The purpose of this study was to identify possible predictors of individual differences in mindfulness as measured by Mindful Attention Awareness Scale by using three variables: trait anxiety (TA), attentional control (AC) and conscientiousness. A convenience sample of 89 psychology students (both sexes) completed self-report measures related to the study. A one-tailed cross-sectional correlational design was employed to examine the data following by a multiple linear regression of mindfulness onto TA, AC and conscientiousness. Results showed that TA was the only significant and negative predictor of the outcome variable. However, a significant trend emerged for AC that was positively correlated with mindfulness. Conscientiousness failed to predict the criterion variable. Findings also indicated a high negative association between TA and AC signifying indirect mediating effect of AC on mindfulness. The interaction between those predictors and their attribution to mindfulness will be discussed.
Mindfulness is often defined as an enhanced moment-to-moment awareness of perceiving all mental contents as they appear, in non-evaluative way via direct and immediate experience. According to Kabat-Zinn(1990) mindfulness refers to the activity of purposeful attention to what is happening in the present. In other words it is a process of controlling attention that brings non-elaborative awareness to present experience with openness, curiosity and acceptance (Bishop et al.2004).The construct of mindfulness originates from contemplative Buddhist tradition that encourages meditation as a way of increasing consciousness (Hayes,2003; Brown & Ryan, 2003).
Best services for writing your paper according to Trustpilot
Mindfulness has become an essential form of treatment for some popular clinical interventions (e.g. Seagal, Williams, & Teasdale,2002; Kabat-Zinn,1982; Shapiro, Schwartz & Bonner,1998; Grossman, Niemann, Schmidt & Walachl,2004). However, the lack of applicable measures appeared as one of the major problems for testing certain assumptions about mindfulness and its practical implications on health and well-being (Walsh, Balint, Smolira SJ, Fredericksen & Madsen,2009).
Brown and Ryan (2003) devised MAAS: a self-reported measure that is currently the most popular scale. The underpinning assumption of the construct is that mindfulness could be potentially considered as a naturally occurring characteristic. Brown and Ryan (2003) emphasised that future research into the antecedents is needed. In this study MAAS was employed to measure mindfulness.
Walsh et al. (2009) designed two studies as an initial step following this direction and examined possible predictors of individual differences in mindfulness. Results of the regressed mindfulness, as measured by MAAS, onto four different predictors in the first study indicated that trait anxiety (TA) was the most significant predictor. FMI-Freiburg Mindfulness Inventory (Walach et al, 2006) was employed to measure mindfulness in their second study. Findings revealed again that TA appears as the most significant and negative predictor, followed by AC.
The aim of this study is to provide a precise estimate of the relationship between Mindfulness and AC, TA and conscientiousness and investigate whether those variables can give a raise for predicting individual differences in mindfulness. This would allow obtaining a greater inside into psychological mechanisms related to mindfulness. Three variables were chosen to explore their predictive utility of individual differences in mindfulness.
Trait anxiety
Since TA is closely related to Neuroticism, one of the ‘big five’ personality factors and it emerged in the previous studies (Walsh et al., 2009) as the strongest negative predictor of mindfulness, it is expected that TA will confirm this findings (H1). The association of TA with mindfulness can be possibly attributed to the fact that anxious individuals are more likely to have a tendency towards information processing biases (Derryberry & Reed,2002, Mathews & MacLeod,2002). As an effect they seem to have more difficulties to disengage from threatening stimuli and shift their focus to the information available at the present. This is opposite to the mindfulness state as being fully aware of the here and now moment.
Attentional Control
Since most of the explanations of ‘mindfulness’ refer to ‘attention’ as a core component we expect to find a significant correlation between those two variables. Mindfulness relates to bringing awareness to present by regulating attention so thoughts and feelings can be detected as they arise in the stream of consciousness. Therefore mindfulness can be conceptualised as the self-regulation of attention and explained as a megacognitive skill (Bishop et al. 2004). Derryberry and Reed (2002) proposed that there are individual differences in AC and that it has negative correlation with anxiety. Findings of their study indicated an attentional bias effect, meaning that for anxious individuals it is more difficult to disengage from threats. This suggested indirect association of AC with Mindfulness via TA. It was proposed that individual differences in AC modulate this effect. Results from Walsh et al. (2009) showed that AC is statistically significant positive predictor of mindfulness. Considering direct and indirect associations, AC is expected to be a positive predictor of mindfulness and this constituted our second hypothesis (H2).
Conscientiousness
Conscientiousness, as a big-five factor has been linked to self-disciplined and control (Costa & McCrae, 1992). These attributes are likely to become apparent acquaintances with mindfulness. Shapiro et al.(2006) speculated that positive effects of mindfulness could be linked with greater skill to self-regulate by mindful individuals, while less conscientious people can be careless and easily distracted. According to Kabat-Zinn (1990) mindfulness is a way of reflective responses to stressors without reacting to them habitually. Thus mindfulness seems to correspond with one of the conscientiousness’ facet, namely deliberateness – tendency to think carefully before acting. Giluk (2009) found strong positive correlation of those constructs suggesting that self-discipline can be associated to a present-focus attention, which requires strong self-regulation. Based on the above findings conscientiousness is expected to be a positive predictor of mindfulness (H3).
Method
Participants:
Eighty-nine UEL psychology students participated in this study (70 females, 17 males, 2 participants did not reveal their gender). Ages ranged from 20 to 50, 5 participants did not disclose their age (M=30.77 SD=7.10). Forty-nine participants classified their ethnicity as white, 14 as black and 24 considered themselves as ‘other’, 2 failed to provide their ethnic background.
Design and Measures
A one-tailed cross-sectional and correlational design with 1 outcome variable (individual differences in mindfulness) and 3 predictor variables (TA, AT, Conscientiousness) were used. Each of the variables was measured as follows by individual self-report questionnaire (See Appendix 3).
Mindfulness
Mindful Attention Awareness Scale (MAAS) was employed (Brown & Ryan, 2003). An example item is: I snack without being aware that I’m eating. Each item has a six-point Likert scale from “almost always” to “almost never” with higher scores indicating greater level of mindfulness.
Trait anxiety
Twenty item Spielberger’s (1983) State Trait anxiety Inventory with four-point Likert scale was used. Questions 1,3,6,7,10,13,14,16 and 19 required reversing, such that a higher scores implied a greater level of anxiety. An example item is: I feel secure.
Attentional control
Attentional Control Scale (Derryberry & Reed’s,2002), comprising 20 items, was used. Each item had a four-point Likert scale with higher scores implicating greater AC. An example question was: I have trouble carrying on two conversations at once. Scores from the following questions were reversed: 2,3,6,7,8,11,12,15,16&18.
Contentiousness
Conscientiousness was assessed by 12 item scale from the NEO Five Factor Inventory (Costa & McCrae, 1992). Each item had a five-point Likert scale with possible response ranging from ‘strongly agree’ to ‘strongly disagree’. The higher score signified greater conscientiousness. Four items required reversing (3,6,9&11). The example item is: I work hard to accomplish my goals.
Procedure
During Research Methods lecture students were given a written invitation to the study along with a consent form, which was completed anonymously and collected prior to receiving the four self-measures (see Appendix 1, 2 and 3) and personal details sheet. Scoring instructions were handed and participants scored their own questionnaires on separate sheets, reversing scores where needed. (See Appendix 4 and 5). Participants were given approximately 15-minutes to answer and 10-minutes to score the questionnaires.
Ethics
The invitation letter informed that the study is voluntary, anonymous and that it attempts to measure individual differences so they could make a choice about participating. Consent forms were collected separately from the questionnaires to guarantee anonymity. Participants were advised to write a memorable number in case they decide to withdraw their data.
Results
The raw data was collected from the questionnaires completed by participants. Descriptive statistics shown in Table 1. represent the means and standard deviations for criterion and predictor variables. AC scores ranged from 36 to 68 with possible range 20-80 (M=52.69,SD=7.44). TA scores ranged from 25 to 72 with potential range 20-80 (M=44.06,SD=10.15). Scores from the conscientiousness measure stretched from 23 to 59 with potential range 12-60 (M=44.48,SD=7.64). The minimum score on mindfulness was 26 and maximum 87 with the possible range 15-90 (M=58.28,SD=11.78). The scores presented in Table 1. indicate that there were participants on extremes of the mindfulness scale with one participant perceiving him/herself as very mindful (scoring 89), and another as really low on the mindfulness scale (scoring 23).
Table1.
Mean scores and standard deviations of mindfulness (DV) and three predictor variables.
N
Minimum
Maximum
Mean
Std. Deviation
Attentional Control
89
36
68
52.69
7.44
Trait Anxiety
89
25
72
44.06
10.15
Conscientiousness
89
23
59
44.48
7.64
Mindfulness
89
26
87
58.28
11.78
One-tailed cross sectional correlational analysis was conducted to explore whether predictors correlate significantly with the outcome variable. The alpha level I±=.05 was used for conscientiousness and I±=.01 for TA and AC. Table 2. presents intercorrelations between variables.
Table 2.
Inter-correlations between variables in the study (N=89)
1
2
3
4
1.Mindfulness
Pearson Correlation
Sig.(1-tailed)
–
2.Trait Anxiety
Pearson Correlation
Sig.(1-tailed)
-.54**
.000
–
3.Attentional Control
Pearson Correlation
Sig.(1-tailed)
.34**
.001
-.33**
.001
–
4.Conscientiousness
Pearson Correlation
Sig.(1-tailed)
.21*
.026
-.27**
.005
.37**
.000
–
**. Correlation is significant at the 0.01 level (1-tailed).
*. Correlation is significant at the 0.05 level (1-tailed).
As expected, the outcome variable correlated significantly with all the predictor variables. The strongest being a negative correlation between mindfulness and TA (r(87)=-.54,p<.01), whereas AC (r(87)=.34,p<.01) and conscientiousness (r(87)=.21,p<.05) were both positively correlated with the criterion variable. The predictor variables significantly correlated with each other, TA was negatively associated with both AC (r(87)=-.33,p<.01) and conscientiousness (r(87)=-.27,p<.01). Whereas AC and conscientiousness indicated significant positive correlation (r(87)=.37,p<0.01). A multiple linear regression analysis was then carried out to examine if the predictors account for significant variation in mindfulness, and if so, to determine the predictive power of each of the predictors. Given that all the variables were correlated with each other, the variance inflator factors (VIF) were included in analysis to eliminate multicollinearity problem.
Table 3.
Regression of Mindfulness onto AC, TA and Conscientiousness.
Predictor
Unstandardized Coefficients
Standardized Coefficients
t
Sig.
Collinearity Statistics
B
Std. Error
Beta
Tolerance
VIF
Constant/Intercept
67.81
11.65
5.82
.00
Attentional Control
.28
.16
.18
1.77
.08
.80
1.25
Trait Anxiety
-.56
.11
-.49
-5.06
.00
.86
1.16
Conscientiousness
.01
.15
.01
.09
.93
.83
1.20
a. Dependent Variable: Mindfulness
Analysis of the standardised regression coefficients as presented in Table 3. indicated that TA was the only one significant predictor (t(88)=-5.06, p<.01) of mindfulness, although AC indicated a significant trend (t(88)=1.77,p=.08). Conscientiousness, on the other hand, failed to predict mindfulness independently (t(88)=0.09,p=.93).
Table 3. shows Constant value of 67.81 (t(88)=5.82,p<.01) that is significant and correlates positively with the criterion variable.
Analysis of the data using simultaneous multiple regression revealed that the combined predictors explained 33% of the variance in mindfulness (R2=0.33),F(3,88)=13.68,p<0.01. TA was the only significant individual predictor, I?=-.49,p<.001; meaning that for every 1 point increase of TA, Mindfulness declines of 0.49. Neither AC, I?=.18,p=.08, nor conscientiousness, I?=.01,p=.93, were found to be significant individual predictors in the final model. However, results indicated that 2.5% of a unique shared variance in Mindfulness could be attributed to AC, since the results of linear regression indicted a significant trend of AC. The difference between the amount of variation in the outcome variable accounted for by the model (R2=33%) and the total 'unique' variance of both predictors (2.5% AC + 20.3% TA = 22.8%) is 9.8%. Given that the model exhibited 9.8% of shared variance between Mindfulness, TA and AC, as depicted in Figure 1.
2.5% Unique Variance of AC in Mindfulness
Shared Variance 9.8%
20.3% Unique Variance of TA in Mindfulness
Figure 1. Graphic representation of unique and shared variance in Mindfulness accounted for by Attentional Control and Trait Anxiety.
Multicollinearity did not appear to be a problem. Tolerance scores were greater than 0.1 for all predictors (TA=.87,AC=.80,Conscientiousness=.84) and corresponding VIF values were all less than 10(TA=1.16,AC=1.25,Conscientiousness=1.20).
Discussion
The current study examined the role of TA, AC and conscientiousness in predicting individual differences in mindfulness based on the sample of psychology students. The results only partially confirmed initial hypothesis and can be summarized as follows. As expected a significant correlation was found between mindfulness and all the predictor variables. Furthermore, all predictors significantly correlated with each other. Mindfulness, AC and conscientiousness were correlated positively. Unlikely, TA which was negatively interrelated with all other variables. However further investigation through a multiple linear regression suggested that TA was the only significant and powerful predictor of individual differences in Mindfulness (p<.01) supporting hypothesis (H1). Results for AC showed no significant power in predicting mindfulness contrary to hypothesis (H2), though a significant trend was noticeable (p=.08). Results showed that AC accounted for 2.5% of variance in Mindfulness in the final model. Conscientiousness failed to predict mindfulness (p=.93).
Findings for TA were consistent with the previous research by Walsh et al.(2009)
and Derryberry and Reed (2002) indicating that anxious individuals are more likely to have information processing biases causing difficulties to disengage from threatening stimuli. Support for this notion comes also from the study by Mathews and MacLeod (1994;2002). Anxiety-prone individuals are more likely to have a tendency for automatic mechanisms involved in the responses to threat due to the absence of awareness in given situation. This provides rationale for the observed negative association of TA with Mindfulness and corresponds with the notion that TA plays an important role in predicting individual differences in Mindfulness. The application of this findings is in line with the concept that mindfulness based interventions could be effective way in reducing anxiety symptoms (Wong et al., 2011; Evans et al, 2008; Hofman ,Sawyer, Witt & Oh, 2010; Grossman et.al., 2003).
Although AC was significantly correlated with Mindfulness it appeared not to be a significant predictor. Nevertheless an interesting trend was revealed, anticipating that if we had a larger sample we could expect to find significance.
Since there is evidence indicating that anxiety disordered subjects are more likely to show a variety of cognitive biases (Mathews & MacLeod,2002), the negative correlation between AC an TA was not unexpected. According to some theoretical accounts people with a high anxiety tend to exhibit a poor recruitment of AC mechanisms what could be associated with impaired AC (Muris, Jong & Engelen, 2003). Findings from the previous study by Walsh et al. (2009) suggested that AC does have a significant power in predicting mindfulness thus future research should explore AC association with Mindfulness in more depth. Given that AC and TA are negatively and significantly correlated (Derryberry & Reed, 2009) it seems reasonable to speculate that AC could have an indirect effect on predicting individual differences in Mindfulness as found by Walsh et al. (2009). However association of Mindfulness with TA via AC is only partially explainable as suggested by Walsh et al. (2009). This implies that there might be some other factors influencing this correlation. The fact that AC accounted for 2.5% of variance in Mindfulness and that it seems to have a partial mediating power in predicting Mindfulness via TA, suggests an interesting avenue for a future research to examine the 3-way relationship between AC, TA and Mindfulness. A factor possibly influencing the relationship of three constructs is information processing biases. Investigation on clinical subjects throughout longitudinal research could be of enormous benefit in exploring the aforementioned 3 way relationship.
Meanwhile self-report methods and correlational nature of this study seem to limit the extent to which conclusions on cause-effect can be drawn. Examining how these variables relate to each other over time by employing more complicated statistical models allowing capturing dynamic interactions in the time series and across measures.
Although conscientiousness was significantly correlated with Mindfulness it was found to lack a predictive power, contrary to the third hypothesis (H3). According to Giluk (2009) one of the facet of conscientiousness, namely deliberation aligns with non-judgemental dimension of mindfulness implying that the lack of significance between the two variables could be due to the conceptual divergence of both concepts. Employing alternative measures could potentially provide a direction for a further investigation of the relation between conscientiousness and mindfulness. As suggested by some scholars the concept of Mindfulness should be revised due to supposed incongruities of the way that mindfulness is conceptualised and measured (Giluk. 2009 Grossman 2011, Walach et al., 2006).
Various other shortcomings should be admitted. Firstly, the self-report measure used in this study to asses Mindfulness seems to account for attention and awareness only and thus omit some other aspects of mindfulness, like non-judgemental attitude, acceptance, lack of specific ego driven goals (Walach et al 2006). The qualities that MAAS seems to neglect could possibly correspond with Conscientiousness if the scale was differently conceptualised. An interesting avenue for future research would then be to conduct a similar study using other measures (e.g. Kentucky Inventory of Mindfulness Skills; Baer, Smith & Allen, 2004) or apply qualitative methods to obtain greater inside into a psychological mechanisms related to the Mindfulness practise (Grossman, 2009) and establish whether mindfulness is more trait or state-like component.
Secondly, the sample consisted of psychology students that might indicate that participants were representing a fairly homogenous group with certain knowledge, psychological awareness and familiarity with the content of self-measures that could influence results. Thus, not only larger, but also more diverse sample would allow for a closer inspection of Mindfulness and its predictors across general population.
