ORIGINAL PAPER

Does Mindfulness Meditation Enhance Attention? A RandomizedControlled Trial

Randye J. Semple

Published online: 5 June 2010# Springer Science+Business Media, LLC 2010

Abstract Mindfulness-based interventions have been in-corporated into a variety of psychotherapies. Attentionaldisruptions are common in many mental disorders, and itseems generally accepted that practicing mindfulnessenhances attention. We tested the hypothesis that mindful-ness training would enhance four components of attention:sustained vigilance, concentration, inhibition of distraction,and executive control. A randomized three-group designincluded: (1) a mindfulness meditation group, (2) aprogressive muscle relaxation group to control for effectsof physical relaxation on attention, (3) a wait-listed groupto control for practice effects of repeated measures. Fifty-three community adults were randomly assigned to one ofthese groups. Forty-five participants completed the 4-weekprogram. After training and 4 weeks of twice-daily practice,the mindfulness group demonstrated significantly greaterdiscriminability on a signal detection task than did the othergroups. Significant improvements in sustained attentionwere found following mindfulness meditation, which didnot appear to be mediated by relaxation or practice effects.Performances on measures of concentration and inhibitionof distraction did not support the hypothesis. These resultspartially support current considerations of mindfulnessmeditation to enhance attention.

Keywords Attention .Mindfulness . Meditation .

Progressive muscle relaxation . Randomized controlled trial

Introduction

Mindfulness practices and meditation are not synonymous,either in theory or in practice, and research terminologyhas evolved over the past 5 decades. However, mostmindfulness-based interventions utilize some form ofmeditative practices. Many of the techniques now beingintegrated into clinical interventions were adapted fromEastern contemplative practices that are believed to effectpsychological as well as spiritual benefits for its practi-tioners (Walsh and Shapiro 2006). In the 1960s, Westernscientists found that meditation typically produceda hypometabolic autonomic nervous system response thatBenson et al. (1974) named the relaxation response.Reductions in heart rate and respiration, however,are comparatively gross, and objective changes relativeto the subtle and subjective shifts in awareness,emotions, and values that have traditionally been thegoals of meditation practiced as a consciousness disci-pline (Walsh 1996). Accordingly, the rise of information-processing paradigms stimulated interest in understandingbetter the cognitive and affective changes resulting frommeditation and mindfulness practices (Boals 1978).As the word is broadly used in clinical research today,mindfulness is considered to be an intentional deploymentand management of attention (see discussion by Bishopet al. 2004).

Qualities of Mindful Attention

Decidedly difficult to describe, everyone knows exactlywhat to do when told to pay attention. The subjectiveexperience of mindful attention appears to be qualitatively

R. J. Semple (*)Department of Psychiatry and the Behavioral Sciences,Keck School of Medicine, University of Southern California,2250 Alcazar Street, CSC 2200, Los Angeles,CA 90033-9074, USAe-mail: semple@usc.edu

Mindfulness (2010) 1:121130DOI 10.1007/s12671-010-0017-2

SuResaltado

different than ordinary attention. Some attributes of mindfulattention may involve:

a. An intentional focusing of awareness. This awarenessmay be of anything (e.g., situations, physical experi-ences, environments, conversations, tangible objects,etc.) but also includes awareness of intrapsychic events(thoughts, emotions, and body sensations).

b. Awareness that is distinguishable from cognitions,moods, emotions, attitudes, or motivations.

c. A dynamic process that is sustained over time.d. Attention that is not easily distractible.e. Volition. A conscious intention toward mindful atten-

tion is present.f. Decentered focus on present-moment events. Memories

and anticipatory thoughts are identified as events in themind, the contents of which relate to past or future events.

g. Thoughts, emotions, and body sensations are experi-enced as intrapsychic events.

h. Noncritical or nonjudgmental awareness of the objectof attention.

i. Efficient utilization of an optimal level of arousal.1

Considering these qualities in combination suggests thatexperiences may be seen more clearly, so that mindfulnesspractices could effect a reduction in habituated or inappro-priately reactive behaviors. We suggest that on a dimen-sional continuum, ordinary attention lies somewherebetween mindlessness and mindful attention. At oneextreme, mindlessness may be broadly characterized byinconsistent awareness, unstable focus, distractibility, emo-tional reactivity, and attention to past or future experiences.Absent-mindedness, which is looking without actuallyseeing what one is looking at, is one example ofmindlessness. Being caught up in an uncontrollable rageof anger is another. At the other extreme, mindful attentionis present-focused, voluntarily directed, stable, not easilydistracted, and utilizes optimal arousal necessary for eachsituation. Mindful attention means seeing with clarity,stability, minimal distraction, and little affective reactivity.

Buddhist psychology suggests that personal suffering maylessen as one focuses nonjudgmental attention toward fullyexperiencing present-moment events, clearly identifying intra-psychic events as such, and avoiding clinging to that which isimpermanent (Bhikkhu Bodhi 1993). Affective equanimityand increased well-being have long been associated withadvanced meditation practices. Restated as a cognitive model,the cultivation of attention may reduce psychological distressby correcting distorted and maladaptive attentional biases incognitions and perceptions (Blackburn and Davidson 1995).

Kabat-Zinn (1994) suggested that mindfulness allows one toobserve the workings of the mind moment-by-moment, to seethoughts as just thoughts, and allow the thoughts to comeand go without getting caught in the stories they relate.Mindful seeing, without the distortions produced by internalstories, emotions, desires, expectations, and beliefs aboutexperiences should therefore enhance the ability to respondappropriately to intrapsychic, interpersonal, and situationalevents.

Research on Mindful Attention

Deikman (1966) observed that repetition of perceptualpatterns and cognitive categorizations resulted in a cogni-tive state in which conditioned or over-learned responseswere more difficult to suppress. The Stroop effect is onewell-known example of this. Deikman hypothesized thatmeditation practices interrupt automatic responding byreinvesting actions and percepts with attention (p. 329).He referred to this attentional management of over-learnedor habituated responses de-automatization. Deikmanbelieved that de-automatization would produce moreflexible and adaptive cognitive processing. Wenk-Sormaz(2005) tested Deikman's theory and found that comparedwith cognitive and resting control groups, as little as one20-min meditation session, with meditation-nave partic-ipants, reduced the Stroop effect. Wenk-Sormaz concludedthat improved attentional control reduced the automaticityof the responses, supporting Deikman's concept of de-automatization. Moore and Malinowski (2009) compared agroup of experienced meditators with non-meditators andfound a similar reduction of the Stroop effect. They alsofound that self-reported mindfulness and other measures ofattention were higher in meditators than in non-meditators.

Duration and quantity of meditation practice may beassociated with greater increases in mindful attention.Davidson et al. (1976) found increases in attentionalcapacity that significantly correlated with the duration ofthe practitioner's meditation experience. Valentine andSweet (1999) reported that experienced meditators weremore adept at a test of sustained attention than non-meditators, and that long-term meditators showed betterattentional performance than short-term meditators. Anoth-er study found that relative to a non-meditating comparisongroup, participants in an intensive 10-day meditation retreatshowed significant improvements in performance measuresof sustained attention (Chambers et al. 2008).

Different mindfulness practices may differentially affectattention. Jha et al. (2007) compared a group of mindfulness-nave individuals participating in an 8-week Mindfulness-Based Stress Reduction program (MBSR; Kabat-Zinn 1994)with a group of experienced meditators who were partici-

1 One of the most consistent findings in psychology is that optimalarousal is necessary to produce efficient performance for any giventask (Broadbent 1958).

122 Mindfulness (2010) 1:121130

SuResaltado

pating in an intensive 30-day meditation retreat. Results froma psychophysical measure of attention; the AttentionNetwork Test (ANT, Fan et al. 2002), showed thatparticipation in the MBSR program improved orienting,while the retreat participants demonstrated improvements inreceptive attentional skills. Jha and her colleagues suggestedthat mindfulness training might differentially enhance spe-cific components of attention. Results from another study ofself-reported mindfulness correlated with psychophysicalmeasures of attention suggest that mindfulness may beinversely related to exaggerated lapses of attention (Schmertzet al. 2009). Tang et al. (2007) reported that compared with acontrol group, an experimental group given 5 days of briefmeditation training showed greater improvement in conflictscores on the ANT as well as significant increases inimmunoreactivity and decreases in stress-related cortisol.

Studies using imaging techniques have reported neuro-biological changes in brain structure (Lazar et al. 2005) andfunctioning (Davidson et al. 2003) that seem to be relatedto practicing mindfulness. Meditative practices also appearto increase activity in the anterior cingulate cortex, an areaof the brain associated with the self-regulation of attention(Tang and Posner 2009).

Two studies have investigated the relationship betweenmindfulness and attention in children. Rani and Rao (1996)evaluated a small group of children (ages 9 to 11) whomeditated regularly with an age- and gender-matched, non-meditating comparison group. They found that the medita-tion group demonstrated greater attention regulation capacitythan the controls. In a randomized controlled study of clinic-referred children (ages 9 to 13), significant reductions inattention-related problems were found after completion of a12-week mindfulness training program. Following theintervention, program participants demonstrated significantlyfewer attention-related problems than wait-listed controls(Lee et al. 2008; Semple et al. 2010). Results from a mixedgroup of adolescents and adults with Attention Deficit-Hyperactivity Disorder (ADHD) who practiced meditationfor 8 weeks showed reductions in ADHD symptoms andimprovements on tasks of attention and cognitive inhibition(Zylowska et al. 2008).

Only one study was found that reported no relationshipbetween mindfulness and attention. In this study, participantswere randomized to an MBSR group or wait-listed controlgroup. Following the intervention, no significant differencesin performance were found on measures of sustainedattention, switching, elaborative processing, and non-directed attention (Anderson et al. 2007).

The definitions, methodologies, and measurements ofattention vary across these studies, as do the mindfulnesstechniques, modes of training, and levels of experience. Thisobviously makes it difficult to compare studies using meta-analytic methods. Rubia (2009) suggested that one aim of

meditation practices was to reduce or eliminate irrelevantcognitive processes through training of internalized attention.Reviewing the combined evidence from neurobiological andclinical studies seems to show some support for the notionthat mindfulness practices enhance attention.

Current evidence suggests that mindfulness-based clinicalinterventions can be effective in treating a variety of medicaland psychiatric disorders (Allen et al. 2006; Baer 2003;Bondolfi 2005; Grossman et al. 2004). Enhanced attentionmay be one path by which these benefits are realized. Inthis study, we explored the use of meditative practices toenhance attention and examined relationships betweenmindful awareness, relaxation, and well-being. We hypothe-sized that meditation would enhance four components ofattention: sustained vigilance, concentration, inhibition ofdistraction, and executive control. A randomized three-groupdesign was used to control for changes resulting fromnonspecific effects of physical relaxation and repeated-measures practice effects.

Method

Participants

Adult volunteers were recruited through newspaper articlesand flyers posted on community bulletin boards. Seventy-eightvolunteers responded to the notices and underwent brieftelephone screens. Applicants were excluded if they reportedhaving prior experience with any meditation technique. Theywere only initially informed that the research was to studyphysical and psychological effects associated with stress andrelaxation. In order to minimize expectancy effects, the wordsmindfulness and meditationwere avoided (Shapiro 1987).

Fifty-three participants began the study (39 females, 14males). Six females and two males dropped out beforecompleting the program (84.9% retention rate). The 45participants who completed the program were between 23and 56 years of age (M=40.2, SD=8.9). Forty participants(89%) described their ethnicity as Caucasian and five(11%) described themselves as other. Participants attendedtwo individual training and assessment sessions and wererequested to maintain twice-daily home practice sessionsduring the 4 weeks of the program. Each participantreceived an instructional audiotape, written practice instruc-tions, and a practice record log. This study was approvedby the University of Auckland Institutional Review Board.Informed consent was obtained from each participant.

Study Design

The research design was an independent groups, repeated-measures 32 factorial design. Participants were randomly

Mindfulness (2010) 1:121130 123

SuResaltado

assigned to one of three groups (see Fig. 1, CONSORTparticipant flow diagram). Mindfulness meditation (MM; n=15) was referred to as the Benson Technique (Benson et al.1974). The second group (n=14) was assigned a modifiedprogressive muscle relaxation (PMR) procedure (Bernsteinand Borkovec 1973; Jacobson 1925/1987), which waslabeled the Jacobson Technique. The primary purpose ofthe PMR group was to control for nonspecific effects ofphysiological relaxation on attention. The third group (n=16)consisted of wait-listed controls (CT). These individualswere not informed that they were controls, but only that theirtraining would begin in a few weeks. The CT groupcontrolled for practice effects on the performance measuresand for normal and expectable fluctuations in anxiety andmood states over time. CT group assessments werecompleted on schedules equivalent to both training groups.

Individual pretest and posttest sessions, 1 month apart,were conducted with all participants. At pretest, baseline datawere collected, and participants were taught and practicedeither MM or PMR or placed on the wait-list for futuretraining. Each training group participant received an audiotapeof pre-recorded MM or PMR instructions. Participants wereasked to practice the 20-min technique twice daily and record

the date, time, and duration of each practice. At the individualposttest sessions, a 20-min mindfulness or relaxation practicewas conducted before posttest data were collected. With oneexception, test administrations were counterbalanced betweengroups and across sessions. As the principal measure ofattention, all participants at posttest were administered acontinuous performance test (CPT) immediately following themindfulness or relaxation practice. Given that the participantswere novices, state effects of mindfulness and/or relaxation onattention were expected to decline quicker than what might beexpected from more experienced practitioners.

Measures

Assessments included measures of attention and measuresof well-being. Participant demographics, characteristics,and health and lifestyle data were also collected. Threepsychophysical instruments were chosen to evaluate someof the theoretical qualities of mindful attention: a Contin-uous Performance Test, Digit Symbol Substitution, and theStroop Color and Word Test. The State-Trait AnxietyInventory and the Profile of Mood States assessed well-being because of its potential relationship with attention.

Excluded (n = 25) Did not meet inclusion criteria (n = 19)

Refused to participate (n = 6)

Randomized (n = 53)

Allocated to Mindfulness Meditation (n = 19) Received Intervention (n = 15)

Discontinued (n = 4)

Analyzed (n = 15) Excluded from analyses (n = 0)

Assessed for eligibility (n = 78)

Allocated to Wait-listed Control (n = 16)

Baseline (n = 16) Discontinued (n = 0)

Posttest Assessment (n = 15) Meditation completers lost to

follow-up (n = 0)

Posttest Assessment (n = 16) Wait-listed controls lost to

follow-up (n = 0)

Posttest Assessment (n = 14) Relaxation completers lost to

follow-up (n = 0)

A

naly

ses

Post

test

Allo

catio

nEn

rollm

ent

Allocated to Progressive Relaxation (n = 18) Received Intervention (n = 14)

Discontinued (n = 4)

Analyzed (n = 14) Excluded from analyses (n = 0)

Analyzed (n = 16) Excluded from analyses (n = 0)

Fig. 1 CONSORT participantflow diagram.

124 Mindfulness (2010) 1:121130

Continuous Performance Test

The CPT is a target-detection sustained vigilance taskoriginally developed by Rosvold et al. (1956). A computer-ized AX version of the CPT was used. Software used tocontrol the display presentation and timing of the stimuli wasdeveloped at the University of Auckland. Amber-coloredblock letters, 4.5 cm high, were presented one at a time at thecenter of a black monitor screen. Eleven letters were used, allhaving angular shapes (A, E, F, H, L, N, T, V, X, Y, Z). Eachstimulus was presented at the center of the screen for 300 mswith a fixed interval of 400 ms between presentations. Thetarget was the letter X, but only if it appeared immediatelyfollowing the letter A. Participants were instructed to press ablue button when they saw the target and a red button afterevery other non-target letter. Six hundred letters werepresented in pseudo-random sequence, with the AX targetsequence programmed to appear on 12% of the presenta-tions. The combinations A then * (A preceding any otherletter) and * then X (any other letter preceding X) eachoccurred on 12% of the presentations. Test administrationlasted 7 min. The order of presentation was randomly variedon the posttest administrations.

Interpretation of the CPT

Response events included targets correctly detected (hits);targets missed (misses); positive responses to non-signals(false alarms); and correctly rejected non-targets (correctrejections). Signal detection methods permit a statisticalestimation of errors of commission (response bias or ) anderrors of omission (discriminability or log d) of a visualsignal. Limited attentional capacity seems to effect rapidreductions in perceptual sensitivity (Nuechterlein et al.1983), so the 7-min vigilance task was considered to beadequately sensitive to detect changes in attention. Ofprimary interest was the measure of discriminability (log d),which is an estimate of the participant's ability todiscriminate signal from noise. Executive control, sustainedalertness, orientation to cues, and resistance to distractionduring the entire 7-min trial was required to produce higherlog d scores. Higher measures of discriminability wereconsidered to represent a greater degree of mindfulattention. Reaction times are influenced by factors not ofinterest in this study and therefore were not recorded orexamined.

Digit Symbol Substitution

Digit Symbol Substitution (DSS) is a performance subtestfrom the Wechsler Adult Intelligence Scale, Third Ed.(Wechsler 1997). A coding key pairs ten simple symbolswith ten single-digit numbers. Participants substitute the

matching symbol for a random succession of numbers.Scoring is based on the number of accurate writtensubstitutions completed in 90 s. Performance is relativelyunaffected by intellectual prowess, memory, or learning.The DSS is considered a test of focused concentration andfreedom from distractibility. It also appears to be a sensitivemeasure of selective attention (Lezak et al. 2004). Advan-tages of the DSS include the short administration time andavailability of reliable normative data.

Stroop Color and Word Test

The Golden version (1978) of the Stroop assesses atten-tional conflict, orientation to cues, and the ability to wardoff distractions (MacLeod 1991). Selective responding tocompeting stimuli activates executive control functionsnecessary to resolve the conflicts. The interference phe-nomenon known as the Stroop effect appears to decreasewith age. Using age-adjusted norms, pure interferencescores were calculated which corrected for the base readinglevel of each participant. This method also controls for thesuperior color-naming skills often seen in women (Golden1978). Testretest reliability coefficients range from 0.69 to0.86 (Jensen and Rohwer 1966). Administration time isapproximately 5 min.

Spielberger State-Trait Anxiety Inventory

The Spielberger State-Trait Anxiety Inventory (Spielberger1983) is a self-report instrument that provides scores forstate anxiety (SA), which measures current apprehension,tension, worry, and nervousness, and trait anxiety (TA),which reflects relatively stable dispositional tendencies toperceive stressful situations as threatening. The SA and TAscales each consist of 20 short self-statements that are ratedon 4-point scales of frequency. State anxiety is variable, soa reported testretest reliability of 0.33 on the SA scale isacceptable. The TA scale has shown high testretestreliability at 0.77 (Spielberger 1983). Appropriate genderand age-stratified norms are available. Administration timeis approximately 6 min.

Profile of Mood States

The Profile of Mood States (POMS) (McNair et al. 1992) isa 65-item self-report measure of general mood states.Scoring consists of six mood factor scores (tensionanxiety,depressiondejection, angerhostility, vigoractivity, fa-tigueinertia, confusionbewilderment) and a total mooddisturbance (TMD) score. TMD is intended to be a reliableglobal estimate of current affective state. Gender-stratifiednorms are available. Administration time for the POMS isapproximately 5 min.

Mindfulness (2010) 1:121130 125

Participant Characteristics

Data were collected on seven health and lifestyle variables:general physical health, general mental health, ability tohandle stress, amount of cigarette smoking, amount of alcoholconsumption, amount of exercise, and amount and quality ofsleep. Expectations of benefits to be gained from mindfulnesspractice appear to be related to the frequency of practice(Delmonte 1987), so ratings of motivation (desire) and self-confidence (efficacy) were included. Each participant char-acteristic was self-rated on a 7-point, Likert-type scale.

Procedures

At pretest, demographic data and baseline assessments werecollected. Training and on-site practice (90 min) in therespective mindfulness or relaxation technique was fol-lowed by 1 month of twice-daily home-based practice. Forboth training groups, pretest measures were completedbefore the training began. The pretest session for thecontrol group did not include an instruction and practiceperiod and lasted just under 1 h. The author contacted eachparticipant via telephone twice during the 1-month periodof home practice (approximately every 10 days). Thepurpose of this follow-up was to encourage maintenanceof the twice-daily practice, to offer advice regarding dailyscheduling issues, and respond to questions about thetechniques. Posttest procedures were similar. For bothtraining groups, the posttest session began with a briefreview of the training technique followed by a 20-minpractice session. Posttest data were collected immediatelyafter the practice session. Participants completed a programevaluation, were debriefed about the purpose of the study,and thanked for their participation. CT group participantswere advised that they had been controls and offered anopportunity to learn a relaxation technique. All acceptedand then received training and instructional materials in onetechnique of their choice.

Results

Repeated-measure ANOVAs were conducted appropriate tothe 32 factorial design. Three groups (mindfulness,relaxation, and control) were evaluated at pre- (time 1)and post- intervention (time 2). Participant variables wereanalyzed using one-way ANOVAs for interval scale dataand KruskalWallis H tests for ordinal scale data. Nosignificant group differences were found for any participantvariable at pretest, so no covariates were included in theinitial analyses. Alpha of 0.05 was used for all tests. CPTdata were corrected for extreme proportions using a loglinear correction (Hautus 1995). Average log d estimates

were calculated to provide a less-biased group total thanthat obtained by pooling the raw data prior to estimatingpopulation log d for each group (Hautus 1997).

Figure 2 shows the mean changes between sessions fordiscriminability (log d) on the CPT. Mean discriminabilityfor the MM group increased markedly from pretest toposttest session, while the PMR and CT groups achievedmodest increases across sessions. The increase observedacross all groups suggests that practice improves CPTperformance. Discriminability across groups differed signif-icantly [F (2, 42)=3.71, p< 0.05]. Scheff post hoc analysisshowed that the MM group achieved a significantly greaterimprovement in discriminability than either the PMR or CTgroup. Significant changes in log d were found in thesessiongroup interaction [F (2, 42)=6.79, p< 0.01]. Thisfinding supports the primary research question, propos thecapability of mindfulness practice to enhance attentionindependent of the potential influence of relaxation onattention.

Mean age between groups was not significantly different,but age proved to be a significant correlate of CPTperformance (rs=0.35, p

Both training groups reported significant declines in meanstate (SA) and trait anxiety (TA) scores [F (2, 42)=18.61, p /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /Warning /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 150 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 600 /MonoImageMinResolutionPolicy /Warning /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False

/Description > /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ > /FormElements false /GenerateStructure false /IncludeBookmarks false /IncludeHyperlinks false /IncludeInteractive false /IncludeLayers false /IncludeProfiles true /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /NA /PreserveEditing false /UntaggedCMYKHandling /UseDocumentProfile /UntaggedRGBHandling /UseDocumentProfile /UseDocumentBleed false >> ]>> setdistillerparams> setpagedevice