Counting task, potentially suggesting a less vigilant state

(Oken et al., 2006), although drowsiness does not explain the N2/P3 difference between

groups.

Meditation Questionnaire. Finally, we assessed group differences on the

(Meditation Depth Index = MEDI) using t-tests. After the Breath Counting session, naïve

participants reported the lowest scores (M = 24), novices were in the middle (M = 27),

and experienced meditators had the highest MEDI scores (M = 28). The difference

between the naïve participants and experienced meditators was significant, p = .024,

and the naïve/novice difference bordered on significance, p = .054.

Behavioral Measures. We assessed accuracy on the Tones task and found a

ceiling effect. The average correct identification rate for target tones was 98% for

meditators and 97% for naïve participants, as measured by button press responses.

Misses due to time outs and false alarms were both taken into account in error rates.

There were no significant group differences in any of these factors. We did not assess

reaction time for the Tones task because participants were not instructed to press the

button as quickly as they could. One naïve group participant had to be excluded after

behavioral analyses revealed that they did not follow instructions. This person

responded to the infrequent low tones as targets in addition to high tones as targets,

thus their data had to be removed from all ERP analyses since the goal of the priming
 17

task was compromised. Importantly, P3 data were only calculated from correct

hits/button presses on the Tones task.

Discussion

This experiment attempted to identify an objective measure of mindfulness

meditation competence through event-related potential indices of attention. To

accomplish this, we compared N2 and P3 component responses to target tones when

they were relevant versus irrelevant in order to capture mindful cognitive states across

naïve, novice, and experienced meditation groups. The Breath Counting session

accomplished a controlled state that was meditative as assessed by the self-rated MEDI

scale.

Previous work strongly suggests that attentional control increases with meditation

experience (Cahn & Polich, 2006). On average, meditators in this experiment had

greater N2 and P3 amplitudes than non-meditators while attending to targets during the

Tones task, and greater attenuation of N2 and P3 while engaged in the breath counting

meditation and ignoring the primed tones. This strongly suggests that meditators had

greater attentional control both while they were specifically attending to target tones as

well as while ignoring them during the Breath Counting meditative task.

The lack of differences between the novice and experienced groups might

indicate that proficiency in the attentional training aspect of mindfulness can be

achieved relatively early on in practice, although the benefits of longer term meditative

practices may not be sensitive to this particular attentional aspect of mindfulness, or

perhaps the group definitions and/or sample sizes weakened the results. The findings

presented here are not conclusive on this matter. The novices in this experiment were
 18

actually quite practiced in meditation with an average of 2.4 years of experience, and

some novices had rather extensive formal training. Perhaps greater differences would

be more apparent with less experienced novices.

Drowsiness was also an a priori concern in all of our groups since drowsiness

attenuates P3, and increased theta is associated with drowsiness. However, our

drowsiness analyses provided some evidence that the decreased P3 amplitude in

meditators during Breath Counting is not due to drowsiness.

This study had some limitations. All novice meditators reported practicing

meditation on breath, and for most, it was their primary practice. In contrast, meditation

on breath was just one among many practices for the experienced meditators. An

experienced Zen meditator participant also provided valuable feedback that the Breath

Counting meditation was essentially a dual performance task since, in Zen meditation,

eyes are usually open without a focus on breath counting. Thus, although we were

constrained by the experimental need to create a replicable and simple meditation task,

Breath Counting may not have captured the more profound meditation states that

experienced meditators may achieve. This potential meditation state difference between

novice and experienced mediators during the Breath Counting meditation was

documented by self-report, but this issue may have contributed to the lack of differences

between novice and experienced meditators. Additionally, experienced meditators may

have other habitual conditions used for their meditation practice that were not present in

this study, and these issues may have also contributed to the lack of differences

between experienced and novice meditators in our study. In the future, it may be

worthwhile to work with more homogenous meditation groups.

 19

In summary, this study set out to identify objective markers of mindfulness

meditation competence. Using ERP techniques, we found that meditators of all

experience levels were better able to direct their attention than naïve participants. This

was reflected by self-rated (MEDI) scores as well as increased N2 and P3 component

responses to target tone stimuli when they were relevant to the task and decreased

responses when irrelevant. Intriguingly, meditators demonstrated these effects

regardless of experience level. The results of this study have important implications for

improving mindfulness meditation interventions by introducing an initial step toward

objective measurement of mindfulness competence.

Acknowledgments: This study was funded by NIH-NCCIH T32 AT002688. Roger

Ellingson, MS, provided programming and engineering support.

References

1. Ahani A, Wahbeh H, Nezamfar H, Miller M, Erdogmus D, Oken B. Quantitative

change of EEG and respiration signals during mindfulness meditation. Journal of

neuroengineering and rehabilitation 2014;11:87

2. Alperin BR, Mott KK, Holcomb PJ, Daffner KR. Does the age-related “anterior

shift” of the P3 reflect an inability to habituate the novelty response?

Neuroscience Letters 2014;577;6-10.

3. Andresen EM, Malmgren JA, Carter WB, Patrick DL. Screening for depressionin

well older adults: Evaluation of a short form of the CES-D. American journal of

preventative medicine 1994;10:77-84.

 20

4. Brefczynski-Lewis JA, Lutz A, Schaefer HS, Levinson DB, Davidson RJ. Neural

correlates of attentional expertise in long-term meditation practitioners. Proc Natl

Acad Sci U S A 2007;104:11483-11488.

5. Cahn BR, Polich J. Meditation states and traits. EEG, ERP, and neuroimaging

studies. Psychological Bulletin 2006;132:180-211.

6. Cahn BR, Polich J. Meditation (Vipassana) and the P3a event-related brain

potential. International Journal of Pscyhophysiology 2009;72:51-60.

7. Chan D, Woollacott M. Effects of level of meditation experience on attentional

focus: is the efficiency of executive or orientation networks improved? J Altern

Complement Med 2007;13:651-657.

8. Chiesa A, Calati R, Serretti A. Does mindfulness training improve cognitive

abilities? a systematic review of neuropsychological findings. Clinical Psychology

Review 2011;32:449-464.

9. Chiesa A, Serretti A. A systematic review of neurobiological and clinical features

of mindfulness meditations. Psychological Medicine 2010;40:1239-1252.

10. Delgado-Pastor LC, Perakakis P, Subramanya P, Telles S, Vila J. Mindfulness

(Vipassana) meditation: Effects on P3b event-related potential and heart rate

variability. Internation Jounral of Psychophysiology 2013;90:207-214.

11. Dickenson J, Berkman ET, Arch J, Lieberman MD. Neural correlates of focused

attention during a brief mindfulness induction. Soc Cogn Affect Neurosci

2013;8:40-47.

12. Fox KC, Nijeboer S, Dixon ML, Floman JL, Ellamil M, Rumak SP, Sedlmeier P,

Christoff K. Is meditation associated with altered brain structure? A systematic

 21

review and meta-analysis of morphometric neuroimaging in meditation

practitioners. Neurosci Biobehav Rev 2014;43:48-73.

13. Grossman P, Niemann L, Schmidt S, Walach H. Mindfulness-based stress

reduction and health benefits. A meta-analysis. J Psychosom Res 2004;57:35-

43.

14. Herscovitch J, Broughton R. Sensitivity of the Stanford Sleepiness Scale to the

effects of cumulative partial sleep deprivation and recovery oversleeping. Sleep

1981;4:83-92

15. Kabat-Zinn J. Wherever you go, there you are: Mindfulness meditation in

everyday life. New York: Hyperion.

16. Kabat-Zinn J, Massion AO, Kisteller J, Peterson LG, Fletcher KE, Pbert L,

Lenderking WR, Santorelli SF. Effectiveness of a meditation-based stress

reduction program in the treatment of anxiety disorders. Am J Psychiatry.

1992;149:936-43.

17. Kilpatrick LA, Suyenobu BY, Smith SR, Bueller JA, Goodman T, Creswell JD,

Tillisch K, Mayer EA, Naliboff BD. Impact of Mindfulness-Based Stress Reduction

training on intrinsic brain connectivity. Neuroimage 2011;56:290-298.

18. Kopp B, Lange F, Howe J, Wessel K. Age-related changes in neural recruitment

for cognitive control. Brain and Cognition 2014;85;209-219.

19. Luck, SJ. 2014. An introduction to the event-related potential technique, 2nd Ed.

Cambridge: The MIT Press.

 22

20. Levinson DB, Stoll EL, Kindy SD, Merry HL, Davidson RJ. A mind you can count

on: validating breath counting as a behavioral measure of mindfulness. Frontiers

in psychology 2014;5:1202.

21. Lutz A, Slatger HA, Dunne JD, Davidson RJ. Attention regulation and monitoring

in meditation. Trends Cogn Sci 2008;12:163-169.

22. Lutz A, Slagter HA, Rawlings NB, Francis AD, Greischar LL, Davidson RJ.

Mental training enhances attentional stability: neural and behavioral evidence. J

Neurosci 2009;29:13418-13427.

23. Miller JJ, Fletcher K, Kabat-Zinn J. Three-year follow-up and clinical implications

of a mindfulness meditation-based stress reduction intervention in the treatment

of anxiety disorders. Gen Hosp Psychiatry. 1995;17:192-200.

24. Milz P, Faber PL, Lehmann D, Kochi K, Pascual-Marqui RD. sLORETA

intracortical lagged coherence during breath counting in meditation-naïve

participants. Frontiers in Human Neuroscience 2014;8:303.

25. Oken BS, Chiappa KH. Short-term variability in EEG frequency analysis.

Electroencephalography and clinical Neurophysiology 1988;69:191-98.

26. Oken BS, Kaye JA. Electrophysiological function in the healthy, extremely old.

Neurology 1992;42:519-526.

27. Oken BS, Salinsky MC, Elsas SM. Vigilance, alertness, or sustained attention:

physiological basis and measurement. Clinical Neurophysiology 2006;117:1885-

1901.

28. Ospina MB, Bond K, Karkhaneh M, Tjosvold L, Vandermeer B, Liang Y, Bialy L,

Hooton N, Buscemi N, Dryden DM, Klassen TP. Meditation practices for health:
 23

state of the research (AHRQ). Evid Rep Technol Assess (Full Rep) 2007;155:1-

263.

29. Piron H. The meditation depth index (MEDI) and the meditation depth

questionnaire (MEDEQ). Journal for meditation and meditation research

2001;1:69-92.

30. Tang YY, Ma Y, Fan Y, Feng H, Wang J, Feng S, Lu Q, Hu B, Lin Y, Li J, Zhang

Y, Wang Y, Zhou L, Fan M. Central and autonomic nervous system interaction is

altered by short-term meditation. Proc Natl Acad Sci U S A 2009;106:8865-8870.

31. Teasdale JD, Segal ZV, Williams JM, Ridgeway VA, Soulsby JM, Lau MA.

Prevention of relapse/recurrence in major depression by mindfulness-based

cognitive therapy. J Consult Clin Psychol. 2000;68:615-23.

32. Wahbeh H, Elsas S-M, Oken B. Mind-body interventions: applications in

neurology. Neurology 2008;70:2321-2328.

33. Welsh KA, Breitner JC, Magruder-Habib KM. Detection of demetia in the elderly

using telephone screening of cognitive stats. Cognitive and behavioral neurology

1993;6:103-10.

34. Zeidan F, Gordon NS, Merchant J, Goolkasian P. The effects of brief mindfulness

meditation training on experimentally induced pain. J Pain. 2010;3:199-209.

Figure 1

Figure 1. P3 responses to targets, Cz

24.00

20.00
Peak to trough amplitudes μV

16.00

12.00 Controls
Meditators

8.00

4.00

0.00
Tones Breath Counting

*For target tones this interaction was significant, p = .039, with meditators showing a
greater change in peak-to-trough P3 amplitudes from the Tones to the Breath Counting
task.
* With novice and experienced groups combined, meditators showed a greater peak-

to-trough change in P3 and N2 across tasks in comparison to non-meditators when

responding to infrequent target tones. Data here are reported from the Cz electrode

site. Please note that the ERP component labels are approximate denotations of

general areas that may not apply to both overlapping conditions.

Figure 3
Figure 4
Figure 5

Figure 5. N2 responses to targets, Cz

-1.00
Peak to trough amplitudes μV

-3.00

-5.00

-7.00
Controls
Meditators
-9.00

-11.00

-13.00

-15.00
Tones Breath Counting
*
*This interaction was significant at the Cz site, p = .017, with a more pronounced effect
than at the Pz site and meditators exhibiting a greater difference in N2 peak-to-trough
amplitudes from the Tones Task to Breath Counting.
Figure 6

Figure 6. N2 reponses to targets, Pz

-1.00

-3.00
Peak to trough amplitudes μV

-5.00

-7.00
Controls
Meditators
-9.00

-11.00

-13.00

-15.00
Tones Breath Counting

*This interaction was significant at the Pz site, p = .044, with meditators on average
showing greater peak-to-trough N2 amplitudes during the Tones task and attenuated N2
amplitudes during the Breath Counting task in comparison to controls.
Figure 7

*For frequent standard tones there were significant changes in N1 and P2 peak-to-
trough values from the Tones task to the Breath Counting task, but there were no group
effects, as expected.
Figure 8

*At the Cz site, there were significant changes in amplitude in N1 and P3 responses to
infrequent non-targets as well as a group effect for N1 responses, with meditators
showing higher amplitudes during the Tones task, p = .02. There was an additional
group effect for P2 with meditators showing higher amplitudes during the Breath
Counting task, p = .04.
 24

Table 1. Participant Characteristics

All Non- Novice Experienced
Participants Meditators Meditators Meditators
(SD) (SD) (SD) (SD)

N 42 13 15 14

Age (M) 49 (13) 48 (11) 50 (13) 49 (15)

Female 62% 69% 73% 43%

Minority 24% 23% 27% 21%

Practice in -- -- 2.4 years (2.5) 22.6 years (13.2)

Years (M)

*There were no significant demographic differences among groups.

Table 2. Tone and Breath Counting Tasks

Tones Tasks Breath Counting Task

Length: 15 minutes Length: 15 minutes
Stimuli: ~45 target tones Stimuli: Same as Tones Task
10% High Frequency Target Tone Instructions: Sit still with eyes closed, ignore all
10% Low Frequency Distracter Tone tones, and count breaths. If count is lost, press a
80% Standard Tone button and resume counting.
Instructions: Sit still with eyes closed and press a Hypothesis: N2 and P3 amplitudes elicited by High
button for the High Frequency Target Tones only. Frequency Tones (previous target) would be most
Hypothesis: Participants’ neurophysiological attenuated in experienced meditators, less
responses (N2 and P3 amplitudes) to the Target attenuated in novice meditators, and least
Tones would correlate with amount of attention attenuated in naive subjects.
training—the more training, the higher the
amplitude.
 25

• Meditators were better able to direct their attention while on-task

• Meditators were better able to redirect their attention after priming
• Identified event-related potential markers of mindfulness competence
• This information can be used to improve mindfulness training and interventions