Participants and procedure
Research design for the present study was generally based on preceding cross-sectional studies involving yoga practitioners  and speed-reading trainees . Thirty-three healthy practitioners of martial arts (11 females and 22 males; age 22–69 years; mean age = 44.9 years, standard deviation (SD) = 12.9) participated as a practitioner group. All these practitioners were members of either a general incorporated foundation or a non-profit organization, both of which were located in Tokyo, Japan, and were founded by the same originator (Hiroyuki Aoki) to promote multiple styles of martial arts practice based on Japanese traditions. According to the self-reports of these practitioners, 22 held the Dan levels and another six held the Kyu levels within the original grading systems of the organizations. Specific content of their daily practice included swordsmanship, karate, bojutsu, etc. and combinations of these. The practitioners reported that they had practiced martial arts for 0.6–35.0 years (mean = 14.9 years; SD = 11.8). The practitioners also self-reported that the mean frequency of the practice of martial arts was 2.0 days per week (SD = 1.7), which corresponded to 188.7 min per week on average (SD = 152.5). Representatives of the two groups and/or the researcher (Hiromitsu Miyata) asked all members of the two groups to participate in the study, unless these members no longer engaged in practice of martial arts at the time of the survey. Distribution and collection of the questionnaires were carried out either by using questionnaires printed on A4 paper or via an e-mail.
In addition, 66 healthy non-practitioners of martial arts (22 females and 44 males; age, 22–67 years; mean age = 44.5 years, SD = 11.6) participated as a control group. Data collection for these participants were conducted after the survey for the practitioners had been completed. On the basis of their self-reports, none of these non-practitioners practiced martial arts nor engaged in other relevant contemplative/mind-body practices including Zen, yoga, etc. According to previous studies [10, 12], data collection for the non-practitioners was performed by using “i Research,” i.e., an online survey system by NEO Marketing Inc., Tokyo, Japan. In this system, there were approximately 6,390,000 monitors in Japan who had agreed to participate in multiple online questionnaire surveys by providing demographic information. Non-practitioners of marital arts were matched with practitioners of martial arts on demographic variables including sex (33% females, 67% males), age range (12% in their twenties, 27% in their thirties, etc.), marital status (44% married, 56% unmarried), annual household income level (18% below 2,000,000 Japanese yen, 36% 2,000,000–4,990,000 Japanese yen, 30% 5,000,000–8,990,000 Japanese yen, 3% above 9,000,000 Japanese yen), and living area (Tokyo and the neighboring prefectures, Gunma, Shizuoka, Ishikawa, or Kyoto prefectures). Regarding household income level, the remaining 12% of the participants for both groups answered that they do not know their household income level. The survey included this option in order not to force all participants to report their income. Question items for the non-practitioners were the same as those for the practitioners. Non-practitioners gave answers to the questions that appeared on an online survey display by checking relevant checkboxes. In the survey for the non-practitioners, answers were required for all the question items, so that there were no items left unanswered.
The survey for both groups included the Japanese versions of the psychological scales listed below. All these psychological scales had been developed and validated in the preceding studies, and had also been used in the preceding studies involving yoga practitioners  and speed-reading trainees . Participants from both groups were notified that the aim of the study involved assessment of the normal psychological status of each participant and had nothing to do with evaluation or superiority/inferiority of any individual. Also, participants from neither group were notified that the study compared practitioners and non-practitioners of martial arts. In addition, the survey instructed participants to always give plain and honest answers. Another psychological scale on state-trait anxiety was included in the survey, the data for which were not analyzed due to technical errors in the instructions/question items.
Five facet mindfulness questionnaire (FFMQ)
The FFMQ  is one of the most commonly used psychological scales to measure dispositional mindfulness. The scale involves five core dimensions of mindfulness (facets/subscales), i.e., observing, describing, acting with awareness, non-judging of inner experience, and non-reactivity to inner experience. Each of the 39 items is rated on a five-point scale from 1 to 5. Sugiura et al.  developed and validated a Japanese version of the FFMQ by involving Japanese students, which was used in the present study. The present study expected that measures pertaining to martial arts should predict higher dispositional mindfulness, because martial arts are supposed to have similar correlates to mindfulness and/or Zen as well as sports.
Subjective well-being scale (SWBS)
The SWBS  is a psychological scale developed and validated in Japanese to measure core components of subjective well-being (see also [28, 29] for former versions of the scale). The SWBS has 15 items, each rated on a four-point scale from 1 to 4. These items are classified into the five subscales each containing three items: general well-being – positive affect, confidence in coping, expectation-achievement congruence, general well-being – negative affect, and transcendence. The present study again expected that practice of martial arts should predict higher subjective well-being, based on the preceding studies suggesting that dispositional mindfulness is positively associated with psychological well-being ([30, 31]; see also [10, 12]),
Beck depression inventory (BDI)
The BDI [32, 33] is one of the common classical measures of depression. The BDI has 21 items, each concerning core symptoms of depression, which are summed to calculate a total score. Each item has four self-evaluative sentences as options, each being scored from 0 to 3. For each item, participants were instructed to select a sentence that best described how they recently felt. The present study used a validated Japanese version of the BDI [34, 35]. Similar to the above, we expected that practice of martial arts should predict lower depression, because increase in dispositional mindfulness has been suggested to improve psychological health [31, 36].
Statistical analysis methods
All statistical analyses were conducted by using SPSS Statistics 25.0 software. Data for all participants from the two groups were included in the analysis. The initial analyses involved a description of outcomes from the psychological scales and comparisons between the groups. After calculation of total scores for the FFMQ, SWBS, and BDI as well as subscale (facet) scores for the FFMQ and SWBS for each group, independent samples t-tests were used to compare the two groups for all these total/subscale scores. As a measure of internal consistency, Cronbach’s alphas (αs) were examined for each group and total/subscale score. Items that were left blank for these psychological scales accounted for 0.16–3.64% for the practitioners and 0.00% for the non-practitioners. Unavailable data for the practitioners were not included in the analysis. Also, to indicate how scores from each scale are associated with each other, Pearson’s correlation coefficients (r values) were calculated between the total scores from the FFMQ, SWBS, and BDI. These correlation analyses were conducted separately for each group.
Next, we examined whether and how practice and/or expertise in martial arts would be correlated with psychological outcomes among the group of practitioners. To indicate practice/expertise in martial arts, the following self-reported measures were considered. (1) The Dan/Kyu rank refers to the rank or grade used to indicate each practitioner’s degree or level of expertise within each practice of martial arts. The organizations to which participants in the present study belonged had multiple original Dan/Kyu grading systems (e.g., 5th to 1st Dans and 1st to 10th Kyus from higher to lower ranks in bojutsu), and multiple practitioners self-reported that they held two or more different Dans and/or Kyus at the time of the survey. Thus, in the present study these practitioners were categorized according to whether s/he held the Dan rank(s) in either of the ranking systems, held only the Kyu rank(s), or held neither of these ranks (2 = Dan holder, 1 = Kyu holder, 0 = non-Dan/Kyu holder). Spearman’s rank correlation coefficients (rs values) were calculated to indicate zero-order correlations between the Dan/Kyu rank and total/subscale scores from the psychological scales. (2) The practice period indicates the length of time (months) elapsed since the practitioner started to practice marital arts. (3) The practice frequency is the number of days in which each participant engaged in the practice of martial arts per week. Finally, (4) the practice time points to the total length of practice of marital arts in minutes per week for each practitioner. Regarding the latter three measures, Pearson’s correlation coefficients were examined among the practitioners to look for correlations between practice in martial arts and psychological outcomes. When reporting practice frequency and practice time, four practitioners noted that meditation was included in the specific content of daily practice (frequency or time for meditation were not reported separately). The remaining content of practice for all the practitioners, i.e., swordsmanship, karate, bojutsu, etc., all involved body movements as in sports. To analyze elements of practice comparable to sports, values for these two measures from these four practitioners were excluded from analysis. Distributions for these three measures were positively skewed (skewness = 0.36 for practice period, 1.48 for practice frequency, and 1.31 for practice time), showing that numbers of relatively extensive practitioners were small. Thus, data for these measures were log transformed (base 10) for this and subsequent statistical analyses to reduce the skewness.
Because multiple measures pertaining to martial arts showed statistically significant correlations with the total/subscale scores from the psychological scales, multiple regression analyses using a stepwise method were further conducted with each total/subscale score as a dependent variable. These analyses again involved the group of practitioners, and were conducted for scales/subscales in which at least one measure pertaining to martial arts showed a statistically significant correlation with the scores. These analyses primarily intended to consider the potential influence of demographic variables on the scores from the psychological scales. In addition to the abovementioned four measures relevant to martial arts, we entered the following demographic variables as independent variables in a stepwise manner: sex (0 = male, 1 = female), age (years), marital status (0 = married, 1 = unmarried), and annual household income level (1 = below 2,000,000 Japanese yen, 2 = 2,000,000–4,990,000 Japanese yen, 3 = 5,000,000–8,990,000 Japanese yen, 4 = above 9,000,000 Japanese yen).