Is There a Dark Side to Mindfulness? Relation of Mindfulness to Criminogenic Cognitions

Abstract

In recent years, mindfulness-based interventions have been modified for use with inmate populations, but how this might relate to specific criminogenic cognitions has not been examined empirically. Theoretically, characteristics of mindfulness should be incompatible with distorted patterns of criminal thinking, but is this in fact the case? Among both 259 male jail inmates and 516 undergraduates, mindfulness was inversely related to the Criminogenic Cognitions Scale (CCS) through a latent variable of emotion regulation. However, in the jail sample, this mediational model also showed a direct, positive path from mindfulness to CCS, with an analogous, but nonsignificant trend in the college sample. Post hoc analyses indicate that the Nonjudgment of Self scale derived from the Mindfulness Inventory: Nine Dimensions (MI:ND) largely accounts for this apparently iatrogenic effect in both samples. Some degree of self-judgment is perhaps necessary and useful, especially among individuals involved in the criminal justice system.

Keywords

mindfulness nonjudgment criminogenic cognitions jail inmate emotion regulation

The United States has the highest rate of incarceration in the world, admitting approximately 12.9 million adults into jails during a 12-month period from 2009 to 2010 (Minton, 2011). Recidivism rates are remarkably high, with 68% of released inmates being rearrested within 3 years (Langan & Levin, 2002). Criminologists and psychologists have begun studying a set of empirically identified “criminogenic needs”—malleable factors that offer promising points of intervention (Andrews, Bonta, & Hoge, 1990). Criminal thinking is a key criminogenic need represented by cognitive distortions that serve to rationalize and perpetuate a lifelong pattern of criminal behavior, and in fact are reliable predictors of reoffense (Tangney, Mashek, & Stuewig, 2007; Walters, 2012).

Mindfulness-based interventions have recently been modified for use with inmate populations, with some promising results (Shonin, Gordon, Slade, & Griffiths, 2013), but the mechanisms by which mindfulness might reduce recidivism remains unclear. In fact, the relationship of mindfulness to criminogenic needs has not been examined empirically. The current article breaks new ground by examining the relationship of mindfulness to a key criminogenic need—criminal thinking—in two independent studies.

Mindfulness

A concept originating from Buddhism, mindfulness has been adopted by psychologists as an approach to increase awareness, learn adaptive responses to emotional distress, and decrease maladaptive behaviors (Bishop et al., 2004). Mindfulness involves deliberately experiencing current thoughts and emotions, even when unpleasant, with the choice to respond adaptively and maintain awareness of potential consequences (Cohen, Wupperman, & Tau, 2013; Hayes & Feldman, 2004). A key component of mindfulness is an attitude of nonjudgment and acceptance toward one’s own experiences (Bishop et al., 2004; Harty et al., 2009).

Mindfulness is examined empirically in this article using the Mindfulness Inventory: Nine Dimensions (MI:ND; Harty et al., 2009) assessing six core components of mindfulness and three closely linked proximal outcomes associated with emotion regulation. The core components of mindfulness are as follows:

Present Awareness—The ability to consciously register internal and external stimuli occurring in the present moment (Brown, Ryan, & Creswell, 2007);

Metacognition—The skill to observe one’s own thoughts and rationally interpret these thoughts (Brown et al., 2007);

Quiet Mind—The capability to clear one’s head of thoughts; literally a quiet/peaceful mind;

Willingness and Acceptance—The ability to clearly recognize all emotions, even painful emotions, and accepting them as part of life;

Nonjudgment of Self—An overall acceptance of oneself; not criticizing one’s own thoughts and actions.

Nonjudgment of Others—An overall acceptance of others; not criticizing or judging other people and their actions.

The three proximal outcomes of mindfulness centered on emotion regulation include the following:

Distress Tolerance—The ability to endure and work through emotional pain when necessary;

(Lack of) Experiential Avoidance—A conscious effort to avoid emotional discomfort;

(Lack of) Emotion-Driven Impulsivity—A maladaptive reaction to strong emotions resulting in impulsive behaviors.

In mindless states, emotions can rule our behavior or even occur outside of our present awareness (Brown & Ryan, 2003). Emotional avoidance and/or overengagement with emotions are believed to be detrimental to the self, physically and mentally. Mindful training promotes emotional balance and is found to reduce experiential avoidance and rumination (Hayes & Feldman, 2004).

Criminogenic Cognitions

Criminogenic cognitions are thought patterns used to reduce dissonance between moral standards and behavior, rationalize deviant behavior, and minimize negative consequences (Sykes & Matza, 1957; Tangney et al., 2012; Walters, 2012). Criminogenic thinking has emerged as a robust predictor of recidivism (Caudy et al., 2015; Walters, 2012). Criminogenic cognitions have also been linked to aggression, history of criminal justice involvement, and misconduct during incarceration (Tangney et al., 2012; Walters & Elliott, 1999). Patterns of criminal thinking have been identified as dynamic risk factors, amenable to change, and appropriate targets for treatment (Gendreau, Little, & Goggin, 1996).

Criminogenic cognitions are studied in this article using the Criminogenic Cognitions Scale (CCS; Tangney, Meyer, Furukawa, & Cosby, 2002) looking at five constructs to measure cognitive distortions specific to criminal offenders:

Notions of Entitlement—The belief of being more deserving and respected than other people;

Failure to Accept Responsibility—A general lack of responsibility for one’s actions, blaming external factors (e.g., family, past events, authority figures) for one’s behavior;

Short-Term Orientation—An overall spontaneous approach to living life with no regard for the future;

Insensitivity to the Impact of Crime—Using rationalizations to minimize the effects of their crimes;

Negative Attitudes Toward Authority—An overall distrust of authority figures and rules/laws.

Mindfulness and Criminogenic Cognitions

We theorize that individuals who are more mindful, either by training or nature, will engage in fewer criminal thinking patterns. In addition, we believe this relationship will be partially mediated via a higher propensity to regulate one’s emotions. Emotion dysregulation has been associated with greater amounts of cognitive discomfort leading to maladaptive behaviors and cognitive distortions (Bishop et al., 2004). In a criminal population, emotional and cognitive discomfort may lead to cognitive distortions, termed criminogenic cognitions, aimed at justifying the criminal behaviors that lead to incarceration and protecting one’s self-image. Beyond the proposed indirect effect through emotion regulation, we hypothesize that individuals who are more mindful will continue to show a direct negative relationship with criminogenic cognitions; those individuals who have the ability to accept and interpret their thoughts and rationally determine how to act on those thoughts will not need to engage in cognitive distortions, including criminogenic thinking.

Study 1

The first study draws on a sample of jail inmates to examine bivariate correlations between the MI:ND and CCS, followed by testing a partial mediation model examining how much emotion regulation mediates the relationship between mindfulness and criminal thinking.

Method

Participants

Participants were 259 adult male jail inmates recruited from a local, suburban detention center as part of a randomized controlled trial (RCT) of a restorative justice intervention (Folk et al., 2016). The sample is 100% male as there were too few females incarcerated at any given time to allow for randomization into the group intervention. The sample’s race was diverse, reflective of the local community (45% Black or African American, 38% Caucasian or White, 3% Asian, and 14% Other), with 10% identifying their ethnicity as Hispanic. Age was also diverse across the sample (M = 32.59, SD = 10.6, range = 18-65). Inmates were considered eligible to participate in the study if they had already been sentenced and the remainder of their time was to be served at the host jail and were not likely to be released early into the community to allow time for treatment. Only inmates housed in the general population areas of the jail were considered eligible, excluding those inmates with severe psychopathology or medical problems. Eligible inmates were informed that participation was voluntary, would have no effect on their sentence, and that data were confidential, protected by a Certificate of Confidentiality from the Department of Health and Human Services (DHHS).

Of the 261 participants who completed baseline measures, one participant’s data were considered invalid and subsequently excluded from any analyses based on scores greater than two standard deviations above the mean on the inconsistency and infrequency scales as outlined by An Interpretive Guide to the Personality Assessment Inventory (Morey, 1991). During analyses, another participant’s data were noted as an extreme outlier in all scales and removed from any analyses, leaving a final sample of 259.

Measures

Participants completed all measures on a touch screen computer with visual and auditory prompts. Participants were able to omit any questions they did not want to answer. Table 1 presents descriptive statistics for variables analyzed here. All measures used for this article were given before participants were randomly assigned to treatment condition; thus, we used the full sample for all analyses.

Table 1.

Univariate Statistics.

	n	M	SD	α	Possible range	Actual range	Skewness	Kurtosis
IVs
Total mindfulness	259	2.80	0.31	.72	1.0-4.0	1.9-3.6	.00	−.19
Total mindfulness	516	2.53	0.29	.73	1.0-4.0	1.6-3.5	−.22	.22
Present awareness	259	3.15	0.47	.56	1.0-4.0	1.8-4.0	.21	−.45
Present awareness	515	2.76	0.49	.58	1.0-4.0	1.4-4.0	.04	−.01
Metacognition	259	2.77	0.43	.05	1.0-4.0	1.6-4.0	.17	.07
Metacognition	516	2.63	0.36	.06	1.0-4.0	1.4-4.0	.29	.22
Quiet mind	259	2.22	0.64	.66	1.0-4.0	1.0-4.0	.06	−.20
Quiet mind	516	2.06	0.60	.70	1.0-4.0	1.0-3.8	.15	−.51
Willingness/acceptance	259	3.18	0.58	.72	1.0-4.0	1.4-4.0	−.47	−.36
Willingness/acceptance	516	2.78	0.58	.75	1.0-4.0	1.0-4.0	−.13	−.04
Nonjudgment of self	259	2.54	0.62	.63	1.0-4.0	1.2-4.0	.04	−.51
Nonjudgment of self	516	2.36	0.59	.67	1.0-4.0	1.0-4.0	.00	−.21
Nonjudgment of others	258	2.91	0.69	.75	1.0-4.0	1.0-4.0	−.52	−.13
Nonjudgment of others	516	2.60	0.64	.75	1.0-4.0	1.0-4.0	−.29	−.25
Mediators
Distress tolerance	259	3.34	0.51	.64	1.0-4.0	1.6-4.0	−.65	−.07
Distress tolerance	516	2.99	0.54	.65	1.0-4.0	1.2-4.0	−.41	.03
Experiential avoidance	259	2.20	0.68	.71	1.0-4.0	1.0-4.0	.22	−.46
Experiential avoidance	516	2.07	0.63	.75	1.0-4.0	1.0-4.0	.35	−.20
Emotion-driven impulsivity	259	2.14	0.80	.84	1.0-4.0	1.0-4.0	.44	−.58
Emotion-driven impulsivity	514	2.08	0.68	.79	1.0-4.0	1.0-4.0	.40	−.21
DVs
CCS total	228	2.17	0.32	.77	1.0-4.0	1.3-3.1	.01	.06
CCS total	514	2.07	0.31	.80	1.0-4.0	1.2-2.9	.02	−.12
Notions of entitlement	228	2.25	0.50	.62	1.0-4.0	1.0-4.0	.24	.13
Notions of entitlement	516	2.27	0.50	.68	1.0-4.0	1.0-4.0	.02	.03
Failure to accept responsibility	228	2.01	0.49	.42	1.0-4.0	1.0-3.6	.24	−.12
Failure to accept responsibility	515	1.74	0.46	.55	1.0-4.0	1.0-3.4	.44	.05
Short-term orientation	228	1.96	0.46	.49	1.0-4.0	1.0-3.2	.18	−.14
Short-term orientation	508	2.25	0.49	.50	1.0-4.0	1.0-3.5	.06	−.01
Insensitivity to the impact of crime	228	2.07	0.51	.57	1.0-4.0	1.0-3.4	.25	−.18
Insensitivity to the impact of crime	173	1.98	0.50	.57	1.0-4.0	1.0-3.2	−.10	−.65
Negative attitudes to authority	228	2.57	0.52	.68	1.0-4.0	1.2-4.0	.19	.51
Negative attitudes to authority	514	2.28	0.45	.72	1.0-4.0	1.0-3.8	.41	.86

Note. Study 1 is recorded in bold; Study 2 is recorded directly below Study 1 and is highlighted. IVs = independent variables; DVs = dependent variables; CCS = Criminogenic Cognitions Scale.

Mindfulness

Mindfulness was assessed using the MI:ND (Harty et al., 2009), a 45-item self-report measure designed to explore the dimensions of mindfulness. The measure assesses six core components of mindfulness : Present Awareness, Metacognition, Quiet Mind, Willingness/Acceptance, Nonjudgment of Self, and Nonjudgment of Others as well as three proximal dimensions of emotion regulation that are theoretically outcomes of mindfulness: Distress Tolerance; (lack of) Experiential Avoidance; and (lack of) Emotion-Driven Impulsivity (Brown et al., 2007; Nightingale, Witt, Malouf, & Tangney, 2012). Items were rated on a 4-point scale with 1 = false, not at all true, 2 = slightly true, 3 = mainly true, and 4 = very true. Items were averaged to create each of the nine scales. A total mindfulness score was composed of the average of the six core components. Alpha estimates of reliability were largely acceptable, except for the poor reliability observed in the Metacognition scale (Table 1).

The total mindfulness score had a significant relationship with the proximal dimensions of mindfulness and emotion regulation. The mindfulness total score was positively correlated with distress tolerance (r = .56, p < .01), whereas it was negatively related with experiential avoidance (r = −.32, p < .01) and emotion-driven impulsivity (r = −.52, p < .01).

Criminogenic cognitions¹

The CCS (Tangney et al., 2002) is a 25-item self-report measure designed to explore five dimensions: (a) Notions of Entitlement, (b) Failure to Accept Responsibility, (c) Short-Term Orientation, (d) Insensitivity to the Impact of Crime, and (e) Negative Attitudes Toward Authority. Items are rated on a 4-point scale with 1 = strongly disagree, 2 = disagree, 3 = agree, and 4 = strongly agree. Items were averaged to create each of the five dimensions as well as a total criminogenic cognitions score. Alpha estimates of reliability for the Total CCS scale were acceptable (α = .77). Estimates of reliability for the five-item domains were lower (see Table 1) ranging from .42 to .68. Longitudinal stabilities from initial assessments (conducted shortly upon incarceration) to prerelease (an average of about 7 months) ranged from .52 to .66, indicating higher reliability than the internal consistency estimates suggest, given that stabilities are attenuated substantially over such lengthy periods of time.

Results

Mindfulness and criminogenic cognitions

Inmates’ mindfulness total scores were significantly, negatively correlated with CCS total scores (r = −.20, p < .01). Individuals higher in core components of mindfulness tended to be less likely to think in ways that rationalize, excuse, and justify criminal behavior. When examining the five dimensions used to create the CCS total score (see Table 2), the mindfulness total score was significantly associated with Notions of Entitlement (r = −.20, p < .01) and Failure to Accept Responsibility (r = −.30, p < .01). Although negative, there was no significant relationship between the mindfulness total score and Short-Term Orientation, Insensitivity to the Impact of Crime, or Negative Attitudes Toward Authority.

Table 2.

Correlations Between the MI:ND and CCS.

	Total mindfulness	Present awareness	Metacognition	Quiet mind	Willingness/acceptance	Nonjudgment of self	Nonjudgment of others	Distress tolerance	Experiential avoidance	Emotion-driven impulsivity
CCS total	−.20** (n = 228)	−.07 (n = 228)	−.14* (n = 228)	−.16* (n = 228)	−.07 (n = 228)	−.06 (n = 228)	−.14* (n = 227)	−.27** (n = 228)	.24** (n = 228)	.34** (n = 228)
CCS total	−.22** (n = 514)	−.17** (n = 513)	−.14** (n = 514)	.03 (n = 514)	−.16** (n = 514)	−.02 (n = 514)	−.25** (n = 514)	−.32** (n = 514)	.26** (n = 514)	.34** (n = 512)
Notions of entitlement	−.20** (n = 228)	.11 ^† (n = 228)	−.08 (n = 228)	−.07 (n = 228)	.04 (n = 228)	−.24** (n = 228)	−.33** (n = 227)	−.21** (n = 228)	.20** (n = 228)	.22** (n = 228)
Notions of entitlement	−.19** (n = 516)	.01 (n = 515)	−.07 (n = 516)	.00 (n = 516)	−.11* (n = 516)	−.03 (n = 516)	−.36** (n = 516)	−.19** (n = 516)	.19** (n = 516)	.20** (n = 514)
Failure to accept responsibility	−.30** (n = 228)	−.15* (n = 228)	−.20** (n = 228)	−.19** (n = 228)	−.15* (n = 228)	−.14* (n = 228)	−.18** (n = 227)	−.33** (n = 228)	.33** (n = 228)	.39** (n = 228)
Failure to accept responsibility	−.32** (n = 515)	−.25** (n = 514)	−.15** (n = 515)	.07^† (n = 515)	−.24** (n = 515)	−.14** (n = 515)	−.19** (n = 515)	−.37** (n = 515)	.25** (n = 515)	.36** (n = 513)
Short-term orientation	−.12 ^† (n = 228)	−.13 ^† (n = 228)	−.07 (n = 228)	−.13 ^† (n = 228)	−.12 ^† (n = 228)	.01 (n = 228)	.01 (n = 227)	−.27** (n = 228)	.08 (n = 228)	.27** (n = 228)
Short-term orientation	.03 (n = 508)	−.00 (n = 507)	−.03 (n = 508)	.06 (n = 508)	.02 (n = 508)	.07 (n = 508)	−.05 (n = 508)	−.11** (n = 508)	.13** (n = 508)	.15** (n = 506)
Insensitivity to impact of crime	−.01 (n = 228)	−.02 (n = 228)	−.01 (n = 228)	−.06 (n = 228)	.01 (n = 228)	.04 (n = 228)	.05 (n = 227)	−.10 (n = 228)	.09 (n = 228)	.12 ^† (n = 228)
Insensitivity to impact of crime	−.09 (n = 173)	−.15^† (n = 173)	.01 (n = 173)	.03 (n = 173)	−.12 (n = 173)	.08 (n = 173)	−.12 (n = 173)	−.11 (n = 173)	.11 (n = 173)	.18* (n = 172)
Negative attitudes to authority	−.02 (n = 228)	−.04 (n = 228)	−.08 (n = 228)	−.08 (n = 228)	.01 (n = 228)	.11 ^† (n = 228)	.01 (n = 227)	.04 (n = 228)	.06 (n = 228)	.10 (n = 228)
Negative attitudes to authority	−.14** (n = 514)	−.15** (n = 513)	−.13** (n = 514)	.04 (n = 514)	−.08^† (n = 514)	−.05 (n = 514)	−.11* (n = 514)	−.24** (n = 514)	.17** (n = 514)	.23** (n = 512)

Note. Study 1 is recorded in bold, Study 2 is recorded directly below Study 1 and is highlighted. The sample size for each individual correlation appears in parentheses underneath the correlation coefficient. MI:ND = Mindfulness Inventory: Nine Dimensions; CCS = Criminogenic Cognitions Scale.

†

p < .10. *p < .05. **p < .01.

Emotion regulation and criminogenic cognitions

The emotion regulation scales (Distress Tolerance, Experiential Avoidance, and Emotion-Driven Impulsivity) hypothesized as the proximal outcomes of mindfulness all showed significant correlations with the CCS total score (Table 2). Distress Tolerance was negatively correlated with the CCS total score (r = −.27, p < .01), whereas Experiential Avoidance and Emotion-Driven Impulsivity were positively correlated with the CCS total score (r = .24, r = .34, ps < .01). These correlations were largely due to the CCS dimensions of Notions of Entitlement, Failure to Accept Responsibility, and Short-Term Orientation (Table 2).

Mediation model

We next tested whether inmates’ emotion regulation mediates the relationship between mindfulness and criminogenic cognitions. Theoretically, individuals who are more mindful are apt to have better emotion regulation; in turn, individuals high in emotion regulation are hypothesized to show low levels of criminogenic thinking patterns because they are better able to cope with distressing emotions and do not need to turn to cognitive distortions aimed at protecting one’s self-image. In addition, a direct (negative) effect from mindfulness to criminogenic thinking was anticipated because mindful individuals presumably have a greater ability to accept responsibility without the need to engage in defensive cognitive distortions.

Mplus statistical software (Muthén & Muthén, 2010) was used to test this partial mediational hypothesis. A latent variable representing emotion regulation was defined by the three scales of emotion regulation: Distress Tolerance, Experiential Avoidance, and Emotion-Driven Impulsivity. We conceptualized both mindfulness and criminogenic cognitions as “formative” (vs. “reflective”) constructs (Edwards & Bagozzi, 2000). That is, mindfulness represents a state characterized by the intersection of the six core components described above (e.g., present awareness, metacognition, quiet mind, etc.), yet these characteristics are not necessarily tightly correlated. For example, some individuals may engage in little self-judgment, but not experience a “quiet mind.” As such, the total score of each was used in analyses.

The model fit the data reasonably well, χ²(4) = 10.60, p = .03, root mean square error of approximation (RMSEA) = .08, comparative fit index (CFI) = .98, and standardized root mean square residual (SRMR) = .03. As hypothesized, mindfulness indirectly and negatively predicted criminal thinking via emotion regulation (see Figure 1; indirect effect, β = −.48, p < .01; 95% confidence interval (CI) = [−.68, −.25]), but contrary to our hypothesis, there was now a positive, direct link between mindfulness and criminogenic cognitions (β = .27, p < .05; 95% CI = [.03, .51]). In contrast to the negative bivariate relationship between mindfulness and criminogenic cognitions, once emotion regulation and its negative pathway is included, those higher in mindfulness now also show more criminogenic thinking. This pattern of results is an example of what MacKinnon and his colleagues (MacKinnon, Fairchild, & Fritz, 2007; MacKinnon, Krull, & Lockwood, 2000) termed “inconsistent mediation”—that is, a special case of partial mediation in which the direct effect and the indirect effect via a mediator are opposite in sign.

Figure 1.

Study 1 partial mediation model.

To further investigate this inconsistent mediation, each individual scale of the MI:ND was substituted into the model as the independent variable instead of the mindfulness total score (see Table 3). Column 2 shows the standardized beta coefficients for path “a” (from mindfulness to emotion regulation), column 3 shows the results for path “b” (from emotion regulation to criminogenic cognitions), column 4 shows the direct path “c’” (from mindfulness to criminogenic cognitions, taking into account the indirect effect via emotion regulation), and column 5 shows “c,” the simple bivariate correlation between mindfulness to criminogenic cognitions. Column 6 shows the variance explained in emotion regulation by mindfulness. Column 7 shows the total variance explained in criminogenic cognitions by the full model. The remaining columns present fit indices for each model. Column 4 shows that it was the Nonjudgment of Self scale that was largely responsible for the inconsistent mediation in the full model. Figure 2 presents the model with Nonjudgment of Self as the independent variable. The indirect effect was negative and significant, whereas the direct effect of nonjudgment of self on criminogenic cognitions was positive and significant. Those who were high on the mindfulness characteristic of non-self-judgment tended to have higher criminogenic cognitions once the pathway through emotion regulation was taken into account.

Table 3.

Partial Mediation Results With the Total Mindfulness Score, and Its Subsequent Subscales, as the IV.

Scale	a (MI:ND → EmReg)	b (EmReg → CCS)	c’ (MI:ND → CCS)	c	EmReg R²	CCS R²	χ²	p	RMSEA	CFI
Mindfulness total	.74**	−.63**	.27*	−.20**	.55	.22	10.60	.03	.08	.98
Mindfulness total	.60**	−.48**	.07	−.22**	.36	.19	14.40	.01	.07	.98
Present awareness	.36**	−.41**	.08	−.07	.13	.15	49.33	.00	.21	.79
Present awareness	.25**	−.41**	−.07	−.17**	.06	.19	36.18	.00	.13	.93
Metacognition	.21**	−.42**	−.05	.14*	.04	.19	9.29	.05	.07	.97
Metacognition	.31**	−.43**	.00	−.14**	.10	.19	14.20	.01	.07	.98
Quiet mind	.45**	−.45**	.05	−.16*	.20	.20	5.57	.24	.04	.99
Quiet mind	.18**	−.46**	.11**	.03	.03	.21	13.79	.01	.07	.98
Willingness/acceptance	.42**	−.43**	.11	−.07	.18	.16	36.71	.00	.18	.85
Willingness/acceptance	.48**	−.43**	.05	.16**	.23	.17	27.44	.00	.11	.95
Nonjudgment of self	.56**	−.59**	.28**	−.06	.31	.24	9.61	.05	.07	.98
Nonjudgment of self	.39**	−.52**	.19**	−.02	.15	.23	13.05	.01	.07	.98
Nonjudgment of others	.50**	−.49**	.11	−.14*	.25	.20	5.29	.26	.04	.99
Nonjudgment of others	.35**	−.41**	−.10*	−.25**	.12	.21	19.14	.00	.09	.97

Note. Study 1, n = 258 to 259; Study 2, n = 515 to 516. Study 1 is recorded in bold, Study 2 is recorded directly below Study 1 and is highlighted. The degrees of freedom are 4 for all of the models presented. The scale presented in the first column is signified by “MI:ND” in the column headings. IV = independent variable; MI:ND = Mindfulness Inventory: Nine Dimensions; EmReg = emotion regulation; CCS = CCS total score. CCS = Criminogenic Cognitions Scale; RMSEA = root mean square error of approximation; CFI = comparative fit index.

p < .05. **p < .01.

Figure 2.

Study 1 partial mediation model with nonjudgment of self as the independent variable.

Discussion

Mindfulness-based interventions have attracted considerable attention as the criminal justice system continues to struggle with overcrowding and an extraordinarily high recidivism rate. The current findings underscore that mindfulness is a complex construct that can have positive as well as negative implications for offender rehabilitation.

On one hand, mindfulness negatively predicted criminogenic cognitions via its impact on emotion regulation. This indirect effect, however, was offset by a positive direct link between mindfulness and criminogenic cognitions. This iatrogenic effect of mindfulness on criminal thinking was primarily driven by nonjudgment of self. It appears that some evaluative scrutiny of one’s thoughts, feelings, and behaviors in the context of criminal thinking may be adaptive. This finding raises questions about the utility of various components of mindfulness when considering externalizing disorders or behavior. In noncorrectional settings, mindfulness interventions are most often implemented with clients suffering from internalizing disorders (e.g., anxiety, depression)—disorders that are typically characterized by excessive negative evaluations of thoughts, emotions, and behavior. Individuals suffering from internalizing disorders may benefit from the nonjudgmental focus of mindfulness. Individuals struggling with externalizing disorders or behaviors, however, are less apt to engage in serious scrutiny of self-experience and behavior. The emphasis on nonjudgment in such populations and settings may actually be counterproductive. Such components of mindfulness interventions may exacerbate patterns of criminal thinking, such as blaming others for their behavior by allowing people to not address their shortcomings and to rationalize their behavior. In short, some self-judgment may actually be beneficial, especially when engaging in treatment with incarcerated populations.

A key question is whether this pattern of inconsistent mediation (MacKinnon et al., 2007; MacKinnon et al., 2000) showing both positive and negative effects of nonjudgment of self is unique to jail inmates, or whether this cautionary message about the nonjudgmental component of mindfulness pertains to nonoffenders—people in the community.

Study 2

Study 2 is an attempt to replicate the results from Study 1 with a diverse sample of college students. Psychologists have paid little attention to cognitions that maintain criminal and antisocial behavior in nonoffender populations (Boren, Moore, Daylor, Tangney, & Stuewig, under review). It is likely that criminogenic cognitions do exist in community populations with those individuals more likely to engage in immoral, deviant, or maladaptive behaviors that place them at higher risk of entering the criminal justice system (McCoy et al., 2006; Schwartz, Fremouw, Schenk, & Ragatz, 2012). Criminogenic cognitions in community samples may be a relevant point of intervention to prevent future criminal justice system involvement in nonoffender populations (Boren et al., under review). Understanding and replicating the findings from Study 1, especially the inconsistent mediation, in a separate sample can help us understand the generalizability and robustness of the results. Bivariate correlations between the MI:ND and the CCS were examined and the mediational model from Study 1 was evaluated. Specifically, we examined the degree in which mindfulness negatively predicts criminal thinking via emotion regulation, and the degree to which mindfulness (especially nonjudgment of self) exerts a countervailing direct effect on criminal thinking.

Method

Participants

Participants were 516 undergraduate students (73% female) from a large, east coast university. The sample’s race was diverse (59% Caucasian or White, 20% Asian, 10% Black or African American, and 11% Other) with 11% identifying their ethnicity as Hispanic. Age was diverse for a university sample (M = 20.4, SD = 3.9, range = 18-58). Students were eligible to participate if they had access to the university’s research participation website to fulfill course requirements and if they were at least 18 years of age.

Of the 595 individuals who consented, 5% (n = 28) were dropped due to skipping or answering a 5 or lower on the following validity question: “On a scale of 1-10, how well did you understand the questions in this survey with 1 being not at all to 10 being completely understood.” Seven percent (n = 44) of participants were dropped due to invalid scores on the negative and positive impression scales as outlined by An Interpretive Guide to the Personality Assessment Inventory (Morey, 1991). Seven individuals were dropped because they finished the survey more than a week after starting it online leaving a final sample of 516.

Measures

Participants completed all questionnaires online, accessed through participation in the psychology department subject pool. They self-reported on gender, age, race, and ethnicity. As in Study 1, participants completed the 45-item MI:ND (Harty et al., 2009) and the 25-item CCS (Tangney et al., 2002). Descriptive statistics are shown in Table 1. Alpha estimates of reliability for the MI:ND were largely acceptable with the exception of the Metacognition scale. Similar to Study 1, the total mindfulness score had a significant relationship with the proximal dimensions of mindfulness and emotion regulation. The mindfulness total score was positively correlated with distress tolerance (r = .49, p < .01) and negatively related to experiential avoidance (r = −.27, p < .01) and emotion-driven impulsivity (r = −.42, p < .01).

Alpha estimates of reliability for the Total CCS scale were good (α = .80). Estimates of reliability for the five-item domains were lower (see Table 1) ranging from .50 to .72. As indicated above, longitudinal stabilities in the jail sample over an average of about 7 months indicated higher reliability, given that stabilities are attenuated substantially over such lengthy periods of time.

Results

Mindfulness and criminogenic cognitions

As in the Study 1 jail sample, Study 2 college students’ mindfulness total scores were significantly, negatively correlated with CCS total scores (r = −.22, p < .01²). Similar to Study 1 (Table 2), when examining the five dimensions used to create the CCS total score, the mindfulness total score was significantly associated with Notions of Entitlement (r = −.19, p < .01) and Failure to Accept Responsibility (r = −.32, p < .01). In addition, there was a significant relationship between college students’ mindfulness total score and the Negative Attitudes Toward Authority scale (r = −.14, p < .01), an effect not observed among Study 1 jail inmates. As observed among jail inmates, there was no significant relationship between the mindfulness total score and the Short-Term Orientation scale or the Insensitivity to the Impact of Crime scale.

Emotion regulation and criminogenic cognitions

The emotion regulation scales (Distress Tolerance, Experiential Avoidance, and Emotion-Driven Impulsivity) all had strong correlations with the CCS total score similar to Study 1 (Table 2). Distress tolerance was negatively correlated with the CCS total score (r = −.32, p < .01), whereas Experiential Avoidance and Emotion-Driven Impulsivity were positively correlated with the CCS total score (r = .26, r = .34, ps < .01). This significant relationship was consistent across most of the subscales of the CCS with the exception of the Insensitivity to the Impact of Crime scale. The Negative Attitudes Toward Authority scale was significantly related to the emotion regulation scales in Study 2, an effect that was not present in Study 1.

In general, the correlations between dimensions of mindfulness, emotion regulation, and criminogenic cognitions were remarkably similar across these two different samples. A series of z tests of independent correlations showed that, of the 60 pairs of correlations in Table 2, six were significantly different from one another: the correlations between Quiet Mind (MI:ND) and the CCS total score (p < .05), Quiet Mind and Short-Term Orientation (CCS; p < .05), Nonjudgment of Self (MI:ND) and Notions of Entitlement (CCS; p < .05), Nonjudgment of Self and Negative Attitudes toward Authority (CCS; p < .05), Distress Tolerance (MI:ND) and Short-Term Orientation (p < .05), and finally Distress Tolerance and Negative Attitudes Toward Authority (p < .01). When applying a Benjamini–Hochberg correction (Benjamini & Hochberg, 1995), only one correlation remained significantly different, the relationship between Distress Tolerance and Negative Attitudes toward Authority. Among jail inmates in Study 1, there was no significant relationship between Distress Tolerance and Negative Attitudes Toward Authority (r = .04, p = .59), but a significant, negative relationship was observed in Study 2 (r = −.24, p < .01). It may be that college students higher in distress tolerance have less negative attitudes toward authority because they have less exposure to the criminal justice system and are not currently interacting with those types of authority.

Mediation model

Parallel to Study 1, a partial mediation model was tested. The model fit the data reasonably well,³ χ²(4) = 14.40, p = .01, RMSEA = .07, CFI = .98, and SRMR = .02. As hypothesized and consistent with Study 1, the indirect effect via emotion regulation was significant and negative (β = −.29, p < .01; 95% CI = [−.37, −.20]). Those higher in mindfulness had higher levels of emotion regulation, which in turn was related to lower levels of criminogenic thinking. The direct effect between the mindfulness total score and the CCS total score was positive, but not significant (Figure 3). Each scale of the MI:ND measure was again analyzed as the independent variable (Table 3). Consistent with the findings in Study 1, the Nonjudgment of Self scale again exhibited inconsistent mediation with a significant, negative indirect effect from the Nonjudgment of Self scale to the CCS total score via emotion regulation and a positive, direct effect (Figure 4).⁴

Figure 3.

Study 2 partial mediation model.

Figure 4.

Study 2 partial mediation model with nonjudgment of self as the independent variable.

Discussion

As in Study 1 of jail inmates, Study 2 of college students showed a robust indirect effect of mindfulness negatively predicting criminogenic thinking via emotion regulation. However, when considering the total Mindfulness scale, the model for the college sample indicated a nonsignificant direct path from mindfulness to criminogenic thinking. Notably, the model considering nonjudgment of self replicated the intriguing pattern of inconsistent mediation observed among jail inmates. Specifically, the Nonjudgment of Self scale maintained an inhibitory indirect effect on criminogenic cognitions via emotion regulation and an iatrogenic direct effect, even among college students. These results suggest that maintaining some level of self-scrutiny and self-judgment may in fact be useful across populations, not just incarcerated individuals.

General Discussion

In two separate studies, one with jail inmates and another with college students, the relation of mindfulness and criminal thinking was complex, suggesting competing paths of influence. On one hand, mindfulness appeared to inhibit criminogenic cognitions via emotion regulation. On the other hand, mindfulness—especially the nonjudgment of self subscale—showed a direct positive relation to criminal thinking, suggesting a competing iatrogenic effect.

In mindfulness practices, acceptance of oneself and a noncritical reaction toward one’s thoughts, feelings, and behaviors are encouraged. The notion is that people suffer from unnecessarily evaluating and negatively judging experiences of the self. This may be true for people suffering from internalizing disorders, such as anxiety and depression—the kinds of disorders typically addressed by mindfulness interventions. This level of nonjudgment and acceptance may not be as beneficial for people suffering from externalizing disorders marked by higher levels of impulsivity. Externalizing disorders are common among an incarcerated population who may benefit from some self-scrutiny in reducing patterns of criminal thinking and behavior (Malouf, Youman, Stuewig, Witt, & Tangney, 2017). The attenuated effect found in the college sample could be due to a higher prevalence of internalizing disorders and a lower prevalence of externalizing disorders in the college sample.

Limitations

Data in both studies were cross-sectional, which limits the ability to determine the directionality of the model. Thus, future research should utilize a longitudinal design to more effectively test out evidence for causality in either or both directions. In addition, participants no doubt varied in their familiarity with mindfulness concepts. People’s interpretations, and thus responses, to items assessing mindfulness may differ depending on their experience with mindfulness-based practices. For example, the construct of “judgment” may have a different connotation for someone with little exposure to the concept and practice of mindfulness compared with someone who practices mindfulness regularly.

Multiple limitations stem from the challenges inherent in measuring mindfulness (see Baer, 2011, for a review). For example, whereas the actual practice of mindfulness consists of moment-to-moment awareness and nonjudgment of internal experiences, retrospective items of typical attitudes and experiences may result in trait measures capturing something somewhat distinct from state measures of mindfulness.

Most relevant to the current study concerns potential ambiguities in the measurement of the nonjudgment of self using the MI:ND. The MI:ND subscale assessing nonjudgment of the self includes items regarding nonjudgment of thoughts and emotions (e.g., “I make judgments about whether my thoughts are good or bad”—reversed) as well as items regarding nonjudgment of the self as a whole (e.g., “I am judgmental of myself”—reversed). Many forms of mindfulness practice emphasize the former much more than the latter.

Yet, it is notable that these items taken together formed a coherent subscale; internal consistencies were substantial in both samples given its brevity (five items). A question for theoreticians, practitioners, and future researchers is the degree to which nonjudgment of internal experiences (e.g., thoughts, feelings) can (and should) be divorced from nonjudgment of behavior and self. Continued improvements in the measurement of trait mindfulness are needed, particularly to maintain fidelity to the complex concepts that underlie the construct of mindfulness. As the MI:ND measure was recently developed, further psychometric analyses and comparisons with other measures of mindfulness and emotion regulation are needed to better determine the reliability and validity of the measure. Future research should attempt to replicate the current study’s findings with other measures of mindfulness and other measures of criminal thinking.

Applied Implications

The findings reported here suggest that caution should be utilized in discouraging self-judgment, especially with jail inmates. Recent results from a pilot RCT of a values-based mindfulness intervention for jail inmates (Malouf et al., 2017) further underscore the complexity of assessing and interpreting non-self-judgment. Relative to a control group, Malouf et al.’s mindfulness treatment group showed robust increases in Willingness/Acceptance on the MI:ND. But the treatment group also showed significant increases in self-judgment (i.e., decreases in nonjudgment of self) and marginally significant increases in shame. These findings may appear counterintuitive as mindfulness emphasizes nonjudgment and self-compassion. But this shift may be the natural ramification of increased willingness/acceptance in a population facing difficult, typically self-imposed, circumstances. All study participants were incarcerated for serious (i.e., felony) charges, a situation apt to evoke self-judgment and shame. Absent intervention, many inmates may be unwilling to experience the discomfort associated with self-judgment, prompting experiential avoidance and outright denial. Researchers have observed high rates of denial and externalization of blame among criminal offenders (e.g., Maruna, 2001). The observed enhanced willingness/acceptance in the treatment group may have allowed participants to overcome this defensiveness to openly reflect on past behavior and take a good hard (somewhat negative) look at the self. Ideally, such negative self-judgment would be steered toward feelings of guilt about specific behaviors, rather than feelings of shame about the self (Tangney, Stuewig, & Mashek, 2007), at the same time as nonjudgment of feelings themselves is encouraged.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) declared receipt of the following financial support for the research, authorship, and/or publication of this article: Portions of this research were supported by Grant #R01 DA14694 from the National Institute on Drug Abuse to June P. Tangney and Jeffrey Stuewig.

Supplemental Material

Supplementary material is available online with this article.

Notes

References

Andrews

D. A.

Bonta

Hoge

R. D.

(1990). Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior, 17, 19-52.

Baer

R. A.

(2011). Measuring mindfulness. Contemporary Buddhism, 12, 241-261.

Benjamini

Hochberg

(1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society B, 57, 289-300.

Bishop

S. R.

Lau

Shapiro

Carlson

Anderson

N. D.

Carmody

. . . Devins

(2004). Mindfulness: A proposed operational definition. Clinical Psychology: Science and Practice, 11, 230-241.

Boren

E. A.

Daylor

Moore

Stuewig

Tangney

J. P

. (under review). We are all a little criminogenic: There is meaningful variance in criminogenic cognitions in a college sample.

Brown

K. W.

Ryan

R. M.

(2003). The benefits of being present: Mindfulness and its role in psychological well-being. Journal of Personality and Social Psychology, 84, 822-848.

Brown

K. W.

Ryan

R. M.

Creswell

J. D.

(2007). Mindfulness: Theoretical foundations and evidence for its salutary effects. Psychological Inquiry, 18, 211-237.

Caudy

M. S.

Folk

J. B.

Stuewig

J. B.

Wooditch

Martinez

Maass

. . . Taxman

F. S.

(2015). Does substance misuse moderate the relationship between criminal thinking and recidivism? Journal of Criminal Justice, 43, 12-19.

Cohen

M. G.

Wupperman

Tau

(2013). Mindfulness in the treatment of adolescents with problem substance use. Adolescent Psychiatry, 3, 172-183.

10.

Edwards

J. R.

Bagozzi

R. P.

(2000). On the nature and direction of relationships between constructs and measures. Psychological Methods, 5, 155-174.

11.

Folk

J. B.

Blasko

B. L.

Warden

Schaefer

Ferssizidis

Stuewig

Tangney

J. P.

(2016). Feasibility and acceptability of an impact of crime group intervention with jail inmates. Victims & Offenders, 11, 436-454. doi:10.1080/15564886.2014.982777

12.

Folk

J. B.

Stuewig

J. B.

Blasko

B. L.

Caudy

Martinez

A. G.

Maass

. . . Tangney

J. P.

(2017). Do demographic factors moderate how well criminal thinking predicts recidivism? International Journal of Offender Therapy and Comparative Criminology. Advance online publication. doi:10.1177/0306624x17694405

13.

Gendreau

Little

Goggin

(1996). A meta-analysis of the predictors of adult offender recidivism: What works! Criminology, 34, 575-608.

14.

Harty

Youman

Malouf

Appel

Hall

Pilafova

Tangney

J. P.

(2009). Mindfulness Inventory: Nine Dimensions (MIND). Fairfax, VA: George Mason University.

15.

Hayes

A. E.

Feldman

(2004). Clarifying the construct of mindfulness in the context of emotion regulation and the process of change in therapy. Clinical Psychology: Science and Practice, 11, 255-262.

16.

Langan

P. A.

Levin

D. J.

(2002). Recidivism of prisoners released in 1994. Federal Sentencing Reporter, 15, 58-65.

17.

MacKinnon

D. P.

Fairchild

A. J.

Fritz

M. S.

(2007). Mediation analysis. Annual Review of Psychology, 58, 593-614.

18.

MacKinnon

D. P.

Krull

J. L.

Lockwood

C. M.

(2000). Equivalence of the mediation, confounding, and suppression effect. Prevention Science, 1, 173-181.

19.

Malouf

E. T.

Youman

Stuewig

Witt

E. A.

Tangney

J. P.

(2017). A pilot RCT of a values-based mindfulness group intervention with jail inmates: Evidence for reduction in post-release risk behavior. Mindfulness, 8, 603-614.

20.

Maruna

(2001). Making good: How ex-convicts reform and rebuild their lives. Washington, DC: American Psychological Association.

21.

McCoy

Fremouw

Tyner

Clegg

Johansson-Love

Strunk

(2006). Criminal thinking styles and illegal behavior among college students: Validation of the PICTS. Journal of Forensic Sciences, 51, 1174-1177.

22.

Minton

T. D.

(2011). Jail inmates at midyear 2010: Statistical tables. Washington, DC: Bureau of Justice Statistics.

23.

Morey

L. C.

(1991). The personality assessment inventory professional manual. Odessa, FL: Psychological Assessment Resources.

24.

Muthén

L. K.

Muthén

B. O.

(2010). Mplus user’s guide (6th ed.). Los Angeles, CA: Author.

25.

Nightingale

Witt

E. A.

Malouf

E. T.

Tangney

J. P

. (2012, July). Reliability and validity of the Mindfulness Inventory Nine Dimensions (MIND): A comprehensive assessment and its mechanisms of action. In Malouf

E. T.

(Chair), Mindfulness and acceptance-based treatment in a jail setting: Theory, measurement, and treatment evaluation from the GMU Inmate Studies. Symposium presented at the Association for Contextual Behavioral Science World Congress, Washington, DC.

26.

Schwartz

R. L.

Fremouw

Schenk

Ragatz

L. L.

(2012). Psychological profile of male and female animal abusers. Journal of Interpersonal Violence, 27, 846-861.

27.

Shonin

Gordon

W. V.

Slade

Griffiths

M. D.

(2013). Mindfulness and other Buddhist-derived interventions in correctional settings: A systematic review. Aggression and Violent Behavior, 18, 365-372.

28.

Sykes

G. M.

Matza

(1957). Techniques of neutralization: A theory of delinquency. American Sociological Review, 22, 664-670.

29.

Tangney

J. P.

Mashek

Stuewig

(2007). The social-clinical-community-criminology interface: The GMU inmate study. Journal of Social and Clinical Psychology, 26, 1-21.

30.

Tangney

J. P.

Meyer

Furukawa

Cosby

(2002). Criminogenic Beliefs and Attitudes Scale. Fairfax, VA: George Mason University.

31.

Tangney

J. P.

Stuewig

Furukawa

Kopelovich

Meyer

P. J.

Cosby

(2012). Reliability, validity, and predictive utility of the 25-item Criminogenic Cognitions Scale (CCS). Criminal Justice and Behavior, 39, 1340-1360. doi:10.1177/0093854812451092

32.

Tangney

J. P.

Stuewig

Mashek

D. J.

(2007). Moral emotions and moral behavior. Annual Review of Psychology, 58, 345-372. doi:10.1146/annurev.psych.56.091103.070145

33.

Walters

G. D.

(2012). Criminal thinking and recidivism: Meta-analytic evidence on the predictive and incremental validity of the Psychological Inventory of Criminal Thinking Styles (PICTS). Aggression and Violent Behavior, 17, 272-278.

34.

Walters

G. D.

Elliott

W. N.

(1999). Predicting release and disciplinary outcome with the Psychological Inventory of Criminal Thinking Styles: Female data. Legal and Criminological Psychology, 4, 15-21.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB

Is There a Dark Side to Mindfulness? Relation of Mindfulness to Criminogenic Cognitions

Abstract

Keywords

Mindfulness

Criminogenic Cognitions

Mindfulness and Criminogenic Cognitions

Study 1

Method

Participants

Measures

Mindfulness

Criminogenic cognitions 1

Results

Mindfulness and criminogenic cognitions

Emotion regulation and criminogenic cognitions

Mediation model

Discussion

Study 2

Method

Participants

Measures

Results

Mindfulness and criminogenic cognitions

Emotion regulation and criminogenic cognitions

Mediation model

Discussion

General Discussion

Limitations

Applied Implications

Footnotes

Declaration of Conflicting Interests

Funding

Supplemental Material

Notes

References

Supplementary Material

Criminogenic cognitions¹