Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You and Your Team.

Learn More →

Meditation affects word recognition of meditation novices

Meditation affects word recognition of meditation novices This work represents one of the first attempts to examine the effects of meditation on the processing of written single words. In the present longitudinal study, participants conducted a lexical decision task and rated the affective valence of nouns before and after a 7-week class in mindfulness meditation, loving-kindness meditation, or a control intervention. Both meditation groups rated the emotional valence of nouns more neutral after the interventions, suggesting a general down-regulation of emotions. In the loving-kindness group, positive words were rated more positively after the intervention, suggesting a specific intensification of positive feelings. After both meditation interventions, response times in the lexical decision task accelerated significantly, with the largest facilitation occurring in the loving-kindness group. We assume that meditation might have led to increased attention, better visual discrimination, a broadened attentional focus, and reduced mind-wandering, which in turn enabled accelerated word recognition. These results extend findings from a previous study with expert Zen meditators, in which we found that one session of advanced meditation can affect word recognition in a very similar way. Introduction symptoms of anxiety (Hofmann et al., 2010; Tang et al., 2007), stress (Goyal et al., 2014), and depression (Grecucci Research on meditation has sparked substantial scientific et  al., 2015; Tang et al., 2007). At work, meditation can interest in recent years. The effects of meditation practice lead to reduced emotional exhaustion and increased job sat- on increased attention have generated particular interest in isfaction (Hülsheger et al., 2013). In meditators, brain gray the research community (Jha et al., 2007; Pagnoni & Cekic, matter density is increased in the left hippocampus (Hölzel 2007; Slagter et al., 2007). Meditation practice is associated et al., 2011), which is associated with emotion regulation with improved efficiency in attentional processing (van den (Corcoran et al., 2005). Hurk et al., 2010) and increased sustained attention (Cham- In recent years, a new field of meditation research has bers et al., 2008; MacLean et al., 2010). Experienced medi- been explored. A study by Tarrasch et al. (2016) was one tators show reduced interference in the Stroop task com- of the first to examine the effects of meditation on reading pared to control subjects (Chan & Woollacott, 2007) and can performance. They found that subjects with developmen- decelerate binocular rivalry switching (Carter et al., 2005). tal dyslexia and ADHD demonstrated significantly fewer Further, meditation practice can lead to a reduced thought reading errors after a 2-month mindfulness course. Their distraction and a strengthened present focus (Kok & Singer, error rate dropped by 19% compared to their original perfor- 2017). Tang et al. (2007) found that already after a 5-day mance. In addition, they showed increased sustained atten- meditation training, participants achieved more improved tion. Another study found that intensive meditation practice performance on the attention network test compared to a can reduce mindless reading (Zanesco et al., 2016). Already control group. after a 5-day workshop in mindfulness-based stress reduc- Substantial scientific research has focused on the effects tion, participants of a pilot study demonstrated a significant of meditation on emotion regulation. Meditation practice increase in reading speed (Rice et al., 2020). This promising can reduce negative affect (Sears & Kraus, 2009), regulate field of research is still widely unexplored. Since texts often contain words with an affective content it is also important to examine the effects of meditation practice on emotional * Larissa Lusnig word processing. In particular, it is not clear if different med- larissa.lusnig@uni-wuppertal.de itation styles have dissociable effects on reading speed and how fast single words with different emotional valences are Department of Psychology, University of Wuppertal, Wuppertal, Germany Vol.:(0123456789) 1 3 Psychological Research processed. Basic scientific research on this topic is needed. Our previous study In the current literature, we find evidence that meditation can influence the processing of emotional images. Medita- In a prior study, we examined if meditation has an impact tors demonstrate a reduced emotional reactivity to affective on the responses to affective and neutral words in an LDT images. In a study by Desbordes et al. (2012) participants and on the valence ratings of these words (Lusnig et al., showed reduced amygdala activation while watching emo- 2020). Experienced Zen meditators rated the valence of tional images. A decreased amygdala activation suggests low-arousal positive and low- and high-arousal negative that the affective images had a decreased emotional impact words more neutral, subsequent to a 90-min meditation (Davis & Whalen, 2001). Subjects with extensive meditation session. In an age- and gender-matched comparison group, experience showed less interference from affective images valence ratings did not change significantly after the com- (Ortner et al., 2007). Chau et al. (2018) found that older parison intervention. In the LDT, the Zen group showed an subjects rate affective pictures more neutral subsequent to a accelerated word recognition subsequent to the meditation meditation intervention. While these studies show that medi- intervention. The comparison group did not show a signifi- tation can affect emotional picture evaluation, the impact of cant change in response times (RTs) after the comparison meditation on affective word processing has not yet been intervention. Because the effect of learning to meditate well explored. could not be investigated in these expert meditators, a lon- gitudinal study, examining participants before and after they learned to meditate, is needed. Interindividual differences and word recognition The emotional connotation of experimental word stimuli is The present study usually assessed by valence ratings (Citron, 2012; Jacobs et al., 2015; Kissler et al., 2007; Vo et al., 2009). Based on In the present work, we adapted the design of our previous such ratings researchers can select word material, for exam- study (Lusnig et al., 2020) for a longitudinal study. Two ple, for lexical decision tasks (LDT). During the LDT, par- experimental groups and a control group (CG) participated ticipants are presented with letter strings, which either form in a meditation or control class. Before and after these a word or a meaningless stimulus. Subjects must choose interventions, all groups performed an LDT and a valence- under pressure of time if a word is displayed or not (Meyer rating task, along with a brief mood-states assessment. & Schvaneveldt, 1971; Rubenstein et al., 1970). Word stimu- At baseline, subjects completed assessments on concen- lus properties like arousal, frequency, imageability, valence, tration capacity, intelligence, and personality traits. One and many others have been shown to have an impact on word experimental group participated in a 7-week class in mind- recognition (Brysbaert et al., 2011; Hofmann et al., 2007; fulness meditation. Mindfulness meditation is a widely Kuchinke et al., 2007; New et al., 2006). used meditation style, which is similar to Zen meditation. Interindividual differences can influence single word rec- During mindfulness meditation, the practitioner develops ognition as well. Self-secure subjects demonstrated faster an effortless and non-judgmental awareness of all present- identification of words expressing positive interpersonal moment experiences (Kabat-Zinn, 1990). The second outcomes compared to participants that are more insecure. experimental group practiced loving-kindness meditation Insecure subjects recognized words faster that expressed (LKM) for 7 weeks. LKM contains the practice of empa- negative interpersonal outcomes (Baldwin et al., 1993). In a thy, first the feeling of loving kindness is directed towards valence identification task, dysphoric subjects needed signif- oneself and then towards others (Lutz et al., 2007). The icantly more time to recognize the valence of positive words control group participated for 7 weeks in a study group. than they needed to identify the valence expressed by nega- As in the previous study, we used the low-arousal positive, tive words when contrasted with non-dysphoric participants. low- and high-arousal negative and neutral word stimuli In an LDT, non-dysphoric participants demonstrated faster of Hofmann et al. (2009) and Lusnig et al. (2020). The recognition of negative in comparison to neutral words. quantity of these word stimuli was doubled for the current Dysphoric subjects, on the other hand, responded slower to study. negative than to neutral words (Siegle et al., 2002). Mueller In our prior work (Lusnig et al., 2020), the Zen group and Kuchinke (2016) linked goal-directed behavior to slower responded in an LDT globally faster to the same stimulus processing of fear words. Further, they tied the spontaneous words after the meditation intervention than before the eye-blink rate, which indexes dopaminergic levels, to a pro- meditation. For the LDT of the present study, we expected cessing advantage for happy words. If and to what extent the both meditation groups to process word stimuli faster processing of words may also be influenced by meditation, after the meditation classes than at baseline. Particularly is so far not well understood. 1 3 Psychological Research mindfulness meditation has been associated with increased online advertisements. Subjects were randomly assigned attention (Valentine & Sweet, 1999); therefore, we antici- to the three groups. None of the participants reported prior pated the greatest change in RTs in the mindfulness group experience with meditation or yoga. None of them had a (MG). We predicted that RTs in the control group would history of psychiatric disorders or any reading and writ- not differ significantly when compared before and after the ing difficulties. All participants received course credits for control intervention. In the initial study by Hofmann et al. participation. (2009), subjects responded slower to low-arousal negative words than to neutral words. In our previous meditation work, we replicated this result in the Zen- and the control Covariates group. Consequently, for the present study, the expectation was that all three groups would process low-arousal nega- Participants completed, at first, the “Aktuelle Stim- tive words slower than neutral words. mungsskala” (ASTS) to examine possible changes in mood In Lusnig et  al.’s (2020) valence-rating task, the Zen states provoked by the interventions in every group. The group rated positive, low- and high-arousal negative words ASTS measures subjects’ current mood states; the scales are more neutral after a single 90-min meditation session. The “positive mood”, “sorrow”, “desperateness”, “fatigue” and comparison group did not demonstrate significantly different “anger”. The internal consistency has Cronbach’s α values valence ratings before and after the comparison interven- between α = 0.83 and 0.94. Factor analysis provided one, tion. Based on these results, we expected that both medita- two and four factor-based approaches. The author provides tion groups would rate emotion words more neutral after evidence for convergent and differential validity (Dalbert, the meditation classes. Especially LKM is associated with 1992a, 1992b). Participants conducted this test at baseline an increased experience of positive emotions (Zeng et al., and after the interventions. Possible significant group dif- 2015); therefore, we anticipated more positive valence rat- ferences in personality traits, sustained attention, and intel- ings in the loving-kindness group (LKG) than in the MG. ligence were examined by covariate tests, which subjects We also predicted that the control group would not rate the conducted at baseline. Subjects performed the d2-Revision valence of emotion words significantly different after the test (d2-R), an assessment of sustained attention and the control intervention. Participants performed assessments on ability to focus on task. In the d2-R, subjects have to find personality traits, concentration capacity, and intelligence the letter “d” with two marks while not getting distracted to control for interindividual differences between the three by similar-looking stimuli. Cronbach’s alpha values are groups. We did expect, however, that mood states would for the scales “number of processed target objects” and for change after the meditation interventions in the LKG and “concentration capacity” are between α = 0.89 and 0.95, for the MG, but not in the CG. “percentage of errors” between α = 0.80 and 0.91, depending on the age group. Authors provide empirical evidence for criterion and construct validity (Brickenkamp et al., 2010). Methods and materials This was followed by the Multiple-Choice Vocabulary Intel- ligence Test (MWT-B). Here, participants have to point out Participants a correct German word among four similar nonwords in a multiple-choice procedure. It contains 37 items, which are Thirty-nine German native speakers participated in the sorted by level of difficulty. For the MWT-B retest reliabil- present study. Thirteen of them took part in the LKG (12 ity is reported with a correlation of r = 0.95 after 6 months female, 19–39 years of age, mean age = 22.2), thirteen in and r = 0.87 after 14 months. The author provides empirical the MG (11 female, 19–44 years of age, mean age = 24.5), evidence for criterion validity (Lehrl, 2005). At last, subjects and thirteen in the CG (12 female, 19–42 years of age, performed the Big Five personality test (B5T), which meas- mean age = 21.8). The formula of Westfall et al. (2014) ures personality traits as specified in the Big Five model of was used to calculate an appropriate sample size. Vari- personality. The B5T had in the original version five scales ance partitioning coefficients were estimated based on “neuroticism” (Cronbach’s alpha, α = 0.90), “conscientious- the values of our previous study (Lusnig et al., 2020). To ness” (α = 0.77), “extraversion” (α = 0.87), “agreeableness” obtain a medium effect size of d = 0.5 and a power of 0.8 (α = 0.76) and “openness to experience” (α = 0.76). For while using 400 word-stimuli per participant, a sample the revised version, which was used in the present study, size of 11.8 participants per group is required. As men- three basic requirements “need for power and influence” tioned above, we worked with a sample size of 13 par- (α = 0.78), “need for safety and peace” (α = 0.84), and “need ticipants per group. Thirty-eight participants were right- for achievement and performance” (α = 0.82) were added. handed one was left-handed. All subjects were students The test demonstrates factorial validity (Satow, 2011). of the University of Wuppertal; they were recruited via 1 3 Psychological Research Stimuli Procedure The stimulus set included 400 words (German nouns) and Pre‑test 400 altered words (nonwords). Half of the item set was taken from Hofmann et al. (2009), while the other half was Subjects first completed the ASTS, then the d2-R, the MWT- generated using the same construction rules. All stimulus B, and the B5T. Subsequently, subjects took part in the main words were part of the revised Berlin Affective Word List experiment. Stimuli were presented on a 21-inch TFT dis- (BAWL-R; Vo et al., 2009). The stimulus set contained four play running at 70 Hz, with the distance between eyes and stimulus conditions: low-arousal negative, high-arousal monitor held constant at approximately 65 cm. negative, (low-arousal) positive, and neutral words. The Participants with even participant numbers were pre- stimulus condition “High-arousal positive words” could not sented with Subset A of the stimuli set. They were asked to be generated because the BAWL-R did not provide enough press key “2”, using the left index finger when identifying a high-arousal positive nouns (cf. Hofmann et al., 2009; Lus- nonword, and to press key “8”, using the right index finger nig et al., 2020; see Fig. 1 in Vo et al., 2009). Each of the when they recognized the stimulus as a word. After par- four stimulus conditions of the whole item set contained ticipants had processed the first half of Subset A, they were 100 words. Low-arousal negative, positive and neutral words presented with the second half of subset A. To balance for were matched for arousal. High-arousal negative words were RT differences due to hand dominance, subjects now pressed matched for valence to low-arousal negative words, but their key “2”, using the left index finger when identifying a word, arousal values were maximized. Nouns were matched for the and pressed key “8”, using the right index finger when they following psycholinguistic variables, which can influence identified a nonword. This way we also excluded the risk LDTs: arousal, word frequency, emotional valence, number that possible effects were caused by the composition of one of orthographic neighbors, number of letters, imageability, of the subsets, for instance by mood induction due to many mean bigram frequency (type), and mean bigram frequency words of the same emotion category (Niedenthal & Setter- (token) (see Table 1, and Table 1 in Hofmann et al., 2009). lund, 1994). Participants with uneven participant numbers The selected nouns were modified to create nonwords. For were presented first with the first half of Stimulus Subset B, this purpose, a vowel of a word stimulus was replaced by pressing the left index finger for a word. In the second half a consonant or another vowel. Stimuli were subdivided in of Subset B, hand assignment was reversed. In any case, Subset A and Subset B, each consisting of 200 words. To responses were to be executed as quickly and accurately as test whether the materials were matched for these variables, possible. we conducted 2 × 4 ANOVAs (subset × emotion condition; Before the LDT, participants were made familiar with all Fs < 1). the test by responding to five practice stimuli. The word stimuli appeared in black uppercase letters on a light gray background, in 20 pt Times New Roman font using presen- tation-software PsychoPy, version 1.82.01 (Peirce, 2007). Stimuli were pseudorandomized. At most three words or nonwords were presented consecutively. During each trial, for 700 ms, a fixation cross (+) was displayed, followed by Table 1 Mean values and standard errors of the manipulation and control variables for high-arousal negative, low-arousal negative, positive, and neutral words Control and manipulation variables High-arousal negative Low-arousal negative Neutral words Positive words words words M SE M SE M SE M SE Emotional valence − 1.31 0.05 − 1.30 0.04 0.06 0.04 1.17 0.07 Arousal 3.90 0.03 2.83 0.04 2.82 0.03 2.82 0.03 Imageability 4.50 0.17 4.11 0.20 4.24 0.16 4.26 0.16 Number of letters 6.42 0.17 6.06 0.21 6.16 0.18 6.18 0.19 Word frequency 12.94 0.31 13.20 0.38 12.62 0.29 12.56 0.27 Number of orthographic neighbors 0.98 0.24 1.50 0.30 1.20 0.24 1.18 0.27 Mean bigram frequency (type) 3113.55 284.41 3408.68 245.04 3525.47 272.50 3223.81 277.65 Mean bigram frequency (token) 167,061.37 19,957.41 189,351.71 21,027.04 190,835.14 19,536.51 178,481.74 20,794.76 1 3 Psychological Research a word or nonword for 1000 ms. A white screen was shown feelings and emotions and situations of others, also using for 500 ms and then for 1500 ms a mask (#####) (see Fig. 1 visual imagery. in Lusnig et al., 2020). After a 5-min break, subjects rated the valence of the words, which they had seen before in the Control intervention LDT. One word at a time was presented on the screen. Below every word, a seven-point grading scale from − 3 to 3 was The study group aimed to involve subjects of the CG in a displayed (0 = neutral, −  3 = ver y negative, 3 = ver y posi- silent active control intervention very close to their usual tive). Subjects gave their responses by clicking the respec- activities. All participants were college students; therefore, a tive number with the cursor. After pressing the space bar, the study group was selected as an adequate intervention. Partic- next word appeared on the screen. The duration of the LDT ipants were instructed to study silently for a class they were was about 20 min followed by a 5-min break, the valence rat- currently attending. An undergraduate assistant supervised ing task lasted for approximately 12 min. The whole experi- the study group. ment, including also covariate tests and instructions, lasted approximately 75 min. Data analysis Post‑test For the LDT, RTs of the correctly given answers were ana- lyzed using linear mixed-effects models (LMEs). 14.95% of Subjects took part in the post-test about 1 week after having all responses were incorrect, they were excluded from the finished one of the meditation classes or the study group. data analysis. The models were calculated with the statis- Participants first finished the ASTS, other covariate tests tical software environment R, (version 3.4.2, http:// cran.r- were not conducted during the post-test. The instructions proje ct. org). Specifically, we used the lme4 library with the and procedures for the LDT were the same as the ones given lmer function (version 1.1–14, Bates et al., 2015). The lmer in the pre-test. Participants with even numbers were pre- function fits an LME to the data. An LME data analysis sented with stimulus Subset B, subjects with uneven par- considers participant and item variance concurrently in a ticipant numbers completed Subset A. Participants took a non-hierarchical approach. Averaging at first level treats the 5-min break. Then, they rated the valence of the words they error variance of items as fixed effects. Separate random had seen during the post-LDT. intercepts for subjects and items result in treating subject and item variance more sensitively. “Subjects” and “items” were Meditation or comparison interventions fitted as random effects. As fixed effects, we fitted “groups” (LKG/MG/CG), “time” (before/after intervention) and The week after the pre-test, all subjects took part in a 1.5-h “emotional valence” (positive words/low-arousal negative/ intervention in the morning on the same day of the week. high-arousal negative). Fixed effects were represented with The training took place in 8 consecutive weeks. In the fourth the use of effects coding. Low-arousal negative, high-arousal week, training was suspended due to a national holiday. A negative, and positive words were confronted with neutral local experienced meditation-trainer led both meditation words. We calculated a time slope for the random effect groups. “item” because the time-series effect might be different for different items (Baayen et al., 2008). For every group, sepa- Mindfulness meditation intervention rate LMEs were calculated to examine the origin of signifi- cant interactions. The dependent variable “response time” Participants of the MG learned at first which sitting postures was log-transformed to satisfy the assumption of normality are adequate for meditation, how to relax their body, and of the residuals, which were verified by qqplots. Estimates of how to breathe calmly and naturally. Then, they practiced the regression coefficients, their standard errors, and t values observing their thoughts, emotions, and physical feelings are reported for the LMEs. P values are reported on the basis and tried to let them go instead of being caught up in them. of the Satterthwaite approximation, which is implemented in the lmerTest package, (Version 2.0-36, Kuznetsova et al., Loving‑kindness meditation intervention 2017). Regarding the valence rating experiment, we examined First, subjects of the LKG practiced suitable sitting postures responses given on a seven-point grading scale ranging from for meditation, how to exercise a natural and calm breath, − 3 to 3. An LME was used again to analyze the experimen- and technics to relax their body. They learned how to be tal data. The procedure of the data analysis was the same aware of their thoughts, emotions, and feelings and trained as for the response time experiment. Data points, which to develop equanimity and self-empathy regarding these were not in a range between – 3 and 3 standard deviations states. Gradually subjects broadened their empathy to their of the residual error, were discarded from the calculations 1 3 Psychological Research (approximately 1% of the data). To analyze the results of and sorrow; for the discussed effects of the present study, the ASTS we conducted a 3 × 2 × 5 ANOVA with the factors however, it does not have relevance. In addition, subjects “group”, “time” and “mood conditions”. were tested at baseline for group differences in personal- ity traits, intelligence, and concentration capacity to control for influences of these factors on the RT results. In these Results comparisons, no significant group differences were found. All statistical values for group differences, mean values, Covariates and standard deviations of the baseline covariate tests are reported in Table 3. To control for a possible change in mood induced by the interventions, subjects completed the ASTS at baseline and Lexical decision and valence rating data after interventions. Mood did not change significantly over time points in any of the groups. Mood conditions (posi- Figure 1 shows that after the loving-kindness- and the mind- tive mood, sorrow, desperateness, fatigue, and anger) were fulness-meditation sessions affective valence ratings were rated significantly different (see Table  2). This effect dem- more neutral, except for valence ratings to positive words in onstrates, for example, the difference between positive mood the LKG, which became more positive. Figure 2 shows that RTs in the LDT were shorter after both meditation inter- ventions, but not after the control intervention. The LME Table 2 Statistical group differences and mean squared errors (MSE) analysis revealed for the valence rating as well as for lexical of ASTS (tested at baseline and after intervention) decision data significant interactions of “group” and “time” Statistical group differences (see Table  4 for entire analysis). The valence rating data showed for “time” as well as for “group” significant two- Time F(1, 36) = 2.31, p = 0.14, way interactions with “positive”, “high-arousal negative”, MSE = 4.13 and “low-arousal negative” words. Further, we found signifi- Time: Group F(2, 36) = 2.39, p = 0.11 cant main effects for “high-arousal negative”, “low-arousal Mood conditions F(1, 36) = 496.85, p = 0.00, MSE = 54.32 negative” and “positive” words. The analysis of the lexical Mood conditions: Group F(2, 36) = 1.04, p = 0.39 decision data revealed for "low-arousal negative" words a Time: Mood conditions F(2, 36) = 0.29, p = 0.78, significant main effect. MSE = 26.47 Separate LMEs were subsequently calculated for all three Time: Mood conditions: Group F(2, 36) = 0.61, p = 0.66 groups to resolve the significant interactions of “time” and “group” (cf. Table  5). For the valence rating and lexical Time (before/after intervention), Group (LKG/MG/CG), mood condi- tions (positive mood, sorrow, desperateness, fatigue, and anger) Table 3 Baseline covariate tests: statistical group differences, mean squared errors (MSE), mean values, and standard deviations Covariate test Statistical group differences LKG MG CG M SD M SD M SD Big Five personality test Neuroticism F(2, 36) = 2.37, p = 0.11, MSE = 2.01 4.85 1.86 4.77 1.34 3.77 1.01 Extraversion F(2, 36) = 0.46, p = 0.63, MSE = 3.16 5.69 1.60 5.77 2.39 6.31 1.11 Conscientiousness F(2, 36) = 1.30, p = 0.29, MSE = 4.39 4.23 1.92 5.54 2.15 4.69 2.21 Agreeableness F(2, 36) = 1.18, p = 0.32, MSE = 2.41 5.15 1.68 5.92 1.71 6.00 1.23 Need for safety and peace F(2, 36) = 2.96, p = 0.06, MSE = 1.74 4.62 1.19 5.46 1.56 4.23 1.16 Need for power and influence F(2, 36) = 0.71, p = 0.50, MSE = 2.86 4.54 1.81 4.31 1.55 5.08 1.71 Openness F(2, 36) = 2.83, p = 0.07, MSE = 2.07 5.77 1.59 5.38 1.04 4.46 1.61 Need for achievement and performance F(2, 36) = 0.01, p = 0.99, MSE = 3.03 5.00 1.87 5.08 1.89 5.00 1.41 d2-Revision, test concentration capacity Number of processed target objects F(2, 36) = 2.18, p = 0.13, MSE = 465.78 163.77 22.06 179.00 21.72 179.15 20.96 Concentration capacity F(2, 36) = 1.91, p = 0.16, MSE = 666.45 143.38 28.31 161.38 28.47 159.53 19.68 Percentage of errors F(2, 36) = 1.89, p = 0.17, MSE = 22.52 11.37 4.72 8.24 4.35 11.37 5.14 Intelligence test (MWT-B) F(2, 36) = 2.05, p = 0.14, MSE = 6.78 22.23 2.52 23.37 2.36 21.31 2.89 1 3 Psychological Research LKG MG CG 1.5 0.5 -0.5 -1 -1.5 -2 A B C D A B C D A B C D A= high-arousal negative words, C= neutral words before after B = low-arousal negative words, D = positive words Fig. 1 Results of the valence rating experiment. Error bars indicate standard errors decision data, we found significant main effects for “time” The valence rating experiment in the LKG and the MG, but not in the CG. Valence ratings differed significantly after the interventions in both meditation groups. After the control intervention, valence ratings did not change. Figure 1 demonstrates that Discussion the MG and the LKG rated words more neutral after the meditation classes. The LKG, however, rated positive words In the present study, we did not find any baseline differ - more positively after the meditation intervention. These ences in intelligence, concentration capacity, and personal- results are in line with previous work on meditation and ity traits between the three groups. Contrary to our expec- emotion regulation. Several studies showed that particularly tations, mood changes could not be detected in the ASTS mindfulness meditation can down-regulate negative emo- after any of the three interventions. Looking at word items, tions such as anxiety (Hofmann et al., 2010; Tang et al., both meditation groups rated valence more neutral after the 2007), stress (Goyal et al., 2014), and depression (Grecucci meditation interventions. Participants in the LKG, however, et al., 2015; Tang et al., 2007). The practice of LKM has rated positive words more positively after the intervention. also been associated with increased positive emotions (Fre- In the control group, valence ratings did not differ signifi- drickson et al., 2008; Hutcherson et al., 2008; Zeng et al., cantly after control intervention. Concerning the LDT, both 2015). However, contrary to our expectations, we did not meditation groups demonstrated faster word recognition find evidence for changes in mood states after any of the after the interventions. This effect was most pronounced for three interventions. The reason for this could be that the participants in the LKG. The control group did not show mood assessment we used (ASTS) might not be sensitive significantly different RTs after the intervention. Hofmann or specific enough to capture the emotion regulation pro- et al. (2009) and our previous study found that low-arousal duced by meditation. On the other hand, these results may negative words were processed slower than neutral words. indicate that meditation practice does not always lead to We replicated this effect in all three groups. a reduced experience of emotions. The practitioners may 1 3 Valence Ratings Psychological Research LKG MG CG A B C D A B C D A B C D A= high-arousal negtive words, C= neutral words before after B = low-arousal negative words, D = positive words Fig. 2 Results of the lexical decision experiment. Error bars indicate standard errors Table 4 Valence rating and LDT Experiments: estimates of regression coefficients, their standard errors, t values, p values, and Cohen’s d effect sizes of the overall analyses Valence ratings Lexical decision task B SE t p d B SE t p d Time 0.013 0.024 0.54 0.592 0.01 – 0.002 0.004 – 0.55 0.586 – 0.02 Group – 0.009 0.032 – 0.28 0.783 – 0.15 0.006 0.016 0.41 0.688 0.13 High-arousal negative – 1.273 0.052 – 24.57 0.001*** – 1.91 0.001 0.008 0.01 0.997 0.01 Low-arousal negative – 0.933 0.052 – 18.01 0.001*** – 1.40 0.026 0.008 3.28 0.001** 0.27 Positive 1.739 0.052 33.59 0.001*** 2.64 – 0.006 0.008 – 0.82 0.412 – 0.07 Time:group 0.039 0.015 2.70 0.007** 0.04 – 0.022 0.003 – 8.21 0.001*** – 0.15 Time:high-arousal negative 0.084 0.042 2.00 0.045* 0.08 – 0.006 0.006 – 0.96 0.335 – 0.04 Group:high-arousal negative 0.128 0.018 7.08 0.001*** 0.09 – 0.001 0.003 – 0.35 0.730 – 0.01 Time:low-arousal negative 0.106 0.042 2.53 0.011* 0.09 – 0.001 0.006 – 0.14 0.890 – 0.01 Group:low-arousal negative 0.102 0.018 5.66 0.001*** 0.07 – 0.003 0.003 – 0.85 0.394 – 0.02 Time:positive – 0.106 0.042 – 2.53 0.011* – 0.11 – 0.002 0.006 – 0.31 0.758 – 0.01 Group:positive – 0.180 0.018 – 9.95 0.001*** – 0.13 0.001 0.003 0.30 0.768 0.01 Time:Group:high-arousal negative 0.017 0.026 0.68 0.499 0.01 0.004 0.005 0.89 0.373 0.02 Time:Group:low-arousal negative – 0.021 0.026 – 0.82 0.412 – 0.01 0.001 0.005 0.08 0.933 0.01 Time:Group:positive 0.037 0.026 1.44 0.149 0.02 – 0.001 0.005 – 0.28 0.777 – 0.01 “***”p < 0.001, “**”p < 0.01, “*”p < 0.05 1 3 Response Times (in ms.) Psychological Research Table 5 Valence rating and LDT experiments: estimates of regression coefficients, their standard errors, t values, p values, and Cohen’s d effect sizes for the LKG, MG, and CG Valence ratings Lexical decision task B SE t p d B SE t p d LKG Time 0.078 0.026 3.08 0.002** 0.31 – 0.048 0.004 – 12.41 0.001*** – 0.41 High-arousal negative – 0.980 0.051 – 19.22 0.001*** – 1.92 – 0.003 0.008 – 0.32 0.747 – 0.03 Low-arousal negative – 0.728 0.051 – 14.27 0.001*** – 1.43 0.016 0.008 1.98 0.048* 0.20 Positive 1.361 0.051 26.69 0.001*** 2.67 – 0.004 0.008 – 0.55 0.585 – 0.06 Time:high-arousal negative 0.094 0.044 2.12 0.035* 0.21 – 0.001 0.007 – 0.21 0.833 – 0.01 Time:low-arousal negative 0.053 0.044 1.20 0.231 0.12 0.006 0.007 0.81 0.417 0.03 Time:positive – 0.011 0.044 – 0.25 0.801 – 0.03 – 0.004 0.007 – 0.57 0.569 – 0.02 MG Time 0.125 0.025 4.91 0.001*** 0.19 – 0.022 0.004 – 5.96 0.001*** – 0.62 High-arousal negative – 1.172 0.054 – 21.78 0.001*** – 2.14 0.001 0.008 0.01 0.993 0.01 Low-arousal negative – 0.768 0.054 – 14.28 0.001*** – 1.41 0.027 0.008 3.49 0.001*** 0.35 Positive 1.554 0.054 28.90 0.001*** 2.85 – 0.005 0.008 – 0.70 0.484 – 0.07 Time:high-arousal negative 0.157 0.044 3.56 0.004*** 0.14 – 0.005 0.006 – 0.78 0.435 0.08 Time:low-arousal negative 0.115 0.044 2.60 0.009** 0.10 – 0.013 0.006 – 1.94 0.053 – 0.19 Time:positive – 0.194 0.044 – 4.40 0.001*** – 0.17 – 0.002 0.006 – 0.27 0.785 – 0.03 CG Time – 0.026 0.027 – 0.95 0.342 – 0.03 – 0.001 0.004 – 0.33 0.739 – 0.03 High-arousal negative – 1.215 0.056 – 21.65 0.001*** – 2.14 – 0.003 0.008 – 0.33 0.745 – 0.03 Low-arousal negative – 0.930 0.056 – 16.57 0.001*** – 1.64 0.023 0.008 2.78 0.006** 0.29 Positive 1.707 0.056 30.40 0.001*** 3.00 – 0.004 0.008 – 0.53 0.594 – 0.06 Time:high-arousal negative 0.047 0.047 1.01 0.311 0.03 – 0.008 0.007 – 1.23 0.221 – 0.13 Time:low-arousal negative 0.077 0.047 1.66 0.098 0.05 0.006 0.007 0.90 0.368 0.09 Time:positive – 0.050 0.047 – 1.06 0.287 – 0.03 – 0.003 0.007 – 0.52 0.603 – 0.06 “***”p < 0.001, “**”p < 0.01, “*”p < 0.05 still experience the emotions the way they did before but do et al. (2007), LKM can be seen as a special case of OMM not judge them and do not get carried away by them. The because it contains the “cultivation of objectless aware- fact that we did not find changes in mood states, however, ness” and “non-referential compassion”. However, it con- ensures that the neutralized valence ratings occurred due to tains also phases of focused attention meditation (FAM), the meditation interventions and were not influenced by a during which the meditator keeps the attention all the time momentary mood change. on one object. In the case of LKM, this object is the feel- In our previous study (Lusnig et al., 2020), adept Zen ing of loving-kindness, which is directed towards oneself meditators assigned to words significantly more neutral or other single persons (Lutz et al., 2007; Vago & Silber- valence ratings after a 90-min Zen meditation. In the pre- sweig, 2012). In the present work, the LKG rated positive vious comparison group, valence ratings did not change words more positively after meditation. In the MG and the after the comparison intervention. These results are in Zen group of the previous study, we did not find such an line with our findings for the MG in the present study. effect. LKM differs from mindfulness- and Zen meditation The similarity of these results was to be expected because because it contains the practice of empathy and positive Zen meditation and mindfulness meditation are compa- feelings towards others. This difference in meditation prac- rable styles of meditation. Both meditation styles belong tice may have led to the more positive valence ratings in to the category of open monitoring meditation (OMM) the LKG. A study by Hunsinger et al. (2013) found results (Lutz et al. 2008), during which the meditator monitors, similar to our study. In their work, loving-kindness nov- in a non-judgmental way, everything that occurs in his ices associated significantly more positivity with neutral moment-to-moment experience, such as sounds, thoughts stimuli after a meditation intervention compared to control that pass the mind, smells, or feelings. According to Lutz participants. 1 3 Psychological Research have helped the participants of the meditation groups to The lexical decision experiment focus more closely on the current word stimulus, enabling faster responses. Half of the stimulus set used in the present study was identical to the one used by Hofmann et al. (2009). They Improved visual discrimination could be another pro- cess that may have contributed to faster RTs in the MG and found, among other results, low-arousal negative words being processed slower than neutral words. In the pre- the LKG. Expert meditators demonstrate visual attentional processing, which is more accurate and flexible in contrast sent study, we replicated this effect in all three groups. In our previous work (Lusnig et al., 2020), we obtained to control subjects’ visual processing. For example, medi- tators notice changes in flickering scenes faster than con- this result in the Zen and the control group. In the cur- rent literature, it is discussed if positive or negative visual trols (Hodgins & Adair, 2010). Brown et al. (1984) tested Buddhist meditators before and after a 3-month meditation stimuli are processed more rapidly. For example, Öhman et al. (2001) found that threatening faces are processed retreat for visual sensitivity. After the meditation interven- tion, meditators noticed shorter single-light flashes and faster than friendly faces. On the other hand, a study by Becker et  al. (2012) demonstrated that dynamic happy could differentiate better successive light-flashes than before the retreat. A control group did not show any such changes facial expressions are detected faster than dynamic angry facial expressions. Hofmann et al. (2009) found that the in visual sensitivity. In a study by MacLean et al. (2010), meditation novices improved after a 3-month meditation arousal level affects the processing speed of emotional single words. In their study, high-arousal negative words training visual discrimination, perceptual sensitivity, and increased vigilance during visual attention. This evidence and positive words are processed faster than low-arousal negative words. In the present study, we found the same points to the possibility that in the present study meditation training might have led to improved visual sensitivity and descriptive result pattern in all three groups (see Fig. 2). The MG, however, demonstrates after the meditation inter- discrimination performance, which in turn allowed for faster responses in the LDT. vention no difference in processing speed for high-arousal negative, low-arousal negative, and positive words (see We expected that the MG would show the largest decrease in RTs after intervention because especially mindful- Fig.  2). This might be because the profound practice of equanimity in mindfulness meditation minimizes the dif- ness meditation has been associated with increased atten- tion (Semple, 2010; Valentine & Sweet, 1999). There was ference in arousal level for negative words. In both meditation groups of the present study, but not indeed a substantial response acceleration, but in our data, this effect was even more pronounced in the LKG. These the control group, RTs changed significantly after the interventions. As illustrated in Fig.  2, RTs to emotional results might be explained considering the association between meditation styles and narrow or broad attentional words were faster after both meditation interventions. These results are also in accordance with those of our focus. Lippelt et al. (2014) argued that FAM, which con- tains mainly a strong concentration on a single object, leads previous study, in which the Zen group demonstrated a significantly faster word recognition after a 90-min medi - to a narrow focus of attention. During OMM the meditator monitors all experiences non-judgmentally. This medita- tation session. Meditation is associated with increased attention (Carter et  al., 2005; Chambers et  al., 2008; tion style is, therefore, thought to lead to a broad attentional focus. Such a broadened attentional focus was shown to MacLean et al., 2010; Chan & Woollacott, 2007; van den Hurk et al., 2010). Hence, it appears plausible to conclude promote better performance on an attention task (Willems & Martens, 2016). In the present study, a broad attentional that increased attentional resources in the meditation groups may have led to accelerated word recognition. As focus, induced by the mindfulness meditation, might have led to a more effective and therefore faster word process- an alternative account, the shorter RTs could be associated with reduced mind-wandering as a result of meditation. ing. The practice of LKM contains open monitoring, the main goal of this meditation style is to broaden the feeling Using functional MRI, Brefczynski-Lewis et al. (2007) found that expert meditators showed less brain activation of loving-kindness starting from a person we like to every- one. This broadened attentional focus combined with the in the default mode network. The default mode network is associated with discursive thoughts. Similarly, a study strong cultivation of loving-kindness might have helped the LKG to process the emotional words especially fast. Since by Pagnoni et al. (2008) found in meditators decreased neural activity in default mode network regions. These positive affect is also associated with a broader attentional focus (Fredrickson & Branigan, 2005), the feeling of loving- authors propose that meditators may be able ‘to control the automatic cascade of semantic associations’ better than kindness might have given an additional speed boost in the LKG. It would be very interesting for subsequent research control subjects (Pagnoni & Cekic, 2007, p. 1). Therefore, spontaneous mind-wandering could be regulated more eas- to compare not just the influence of mindfulness meditation and LKM on word processing but also the effects of FAM. ily. In the present study, regulated mind-wandering may 1 3 Psychological Research This way the effects of narrowed and broadened attentional an object like the breath and calm themselves down. On focus could be compared. the other hand, Richards et  al. (2014) argued that peo- ple with anxiety disorders show, in confrontation with a Future directions and limitations specific threatening stimulus, a narrow attentional focus. OMM could help persons, in such situations, to deliber- Concerning the sequence of experimental and covariate tests ately broaden their attention and thereby allow them to in the present study, it might have been better to give to the detach their attention from the threatening stimulus. participants first the ASTS, then the LDT and the valence rating task, and at last the remaining covariate tests. This way the mood states, which were measured with the ASTS could not have been changed through tiring covariate tests. Conclusions However, since in the present study the ASTS results were not influenced by the meditation intervention, the test plays In a previous study Lusnig et  al. (2020), we found that a minor role in the interpretation of our study, and we do not advanced Zen meditation can neutralize valence ratings of see problems with the sequence of the tests. emotional words and accelerated RTs to these words. In the For future studies on the influence of meditation and present study, we were able to obtain similar results with a visual word processing it might also be beneficial to use longitudinal study design. Subjects, which participated in assessments on emotion regulation and increased attention a 7-week loving-kindness- or mindfulness course, demon- not just before the meditation intervention, as in the present strated significantly more neutral valence ratings after the study, but also after it. With such an experimental design it interventions. Positive words were rated more positively could be extensively examined which of these underlying after the LKM course. These results suggest that different mechanisms of meditation influences altered word process- meditation styles can contribute to the down- and up-regu- ing in meditation practitioners the most. It would be espe- lation of emotions. However, contrary to our expectations, cially informative to investigate further the role of emotional mood states did not appear to change after meditation inter- variability. We would suggest that subjects should conduct ventions. In both meditation groups, RTs were faster after adequate emotion assessments at baseline, two times during the interventions than before, with the largest changes occur- the intervention and after it. ring in the LKG. Improved increased attention, visual dis- The present study and our previous study (Lusnig et al., crimination, and reduced mind-wandering, caused by medi- 2020) focused on the question if meditation practice can tation, may have enabled accelerated word recognition. The influence visual single word processing. Since we found results of the present study could help to understand better that meditation practice can accelerate the responses to the influence of meditation in text processing of affectively single words and neutralize the valence ratings of emotion loaded content. words, it would be interesting to examine in the future if meditation might have similar effects on the processing of Author contributions LL, MH, and RR planned the study. LL con- entire written texts. If the effects of meditation practice on ducted the experiments. MH and LL planned and performed the analy- written texts are similar to those on single words, it can be ses. LL, MH, and RR wrote the paper. All authors read and approved assumed that meditation can accelerate the reading speed the final manuscript. of practitioners, as suggested also by a pilot study by Rice et al. (2020). It seems also important to examine clearly Funding Open Access funding enabled and organized by Projekt DEAL. Financial disclosure statement. This work was partly sup- how meditation styles lead to a narrow or broad attentional ported by a grant of the Deutsche Forschungsgemeinschaft to MH focus and how such a focus affects word and text process- (HO 5139/2-2) and by a doctoral studies grant of the Evangelisches ing. Lippelt et al. (2014) proposed that FAM leads to a nar- Studienwerk e.V. to LL. The funders had no role in study design, data row attentional focus and that OMM induces a broadened collection and analysis, decision to publish, or preparation of the manu- script. There was no additional external funding received for this study. attentional focus. It seems important to examine clearly in experimental studies how these attentional foci affect Availability of data and material The datasets analyzed dur- word and text processing. The further examination of nar- ing the present study are available from the corresponding row and broad attentional foci, triggered by meditation, author on reasonable request. could also have important implications for the treatment of anxiety disorders. Richards et al. (2014) claimed that Code availability The code used for the data analysis of the anxious persons, who show hypervigilance, demonstrate a present study is available from the corresponding author on broadened attentional focus that scans for potential threats reasonable request. in the environment. The practice of FAM could be helpful for these individuals to deliberately narrow their focus on 1 3 Psychological Research Chambers, R., Lo, B. C. Y., & Allen, N. B. (2008). The impact of inten- Declarations sive mindfulness training on attentional control, cognitive style, and affect. Cognitive Therapy and Research, 32, 303 Conflict of interest/competing interests The authors declare that they Chan, D., & Woollacott, M. (2007). Effects of level of meditation have no conflict of interest. experience on attentional focus: Is the efficiency of executive or orientation networks improved? Journal of Alternative and Com- Ethics approval The present study was carried out following the ethi- plementary Medicine, 13(6), 651–657 cal standards provided by the Declaration of Helsinki and the German Chau, B. K. H., Keuper, K., Lo, M., So, K.-F., Chan, C. C. H., & Lee, Society for Psychology (DGPs). The Ethics Committee of the Univer- T. M. C. (2018). Meditation-induced neuroplastic changes of the sity of Wuppertal approved the study. prefrontal network are associated with reduced valence perception in older people. Brain and Neuroscience Advances. https://doi. or g/ Consent to participate All subjects gave their written informed con- 10. 1177/ 23982 12818 771822 sent. Citron, F. M. M. (2012). Neural correlates of written emotion word processing: A review of recent electrophysiological and hemo- Consent for publication We did not use participant identifying infor- dynamic neuroimaging studies. Brain and Language, 122, mation. 211–266 Corcoran, K. A., Desmond, T. J., Frey, K. A., & Maren, S. (2005). Hippocampal inactivation disrupts the acquisition and contex- Open Access This article is licensed under a Creative Commons Attri- tual encoding of fear extinction. Journal of Neuroscience, 25, bution 4.0 International License, which permits use, sharing, adapta- 8978–8987 tion, distribution and reproduction in any medium or format, as long Dalbert, C. (1992a). ASTS - Aktuelle Stimmungsskala. In Leibniz- as you give appropriate credit to the original author(s) and the source, Zentrum für Psychologische Information und Dokumentation provide a link to the Creative Commons licence, and indicate if changes (ZPID) (Hrsg.), Elektronisches Testarchiv (PSYNDEX Tests- were made. The images or other third party material in this article are Nr. 9003796). Trier: ZPID. included in the article’s Creative Commons licence, unless indicated Dalbert, C. (1992b). Subjektives Wohlbefinden junger Erwachsener: otherwise in a credit line to the material. If material is not included in Theoretische und empirische Analysen der Struktur und Stabil- the article’s Creative Commons licence and your intended use is not ität. Zeitschrift für Differentielle und Diagnostische Psycholo- permitted by statutory regulation or exceeds the permitted use, you will gie, 13, 207–220 need to obtain permission directly from the copyright holder. To view a Davis, M., & Whalen, P. J. (2001). The amygdala: Vigilance and copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . emotion. Molecular Psychiatry, 6, 13–34 Desbordes, G., Negi, L. T., Pace, T. W., Wallace, B. A., Raison, C. L., & Schwartz, E. L. (2012). Effects of mindful-attention and compassion meditation training on amygdala response to emo- References tional stimuli in an ordinary, non-meditative state. Frontiers in Human Neuroscience, 6, 292 Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects Fredrickson, B. L., & Branigan, C. (2005). Positive emotions modeling with crossed random effects for subjects and items. broaden the scope of attention and thought-action repertoires. Journal of Memory and Language, 59, 390–412 Cognition & Emotion, 19(3), 313–332 Baldwin, M. W., Fehr, B., Keedian, E., Seidel, M., & Thomson, D. Fredrickson, B. L., Cohn, M. A., Coffey, K. A., Pek, J., & Finkel, S. W. (1993). An Exploration of the relational schemata underlying M. (2008). Open hearts build lives: Positive emotions, induced attachment styles. Self-report and lexical decision approaches. through loving-kindness meditation, build consequential per- Personality and Social Psychology Bulletin, 19(6), 746–754 sonal resources. Journal of Personality and Social Psychology, Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting Linear 95(5), 1045–1062 Mixed-Effects Models Using lme4. Journal of Statistical Soft- Goyal, M., Singh, S., Sibinga, E. M. S., Gould, N. F., Rowland- ware, 67, 1 Seymour, A., Sharma, R., et al. (2014). Meditation programs Becker, D. V., Neel, R., Srinivasan, N., Neufeld, S., Kumar, D., & for psychological stress and well-being: A systematic review Fouse, S. (2012). The vividness of happiness in dynamic facial and meta-analysis. JAMA Internal Medicine, 174(3), 357–368 displays of emotion. PLoS ONE, 7(1), e26551 Grecucci, A., Pappaianni, E., Siugzdaite, R., Theuninck, A., & Job, Brefczynski-Lewis, J. A., Lutz, A., Schaefer, H. S., Levinson, D. B., & R. (2015). Mindful emotion regulation: Exploring the neuro- Davidson, R. J. (2007). Neural correlates of attentional expertise cognitive mechanisms behind mindfulness. BioMed Research in long-term meditation practitioners. Proceedings of the National International, 2015, 670724 Academy of Sciences of the United States of America., 104(27), Hodgins, H. S., & Adair, K. C. (2010). Attentional processes and 11483–11488 meditation. Consciousness and Cognition, 19(4), 872–878 Brickenkamp, R., Schmidt-Atzert, L., & Liepmann, D. (2010). d2-R: Hofmann, M. J., Kuchinke, L., Tamm, S., Võ, M.L.-H., & Jacobs, A. Test d2—Revision. Hogrefe. M. (2009). Affective processing within 1/10th of a second: High Brown, D., Forte, M., & Dysart, M. (1984). Visual sensitivity and arousal is necessary for early facilitative processing of nega- mindfulness meditation. Perceptual and Motor Skills, 58(3), tive but not positive words. Cognitive, Affective, & Behavioral 775–784 Neuroscience, 9, 389–397 Brysbaert, M., Buchmeier, M., Conrad, M., Jacobs, A. M., Bölte, J., & Hofmann, M. J., Stenneken, P., Conrad, M., & Jacobs, A. M. (2007). Böhl, A. (2011). The word frequency effect: A review of recent Sublexical frequency measures for orthographic and phono- developments and implications for the choice of frequency esti- logical units in German. Behavior Research Methods, 39(3), mates in German. Journal of Experimental Psychology, 58(5), 620–629 412–424 Hofmann, S. G., Sawyer, A. T., Witt, A. A., & Oh, D. (2010). The effect Carter, O. L., Presti, D. E., Callistemon, C., Ungerer, Y., Liu, G. B., of mindfulness-based therapy on anxiety and depression: A meta- & Pettigrew, J. D. (2005). Meditation alters perceptual rivalry in analytic review. Journal of Consulting and Clinical Psychology, Tibetan Buddhist monks. Current Biology, 15, R412–R413 78(2), 169–183 1 3 Psychological Research Hölzel, B., Carmody, J., Vangel, M., Congleton, C., Yerramsetti, S. from the English Lexicon Project. Psychonomic Bulletin & M., Gard, T., et al. (2011). Mindfulness practice leads to increases Review, 13(1), 45–52 in regional brain gray matter density. Psychiatry Research, 191, Niedenthal, P. M., & Setterlund, M. B. (1994). Emotion congruence 36–43 in perception. Personality and Social Psychology Bulletin, 20(4), Hülsheger, U. R., Alberts, H. J., Feinholdt, A., & Lang, J. W. (2013). 401–411 Benefits of mindfulness at work: The role of mindfulness in emo- Öhman, A., Lundqvist, D., & Esteves, F. (2001). The face in the crowd tion regulation, emotional exhaustion, and job satisfaction. Jour- revisited: A threat advantage with schematic stimuli. Journal of nal of Applied Psychology, 98(2), 310–325 Personality and Social Psychology, 80(3), 381 Hunsinger, M., Livingston, R., & Isbell, L. (2013). The impact of Ortner, C. N. M., Kilner, S. J., & Zelazo, P. D. (2007). Mindfulness loving-kindness meditation on affective learning and cognitive meditation and reduced emotional interference on a cognitive task. control. Mindfulness, 4, 275 Motivation and Emotion, 31(4), 271–283 Hutcherson, C. A., Seppala, E. M., & Gross, J. J. (2008). Loving-kind- Pagnoni, G., & Cekic, M. (2007). Age effects on gray matter volume ness meditation increases social connectedness. Emotion, 8(5), and attentional performance in Zen meditation. Neurobiology of 720–724 Aging, 28, 1623–1627 Jacobs, A. M., Vo, M.L.-H., Briesemeister, B. B., Conrad, M., Hof- Pagnoni, G., Cekic, M., & Guo, Y. (2008). “Thinking about not-think- mann, M. J., Kuchinke, L., Lüdtke, J., & Braun, M. (2015). ing”: Neural correlates of conceptual processing during Zen medi- 10 years of BAWLing into affective and aesthetic processes in tation. PLoS ONE, 3(9), e3083 reading: What are the echoes? Frontiers in Psychology, 6, 714 Peirce, J. W. (2007). PsychoPy—Psychophysics software in Python. Jha, A. P., Krompinger, J., & Baime, M. J. (2007). Mindfulness train- Journal of Neuroscience Methods, 162(1–2), 8–13 ing modifies subsystems of attention. Cognitive, Affective & Rice, V., Enders, L., & Jeter, A. (2020). Visual scanning and reading Behavioral Neuroscience, 7(2), 109–119 speed following 5-days of mindfulness-based stress reduction: A Kabat-Zinn, J. (1990). Full catastrophe living. Delta Publishing. pilot study. In: International Conference on Applied Human Fac- Kissler, J., Herbert, C., Peyk, P., & Junghofer, M. (2007). Buz- tors and Ergonomics (pp. 221–227). Springer, Cham. zwords: early cortical responses to emotional words during Richards, H. J., Benson, V., Donnelly, N., & Hadwin, J. A. (2014). reading. Psychological Science, 18(6), 475–480 Exploring the function of selective attention and hypervigilance Kok, B. E., & Singer, T. (2017). Phenomenological fingerprints for threat in anxiety. Clinical Psychology Review, 34(1), 1–13 of four meditations: differential state changes in affect, mind- Rubenstein, H., Garfield, L., & Millikan, J. A. (1970). Homographic wandering, meta-cognition, and interoception before and after entries in the internal lexicon. Journal of Verbal Learning & Ver- daily practice across 9 months of training. Mindfulness, 8(1), bal Behavior, 9(5), 487–494 218–231 Satow, L. (2011). Big-Five-Persönlichkeitstest (B5T) [Testbe- Kuchinke, L., Võ, M.L.-H., Hofmann, M. L., & Jacobs, A. M. schreibung und Items]. In Zentrum für Psychologische Informa- (2007). Pupillary responses during lexical decisions vary with tion und Dokumentation (ZPID) (Hrsg.), PSYNDEX (Online- word frequency but not emotional valence. International Jour- Datenbanksegment PSYNDEX Tests Dok.-Nr. 9006357). Trier: nal of Psychophysiology, 65(2), 132–140 ZPID. Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. B. (2017). Sears, S., & Kraus, S. (2009). I think therefore I om: cognitive distor- lmerTest package: Tests in linear mixed effects models. Journal tions and coping style as mediators for the effects of mindfulness of Statistical Software, 82, 13 meditation on anxiety, positive and negative affect, and hope. Lehrl, S. (2005). Manual zum MWT-B. [Mehrfachwahl-Wortschatz- Journal of Clinical Psychology, 65(6), 561–573 Intelligenztest]. (5th unchanged). Spitta-Verl. Semple, R. J. (2010). Does mindfulness meditation enhance attention? Lippelt, D. P., Hommel, B., & Colzato, L. S. (2014). Focused atten- A randomized controlled trial. Mindfulness, 1, 121–130 tion, open monitoring and loving kindness meditation: effects Siegle, G., Ingram, R., & Matt, G. (2002). Affective interference: An on attention, conflict monitoring, and creativity–A review. Fron- explanation for negative attention biases in dysphoria? Cognitive tiers in Psychology, 5, 1083 Therapy and Research, 26(1), 73–87 Lusnig, L., Radach, R., Mueller, C. J., & Hofmann, M. J. (2020). Zen Slagter, H. A., Lutz, A., Greischar, L. L., Francis, A. D., Nieuwenhuis, meditation neutralizes emotional evaluation, but not implicit S., Davis, J. M., et al. (2007). Mental training affects distribution affective processing of words. PLoS ONE, 15(2), e0229310 of limited brain resources. PLoS Biology, 5(6), e138 Lutz, A., Dunne, J. D., & Davidson, R. J. (2007). Meditation and the Tang, Y.-Y., Ma, Y., Wang, J., Fan, Y., Feng, S., Lu, Q., et al. (2007). neuroscience of consciousness. In P. Zelazo, M. Moscovitch, & Short-term meditation training improves attention and self-regu- E. Thompson (Eds.), Cambridge handbook of consciousness. lation. Proceedings of the National Academy of Sciences of the (pp. 499–555). Cambridge University Press. United States of America, 104(43), 17152–17156 Lutz, A., Slagter, H. A., Dunne, J. D., & Davidson, R. J. (2008). Tarrasch, R., Berman, Z., & Friedmann, N. (2016). Mindful reading: Attention regulation and monitoring in meditation. Trends in Mindfulness meditation helps keep readers with dyslexia and Cognitive Sciences, 12(4), 163–169 ADHD on the lexical track. Frontiers in Psychology, 7, 578 MacLean, K. A., Ferrer, E., Aichele, S. R., Bridwell, D. A., Zanesco, Vago, D. R., & Silbersweig, D. A. (2012). Self-awareness, self- A. P., Jacobs, T. L., et al. (2010). Intensive meditation training regulation, and self-transcendence (S-ART): a framework for improves perceptual discrimination and sustained attention. understanding the neurobiological mechanisms of mindfulness. Psychological Science, 21(6), 829–839 Frontiers in Human Neuroscience, 6, 296. https://doi. or g/10. 3389/ Meyer, D. E., & Schvaneveldt, R. W. (1971). Facilitation in rec-fnhum. 2012. 00296 ognizing pairs of words: Evidence of a dependence between Valentine, E., & Sweet, P. (1999). Meditation and attention: A com- retrieval operations. Journal of Experimental Psychology, 90(2), parison of the effects of concentrative and mindfulness medita- 227–234 tion on sustained attention. Mental Health, Religion & Culture, Mueller, C. J., & Kuchinke, L. (2016). Individual differences in emo- 2(1), 59–70 tion word processing: A diffusion model analysis. Cognitive, van den Hurk, P. A., Giommi, F., Gielen, S. C., Speckens, A. E., & Affective, & Behavioral Neuroscience, 16(3), 489–501 Barendregt, H. P. (2010). Greater efficiency in attentional pro- cessing related to mindfulness meditation. Quarterly Journal of New, B., Ferrand, L., Pallier, C., & Brysbaert, M. (2006). Reexamining Experimental Psychology, 63(6), 1168–1180 the word length effect in visual word recognition. New evidence 1 3 Psychological Research Vo, M.L.-H., Conrad, M., Kuchinke, L., Urton, K., Hofmann, M. J., Zeng, X., Chiu, C. P., Wang, R., Oei, T. P., & Leung, F. Y. (2015). & Jacobs, A. M. (2009). The Berlin affective word list reloaded The effect of loving-kindness meditation on positive emotions: A (BAWL-R). Behavior Research Methods, 41, 2 meta-analytic review. Frontiers in Psychology, 6, 1693 Westfall, J., Kenny, D. A., & Judd, C. M. (2014). Statistical power and optimal design in experiments in which samples of participants Publisher’s Note Springer Nature remains neutral with regard to respond to samples of stimuli. Journal of Experimental Psychol- jurisdictional claims in published maps and institutional affiliations. ogy General, 143(5), 2020–2045 Willems, C., & Martens, S. (2016). Time to see the bigger picture: Individual differences in the attentional blink. Psychonomic Bul- letin & Review, 23(5), 1289–1299 Xu, R. (2003). Measuring explained variation in linear mixed effects models. Statistics in Medicine, 22(22), 3527–3541 Zanesco, A. P., King, B. G., MacLean, K. A., Jacobs, T. L., Aichele, S. R., Wallace, B. A., & Saron, C. D. (2016). Meditation training influences mind wandering and mindless reading. Psychology of Consciousness: Theory, Research, and Practice, 3(1), 12 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Psychological Research Springer Journals

Meditation affects word recognition of meditation novices

Loading next page...
 
/lp/springer-journals/meditation-affects-word-recognition-of-meditation-novices-0fwv5vaBcz
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
ISSN
0340-0727
eISSN
1430-2772
DOI
10.1007/s00426-021-01522-5
Publisher site
See Article on Publisher Site

Abstract

This work represents one of the first attempts to examine the effects of meditation on the processing of written single words. In the present longitudinal study, participants conducted a lexical decision task and rated the affective valence of nouns before and after a 7-week class in mindfulness meditation, loving-kindness meditation, or a control intervention. Both meditation groups rated the emotional valence of nouns more neutral after the interventions, suggesting a general down-regulation of emotions. In the loving-kindness group, positive words were rated more positively after the intervention, suggesting a specific intensification of positive feelings. After both meditation interventions, response times in the lexical decision task accelerated significantly, with the largest facilitation occurring in the loving-kindness group. We assume that meditation might have led to increased attention, better visual discrimination, a broadened attentional focus, and reduced mind-wandering, which in turn enabled accelerated word recognition. These results extend findings from a previous study with expert Zen meditators, in which we found that one session of advanced meditation can affect word recognition in a very similar way. Introduction symptoms of anxiety (Hofmann et al., 2010; Tang et al., 2007), stress (Goyal et al., 2014), and depression (Grecucci Research on meditation has sparked substantial scientific et  al., 2015; Tang et al., 2007). At work, meditation can interest in recent years. The effects of meditation practice lead to reduced emotional exhaustion and increased job sat- on increased attention have generated particular interest in isfaction (Hülsheger et al., 2013). In meditators, brain gray the research community (Jha et al., 2007; Pagnoni & Cekic, matter density is increased in the left hippocampus (Hölzel 2007; Slagter et al., 2007). Meditation practice is associated et al., 2011), which is associated with emotion regulation with improved efficiency in attentional processing (van den (Corcoran et al., 2005). Hurk et al., 2010) and increased sustained attention (Cham- In recent years, a new field of meditation research has bers et al., 2008; MacLean et al., 2010). Experienced medi- been explored. A study by Tarrasch et al. (2016) was one tators show reduced interference in the Stroop task com- of the first to examine the effects of meditation on reading pared to control subjects (Chan & Woollacott, 2007) and can performance. They found that subjects with developmen- decelerate binocular rivalry switching (Carter et al., 2005). tal dyslexia and ADHD demonstrated significantly fewer Further, meditation practice can lead to a reduced thought reading errors after a 2-month mindfulness course. Their distraction and a strengthened present focus (Kok & Singer, error rate dropped by 19% compared to their original perfor- 2017). Tang et al. (2007) found that already after a 5-day mance. In addition, they showed increased sustained atten- meditation training, participants achieved more improved tion. Another study found that intensive meditation practice performance on the attention network test compared to a can reduce mindless reading (Zanesco et al., 2016). Already control group. after a 5-day workshop in mindfulness-based stress reduc- Substantial scientific research has focused on the effects tion, participants of a pilot study demonstrated a significant of meditation on emotion regulation. Meditation practice increase in reading speed (Rice et al., 2020). This promising can reduce negative affect (Sears & Kraus, 2009), regulate field of research is still widely unexplored. Since texts often contain words with an affective content it is also important to examine the effects of meditation practice on emotional * Larissa Lusnig word processing. In particular, it is not clear if different med- larissa.lusnig@uni-wuppertal.de itation styles have dissociable effects on reading speed and how fast single words with different emotional valences are Department of Psychology, University of Wuppertal, Wuppertal, Germany Vol.:(0123456789) 1 3 Psychological Research processed. Basic scientific research on this topic is needed. Our previous study In the current literature, we find evidence that meditation can influence the processing of emotional images. Medita- In a prior study, we examined if meditation has an impact tors demonstrate a reduced emotional reactivity to affective on the responses to affective and neutral words in an LDT images. In a study by Desbordes et al. (2012) participants and on the valence ratings of these words (Lusnig et al., showed reduced amygdala activation while watching emo- 2020). Experienced Zen meditators rated the valence of tional images. A decreased amygdala activation suggests low-arousal positive and low- and high-arousal negative that the affective images had a decreased emotional impact words more neutral, subsequent to a 90-min meditation (Davis & Whalen, 2001). Subjects with extensive meditation session. In an age- and gender-matched comparison group, experience showed less interference from affective images valence ratings did not change significantly after the com- (Ortner et al., 2007). Chau et al. (2018) found that older parison intervention. In the LDT, the Zen group showed an subjects rate affective pictures more neutral subsequent to a accelerated word recognition subsequent to the meditation meditation intervention. While these studies show that medi- intervention. The comparison group did not show a signifi- tation can affect emotional picture evaluation, the impact of cant change in response times (RTs) after the comparison meditation on affective word processing has not yet been intervention. Because the effect of learning to meditate well explored. could not be investigated in these expert meditators, a lon- gitudinal study, examining participants before and after they learned to meditate, is needed. Interindividual differences and word recognition The emotional connotation of experimental word stimuli is The present study usually assessed by valence ratings (Citron, 2012; Jacobs et al., 2015; Kissler et al., 2007; Vo et al., 2009). Based on In the present work, we adapted the design of our previous such ratings researchers can select word material, for exam- study (Lusnig et al., 2020) for a longitudinal study. Two ple, for lexical decision tasks (LDT). During the LDT, par- experimental groups and a control group (CG) participated ticipants are presented with letter strings, which either form in a meditation or control class. Before and after these a word or a meaningless stimulus. Subjects must choose interventions, all groups performed an LDT and a valence- under pressure of time if a word is displayed or not (Meyer rating task, along with a brief mood-states assessment. & Schvaneveldt, 1971; Rubenstein et al., 1970). Word stimu- At baseline, subjects completed assessments on concen- lus properties like arousal, frequency, imageability, valence, tration capacity, intelligence, and personality traits. One and many others have been shown to have an impact on word experimental group participated in a 7-week class in mind- recognition (Brysbaert et al., 2011; Hofmann et al., 2007; fulness meditation. Mindfulness meditation is a widely Kuchinke et al., 2007; New et al., 2006). used meditation style, which is similar to Zen meditation. Interindividual differences can influence single word rec- During mindfulness meditation, the practitioner develops ognition as well. Self-secure subjects demonstrated faster an effortless and non-judgmental awareness of all present- identification of words expressing positive interpersonal moment experiences (Kabat-Zinn, 1990). The second outcomes compared to participants that are more insecure. experimental group practiced loving-kindness meditation Insecure subjects recognized words faster that expressed (LKM) for 7 weeks. LKM contains the practice of empa- negative interpersonal outcomes (Baldwin et al., 1993). In a thy, first the feeling of loving kindness is directed towards valence identification task, dysphoric subjects needed signif- oneself and then towards others (Lutz et al., 2007). The icantly more time to recognize the valence of positive words control group participated for 7 weeks in a study group. than they needed to identify the valence expressed by nega- As in the previous study, we used the low-arousal positive, tive words when contrasted with non-dysphoric participants. low- and high-arousal negative and neutral word stimuli In an LDT, non-dysphoric participants demonstrated faster of Hofmann et al. (2009) and Lusnig et al. (2020). The recognition of negative in comparison to neutral words. quantity of these word stimuli was doubled for the current Dysphoric subjects, on the other hand, responded slower to study. negative than to neutral words (Siegle et al., 2002). Mueller In our prior work (Lusnig et al., 2020), the Zen group and Kuchinke (2016) linked goal-directed behavior to slower responded in an LDT globally faster to the same stimulus processing of fear words. Further, they tied the spontaneous words after the meditation intervention than before the eye-blink rate, which indexes dopaminergic levels, to a pro- meditation. For the LDT of the present study, we expected cessing advantage for happy words. If and to what extent the both meditation groups to process word stimuli faster processing of words may also be influenced by meditation, after the meditation classes than at baseline. Particularly is so far not well understood. 1 3 Psychological Research mindfulness meditation has been associated with increased online advertisements. Subjects were randomly assigned attention (Valentine & Sweet, 1999); therefore, we antici- to the three groups. None of the participants reported prior pated the greatest change in RTs in the mindfulness group experience with meditation or yoga. None of them had a (MG). We predicted that RTs in the control group would history of psychiatric disorders or any reading and writ- not differ significantly when compared before and after the ing difficulties. All participants received course credits for control intervention. In the initial study by Hofmann et al. participation. (2009), subjects responded slower to low-arousal negative words than to neutral words. In our previous meditation work, we replicated this result in the Zen- and the control Covariates group. Consequently, for the present study, the expectation was that all three groups would process low-arousal nega- Participants completed, at first, the “Aktuelle Stim- tive words slower than neutral words. mungsskala” (ASTS) to examine possible changes in mood In Lusnig et  al.’s (2020) valence-rating task, the Zen states provoked by the interventions in every group. The group rated positive, low- and high-arousal negative words ASTS measures subjects’ current mood states; the scales are more neutral after a single 90-min meditation session. The “positive mood”, “sorrow”, “desperateness”, “fatigue” and comparison group did not demonstrate significantly different “anger”. The internal consistency has Cronbach’s α values valence ratings before and after the comparison interven- between α = 0.83 and 0.94. Factor analysis provided one, tion. Based on these results, we expected that both medita- two and four factor-based approaches. The author provides tion groups would rate emotion words more neutral after evidence for convergent and differential validity (Dalbert, the meditation classes. Especially LKM is associated with 1992a, 1992b). Participants conducted this test at baseline an increased experience of positive emotions (Zeng et al., and after the interventions. Possible significant group dif- 2015); therefore, we anticipated more positive valence rat- ferences in personality traits, sustained attention, and intel- ings in the loving-kindness group (LKG) than in the MG. ligence were examined by covariate tests, which subjects We also predicted that the control group would not rate the conducted at baseline. Subjects performed the d2-Revision valence of emotion words significantly different after the test (d2-R), an assessment of sustained attention and the control intervention. Participants performed assessments on ability to focus on task. In the d2-R, subjects have to find personality traits, concentration capacity, and intelligence the letter “d” with two marks while not getting distracted to control for interindividual differences between the three by similar-looking stimuli. Cronbach’s alpha values are groups. We did expect, however, that mood states would for the scales “number of processed target objects” and for change after the meditation interventions in the LKG and “concentration capacity” are between α = 0.89 and 0.95, for the MG, but not in the CG. “percentage of errors” between α = 0.80 and 0.91, depending on the age group. Authors provide empirical evidence for criterion and construct validity (Brickenkamp et al., 2010). Methods and materials This was followed by the Multiple-Choice Vocabulary Intel- ligence Test (MWT-B). Here, participants have to point out Participants a correct German word among four similar nonwords in a multiple-choice procedure. It contains 37 items, which are Thirty-nine German native speakers participated in the sorted by level of difficulty. For the MWT-B retest reliabil- present study. Thirteen of them took part in the LKG (12 ity is reported with a correlation of r = 0.95 after 6 months female, 19–39 years of age, mean age = 22.2), thirteen in and r = 0.87 after 14 months. The author provides empirical the MG (11 female, 19–44 years of age, mean age = 24.5), evidence for criterion validity (Lehrl, 2005). At last, subjects and thirteen in the CG (12 female, 19–42 years of age, performed the Big Five personality test (B5T), which meas- mean age = 21.8). The formula of Westfall et al. (2014) ures personality traits as specified in the Big Five model of was used to calculate an appropriate sample size. Vari- personality. The B5T had in the original version five scales ance partitioning coefficients were estimated based on “neuroticism” (Cronbach’s alpha, α = 0.90), “conscientious- the values of our previous study (Lusnig et al., 2020). To ness” (α = 0.77), “extraversion” (α = 0.87), “agreeableness” obtain a medium effect size of d = 0.5 and a power of 0.8 (α = 0.76) and “openness to experience” (α = 0.76). For while using 400 word-stimuli per participant, a sample the revised version, which was used in the present study, size of 11.8 participants per group is required. As men- three basic requirements “need for power and influence” tioned above, we worked with a sample size of 13 par- (α = 0.78), “need for safety and peace” (α = 0.84), and “need ticipants per group. Thirty-eight participants were right- for achievement and performance” (α = 0.82) were added. handed one was left-handed. All subjects were students The test demonstrates factorial validity (Satow, 2011). of the University of Wuppertal; they were recruited via 1 3 Psychological Research Stimuli Procedure The stimulus set included 400 words (German nouns) and Pre‑test 400 altered words (nonwords). Half of the item set was taken from Hofmann et al. (2009), while the other half was Subjects first completed the ASTS, then the d2-R, the MWT- generated using the same construction rules. All stimulus B, and the B5T. Subsequently, subjects took part in the main words were part of the revised Berlin Affective Word List experiment. Stimuli were presented on a 21-inch TFT dis- (BAWL-R; Vo et al., 2009). The stimulus set contained four play running at 70 Hz, with the distance between eyes and stimulus conditions: low-arousal negative, high-arousal monitor held constant at approximately 65 cm. negative, (low-arousal) positive, and neutral words. The Participants with even participant numbers were pre- stimulus condition “High-arousal positive words” could not sented with Subset A of the stimuli set. They were asked to be generated because the BAWL-R did not provide enough press key “2”, using the left index finger when identifying a high-arousal positive nouns (cf. Hofmann et al., 2009; Lus- nonword, and to press key “8”, using the right index finger nig et al., 2020; see Fig. 1 in Vo et al., 2009). Each of the when they recognized the stimulus as a word. After par- four stimulus conditions of the whole item set contained ticipants had processed the first half of Subset A, they were 100 words. Low-arousal negative, positive and neutral words presented with the second half of subset A. To balance for were matched for arousal. High-arousal negative words were RT differences due to hand dominance, subjects now pressed matched for valence to low-arousal negative words, but their key “2”, using the left index finger when identifying a word, arousal values were maximized. Nouns were matched for the and pressed key “8”, using the right index finger when they following psycholinguistic variables, which can influence identified a nonword. This way we also excluded the risk LDTs: arousal, word frequency, emotional valence, number that possible effects were caused by the composition of one of orthographic neighbors, number of letters, imageability, of the subsets, for instance by mood induction due to many mean bigram frequency (type), and mean bigram frequency words of the same emotion category (Niedenthal & Setter- (token) (see Table 1, and Table 1 in Hofmann et al., 2009). lund, 1994). Participants with uneven participant numbers The selected nouns were modified to create nonwords. For were presented first with the first half of Stimulus Subset B, this purpose, a vowel of a word stimulus was replaced by pressing the left index finger for a word. In the second half a consonant or another vowel. Stimuli were subdivided in of Subset B, hand assignment was reversed. In any case, Subset A and Subset B, each consisting of 200 words. To responses were to be executed as quickly and accurately as test whether the materials were matched for these variables, possible. we conducted 2 × 4 ANOVAs (subset × emotion condition; Before the LDT, participants were made familiar with all Fs < 1). the test by responding to five practice stimuli. The word stimuli appeared in black uppercase letters on a light gray background, in 20 pt Times New Roman font using presen- tation-software PsychoPy, version 1.82.01 (Peirce, 2007). Stimuli were pseudorandomized. At most three words or nonwords were presented consecutively. During each trial, for 700 ms, a fixation cross (+) was displayed, followed by Table 1 Mean values and standard errors of the manipulation and control variables for high-arousal negative, low-arousal negative, positive, and neutral words Control and manipulation variables High-arousal negative Low-arousal negative Neutral words Positive words words words M SE M SE M SE M SE Emotional valence − 1.31 0.05 − 1.30 0.04 0.06 0.04 1.17 0.07 Arousal 3.90 0.03 2.83 0.04 2.82 0.03 2.82 0.03 Imageability 4.50 0.17 4.11 0.20 4.24 0.16 4.26 0.16 Number of letters 6.42 0.17 6.06 0.21 6.16 0.18 6.18 0.19 Word frequency 12.94 0.31 13.20 0.38 12.62 0.29 12.56 0.27 Number of orthographic neighbors 0.98 0.24 1.50 0.30 1.20 0.24 1.18 0.27 Mean bigram frequency (type) 3113.55 284.41 3408.68 245.04 3525.47 272.50 3223.81 277.65 Mean bigram frequency (token) 167,061.37 19,957.41 189,351.71 21,027.04 190,835.14 19,536.51 178,481.74 20,794.76 1 3 Psychological Research a word or nonword for 1000 ms. A white screen was shown feelings and emotions and situations of others, also using for 500 ms and then for 1500 ms a mask (#####) (see Fig. 1 visual imagery. in Lusnig et al., 2020). After a 5-min break, subjects rated the valence of the words, which they had seen before in the Control intervention LDT. One word at a time was presented on the screen. Below every word, a seven-point grading scale from − 3 to 3 was The study group aimed to involve subjects of the CG in a displayed (0 = neutral, −  3 = ver y negative, 3 = ver y posi- silent active control intervention very close to their usual tive). Subjects gave their responses by clicking the respec- activities. All participants were college students; therefore, a tive number with the cursor. After pressing the space bar, the study group was selected as an adequate intervention. Partic- next word appeared on the screen. The duration of the LDT ipants were instructed to study silently for a class they were was about 20 min followed by a 5-min break, the valence rat- currently attending. An undergraduate assistant supervised ing task lasted for approximately 12 min. The whole experi- the study group. ment, including also covariate tests and instructions, lasted approximately 75 min. Data analysis Post‑test For the LDT, RTs of the correctly given answers were ana- lyzed using linear mixed-effects models (LMEs). 14.95% of Subjects took part in the post-test about 1 week after having all responses were incorrect, they were excluded from the finished one of the meditation classes or the study group. data analysis. The models were calculated with the statis- Participants first finished the ASTS, other covariate tests tical software environment R, (version 3.4.2, http:// cran.r- were not conducted during the post-test. The instructions proje ct. org). Specifically, we used the lme4 library with the and procedures for the LDT were the same as the ones given lmer function (version 1.1–14, Bates et al., 2015). The lmer in the pre-test. Participants with even numbers were pre- function fits an LME to the data. An LME data analysis sented with stimulus Subset B, subjects with uneven par- considers participant and item variance concurrently in a ticipant numbers completed Subset A. Participants took a non-hierarchical approach. Averaging at first level treats the 5-min break. Then, they rated the valence of the words they error variance of items as fixed effects. Separate random had seen during the post-LDT. intercepts for subjects and items result in treating subject and item variance more sensitively. “Subjects” and “items” were Meditation or comparison interventions fitted as random effects. As fixed effects, we fitted “groups” (LKG/MG/CG), “time” (before/after intervention) and The week after the pre-test, all subjects took part in a 1.5-h “emotional valence” (positive words/low-arousal negative/ intervention in the morning on the same day of the week. high-arousal negative). Fixed effects were represented with The training took place in 8 consecutive weeks. In the fourth the use of effects coding. Low-arousal negative, high-arousal week, training was suspended due to a national holiday. A negative, and positive words were confronted with neutral local experienced meditation-trainer led both meditation words. We calculated a time slope for the random effect groups. “item” because the time-series effect might be different for different items (Baayen et al., 2008). For every group, sepa- Mindfulness meditation intervention rate LMEs were calculated to examine the origin of signifi- cant interactions. The dependent variable “response time” Participants of the MG learned at first which sitting postures was log-transformed to satisfy the assumption of normality are adequate for meditation, how to relax their body, and of the residuals, which were verified by qqplots. Estimates of how to breathe calmly and naturally. Then, they practiced the regression coefficients, their standard errors, and t values observing their thoughts, emotions, and physical feelings are reported for the LMEs. P values are reported on the basis and tried to let them go instead of being caught up in them. of the Satterthwaite approximation, which is implemented in the lmerTest package, (Version 2.0-36, Kuznetsova et al., Loving‑kindness meditation intervention 2017). Regarding the valence rating experiment, we examined First, subjects of the LKG practiced suitable sitting postures responses given on a seven-point grading scale ranging from for meditation, how to exercise a natural and calm breath, − 3 to 3. An LME was used again to analyze the experimen- and technics to relax their body. They learned how to be tal data. The procedure of the data analysis was the same aware of their thoughts, emotions, and feelings and trained as for the response time experiment. Data points, which to develop equanimity and self-empathy regarding these were not in a range between – 3 and 3 standard deviations states. Gradually subjects broadened their empathy to their of the residual error, were discarded from the calculations 1 3 Psychological Research (approximately 1% of the data). To analyze the results of and sorrow; for the discussed effects of the present study, the ASTS we conducted a 3 × 2 × 5 ANOVA with the factors however, it does not have relevance. In addition, subjects “group”, “time” and “mood conditions”. were tested at baseline for group differences in personal- ity traits, intelligence, and concentration capacity to control for influences of these factors on the RT results. In these Results comparisons, no significant group differences were found. All statistical values for group differences, mean values, Covariates and standard deviations of the baseline covariate tests are reported in Table 3. To control for a possible change in mood induced by the interventions, subjects completed the ASTS at baseline and Lexical decision and valence rating data after interventions. Mood did not change significantly over time points in any of the groups. Mood conditions (posi- Figure 1 shows that after the loving-kindness- and the mind- tive mood, sorrow, desperateness, fatigue, and anger) were fulness-meditation sessions affective valence ratings were rated significantly different (see Table  2). This effect dem- more neutral, except for valence ratings to positive words in onstrates, for example, the difference between positive mood the LKG, which became more positive. Figure 2 shows that RTs in the LDT were shorter after both meditation inter- ventions, but not after the control intervention. The LME Table 2 Statistical group differences and mean squared errors (MSE) analysis revealed for the valence rating as well as for lexical of ASTS (tested at baseline and after intervention) decision data significant interactions of “group” and “time” Statistical group differences (see Table  4 for entire analysis). The valence rating data showed for “time” as well as for “group” significant two- Time F(1, 36) = 2.31, p = 0.14, way interactions with “positive”, “high-arousal negative”, MSE = 4.13 and “low-arousal negative” words. Further, we found signifi- Time: Group F(2, 36) = 2.39, p = 0.11 cant main effects for “high-arousal negative”, “low-arousal Mood conditions F(1, 36) = 496.85, p = 0.00, MSE = 54.32 negative” and “positive” words. The analysis of the lexical Mood conditions: Group F(2, 36) = 1.04, p = 0.39 decision data revealed for "low-arousal negative" words a Time: Mood conditions F(2, 36) = 0.29, p = 0.78, significant main effect. MSE = 26.47 Separate LMEs were subsequently calculated for all three Time: Mood conditions: Group F(2, 36) = 0.61, p = 0.66 groups to resolve the significant interactions of “time” and “group” (cf. Table  5). For the valence rating and lexical Time (before/after intervention), Group (LKG/MG/CG), mood condi- tions (positive mood, sorrow, desperateness, fatigue, and anger) Table 3 Baseline covariate tests: statistical group differences, mean squared errors (MSE), mean values, and standard deviations Covariate test Statistical group differences LKG MG CG M SD M SD M SD Big Five personality test Neuroticism F(2, 36) = 2.37, p = 0.11, MSE = 2.01 4.85 1.86 4.77 1.34 3.77 1.01 Extraversion F(2, 36) = 0.46, p = 0.63, MSE = 3.16 5.69 1.60 5.77 2.39 6.31 1.11 Conscientiousness F(2, 36) = 1.30, p = 0.29, MSE = 4.39 4.23 1.92 5.54 2.15 4.69 2.21 Agreeableness F(2, 36) = 1.18, p = 0.32, MSE = 2.41 5.15 1.68 5.92 1.71 6.00 1.23 Need for safety and peace F(2, 36) = 2.96, p = 0.06, MSE = 1.74 4.62 1.19 5.46 1.56 4.23 1.16 Need for power and influence F(2, 36) = 0.71, p = 0.50, MSE = 2.86 4.54 1.81 4.31 1.55 5.08 1.71 Openness F(2, 36) = 2.83, p = 0.07, MSE = 2.07 5.77 1.59 5.38 1.04 4.46 1.61 Need for achievement and performance F(2, 36) = 0.01, p = 0.99, MSE = 3.03 5.00 1.87 5.08 1.89 5.00 1.41 d2-Revision, test concentration capacity Number of processed target objects F(2, 36) = 2.18, p = 0.13, MSE = 465.78 163.77 22.06 179.00 21.72 179.15 20.96 Concentration capacity F(2, 36) = 1.91, p = 0.16, MSE = 666.45 143.38 28.31 161.38 28.47 159.53 19.68 Percentage of errors F(2, 36) = 1.89, p = 0.17, MSE = 22.52 11.37 4.72 8.24 4.35 11.37 5.14 Intelligence test (MWT-B) F(2, 36) = 2.05, p = 0.14, MSE = 6.78 22.23 2.52 23.37 2.36 21.31 2.89 1 3 Psychological Research LKG MG CG 1.5 0.5 -0.5 -1 -1.5 -2 A B C D A B C D A B C D A= high-arousal negative words, C= neutral words before after B = low-arousal negative words, D = positive words Fig. 1 Results of the valence rating experiment. Error bars indicate standard errors decision data, we found significant main effects for “time” The valence rating experiment in the LKG and the MG, but not in the CG. Valence ratings differed significantly after the interventions in both meditation groups. After the control intervention, valence ratings did not change. Figure 1 demonstrates that Discussion the MG and the LKG rated words more neutral after the meditation classes. The LKG, however, rated positive words In the present study, we did not find any baseline differ - more positively after the meditation intervention. These ences in intelligence, concentration capacity, and personal- results are in line with previous work on meditation and ity traits between the three groups. Contrary to our expec- emotion regulation. Several studies showed that particularly tations, mood changes could not be detected in the ASTS mindfulness meditation can down-regulate negative emo- after any of the three interventions. Looking at word items, tions such as anxiety (Hofmann et al., 2010; Tang et al., both meditation groups rated valence more neutral after the 2007), stress (Goyal et al., 2014), and depression (Grecucci meditation interventions. Participants in the LKG, however, et al., 2015; Tang et al., 2007). The practice of LKM has rated positive words more positively after the intervention. also been associated with increased positive emotions (Fre- In the control group, valence ratings did not differ signifi- drickson et al., 2008; Hutcherson et al., 2008; Zeng et al., cantly after control intervention. Concerning the LDT, both 2015). However, contrary to our expectations, we did not meditation groups demonstrated faster word recognition find evidence for changes in mood states after any of the after the interventions. This effect was most pronounced for three interventions. The reason for this could be that the participants in the LKG. The control group did not show mood assessment we used (ASTS) might not be sensitive significantly different RTs after the intervention. Hofmann or specific enough to capture the emotion regulation pro- et al. (2009) and our previous study found that low-arousal duced by meditation. On the other hand, these results may negative words were processed slower than neutral words. indicate that meditation practice does not always lead to We replicated this effect in all three groups. a reduced experience of emotions. The practitioners may 1 3 Valence Ratings Psychological Research LKG MG CG A B C D A B C D A B C D A= high-arousal negtive words, C= neutral words before after B = low-arousal negative words, D = positive words Fig. 2 Results of the lexical decision experiment. Error bars indicate standard errors Table 4 Valence rating and LDT Experiments: estimates of regression coefficients, their standard errors, t values, p values, and Cohen’s d effect sizes of the overall analyses Valence ratings Lexical decision task B SE t p d B SE t p d Time 0.013 0.024 0.54 0.592 0.01 – 0.002 0.004 – 0.55 0.586 – 0.02 Group – 0.009 0.032 – 0.28 0.783 – 0.15 0.006 0.016 0.41 0.688 0.13 High-arousal negative – 1.273 0.052 – 24.57 0.001*** – 1.91 0.001 0.008 0.01 0.997 0.01 Low-arousal negative – 0.933 0.052 – 18.01 0.001*** – 1.40 0.026 0.008 3.28 0.001** 0.27 Positive 1.739 0.052 33.59 0.001*** 2.64 – 0.006 0.008 – 0.82 0.412 – 0.07 Time:group 0.039 0.015 2.70 0.007** 0.04 – 0.022 0.003 – 8.21 0.001*** – 0.15 Time:high-arousal negative 0.084 0.042 2.00 0.045* 0.08 – 0.006 0.006 – 0.96 0.335 – 0.04 Group:high-arousal negative 0.128 0.018 7.08 0.001*** 0.09 – 0.001 0.003 – 0.35 0.730 – 0.01 Time:low-arousal negative 0.106 0.042 2.53 0.011* 0.09 – 0.001 0.006 – 0.14 0.890 – 0.01 Group:low-arousal negative 0.102 0.018 5.66 0.001*** 0.07 – 0.003 0.003 – 0.85 0.394 – 0.02 Time:positive – 0.106 0.042 – 2.53 0.011* – 0.11 – 0.002 0.006 – 0.31 0.758 – 0.01 Group:positive – 0.180 0.018 – 9.95 0.001*** – 0.13 0.001 0.003 0.30 0.768 0.01 Time:Group:high-arousal negative 0.017 0.026 0.68 0.499 0.01 0.004 0.005 0.89 0.373 0.02 Time:Group:low-arousal negative – 0.021 0.026 – 0.82 0.412 – 0.01 0.001 0.005 0.08 0.933 0.01 Time:Group:positive 0.037 0.026 1.44 0.149 0.02 – 0.001 0.005 – 0.28 0.777 – 0.01 “***”p < 0.001, “**”p < 0.01, “*”p < 0.05 1 3 Response Times (in ms.) Psychological Research Table 5 Valence rating and LDT experiments: estimates of regression coefficients, their standard errors, t values, p values, and Cohen’s d effect sizes for the LKG, MG, and CG Valence ratings Lexical decision task B SE t p d B SE t p d LKG Time 0.078 0.026 3.08 0.002** 0.31 – 0.048 0.004 – 12.41 0.001*** – 0.41 High-arousal negative – 0.980 0.051 – 19.22 0.001*** – 1.92 – 0.003 0.008 – 0.32 0.747 – 0.03 Low-arousal negative – 0.728 0.051 – 14.27 0.001*** – 1.43 0.016 0.008 1.98 0.048* 0.20 Positive 1.361 0.051 26.69 0.001*** 2.67 – 0.004 0.008 – 0.55 0.585 – 0.06 Time:high-arousal negative 0.094 0.044 2.12 0.035* 0.21 – 0.001 0.007 – 0.21 0.833 – 0.01 Time:low-arousal negative 0.053 0.044 1.20 0.231 0.12 0.006 0.007 0.81 0.417 0.03 Time:positive – 0.011 0.044 – 0.25 0.801 – 0.03 – 0.004 0.007 – 0.57 0.569 – 0.02 MG Time 0.125 0.025 4.91 0.001*** 0.19 – 0.022 0.004 – 5.96 0.001*** – 0.62 High-arousal negative – 1.172 0.054 – 21.78 0.001*** – 2.14 0.001 0.008 0.01 0.993 0.01 Low-arousal negative – 0.768 0.054 – 14.28 0.001*** – 1.41 0.027 0.008 3.49 0.001*** 0.35 Positive 1.554 0.054 28.90 0.001*** 2.85 – 0.005 0.008 – 0.70 0.484 – 0.07 Time:high-arousal negative 0.157 0.044 3.56 0.004*** 0.14 – 0.005 0.006 – 0.78 0.435 0.08 Time:low-arousal negative 0.115 0.044 2.60 0.009** 0.10 – 0.013 0.006 – 1.94 0.053 – 0.19 Time:positive – 0.194 0.044 – 4.40 0.001*** – 0.17 – 0.002 0.006 – 0.27 0.785 – 0.03 CG Time – 0.026 0.027 – 0.95 0.342 – 0.03 – 0.001 0.004 – 0.33 0.739 – 0.03 High-arousal negative – 1.215 0.056 – 21.65 0.001*** – 2.14 – 0.003 0.008 – 0.33 0.745 – 0.03 Low-arousal negative – 0.930 0.056 – 16.57 0.001*** – 1.64 0.023 0.008 2.78 0.006** 0.29 Positive 1.707 0.056 30.40 0.001*** 3.00 – 0.004 0.008 – 0.53 0.594 – 0.06 Time:high-arousal negative 0.047 0.047 1.01 0.311 0.03 – 0.008 0.007 – 1.23 0.221 – 0.13 Time:low-arousal negative 0.077 0.047 1.66 0.098 0.05 0.006 0.007 0.90 0.368 0.09 Time:positive – 0.050 0.047 – 1.06 0.287 – 0.03 – 0.003 0.007 – 0.52 0.603 – 0.06 “***”p < 0.001, “**”p < 0.01, “*”p < 0.05 still experience the emotions the way they did before but do et al. (2007), LKM can be seen as a special case of OMM not judge them and do not get carried away by them. The because it contains the “cultivation of objectless aware- fact that we did not find changes in mood states, however, ness” and “non-referential compassion”. However, it con- ensures that the neutralized valence ratings occurred due to tains also phases of focused attention meditation (FAM), the meditation interventions and were not influenced by a during which the meditator keeps the attention all the time momentary mood change. on one object. In the case of LKM, this object is the feel- In our previous study (Lusnig et al., 2020), adept Zen ing of loving-kindness, which is directed towards oneself meditators assigned to words significantly more neutral or other single persons (Lutz et al., 2007; Vago & Silber- valence ratings after a 90-min Zen meditation. In the pre- sweig, 2012). In the present work, the LKG rated positive vious comparison group, valence ratings did not change words more positively after meditation. In the MG and the after the comparison intervention. These results are in Zen group of the previous study, we did not find such an line with our findings for the MG in the present study. effect. LKM differs from mindfulness- and Zen meditation The similarity of these results was to be expected because because it contains the practice of empathy and positive Zen meditation and mindfulness meditation are compa- feelings towards others. This difference in meditation prac- rable styles of meditation. Both meditation styles belong tice may have led to the more positive valence ratings in to the category of open monitoring meditation (OMM) the LKG. A study by Hunsinger et al. (2013) found results (Lutz et al. 2008), during which the meditator monitors, similar to our study. In their work, loving-kindness nov- in a non-judgmental way, everything that occurs in his ices associated significantly more positivity with neutral moment-to-moment experience, such as sounds, thoughts stimuli after a meditation intervention compared to control that pass the mind, smells, or feelings. According to Lutz participants. 1 3 Psychological Research have helped the participants of the meditation groups to The lexical decision experiment focus more closely on the current word stimulus, enabling faster responses. Half of the stimulus set used in the present study was identical to the one used by Hofmann et al. (2009). They Improved visual discrimination could be another pro- cess that may have contributed to faster RTs in the MG and found, among other results, low-arousal negative words being processed slower than neutral words. In the pre- the LKG. Expert meditators demonstrate visual attentional processing, which is more accurate and flexible in contrast sent study, we replicated this effect in all three groups. In our previous work (Lusnig et al., 2020), we obtained to control subjects’ visual processing. For example, medi- tators notice changes in flickering scenes faster than con- this result in the Zen and the control group. In the cur- rent literature, it is discussed if positive or negative visual trols (Hodgins & Adair, 2010). Brown et al. (1984) tested Buddhist meditators before and after a 3-month meditation stimuli are processed more rapidly. For example, Öhman et al. (2001) found that threatening faces are processed retreat for visual sensitivity. After the meditation interven- tion, meditators noticed shorter single-light flashes and faster than friendly faces. On the other hand, a study by Becker et  al. (2012) demonstrated that dynamic happy could differentiate better successive light-flashes than before the retreat. A control group did not show any such changes facial expressions are detected faster than dynamic angry facial expressions. Hofmann et al. (2009) found that the in visual sensitivity. In a study by MacLean et al. (2010), meditation novices improved after a 3-month meditation arousal level affects the processing speed of emotional single words. In their study, high-arousal negative words training visual discrimination, perceptual sensitivity, and increased vigilance during visual attention. This evidence and positive words are processed faster than low-arousal negative words. In the present study, we found the same points to the possibility that in the present study meditation training might have led to improved visual sensitivity and descriptive result pattern in all three groups (see Fig. 2). The MG, however, demonstrates after the meditation inter- discrimination performance, which in turn allowed for faster responses in the LDT. vention no difference in processing speed for high-arousal negative, low-arousal negative, and positive words (see We expected that the MG would show the largest decrease in RTs after intervention because especially mindful- Fig.  2). This might be because the profound practice of equanimity in mindfulness meditation minimizes the dif- ness meditation has been associated with increased atten- tion (Semple, 2010; Valentine & Sweet, 1999). There was ference in arousal level for negative words. In both meditation groups of the present study, but not indeed a substantial response acceleration, but in our data, this effect was even more pronounced in the LKG. These the control group, RTs changed significantly after the interventions. As illustrated in Fig.  2, RTs to emotional results might be explained considering the association between meditation styles and narrow or broad attentional words were faster after both meditation interventions. These results are also in accordance with those of our focus. Lippelt et al. (2014) argued that FAM, which con- tains mainly a strong concentration on a single object, leads previous study, in which the Zen group demonstrated a significantly faster word recognition after a 90-min medi - to a narrow focus of attention. During OMM the meditator monitors all experiences non-judgmentally. This medita- tation session. Meditation is associated with increased attention (Carter et  al., 2005; Chambers et  al., 2008; tion style is, therefore, thought to lead to a broad attentional focus. Such a broadened attentional focus was shown to MacLean et al., 2010; Chan & Woollacott, 2007; van den Hurk et al., 2010). Hence, it appears plausible to conclude promote better performance on an attention task (Willems & Martens, 2016). In the present study, a broad attentional that increased attentional resources in the meditation groups may have led to accelerated word recognition. As focus, induced by the mindfulness meditation, might have led to a more effective and therefore faster word process- an alternative account, the shorter RTs could be associated with reduced mind-wandering as a result of meditation. ing. The practice of LKM contains open monitoring, the main goal of this meditation style is to broaden the feeling Using functional MRI, Brefczynski-Lewis et al. (2007) found that expert meditators showed less brain activation of loving-kindness starting from a person we like to every- one. This broadened attentional focus combined with the in the default mode network. The default mode network is associated with discursive thoughts. Similarly, a study strong cultivation of loving-kindness might have helped the LKG to process the emotional words especially fast. Since by Pagnoni et al. (2008) found in meditators decreased neural activity in default mode network regions. These positive affect is also associated with a broader attentional focus (Fredrickson & Branigan, 2005), the feeling of loving- authors propose that meditators may be able ‘to control the automatic cascade of semantic associations’ better than kindness might have given an additional speed boost in the LKG. It would be very interesting for subsequent research control subjects (Pagnoni & Cekic, 2007, p. 1). Therefore, spontaneous mind-wandering could be regulated more eas- to compare not just the influence of mindfulness meditation and LKM on word processing but also the effects of FAM. ily. In the present study, regulated mind-wandering may 1 3 Psychological Research This way the effects of narrowed and broadened attentional an object like the breath and calm themselves down. On focus could be compared. the other hand, Richards et  al. (2014) argued that peo- ple with anxiety disorders show, in confrontation with a Future directions and limitations specific threatening stimulus, a narrow attentional focus. OMM could help persons, in such situations, to deliber- Concerning the sequence of experimental and covariate tests ately broaden their attention and thereby allow them to in the present study, it might have been better to give to the detach their attention from the threatening stimulus. participants first the ASTS, then the LDT and the valence rating task, and at last the remaining covariate tests. This way the mood states, which were measured with the ASTS could not have been changed through tiring covariate tests. Conclusions However, since in the present study the ASTS results were not influenced by the meditation intervention, the test plays In a previous study Lusnig et  al. (2020), we found that a minor role in the interpretation of our study, and we do not advanced Zen meditation can neutralize valence ratings of see problems with the sequence of the tests. emotional words and accelerated RTs to these words. In the For future studies on the influence of meditation and present study, we were able to obtain similar results with a visual word processing it might also be beneficial to use longitudinal study design. Subjects, which participated in assessments on emotion regulation and increased attention a 7-week loving-kindness- or mindfulness course, demon- not just before the meditation intervention, as in the present strated significantly more neutral valence ratings after the study, but also after it. With such an experimental design it interventions. Positive words were rated more positively could be extensively examined which of these underlying after the LKM course. These results suggest that different mechanisms of meditation influences altered word process- meditation styles can contribute to the down- and up-regu- ing in meditation practitioners the most. It would be espe- lation of emotions. However, contrary to our expectations, cially informative to investigate further the role of emotional mood states did not appear to change after meditation inter- variability. We would suggest that subjects should conduct ventions. In both meditation groups, RTs were faster after adequate emotion assessments at baseline, two times during the interventions than before, with the largest changes occur- the intervention and after it. ring in the LKG. Improved increased attention, visual dis- The present study and our previous study (Lusnig et al., crimination, and reduced mind-wandering, caused by medi- 2020) focused on the question if meditation practice can tation, may have enabled accelerated word recognition. The influence visual single word processing. Since we found results of the present study could help to understand better that meditation practice can accelerate the responses to the influence of meditation in text processing of affectively single words and neutralize the valence ratings of emotion loaded content. words, it would be interesting to examine in the future if meditation might have similar effects on the processing of Author contributions LL, MH, and RR planned the study. LL con- entire written texts. If the effects of meditation practice on ducted the experiments. MH and LL planned and performed the analy- written texts are similar to those on single words, it can be ses. LL, MH, and RR wrote the paper. All authors read and approved assumed that meditation can accelerate the reading speed the final manuscript. of practitioners, as suggested also by a pilot study by Rice et al. (2020). It seems also important to examine clearly Funding Open Access funding enabled and organized by Projekt DEAL. Financial disclosure statement. This work was partly sup- how meditation styles lead to a narrow or broad attentional ported by a grant of the Deutsche Forschungsgemeinschaft to MH focus and how such a focus affects word and text process- (HO 5139/2-2) and by a doctoral studies grant of the Evangelisches ing. Lippelt et al. (2014) proposed that FAM leads to a nar- Studienwerk e.V. to LL. The funders had no role in study design, data row attentional focus and that OMM induces a broadened collection and analysis, decision to publish, or preparation of the manu- script. There was no additional external funding received for this study. attentional focus. It seems important to examine clearly in experimental studies how these attentional foci affect Availability of data and material The datasets analyzed dur- word and text processing. The further examination of nar- ing the present study are available from the corresponding row and broad attentional foci, triggered by meditation, author on reasonable request. could also have important implications for the treatment of anxiety disorders. Richards et al. (2014) claimed that Code availability The code used for the data analysis of the anxious persons, who show hypervigilance, demonstrate a present study is available from the corresponding author on broadened attentional focus that scans for potential threats reasonable request. in the environment. The practice of FAM could be helpful for these individuals to deliberately narrow their focus on 1 3 Psychological Research Chambers, R., Lo, B. C. Y., & Allen, N. B. (2008). The impact of inten- Declarations sive mindfulness training on attentional control, cognitive style, and affect. Cognitive Therapy and Research, 32, 303 Conflict of interest/competing interests The authors declare that they Chan, D., & Woollacott, M. (2007). Effects of level of meditation have no conflict of interest. experience on attentional focus: Is the efficiency of executive or orientation networks improved? Journal of Alternative and Com- Ethics approval The present study was carried out following the ethi- plementary Medicine, 13(6), 651–657 cal standards provided by the Declaration of Helsinki and the German Chau, B. K. H., Keuper, K., Lo, M., So, K.-F., Chan, C. C. H., & Lee, Society for Psychology (DGPs). The Ethics Committee of the Univer- T. M. C. (2018). Meditation-induced neuroplastic changes of the sity of Wuppertal approved the study. prefrontal network are associated with reduced valence perception in older people. Brain and Neuroscience Advances. https://doi. or g/ Consent to participate All subjects gave their written informed con- 10. 1177/ 23982 12818 771822 sent. Citron, F. M. M. (2012). Neural correlates of written emotion word processing: A review of recent electrophysiological and hemo- Consent for publication We did not use participant identifying infor- dynamic neuroimaging studies. Brain and Language, 122, mation. 211–266 Corcoran, K. A., Desmond, T. J., Frey, K. A., & Maren, S. (2005). Hippocampal inactivation disrupts the acquisition and contex- Open Access This article is licensed under a Creative Commons Attri- tual encoding of fear extinction. Journal of Neuroscience, 25, bution 4.0 International License, which permits use, sharing, adapta- 8978–8987 tion, distribution and reproduction in any medium or format, as long Dalbert, C. (1992a). ASTS - Aktuelle Stimmungsskala. In Leibniz- as you give appropriate credit to the original author(s) and the source, Zentrum für Psychologische Information und Dokumentation provide a link to the Creative Commons licence, and indicate if changes (ZPID) (Hrsg.), Elektronisches Testarchiv (PSYNDEX Tests- were made. The images or other third party material in this article are Nr. 9003796). Trier: ZPID. included in the article’s Creative Commons licence, unless indicated Dalbert, C. (1992b). Subjektives Wohlbefinden junger Erwachsener: otherwise in a credit line to the material. If material is not included in Theoretische und empirische Analysen der Struktur und Stabil- the article’s Creative Commons licence and your intended use is not ität. Zeitschrift für Differentielle und Diagnostische Psycholo- permitted by statutory regulation or exceeds the permitted use, you will gie, 13, 207–220 need to obtain permission directly from the copyright holder. To view a Davis, M., & Whalen, P. J. (2001). The amygdala: Vigilance and copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . emotion. Molecular Psychiatry, 6, 13–34 Desbordes, G., Negi, L. T., Pace, T. W., Wallace, B. A., Raison, C. L., & Schwartz, E. L. (2012). Effects of mindful-attention and compassion meditation training on amygdala response to emo- References tional stimuli in an ordinary, non-meditative state. Frontiers in Human Neuroscience, 6, 292 Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects Fredrickson, B. L., & Branigan, C. (2005). Positive emotions modeling with crossed random effects for subjects and items. broaden the scope of attention and thought-action repertoires. Journal of Memory and Language, 59, 390–412 Cognition & Emotion, 19(3), 313–332 Baldwin, M. W., Fehr, B., Keedian, E., Seidel, M., & Thomson, D. Fredrickson, B. L., Cohn, M. A., Coffey, K. A., Pek, J., & Finkel, S. W. (1993). An Exploration of the relational schemata underlying M. (2008). Open hearts build lives: Positive emotions, induced attachment styles. Self-report and lexical decision approaches. through loving-kindness meditation, build consequential per- Personality and Social Psychology Bulletin, 19(6), 746–754 sonal resources. Journal of Personality and Social Psychology, Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting Linear 95(5), 1045–1062 Mixed-Effects Models Using lme4. Journal of Statistical Soft- Goyal, M., Singh, S., Sibinga, E. M. S., Gould, N. F., Rowland- ware, 67, 1 Seymour, A., Sharma, R., et al. (2014). Meditation programs Becker, D. V., Neel, R., Srinivasan, N., Neufeld, S., Kumar, D., & for psychological stress and well-being: A systematic review Fouse, S. (2012). The vividness of happiness in dynamic facial and meta-analysis. JAMA Internal Medicine, 174(3), 357–368 displays of emotion. PLoS ONE, 7(1), e26551 Grecucci, A., Pappaianni, E., Siugzdaite, R., Theuninck, A., & Job, Brefczynski-Lewis, J. A., Lutz, A., Schaefer, H. S., Levinson, D. B., & R. (2015). Mindful emotion regulation: Exploring the neuro- Davidson, R. J. (2007). Neural correlates of attentional expertise cognitive mechanisms behind mindfulness. BioMed Research in long-term meditation practitioners. Proceedings of the National International, 2015, 670724 Academy of Sciences of the United States of America., 104(27), Hodgins, H. S., & Adair, K. C. (2010). Attentional processes and 11483–11488 meditation. Consciousness and Cognition, 19(4), 872–878 Brickenkamp, R., Schmidt-Atzert, L., & Liepmann, D. (2010). d2-R: Hofmann, M. J., Kuchinke, L., Tamm, S., Võ, M.L.-H., & Jacobs, A. Test d2—Revision. Hogrefe. M. (2009). Affective processing within 1/10th of a second: High Brown, D., Forte, M., & Dysart, M. (1984). Visual sensitivity and arousal is necessary for early facilitative processing of nega- mindfulness meditation. Perceptual and Motor Skills, 58(3), tive but not positive words. Cognitive, Affective, & Behavioral 775–784 Neuroscience, 9, 389–397 Brysbaert, M., Buchmeier, M., Conrad, M., Jacobs, A. M., Bölte, J., & Hofmann, M. J., Stenneken, P., Conrad, M., & Jacobs, A. M. (2007). Böhl, A. (2011). The word frequency effect: A review of recent Sublexical frequency measures for orthographic and phono- developments and implications for the choice of frequency esti- logical units in German. Behavior Research Methods, 39(3), mates in German. Journal of Experimental Psychology, 58(5), 620–629 412–424 Hofmann, S. G., Sawyer, A. T., Witt, A. A., & Oh, D. (2010). The effect Carter, O. L., Presti, D. E., Callistemon, C., Ungerer, Y., Liu, G. B., of mindfulness-based therapy on anxiety and depression: A meta- & Pettigrew, J. D. (2005). Meditation alters perceptual rivalry in analytic review. Journal of Consulting and Clinical Psychology, Tibetan Buddhist monks. Current Biology, 15, R412–R413 78(2), 169–183 1 3 Psychological Research Hölzel, B., Carmody, J., Vangel, M., Congleton, C., Yerramsetti, S. from the English Lexicon Project. Psychonomic Bulletin & M., Gard, T., et al. (2011). Mindfulness practice leads to increases Review, 13(1), 45–52 in regional brain gray matter density. Psychiatry Research, 191, Niedenthal, P. M., & Setterlund, M. B. (1994). Emotion congruence 36–43 in perception. Personality and Social Psychology Bulletin, 20(4), Hülsheger, U. R., Alberts, H. J., Feinholdt, A., & Lang, J. W. (2013). 401–411 Benefits of mindfulness at work: The role of mindfulness in emo- Öhman, A., Lundqvist, D., & Esteves, F. (2001). The face in the crowd tion regulation, emotional exhaustion, and job satisfaction. Jour- revisited: A threat advantage with schematic stimuli. Journal of nal of Applied Psychology, 98(2), 310–325 Personality and Social Psychology, 80(3), 381 Hunsinger, M., Livingston, R., & Isbell, L. (2013). The impact of Ortner, C. N. M., Kilner, S. J., & Zelazo, P. D. (2007). Mindfulness loving-kindness meditation on affective learning and cognitive meditation and reduced emotional interference on a cognitive task. control. Mindfulness, 4, 275 Motivation and Emotion, 31(4), 271–283 Hutcherson, C. A., Seppala, E. M., & Gross, J. J. (2008). Loving-kind- Pagnoni, G., & Cekic, M. (2007). Age effects on gray matter volume ness meditation increases social connectedness. Emotion, 8(5), and attentional performance in Zen meditation. Neurobiology of 720–724 Aging, 28, 1623–1627 Jacobs, A. M., Vo, M.L.-H., Briesemeister, B. B., Conrad, M., Hof- Pagnoni, G., Cekic, M., & Guo, Y. (2008). “Thinking about not-think- mann, M. J., Kuchinke, L., Lüdtke, J., & Braun, M. (2015). ing”: Neural correlates of conceptual processing during Zen medi- 10 years of BAWLing into affective and aesthetic processes in tation. PLoS ONE, 3(9), e3083 reading: What are the echoes? Frontiers in Psychology, 6, 714 Peirce, J. W. (2007). PsychoPy—Psychophysics software in Python. Jha, A. P., Krompinger, J., & Baime, M. J. (2007). Mindfulness train- Journal of Neuroscience Methods, 162(1–2), 8–13 ing modifies subsystems of attention. Cognitive, Affective & Rice, V., Enders, L., & Jeter, A. (2020). Visual scanning and reading Behavioral Neuroscience, 7(2), 109–119 speed following 5-days of mindfulness-based stress reduction: A Kabat-Zinn, J. (1990). Full catastrophe living. Delta Publishing. pilot study. In: International Conference on Applied Human Fac- Kissler, J., Herbert, C., Peyk, P., & Junghofer, M. (2007). Buz- tors and Ergonomics (pp. 221–227). Springer, Cham. zwords: early cortical responses to emotional words during Richards, H. J., Benson, V., Donnelly, N., & Hadwin, J. A. (2014). reading. Psychological Science, 18(6), 475–480 Exploring the function of selective attention and hypervigilance Kok, B. E., & Singer, T. (2017). Phenomenological fingerprints for threat in anxiety. Clinical Psychology Review, 34(1), 1–13 of four meditations: differential state changes in affect, mind- Rubenstein, H., Garfield, L., & Millikan, J. A. (1970). Homographic wandering, meta-cognition, and interoception before and after entries in the internal lexicon. Journal of Verbal Learning & Ver- daily practice across 9 months of training. Mindfulness, 8(1), bal Behavior, 9(5), 487–494 218–231 Satow, L. (2011). Big-Five-Persönlichkeitstest (B5T) [Testbe- Kuchinke, L., Võ, M.L.-H., Hofmann, M. L., & Jacobs, A. M. schreibung und Items]. In Zentrum für Psychologische Informa- (2007). Pupillary responses during lexical decisions vary with tion und Dokumentation (ZPID) (Hrsg.), PSYNDEX (Online- word frequency but not emotional valence. International Jour- Datenbanksegment PSYNDEX Tests Dok.-Nr. 9006357). Trier: nal of Psychophysiology, 65(2), 132–140 ZPID. Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. B. (2017). Sears, S., & Kraus, S. (2009). I think therefore I om: cognitive distor- lmerTest package: Tests in linear mixed effects models. Journal tions and coping style as mediators for the effects of mindfulness of Statistical Software, 82, 13 meditation on anxiety, positive and negative affect, and hope. Lehrl, S. (2005). Manual zum MWT-B. [Mehrfachwahl-Wortschatz- Journal of Clinical Psychology, 65(6), 561–573 Intelligenztest]. (5th unchanged). Spitta-Verl. Semple, R. J. (2010). Does mindfulness meditation enhance attention? Lippelt, D. P., Hommel, B., & Colzato, L. S. (2014). Focused atten- A randomized controlled trial. Mindfulness, 1, 121–130 tion, open monitoring and loving kindness meditation: effects Siegle, G., Ingram, R., & Matt, G. (2002). Affective interference: An on attention, conflict monitoring, and creativity–A review. Fron- explanation for negative attention biases in dysphoria? Cognitive tiers in Psychology, 5, 1083 Therapy and Research, 26(1), 73–87 Lusnig, L., Radach, R., Mueller, C. J., & Hofmann, M. J. (2020). Zen Slagter, H. A., Lutz, A., Greischar, L. L., Francis, A. D., Nieuwenhuis, meditation neutralizes emotional evaluation, but not implicit S., Davis, J. M., et al. (2007). Mental training affects distribution affective processing of words. PLoS ONE, 15(2), e0229310 of limited brain resources. PLoS Biology, 5(6), e138 Lutz, A., Dunne, J. D., & Davidson, R. J. (2007). Meditation and the Tang, Y.-Y., Ma, Y., Wang, J., Fan, Y., Feng, S., Lu, Q., et al. (2007). neuroscience of consciousness. In P. Zelazo, M. Moscovitch, & Short-term meditation training improves attention and self-regu- E. Thompson (Eds.), Cambridge handbook of consciousness. lation. Proceedings of the National Academy of Sciences of the (pp. 499–555). Cambridge University Press. United States of America, 104(43), 17152–17156 Lutz, A., Slagter, H. A., Dunne, J. D., & Davidson, R. J. (2008). Tarrasch, R., Berman, Z., & Friedmann, N. (2016). Mindful reading: Attention regulation and monitoring in meditation. Trends in Mindfulness meditation helps keep readers with dyslexia and Cognitive Sciences, 12(4), 163–169 ADHD on the lexical track. Frontiers in Psychology, 7, 578 MacLean, K. A., Ferrer, E., Aichele, S. R., Bridwell, D. A., Zanesco, Vago, D. R., & Silbersweig, D. A. (2012). Self-awareness, self- A. P., Jacobs, T. L., et al. (2010). Intensive meditation training regulation, and self-transcendence (S-ART): a framework for improves perceptual discrimination and sustained attention. understanding the neurobiological mechanisms of mindfulness. Psychological Science, 21(6), 829–839 Frontiers in Human Neuroscience, 6, 296. https://doi. or g/10. 3389/ Meyer, D. E., & Schvaneveldt, R. W. (1971). Facilitation in rec-fnhum. 2012. 00296 ognizing pairs of words: Evidence of a dependence between Valentine, E., & Sweet, P. (1999). Meditation and attention: A com- retrieval operations. Journal of Experimental Psychology, 90(2), parison of the effects of concentrative and mindfulness medita- 227–234 tion on sustained attention. Mental Health, Religion & Culture, Mueller, C. J., & Kuchinke, L. (2016). Individual differences in emo- 2(1), 59–70 tion word processing: A diffusion model analysis. Cognitive, van den Hurk, P. A., Giommi, F., Gielen, S. C., Speckens, A. E., & Affective, & Behavioral Neuroscience, 16(3), 489–501 Barendregt, H. P. (2010). Greater efficiency in attentional pro- cessing related to mindfulness meditation. Quarterly Journal of New, B., Ferrand, L., Pallier, C., & Brysbaert, M. (2006). Reexamining Experimental Psychology, 63(6), 1168–1180 the word length effect in visual word recognition. New evidence 1 3 Psychological Research Vo, M.L.-H., Conrad, M., Kuchinke, L., Urton, K., Hofmann, M. J., Zeng, X., Chiu, C. P., Wang, R., Oei, T. P., & Leung, F. Y. (2015). & Jacobs, A. M. (2009). The Berlin affective word list reloaded The effect of loving-kindness meditation on positive emotions: A (BAWL-R). Behavior Research Methods, 41, 2 meta-analytic review. Frontiers in Psychology, 6, 1693 Westfall, J., Kenny, D. A., & Judd, C. M. (2014). Statistical power and optimal design in experiments in which samples of participants Publisher’s Note Springer Nature remains neutral with regard to respond to samples of stimuli. Journal of Experimental Psychol- jurisdictional claims in published maps and institutional affiliations. ogy General, 143(5), 2020–2045 Willems, C., & Martens, S. (2016). Time to see the bigger picture: Individual differences in the attentional blink. Psychonomic Bul- letin & Review, 23(5), 1289–1299 Xu, R. (2003). Measuring explained variation in linear mixed effects models. Statistics in Medicine, 22(22), 3527–3541 Zanesco, A. P., King, B. G., MacLean, K. A., Jacobs, T. L., Aichele, S. R., Wallace, B. A., & Saron, C. D. (2016). Meditation training influences mind wandering and mindless reading. Psychology of Consciousness: Theory, Research, and Practice, 3(1), 12 1 3

Journal

Psychological ResearchSpringer Journals

Published: May 9, 2021

There are no references for this article.