Disambiguating the relationship between processing modes and mindfulness in Japan

Abstract

Repetitive thoughts are divided into constructive (concrete-experiential thought) and unconstructive (abstract-analytical thought) in processing mode theory. While previous studies have highlighted the similarities between concrete-experiential thought and mindfulness, no large-scale empirical study has been conducted to investigate this relationship. We conducted a cross-sectional questionnaire survey of 1,030 Japanese adults to explore this relationship. We found that abstract-analytic thought had negative correlations with all mindfulness aspects except observation. Conversely, concrete-experiential thought was positively related to the mindfulness aspects of observing, describing, and non-reacting; it was not correlated with non-judging or acting with awareness. Our study suggests that there are similarities between concrete-experiential thoughts and mindfulness: they both focus on the context and specificity of events and experiences, yet harbor some key differences. This implication may contribute to psychological intervention on repetitive thought and mindfulness.

Keywords:

• repetitive thought
• processing modes
• mindfulness
• rumination

1. Introduction

Over the past three decades, the evidence has suggested that rumination is a risk factor for depression (Nolen-Hoeksema, 2000; Nolen-Hoeksema et al., 2008). Rumination refers to repetitive and negative thoughts about one’s self, feelings, and experienced stress events (Nolen-Hoeksema et al., 2008).

While some important theories about rumination have been constructed, little attention has been paid to why rumination leads to unconstructive results, or why other types of repetitive thoughts contribute to constructive results (Watkins, 2008). The H-EX-A-GO-N model is a theoretical attempt to show that the processing mode is a key factor in examining whether repetitive thoughts are related to constructive results (Watkins & Roberts, 2020). This theory divides thoughts into abstract-analytic thought (AAT) and concrete-experiential thought (CET). AAT refers to evaluative thought that is focused on the causes, meanings, implications, and consequences of one’s life experiences (“why”-type thoughts). Conversely, CET refers to low-level, specific, and contextual thoughts (“how”- type thoughts). The theory on mode of thought was developed based on the evidence of mindset in social psychology. (Watkins, 2008). However, some theories imply that mode of thought is related to emotion or information processing. For example, the States of Mind theory indicates that individuals with excessive top-down mode of thought tend not to be sensitive to alterations in the environment that may improve mood; thus, this type of thought exacerbates negative moods (Herz et al., 2020). According to the H-EX-A-GO-N theory, rumination is categorized as excessive AAT and infrequent CET because it is characterized by abstract and evaluative thoughts about negative feelings and events (Nolen-Hoeksema et al., 2008).

Excessive AAT preserves thoughts that are negative and unlikely to promote problem-solving in difficult situations (Watkins, 2011). Previous studies have revealed that individuals with excessive and predominant AAT tend to have exacerbated negative affect (Watkins et al., 2008), obstructed problem-solving (Kingston et al., 2014; Watkins & Moulds, 2005), and decreased behavioral activation (Kambara et al., 2019b). Conversely, individuals with increased CET tend to have modified depression (Watkins et al., 2009, 2012), improved problem-solving (Kingston et al., 2014; Watkins & Moulds, 2005), and increased goal-striving (Kambara et al., 2019b). In sum, the manner in which rumination exacerbates depression is caused by excessive AAT and insufficient CET.

To the best of our knowledge, only the Mini Cambridge-Exeter Repetitive Thought Scale (Mini-CERTS; Barnard et al., 2007; Douilliez et al., 2014) simultaneously assesses AAT and CET. Previous studies that have used the Mini-CERTS have shown the effect of processing modes on mental health and maladaptive behaviors. For instance, regarding clinical symptoms, some studies have shown that AAT has a positive correlation with depression (Barry et al., 2019; Devynck et al., 2017; Douilliez et al., 2014; Kambara et al., 2019a; Di Schiena et al., 2013, 2011, 2012) while CET has a negative correlation with depression (Devynck et al., 2017; Douilliez et al., 2014). While AAT has shown correlation with maladaptive thoughts such as rumination (Devynck et al., 2017; Douilliez et al., 2014; Kambara et al., 2019a; Maurage et al., 2017) CET has shown correlation with adaptive psychological factors such as mindfulness, which has been measured using the Five Facet Mindfulness Questionnaire (FFMQ; Roux et al., 2019). Further, the FFMQ captures the mechanisms through which psychological interventions reduce mental illness symptoms. Individuals have displayed reduced AAT and improved CET scores after mindfulness training aimed at increasing their focus on the here and now (Deplus et al., 2016; Heeren et al., 2015).

CET and mindfulness states have some similar features; for example, both constructs include a contextual and specific mental representation of emotional and personal events, can decrease depressive rumination, and can increase engagement with the environment (Watkins, 2016b, 2016a). Mindfulness is defined as “paying attention in a particular way: on purpose, in the present moment, and nonjudgmentally” (Kabat-Zinn, 1994, p. 4). Mindfulness is a factor that modifies negative moods (Brockman et al., 2017). Questionnaire research has identified five aspects of mindfulness as “observing” (observing internal and external events), “non-judging” (not evaluating events), “non-reacting” (not emotionally and cognitively reacting to negative events), “describing” (describing what is happening now), and “acting with awareness” (noticing what I am doing; Baer et al., 2006). Thus, CET may be related to observing and describing, as it is focused on the here and now (Watkins, 2008). Moreover, CET may also be associated with non-judging because it does not include thoughts that evaluate the reason for an event’s occurrence (Watkins, 2008). Further, previous studies on a mindfulness-based intervention have implied that increasing mindfulness aspects can improve CET (Deplus et al., 2016; Heeren & Philippot, 2011). Although previous evidence has highlighted the relationship between processing modes and mindfulness, few empirical studies have examined this relationship (Watkins, 2016a). Moreover, to the best of our knowledge, while only one study has investigated the association between processing modes and mindfulness, it has not examined the aspects of mindfulness that are correlated with processing modes (Roux et al., 2019). Additionally, as the aforementioned study sample only includes children and adolescents, it cannot be generalized to the adult population. Therefore, research that investigates the relationship between processing modes and mindfulness aspects in the general population is required. Such an examination would deepen our understanding of the effect of a processing mode intervention on mindfulness, and vice versa (Watkins, 2016a).

In Japan, previous studies have developed a Japanese version of the Mini-CERTS (CERTS-J) to investigate the association between processing modes, other maladaptive psychological factors, and depression (Kambara et al., 2019a, 2019b); however, they have identified issues regarding factor validity owing to cultural differences. Moreover, these studies have only examined criterion-related validity in the factor that is predicted to be negatively correlated with CET (Kambara et al., 2019a). Given that previous findings have revealed that a higher concrete processing mode contributes to the activation of problem-solving, goal-striving, and proactive behavior (Dey et al., 2019; Kambara et al., 2019b), the current study hypothesizes that CET will be positively related to the activation of such behavior in line with the processing mode theory if the CERTS-J has an appropriate structure. Therefore, this study attempts to revise the CERTS-J to examine the association between processing modes and adaptive or maladaptive aspects. Accordingly, we posit the following hypotheses:

H1. If our revision of the CERTS-J is adaptive, then the factor structure is consistent with the original version, and AAT has a positive correlation with rumination and depression, behavioral avoidance, and CET with behavioral activation.

H2. AAT has a negative correlation with all mindfulness aspects because, while the abstract processing mode comprises highly evaluative thoughts, it does not focus on the here and now (Watkins, 2008, 2016a).

H3. CET has a positive association with observing, describing, non-reactivity, and non-judgment aspects because the concrete processing mode merely promotes the contextual description of how events occur without addressing the evaluation and implication of events (Watkins, 2008, 2016a).

2. Materials and methods

2.1. Participants

This study’s participants were selected through an online questionnaire survey conducted by a web-based research company (Macromill, Inc.) to investigate the temporal relationship between thought style and mental health in a cohort of individuals aged 25, 35, 45, and 55 years. We opted for quota sampling to reduce costs and set the age range of the screening criteria for the cohort in this manner for two reasons. First, we aimed to target a distribution of adult participants that reflected the demographic make-up of the Japanese population; second, to avoid overrepresentation of younger respondents who are more likely to respond to web research survey panels in Japan (Terasawa, 2021). A total of 1,030 individuals participated in this study. Among them, 56.6% were women, their mean age was 40.86 years, and SD was 10.38 years. Most participants were Japanese (n = 1016; 98.6%). Table 1 presents the participants’ demographic variables. Approximately 60% of the participants were married, 50% had children, 76% were employed, and 24% were unemployed. Most participants’ annual income ranged between JPY 2 million and 6 million. According to the Japan National Tax Agency (https://www.nta.go.jp/, Results of private sector salary survey, 2022), the average annual income is JPY 4.33 million; thus, our sample represents the average distribution of annual income in Japan. The participants resided throughout Japan. We calculated the required sample size using G*power (Faul et al., 2007), based on 95% power to detect a f2 = 0.02 effect size using multiple regression analysis given a 95% confidence interval. The results of our analysis indicated that at least 652 participants were required. Thus, our sample size can be considered sufficient for empirical statistical analyses. To determine the distribution of depressive symptoms in our sample, we screened the depression scale scores using the cutoff scores (over 21 points) from a previous study examining a good trade-off cutoff ratio related to clinical diagnostic interview scores (Henry et al., 2018). A total of 316 participants (30%) seemed to have clinically relevant depressive symptoms as they had higher scores than the cutoff.

Table 1. Participants’ demographic variables

		N	%
Age	Total	1,030	100
	25 years	169	16.40
	35 years	346	33.60
	45 years	254	24.70
	55 years	261	25.30
Sex	Female	583	56.60
	Male	447	43.40
Married	Yes	614	59.6
	No	416	40.40
Child	Yes	496	48.20
	No	534	41.80
Annual incomea	< USD 2,000	65	6.30
	USD 2,000 to USD 6,000	390	37.86
	> USD 6,000	310	30.01
	No answer	265	25.72
Employee	Yes	783	76.02
	No	247	23.98

a Annual income is approximately JPY 100 = USD 1

2.2. Measures

2.2.1. Mini-CERTS Japanese version-revised

We revised this scale from the CERTS-J (Kambara et al., 2019a). The original version was developed by Barnard et al. (2007) and a short version was later created by Douilliez et al. (2014). In general, the scale measures repetitive thoughts. It begins with the statement “When thoughts about myself, feelings, situations, or events come to mind … ” and is followed by a list of responses for each item. The questionnaire assesses two dimensions. The first is AAT (Items 1, 3, 5, 6, 7, 10, 12, 14, and 15). An example is “My thinking tends to get stuck in a rut, involving only a few themes” (Item 1). The second is CET (Items 2, 4, 8, 9, 11, 13, and 16). An example is “I can grasp and respond to changes in the world around me without having to analyze the details” (Item 2). This factor structure has been confirmed by several studies (Douilliez et al., 2014; Kambara et al., 2019a; Kornacka et al., 2016). The participants rated each item on a four-point scale (1 = almost never, 4 = always). The scale scores ranged from 9‒36 for AAT and 7‒28 for CET. High scores on the AAT and CET subscales indicate a high tendency to engage in each type of thought.

2.2.2. Center for epidemiologic studies depression scale (CES-D)

We assessed depressive symptoms using the Japanese version (Radloff, 1977) of the CES-D (Shima et al., 1985). The CES-D comprises 20 items that evaluate depressive symptoms (e.g., “I was bothered by things that usually don’t bother me”). The participants answered each item in terms of the frequency with which they experienced a particular symptom within a seven-day period. The response options ranged from 0 (rarely or none of the time; less than 1 day) to 3 (most or all of the time; 5–7 days). The total scale score ranged from 0 to 60, with a high score indicating more depressive symptoms. Cronbach’s α of the scale was .87 in the present study.

2.2.3. Ruminative response scale (RRS)

We assessed the tendency to engage in ruminative thought using the RRS (Nolen-Hoeksema & Morrow, 1991). We used the Japanese version, which has demonstrated acceptable reliability and validity (Hasegawa, 2013; Hasegawa et al., 2018). This questionnaire consists of 22 items that assess the frequency of ruminative thought related to feelings of depression or sadness. The participants responded to items on a 4-point scale ranging from 1 (almost never) to 4 (almost always). The RRS also includes two subscales named “brooding” (e.g., “Think ‘Why do I always react this way?”) and “reflection” (e.g., “Write down what you are thinking and analyze it”). Brooding relates to the tendency to think analytically; reflection relates to the tendency to engage in problem-solving. The participants rated each subscale using five items (Hasegawa, 2013; Treynor et al., 2003), and the total score of this scale ranged from 5‒20. Cronbach’s α was .96 for the total score, .88 for brooding, and .83 for reflection in the present study.

2.2.4. Avoidance behavior: Behavioral activation for depression scale (BADS)-short form

We assessed the tendency to use avoidance behaviors using the Japanese version of the BADS (Manos et al., 2011; Yamamoto et al., 2015). This scale consists of nine items that evaluate the general activation of and changes in avoidance behavior on a 7-point scale, which ranges from 0 (not at all) to 6 (almost). The BADS was developed to measure behavioral activation by using two subscales that contain three items on avoidance behavior (e.g., “Most of what I did was to escape from or avoid something unpleasant”) and five items on activation (e.g., “I engaged in many different activities”). The avoidance subscale in the original version includes one item on rumination (Manos et al., 2011). However, the Japanese version does not include this item because it decreases the scale’s reliability (Yamamoto et al., 2015). The scale score ranged from 0‒18 in the avoidance subscale and 0‒30 in the activation subscale. High scores indicate high avoidance or activation tendencies among individuals. Cronbach’s α was .81 for activation and .75 for avoidance in the present study.

2.2.5. Five facet mindfulness questionnaire (FFMQ)

We measured participants’ mindfulness states using the Japanese version (Sugiura et al., 2012) of the FFMQ (Baer et al., 2006). The questionnaire includes 39 items that are rated on a 5-point scale ranging from 1 (never or very rarely true) to 5 (very often or always true). The measured constructs of FFMQ include observing (e.g., “When I’m walking, I deliberately notice the sensation of my body moving”), non-reactivity (e.g., “Usually when I have distressing thoughts or images, I am able just to notice them without reacting”), non-judging (e.g., reversed item, “I criticize myself for having irrational or inappropriate emotions”), describing (e.g., “I’m good at finding the words to describe my feelings”), and acting with awareness (e.g., reversed item, “I find it difficult to stay focused on what’s happening in the present”). The total score of this scale ranged from 39‒195. The subscale scores ranged from 8‒40 for observing, 7‒35 for non-reactivity, 8‒40 for non-judging, 8‒40 for describing, and 8‒40 for acting with awareness. High scores indicate high mindfulness states among individuals. Cronbach’s α was .93 for the total score, .80 for observing, .79 for non-reactivity, .86 for non-judging, .66 for describing, and .85 for acting with awareness in the present study.

2.3. Procedure

First, we first developed a revised version of the CERTS-J. Previous studies on the CERTS-J have revealed a factor structure problem, and have excluded some items because they do not show significant factor loading. We discussed reconsidering the Japanese translation of the items to ensure that the original meanings were clearly conveyed. For example, Item 5 (“I judge myself against my own standards and beliefs”) should be allocated to the AAT factor. However, some Japanese individuals may view it as CET, because it is natural in Japanese culture to adhere to strict personal rules and beliefs. Considering this, we translated this item to emphasize the nuances of evaluating and judging. The details of the item revision process are shown in Table S1. The revised version of the CERTS-J was then sent to Philip Barnard (the developer of the original Mini-CERTS) for review (Barnard et al., 2007), who confirmed that these new items conveyed the meanings of the original scale. We subsequently modified the item sentences included in the first revision to obtain the final revised version of CERTS-J.

Second, we conducted an online questionnaire survey through a web-based research company. Participants were enrolled in the survey if they approved the study purpose and provided their informed consent. They completed the questionnaires within (approximately) 30 min. Our study was approved by the Ethics Committee at the Graduate School of Humanities and Social Sciences, Hiroshima University, Japan.

2.4. Analysis

Our study attempts to confirm the reliability and validity of the CERTS-J. Accordingly, we first examined the factor structure using exploratory factor analysis (EFA). Following the recommendations of Hori (2005), we used parallel analysis and minimum average partial correlations. Thereafter, we conducted an EFA using oblique rotation (promax rotation) to check factor loadings and assessed the correlations between factors using the maximum likelihood method. Subsequently, we conducted confirmatory factor analysis (CFA) in line with the original factor structure. Then, to confirm validity, we checked whether the factor structure was consistent with the original version using model fitting scores, such as the goodness-of-fit (GFI), adjusted goodness-of-fit (AGFI), comparative fit index (CFI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR). An acceptable value for each index is above 0.90 for the GFI, AGFI, and CFI and below 0.08 for the RMSEA and SRMR (Hooper et al., 2008). To obtain the indicators for the reliability of the CERTS-J, we calculated Cronbach’s α coefficients. Finally, we cross-validated the CERTS-J by conducting correlation analyses between the CERTS-J scores and other variables (namely rumination, depression, activation, and avoidance) for external validity. We also conducted multiple regression analysis for dependent prediction by the CERTS-J and RRS to examine the effect of CERTS-J on depression while partialling out the effect of RRS. In this analysis, the predictor variables were grand-mean centered.

To examine the magnitude of the relationships between processing modes and mindfulness, we conducted a correlation analysis between AAT/CET and mindfulness with partialled out CET/AAT, and vice-versa. Moreover, we analyzed partial correlations between AAT/CET and mindfulness subscales with other mindfulness subscales partialled out. The data analyses were conducted using R 3.4.3 (psych and lavaan packages).

3. Results

The parallel analysis and the minimum average partial suggested that the scale had a two-factor structure; thus, we conducted an EFA of the two-factor structure. The CERTS-J index is presented in Table S2. The results reveal that all items have factor loadings above .40, except the second item (.388). Moreover, the items loaded with X1 and X2 factors are consistent with the items in the Mini-CERTS. Therefore, we considered this factor structure to be equal with the original version. The X1 factor included loaded Items 1, 3, 5, 6, 7, 10, 12, 14, and 15. These items were the same as the AAT in the original version of the scale. Accordingly, we named it “AAT.” The X2 factor consisted of Items 2, 4, 8, 9, 11, 13, and 16 to correspond to the original version’s CET items. Therefore, we named it “CET.”

The internal consistency of Cronbach’s α is .88 for AAT and .74 for CET in the present CERTS-J. Given that Cronbach’s αs in the previous version are .65 and .69 for AAT and CET, respectively (Kambara et al., 2019a), the CERTS-J demonstrates adaptive reliability values (Cortina, 1993).

To confirm the model fitting of the present factor structure in our data, we conducted CFA. The model was a two-latent-factor model, and the factors had path toward items loaded with AAT or CET, as indicated by the EFA. The results show that the model has adaptive model fit indices (GFI = .922, AGFI = .896, CFI = .876, RMSEA = .074, and SRMR = .086; Hooper et al., 2008). All paths from the latent factors to the items are significant. Therefore, the analysis implied that the factor structure of the CERTS-J is valid and fits the present data.

To identify the criterion-related validity of the CERTS-J, we computed zero-order correlations between the CERTS-J and the other variables (Table 2). As predicted, AAT is positively correlated with avoidance in the BADS (r = .521, p < .01), brooding in the RRS (r = .618, p < .01), reflection in the RRS (r = .464, p < .01), and depression in the CES-D (r = .592, p < .01). AAT is negatively correlated with activation in the BADS (r = −.166, p < .01). Moreover, as predicted, CET is positively correlated with activation in the BADS (r = .464, p < .01). CET is also negatively correlated with depression in the CES-D (r = −.109, p < .01), which is consistent with our hypotheses. Although almost all the correlations found through our analyses are in line with our hypotheses, CET is not associated with avoidance in the BADS and brooding in the RRS.

Table 2. Mean, SD, and correlations

		1		2		3		4		5		6		7		8		9		10		11		12		13		M	SD
1	CER_AAT																											19.48	5.73
2	CER_CET	.094	**																									14.90	3.82
3	BADS_AC	−.166	**	.464	**																							9.98	5.00
4	BADS_AV	.521	**	−.015		−.137	**																					5.92	3.50
5	FFMQ_OB	.280	**	.364	**	.128	**	.188	**																			18.85	5.60
6	FFMQ_NR	−.100	**	.371	**	.237	**	−.062	*	.509	**																	18.20	5.01
7	FFMQ_NJ	−.548	**	−.056	+	.114	**	−.408	**	−.576	**	−.262	**															28.03	6.06
8	FFMQ_DE	−.299	**	.396	**	.397	**	−.241	**	.230	**	.394	**	.117	**													22.81	5.34
9	FFMQ_AW	−.473	**	−.076	*	.046		−.481	**	−.496	**	−.275	**	.618	**	.181	**											28.86	5.88
10	FFMQ_total	−.496	**	.391	**	.373	**	−.438	**	.222	**	.506	**	.432	**	.776	**	.481	**									116.76	13.36
11	RRS_BR	.618	**	−.002		−.180	**	.507	**	.394	**	−.001		−.624	**	−.241	**	−.562	**	−.462	**							9.61	4.01
12	RRS_RF	.464	**	.167	**	.047		.404	**	.449	**	.104	**	−.488	**	.013		−.452	**	−.187	**	.708	**					8.23	3.31
13	RRS_total	.633	**	.054	+	−.127	**	.554	**	.438	**	.023		−.630	**	−.183	**	−.596	**	−.429	**	.933	**	.859	**			39.98	14.82
14	CESD	.592	**	−.109	**	−.273	**	.556	**	.234	**	−.155	**	−.482	**	−.309	**	−.481	**	−.513	**	.693	**	.581	**	.745	**	17.18	11.09
** p < .01, * p < .05, + p < .10

BADS_AC: activation; BADS_AV: avoidance; CER_AAT: abstract-analytic thought; CER_CET: concrete-experiential thought; CESD: center for epidemiologic studies depression scale; FFMQ_AW: acting with awareness; FFMQ_DE: describing; FFMQ_NJ: non-judging; FFMQ_NR: non-reactivity; FFMQ_OB: observing; FFMQ_total: Five Facet Mindfulness Questionnaire total score; RRS_BR: brooding; RRS_RF: reflection; RRS_total: rumination response scale total score.

We conducted multiple regression analyses for depression as predicted by AAT, CET, brooding, and reflection. The results show that AAT, brooding, and reflection positively predict depression (b = 3.11, standard error = 0.29; b = 4.00, standard error = 0.37; b = 2.47, standard error = 0.33), whereas CET negatively predicts depression (b = −1.90, standard error = 0.23).

In our primary analysis, we conducted a correlation analysis between AAT/CET and mindfulness with partial CET/AAT. The results show that AAT has a negative association with the FFMQ’s aspects of non-judging (rp = −.546, p < .01), non-reactivity (rp = −.146, p < .01), describing (rp = −.368, p < .01), and acting with awareness (rp = −.469, p < .01), but not with observing (rp = .265, p < .01). Conversely, CET is positively correlated with observing (rp = .265, p < .01), non-reactivity (rp = .384, p < .01), and describing (rp = .446, p < .01), but shows no correlation with non-judging (rp = −.006, n.s.) and acting with awareness (rp = −.036, n.s.). Additionally, we analyzed partial correlations between AAT/CET and the mindfulness subscales with other mindfulness subscales partialled out. The results showed that AAT was negatively correlated with all the mindfulness subscales, except observing (observing: rp = .091; p < .01; non-reactivity: rp = −.272; p < .01; non-judging: rp = −.357; p < .01; describing: rp = −.140; p < .01; acting with awareness: rp = −.209; p < .01). CET was positively correlated with all the mindfulness subscales, except acting with awareness (observing: rp = .235; p < .01; non-reactivity: rp = .138; p < .01; non-judging: rp = .111; p < .01; describing: rp = .241; p < .01; acting with awareness: rp = −.014; p > .10). These two partial correlations are consistent without CET and the non-judging subscale. Therefore, our results imply that the processing mode is generally related to mindfulness aspects.

4. Discussion

This study attempts to examine the degree of association between processing modes and mindfulness to clarify the similarities and differences between CET and mindfulness. We first revised the previous version of the CERTS-J to find that the CERTS-J had better reliability and validity than the CERTS-J (Kambara et al., 2019a), which is consistent with H1. Further, AAT is negatively related to mindfulness, except the observing subscale, and CET is positively associated with some aspects of mindfulness (e.g., observing, non-reactivity, and describing), which is partially consistent with H2 and H3. This is the first large-scale investigation to examine the association between processing modes and mindfulness in Japan.

We then revised the CERTS-J and confirmed an adaptive model fit index and good factor structure that was consistent with the original scale (Douilliez et al., 2014). Our results indicate that CERTS-J is better in terms of its structure and ability to capture the original constructs than the previous version (Kambara et al., 2019a). Therefore, our scale can effectively capture the processing modes; accordingly, our study extends the research on repetitive thought in Japan.

Subsequently, we examined the relationship between processing modes and mindfulness, and partialled out the effect of AAT or CET. We found that AAT was correlated with almost all mindfulness aspects except observation. As AAT is defined as a thought with evaluative and implicated content (Watkins, 2008), it is viewed as a contrasting mindset to mindfulness (Roux et al., 2019). However, we found a surprising positive correlation between AAT and observing. This tendency of significantly positive or negative correlations did not change when other mindfulness aspects were partialled out. Previous studies that have used the FFMQ have found the observing subscale to be positively correlated with depression and anxiety (Baer et al., 2006; Carpenter et al., 2019; Sugiura et al., 2012). Moreover, previous research has shown that observing has a positive correlation with rumination (Royuela-Colomer & Calvete, 2016). Observing refers to paying attention not only to external phenomena but also to inner experiences (Sugiura et al., 2012). Thus, observing and rumination have a similar focusing pattern, which increases inner attention (Zhou et al., 2020). Further, previous research has indicated that non-mastery mediators cannot observe their inner state without evaluation and judgment (Royuela-Colomer & Calvete, 2016). In sum, individuals with high AAT are likely to show a low-mindfulness mind state.

CET is related to observing, describing, and non-reactivity, which is consistent with our hypotheses. CET helps to describe events and experiences in one’s life both in context and with specificity (Watkins, 2008), which is similar to the definition of observing and describing (Baer et al., 2006). Given the construct of non-reactivity, engaging in CET represents an inner state in which individuals do not get caught up with their thoughts and feelings. Although the correlations between CET and mindfulness are consistent with our hypotheses, our results do not show a relationship between CET and non-judging. Non-judging includes the “refraining from judgment or self-criticism” aspect of mindfulness (Baer et al., 2006). The content of CET does not cover self-criticism because CET is not an evaluative thought. However, the processing mode theory does not imply that individuals with high CET do not possess evaluative and judgmental thoughts, but indicates that using AAT and CET in a flexible manner is important for improving mental health (Watkins, 2008, 2011). Moreover, when the correlation between CET and non-judging was analyzed with other mindfulness subscales partialled out, CET was positively correlated with non-judging. Thus, the initial non-significant correlation between CET and non-judging may have been caused by aspects other mindfulness subscales that are not related to CET. In sum, our results imply that CET may be similar to mindfulness in terms of the contextual and specific focus on events and experiences; however, CET may differ from the non-judging and acting with awareness aspects of mindfulness.

Our findings highlight important clinical and research implications. Our results can inform the development of interventions related to processing modes and mindfulness. For example, interventions that aim to improve CET may improve problem-solving by increasing some aspect of mindfulness (e.g., focusing on the here and now). However, such an intervention may not directly affect evaluative thought; therefore, it may not increase one’s likelihood of acting with awareness. Meanwhile, interventions that aim to improve mindfulness may address processing mode flexibility by increasing CET. However, this may not work in all cases because mindfulness meditation may not be related to increased CET. Accordingly, therapeutic techniques such as concreteness training (Watkins et al., 2009) and mindfulness mediation (Gu et al., 2015) may be useful in both types of intervention. Moreover, our study revised and upgraded the CERTS-J to measure AAT and CET more accurately. Thus, our findings may be helpful for future studies that examine and assess processing modes.

The study has the following limitations. First, its cross-sectional nature means that causal relationships cannot be determined; hence, further experimental or longitudinal studies will be required to examine the causal relationship between processing modes and mindfulness. In addition, experimental or interventional studies may contribute to the identification of potential relationships. Second, although we reveal the features of processing modes and mindfulness in Japan, it is difficult to examine the differences regarding this phenomenon in other countries because we were unable to directly compare our results with other countries.

Despite these limitations, our study develops a scale that can appropriately measure processing modes in Japan. Moreover, we reveal the relationship between processing modes and mindfulness in an adult population as follows: AAT has a negative correlation with mindfulness, whereas CET has a positive correlation with mindfulness. CET is only associated with the observing, describing, and non-reactivity aspects of mindfulness. Our study contributes to the research that examines the mechanisms of repetitive thought in Japan and implies that some aspects of CET differ from mindfulness. Accordingly, research is needed to further examine the mechanisms of processing modes and mindfulness.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are openly available in OSF at https://osf.io/f79pw/?view_only=e4bd3a792ac04db58c9ac1ac8b149621.

Additional information

Funding

This work was supported by JSPS KAKENHI under Grant Number 20K22285.