- Email: [email protected]
Predicting individual differences in mindfulness: The role of trait anxiety, attachment anxiety and attentional control
Personality and Individual Differences 46 (2009) 94–99
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
Predicting individual differences in mindfulness: The role of trait anxiety, attachment anxiety and attentional control James J. Walsh *, Marc G. Balint, David R. Smolira SJ, Line Kamstrup Fredericksen, Stine Madsen School of Psychology, University of East London, Romford Road, London E15 4LZ, UK
a r t i c l e
i n f o
Article history: Received 8 February 2008 Received in revised form 4 September 2008 Accepted 10 September 2008 Available online 23 October 2008 Keywords: Mindfulness Trait anxiety Attachment anxiety Attentional control
a b s t r a c t Two correlational studies sought to identify possible predictors of individual differences in naturally occurring mindfulness. In study one, trait anxiety and attachment anxiety, but not attachment avoidance, were negatively predictive of mindfulness. In study two, trait anxiety ( ) and attentional control (+), but not openness or parental nurturance, predicted mindfulness. In addition, there was evidence of a partial mediation effect of attentional control on the association between trait anxiety and mindfulness. Key features of trait anxiety such as attentional and interpretative processing biases, as well as those of attachment anxiety such as rumination and hypersensitivity, are at odds with mindfulness characteristics such as attention to what is present coupled with an attitude of openness and acceptance. Thus, whether generalised or specific, anxiety appears to be antagonistic to mindfulness; control over one’s attentional resources may form part of the underlying explanation. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Originating in contemplative traditions such as Buddhism, mindfulness is defined as a state of enhanced attention to, and awareness of, what is taking place in the present (Brown & Ryan, 2003). Such awareness is characterised as open and receptive, but not judgemental (Bishop et al., 2004; Deikman, 1982). Mindfulness appears to be absent when attention is captured by rumination and fantasy (Brown & Ryan, 2003). Building on the original work of Kabat Zinn (1982), mindfulness-based interventions have proliferated over the past twenty years, mostly with very positive outcomes (Baer, 2003; Grossman, Niemann, Schmidt, & Walach, 2004). Underpinning most of these studies is an assumption that mindfulness training increases levels of mindfulness and such increases mediate the observed positive outcomes. However, attempts to validate this assumption have rarely been undertaken, the most likely reason being the absence of appropriate measures. It is only in the last few years that psychometrically valid measures of mindfulness have appeared in the literature (Baer, Smith, & Allen, 2004; Brown & Ryan, 2003; Walach, Buchheld, Buttenmüller, Kleinknecht, & Schmidt, 2006). In developing their measure, Brown and Ryan (2003) argued that mindfulness could be considered a ‘naturally occurring characteristic’ (p. 822) with both inter- and intra-personal variation. Similarly, Walach et al. (2006) argued that mindfulness could be expressed dispositionally and in state form, depending on the
* Corresponding author. E-mail address: [email protected] (J.J. Walsh). 0191-8869/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.paid.2008.09.008
time-frame in question. Given the importance of the construct in terms of physical and psychological well-being, Brown and Ryan (2003) emphasised the need to explore its antecedents. As an initial step in this direction, the two studies reported here were designed to investigate possible predictors of mindfulness. 2. Study 1 Whilst it is likely that individual differences in mindfulness will eventually prove to be multiply determined, a decision was made in the first instance to explore both the developmental and personality domains for factors that might possess predictive utility. Attachment and trait anxiety emerged as strong candidates from their respective fields. 2.1. Attachment Attachment processes developed in early childhood are believed to remain active throughout the lifespan (Bowlby, 1988). Current conceptualisations of attachment suggest a two-dimensional system, namely anxiety and avoidance (Brennan, Clark, & Shaver, 1998; Fraley & Waller, 1998). Low levels of anxiety and avoidance reflect secure attachment (Schachner, Shaver, & Mikulincer, 2005). Insecure attachment, however, can take one of two forms. If proximity-seeking is highly desired, then a ‘hyper-activating’ attachment strategy is adopted. This is characterised by intensive efforts to seek proximity and protection, hypersensitivity to signs of rejection and abandonment, and excessive rumination upon
95
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99
one’s personal deficiencies and threats to one’s relationships (Mikulincer & Florian, 1998). Collectively, these features constitute attachment anxiety (Mikulincer & Shaver, 2003). In contrast, where it is felt that proximity-seeking will not serve to reduce threat, a ‘deactivating’ attachment strategy is adopted. Here, one removes oneself from stimuli that are likely to activate the attachment system. This results in the avoidance of proximity-seeking, the denial of attachment needs and the suppression of signs of vulnerability. This strategy is known as attachment avoidance (Mikulincer & Shaver, 2003). The current study seeks to determine whether attachment anxiety and attachment avoidance predict mindfulness. According to Bishop et al. (2004), mindfulness involves attending to immediate experience and adopting an attitude of curiosity, openness and acceptance. It also includes a ‘decentred’ approach to one’s thoughts which serves to reduce cognitive elaboration and prevents rumination (Teasdale et al., 2000). Finally, instead of observing one’s experiences through various filters of beliefs and expectations, mindfulness involves direct observation of them. Thus, given the hypersensitivity to rejection and rumination on personal deficiencies associated with attachment anxiety, together with the thought suppression, relationship avoidance and person perception biases associated with attachment avoidance, it seems that these dimensions are essentially antithetical to many of the key features of mindfulness. Therefore, it is hypothesised that both attachment anxiety (H1) and attachment avoidance (H2) will be negatively predictive of mindfulness. Furthermore, since low scores on both attachment dimensions are associated with secure forms of attachment, the analysis will examine the interaction of the two main predictors to see if it accounts for additional amounts of criterion variance. 2.2. Trait anxiety Trait anxiety is closely related to one of the ‘big five’ personality factors, namely neuroticism; however, the underlying cognitive architecture of the former is probably better understood (Eysenck, 1992,1997). People high in trait anxiety are thought to possess both attentional and interpretative processing biases (MacLeod, 1990). Specifically, they are more likely than low anxious counterparts to detect threat, regardless of its source (Eysenck, 2000). Similarly, they are more likely to make threatening interpretations of ambiguous stimuli whereas low anxious counterparts are inclined to make neutral interpretations (Eysenck, Mogg, May, Richards, & Mathews, 1991). This bias towards detecting, interpreting and elaborating threat may be contrasted with mindfulness which advocates an open acceptance of what is present. Accordingly, it may be hypothesised that trait anxiety will be negatively predictive of mindfulness (H3).
3. Method 3.1. Design and measures A cross-sectional, correlational design was employed. Attachment-related anxiety, attachment-related avoidance and trait anxiety constituted the predictor variables. The criterion variable was mindfulness. Attachment-related anxiety and attachment-related avoidance were measured by the Experiences in Close Relationships-Revised (ECR-R; Fraley, Waller, & Brennan, 2000) questionnaire. The measure comprises 36-items and employs a six-point Likert scale. Higher scores signified more attachment-related anxiety (a = 0.92) and attachment-related avoidance (a = 0.92). Secure attachment patterns are reflected in low scores on both scales.
Trait anxiety, a ‘facet’ of neuroticism, was measured using an 8item scale from the revised NEO personality inventory (NEO-PI-R; Costa and McCrae (1992)). Respondents were asked to read the item statements and to rate their level of agreement with each using a five-point Likert scale. Higher scores indicated more trait anxiety (a = 0.76). Mindfulness was measured by the Mindful Attention Awareness Scale (MAAS) (Brown & Ryan, 2003). Although each of the 15 items signified the absence of mindfulness, the six-point Likert scale is arranged such that higher scores reflected more mindfulness (a = 0.85). 3.2. Participants Of the 334 questionnaires distributed to psychology students and staff, 127 usable replies were obtained (38%). The sample comprised 100 females and 23 males (four respondents failed to provide demographic details). Ages ranged from 17–60 years, with just over half (53%) of the participants being 30 years or younger. 3.3. Procedure Students and teaching staff were approached on campus and invited to participate in the study. Participants were asked to complete the questionnaire anonymously and return it either to a designated box on campus or post it to the researcher in an accompanying stamped self-addressed envelope. 4. Results The means and standard deviations for the variables, together with their inter-correlations, are presented in Table 1. As expected, mindfulness correlated negatively and significantly with all three hypothesised predictors. Next, mindfulness was regressed onto attachment anxiety, attachment avoidance, their interaction, and trait anxiety (see Table 2). Overall, the model accounted for 18% of the variance in mindfulness and proved to be reliable [F(4,122) = 6.57, p < 0.001]. Individually, attachment anxiety and trait anxiety were significant predictors, the latter emerging as slightly stronger. Neither attachment avoidance, nor its interaction with attachment anxiety, emerged as significant. 5. Discussion In line with the first hypothesis, adult attachment anxiety emerged as a significant independent predictor of individual differences in mindfulness. Examination of the attachment literature offers a solid conceptual basis for such a finding. Firstly, anxiously attached individuals are hypervigilant to threat-related cues, especially cues of rejection and abandonment, and ruminate extensively on distress-related material (Mikulincer & Florian, 1998). In contrast, mindful people do not deliberately monitor the environment for threat signals and, given their tendency not to elaborate incoming sensations and thoughts, they avoid rumination Table 1 Mean scores, standard deviations, and inter-correlations between variables in study 1 (N = 127)
1. 2. 3. 4.
Mindfulness Attachment anxiety Attachment avoidance Trait anxiety
*p 6 0.01; **p 6 0.001.
Mean
SD
1
3.87 2.71 2.68 3.09
0.81 0.96 0.95 0.72
–
0.32** 0.25* 0.33**
2
3
4
– 0.36** 0.31**
– 0.22*
–
96
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99
Table 2 Regression of mindfulness onto attachment anxiety, attachment avoidance, their interaction, and trait anxiety
Attachment anxiety Attachment avoidance Interaction Trait anxiety
B
SE(B)
Beta
0.18 0.11 0.05 0.27
0.08 0.08 0.06 0.10
0.21* 0.13 0.08 0.24**
*p < 0.05; **p < 0.01.
(Brown & Ryan, 2003). Secondly, while those who are anxiously attached resist information that disconfirms their expectations and fail to update existing schemas in light of contradictory evidence (Mikulincer & Arad, 1999), mindful people are much more open to new information and seek to avoid schema-congruent processing (Brown & Ryan, 2003). Finally, attachment anxiety is known to be associated with intolerance of ambiguity (Hazan & Shaver, 1987). In contrast, Brown and Ryan (2003) have shown that mindfulness is associated with greater cognitive flexibility and openness. In sum, the most notable features of mindfulness – attending to one’s immediate experience with an open and accepting attitude, avoiding the tendency to elaborate one’s thoughts and sensations, decentring from self-derogatory thoughts and not ruminating – contrast substantially with attachment anxiety. Attachment avoidance failed to predict mindfulness even though their zero-order correlation was significant. Whilst some features of attachment avoidance (thought suppression, defensive projection and cognitive closure) point towards a negative association with mindfulness, other features (inhibited processing of threat and the non-elaboration of a broad range of cognitions) may be positively related (Mikulincer & Shaver, 2003). This combination of attachment avoidance features appears to have differential, and potentially conflicting, predictive impact on mindfulness and may underlie the observed null result. Developing a multifaceted measure of attachment avoidance that contains independent sub-scales would facilitate a more precise testing of the hypothesised association with mindfulness. An alternative reason why attachment avoidance failed to emerge as significantly predictive may be due to its conceptual overlap with both attachment anxiety and trait anxiety. Trait anxiety emerged as the strongest predictor of mindfulness. According to Costa and McCrae (1992), trait anxiety represents a generalised tendency to be fearful, worried, and apprehensive about the future. Other conceptualisations of trait anxiety suggest the presence of danger- and threat-related schemas in long-term memory which guide information processing (Beck & Emery, 1985). The tendency to cognitively elaborate threat-related stimuli, coupled with schema-driven processing, conflicts with the key elements of mindfulness, namely, the allocation of attention to immediate stimuli and their non-elaboration (Bishop et al., 2004). An attempt will be made in Section 6 to explore more thoroughly this association between trait anxiety and mindfulness.
6. Study 2 Study 2 was designed partly to replicate the main finding from Section 2 concerning the key predictive role of trait anxiety. On this occasion, however, an alternative measure of mindfulness was employed, namely the Freiburg mindfulness inventory (FMI – Walach et al. (2006)). Brown and Ryan’s (2003) measure of mindfulness (MAAS) is known to be unidimensional whereas other measures of mindfulness are predicated upon a multidimensional conceptualisation (Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006). The FMI depicts mindfulness as a general construct, which incorporates
inter-related facets such as non-judgemental acceptance, mindful presence and insight. In a study examining the associations between five measures of mindfulness, the correlation between the MAAS and the FMI (r = .31) was the weakest (Baer et al., 2006). If trait anxiety once again proves to be predictive of mindfulness when an alternative, multidimensional, measure is employed, then this would confirm the findings from Section 2 and consolidate the underlying process-based link. As in Section 2, a negative association between trait anxiety and mindfulness is hypothesised (H1). It should be noted that as well as changing the measure of mindfulness, a different measure of trait anxiety was also used, namely Spielberger’s (1983) STAI-T. This is probably the most widely used measure of trait anxiety in the literature. Three more possible predictors of mindfulness were also examined – openness to experience, parental nurturance and attentional control. Openness to experience is believed to be a key dimension of individual differences and constitutes one of the big five personality factors (Costa & McCrae, 1992). People high in openness are attentive to inner feelings, intellectually curious, and independent in judgement. Low scorers, in contrast, are much more conservative in outlook and prefer the familiar to the novel. Since mindfulness is conceptualised as an open form of attention to, and awareness of, the present (Brown & Ryan, 2003), it seems reasonable to argue that openness would positively predict it (H2). Brown and Ryan (2003) measured this association in two student samples and found surprisingly low coefficients. However, as noted above, it may be that their measure of mindfulness was not broad enough to capture in full the hypothesised association between the two constructs. The use of Walach et al. (2006) multidimensional measure will provide a more comprehensive test of the hypothesis. Parental nurturance (Buri, 1989; Buri, Kirschner, & Walsh, 1987) is thought to be one of the key mechanisms through which individuals develop their attachment styles. In contrast to adult measures of attachment, it relates more explicitly to one’s formative years. It was theorised that experiencing a secure and supportive childhood via nurturing parents would facilitate a tendency to employ an open, explorative and accepting attitude towards the environment (mindfulness) in later life. Therefore, the third hypothesis predicts a positive association between parental nurturance and mindfulness (H3). While most theoretical conceptualisations of mindfulness highlight ‘attention’ as a core feature (Bishop et al., 2004; Brown & Ryan, 2003; Walach et al., 2006), they rarely specify which aspect of attention is implicated. Derryberry and Reed (2002) identified individual differences in attentional control. People high in attentional control are thought to be better at focusing attention, shifting attention and controlling thoughts in a flexible manner than low control counterparts. Such characteristics seem to overlap considerably with those of mindfulness. If so, one would expect to find a significant positive association between attentional control and mindfulness and this constitutes the fourth hypothesis (H4). When validating their measure, Derryberry and Reed (2002) found a correlation of r = 0.55 between trait anxiety and attentional control. The tendency for trait anxious participants to have poorer attentional control may underpin the negative association between trait anxiety and mindfulness observed in Section 2. This suggestion will be explored empirically in the current study. Specifically, it is hypothesised that as well as having a direct effect on mindfulness, attentional control also will mediate the association between trait anxiety and mindfulness (H5). 7. Method 7.1. Participants Undergraduate and graduate students (N = 153) participated in the study. Nurturance from just one parent was reported by 17 stu-
97
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99
dents and four others failed to complete the mindfulness measure; as a result, their data were removed from further analysis. Of the remaining 132 participants, 26 (20%) were male and 105 (79%) female with one respondent failing to indicate their sex. The mean age of participants was 25.9 years (SD = 6.7 years). 7.2. Design and measures
Table 4 Regression of mindfulness onto trait anxiety, openness, attentional control and parental nurturance (mother)
Trait anxiety Openness Attentional control Parental nurturance (Mo)
B
SE(B)
Beta
0.37 0.06 0.20 <0.01
0.07 0.06 0.10 0.04
0.46** 0.07 0.17* 0.001
A cross-sectional, correlational design was employed. Individual differences in mindfulness constituted the criterion variable whilst the predictors included openness, trait anxiety, parental nurturance (maternal and paternal) and attentional control. Each variable was measured as follows.
*p < 0.05; **p < 0.001. R2 = 0.33; Model F(4,127) = 15.39, p < 0.001.
7.3. Mindfulness
The means and standard deviations for the variables, together with their inter-correlations, are presented in Table 3. As predicted, mindfulness correlated positively and significantly with openness, attentional control and parental nurturance (mother), and negatively with trait anxiety. A multiple regression analysis was carried out to determine the predictive power of each of the significant correlates of mindfulness. Examination of the standardised regression coefficients indicates that trait anxiety was the strongest predictor of mindfulness although attentional control was also significant. Openness and parental nurturance (mother), on the other hand, failed to predict mindfulness independently (see Table 4). To test the mediating effect of attentional control on the relationship between trait anxiety and mindfulness, three regressions were carried out in line with Baron and Kenny (1986). In the final model, the mediator proved to be significant and the effects of trait anxiety decreased (beta was reduced from 0.54 to 0.46, a drop of almost 15%) but still remained significant. Thus, a partial mediation effect was noted. The amount of variance accounted for was R2 = 0.32, p < 0.001 (see Table 5).
Walach et al. (2006) Freiburg mindfulness inventory was used. This 14-item measure uses a four-point Likert scale with higher scores indicate greater levels of mindfulness (a = 0.77). 7.4. Trait anxiety The trait component of Spielberger’s (1983) state-trait anxiety inventory, comprising 20 items, was used. A four-point response scale was employed with higher scores indicating more trait anxiety (a = 0.90). 7.5. Openness Openness to experience was assessed using Costa and McCrae’s (1992) 12-item scale from the NEO-FFI (neuroticism, extraversion, openness – five factor inventory) – form S. A five-point Likert scale was used with larger scores indicating greater openness (a = 0.73).
8. Results
7.6. Parental nurturance Parental nurturance was assessed using Buri et al.’s (1987) 24item scale. Items describing positive and negative aspects of parental nurturance were rated on a five-point Likert scale on two occasions, once for mother/maternal carer and once for father/paternal carer. Higher scores indicated greater levels of parental nurturance (a = 0.96 (maternal) and 0.95 (paternal)). 7.7. Attentional control Derryberry and Reed’s (2002) 20 item measure was employed. The four-point Likert scale was scored such that larger scores signified greater attentional control (a = 0.83). 7.8. Procedure Questionnaires were completed anonymously during a research methods class in the following order: attentional control, openness, mindfulness, trait anxiety and parental nurturance.
9. Discussion Once again, trait anxiety emerged as the best predictor of mindfulness, thereby confirming the findings from study one as well as those of Baer et al. (2006) and Brown and Ryan (2003). People high in trait anxiety are believed to possess ‘danger’ schemata which influence information processing in a top–down manner (Beck & Emery, 1985). Being predisposed to attend to threat, and preferentially interpreting threat from ambiguous stimuli, requires the utilization of limited processing resources such that less is available for being openly attentive to, and aware of, current experience. Indeed, there is good evidence that anxiety seems to direct thinking towards future events (Eysenck, Payne, & Santos, 2006). The combination of a relative inability to maintain focus on the present, whilst simultaneously over-interpreting incoming stimuli, may explain why more anxious people are less mindful. It would be fascinating to see if mindfulness-based interventions could reduce trait anxiety and its associated processing biases over time.
Table 3 Mean scores, standard deviations, and inter-correlations between variables in study 2 (N = 132)
1. 2. 3. 4. 5. 6.
Mindfulness Trait anxiety Openness Attentional control Parental nurturance (Mo) Parental nurturance (Fa)
*p < 0.05; **p < 0.01; ***p 6 0.001.
Mean
SD
1
2.69 2.19 3.61 2.58 3.85 3.54
0.45 0.56 0.55 0.40 0.91 0.91
–
0.54*** 0.16* 0.39*** 0.16* 0.07
2
3
4
5
6
– 0.20** 0.07 0.09
– 0.19* 0.16*
– 48***
–
– 0.12 0.46*** 0.27*** 0.30***
98
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99
Table 5 Regression of mindfulness onto trait anxiety with and without the mediator (attentional control) Without mediator B Trait anxiety Attentional control
SE 0.44
–
0.06 –
With mediator Beta
t
0.54 –
7.36 –
Attentional control, incorporating the abilities to focus attention, avoid distraction and switch attentional focus, also predicted mindfulness. This is not surprising given that mindfulness requires a strong attentional focus on what is present together with a facility to switch easily to incoming, replacement, stimuli (Bishop et al., 2004). Using a direct measure of attentional control, the current results endorse those of Baer et al. (2006) who found that mindfulness was negatively associated with poor attentional control. In Derryberry and Reed’s (2002) initial research, as well as in the current study, trait anxiety was negatively associated with attentional control. Exploration of the mediating role of attentional control in the association between trait anxiety and mindfulness revealed a small but interesting effect. The combination of a predisposition to attend to threat on the one hand (high trait anxiety), and an inability to switch attention away from threat on the other (low attentional control), may prevent high anxious people from being mindful. It should be noted, of course, that trait anxiety still has a direct predictive association with mindfulness and so this indirect route via attentional control is only part of the explanation. However, attentional control does constitute a promising predictor of mindfulness alongside personality and developmental factors. Contrary to the second hypothesis, openness failed to predict mindfulness. The low zero-order correlation between the two constructs confirmed the findings of Brown and Ryan (2003). Thus, the facets of openness which the measure taps into were reexamined (Costa & McCrae, 1992). It was felt that openness to fantasy and openness to aesthetics did not depict the kind of openness that was conceived to be similar to mindfulness with its emphasis on current experience. Accordingly, the relevant items were removed from the overall openness score which was recalculated and correlated with mindfulness. The resultant coefficient reduced to 0.10 and became non-significant. It may be concluded that openness to experience and mindfulness are essentially different theoretical constructs, the conceptual breadth of the two rendering them incompatible. While openness constitutes a fairly broad dimension of personality, mindfulness is much narrower in scope and focuses mainly on attention to one’s immediate experience. Such differences in conceptual latitude may account for the very modest association detected. Parental nurturance failed to predict mindfulness. By encouraging participants to think about each statement as it applied to them ‘. . . during your years of growing up’, it may be that the instructions lacked temporal specificity in terms of early, middle or late childhood. Anecdotal reports from the participants confirmed this. Alternatively, it is likely that the temporal gap between childhood and adulthood would have provided participants with ample opportunity to influence the hypothesised association between variables. In essence, exposure to non-nurturing parents may serve to motivate an individual to search for change and to actively pursue psychological repair. The fact that participants had all chosen to study psychology suggests that they were interested in gaining insight into human (and by inference their own) behaviour and experience. It should also be noted that parental nurturance correlated significantly and negatively with trait anxiety and positively with attentional control; these latter two constructs accounted for significant amounts of variance in mindfulness in the regression model.
p
B
SE
Beta
t
p
<0.001 –
0.37 0.21
0.07 0.09
0.46 0.19
5.61 2.28
<0.001 =0.024
10. Conclusions and limitations In the two studies reported here, trait anxiety emerged as a reliable predictor of mindfulness irrespective of how either variable was measured. Key features of trait anxiety such as attentional and interpretative processing biases (Eysenck, 1992) are incompatible with what are believed to be the central components of mindfulness, namely, attention to one’s immediate experiences and an attitude of openness and acceptance (Bishop et al., 2004; Brown & Ryan, 2003; Walach et al., 2006). In addition, attachment anxiety is also negatively predictive of mindfulness. Again, examination of the mechanisms and processes underpinning hyper-activation and attachment anxiety such as effortful proximity-seeking, hypersensitivity and excessive rumination (Mikulincer & Shaver, 2003) provides a clear rationale for the observed association. Thus, regardless of the specificity with which it is assessed, anxiety appears to militate against the expression of mindfulness. While the lack of attentional control associated with anxiety (Derryberry & Reed, 2002) constitutes part of the explanation of its association with mindfulness, other factors are clearly implicated. However, the reliance of the current studies on self-report methods, crosssectional and correlational designs, limits the extent to which conclusions can be drawn about possible pathways between anxiety, attentional control and mindfulness.
References Baer, R. (2003). Mindfulness training as a clinical intervention: A conceptual and empirical review. Clinical Psychology: Science and Practice, 10, 125–143. Baer, R., Smith, G., & Allen, K. (2004). Assessment of mindfulness by self-report: The Kentucky inventory of mindfulness skills. Assessment, 11, 191–206. Baer, R., Smith, G., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using selfreport assessment methods to explore facets of mindfulness. Assessment, 13, 27–45. Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182. Beck, A., & Emery, G. (1985). Anxiety disorders and phobias: A cognitive perspective. New York: Basic Books. Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. D., Carmody, J., et al. (2004). Mindfulness: A proposed operational definition. Clinical Psychology: Science and Practice, 11, 230–241. Bowlby, J. (1988). A secure base: Clinical applications of attachment theory. London: Routledge. Brennan, K., Clark, C., & Shaver, P. (1998). Self-report measurement of adult attachment: An integrative overview. In J. Simpson & W. Rholes (Eds.), Attachment theory and close relationships (pp. 46–76). New York: Guilford Press. Brown, K., & Ryan, R. (2003). The benefits of being present: Mindfulness and its role in psychological well-being. Journal of Personality and Social Psychology, 84, 822–848. Buri, J. (1989). Self-esteem and appraisals of parental behaviour. Journal of Adolescent Behavior, 4, 33–49. Buri, J. R., Kirschner, P. A., & Walsh, J. M. (1987). Familial correlates of self-esteem in American young adults. Journal of Social Psychology, 127, 583–588. Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-R) and NEO five factor inventory (NEO-FFI): Professional manual. Odessa, FL: Psychological Assessment Resources. Deikman, A. J. (1982). The observing self. Boston: Beacon Press. Derryberry, D., & Reed, M. (2002). Anxiety-related attentional biases and their regulation by attentional control. Journal of Abnormal Psychology, 111, 225–236. Eysenck, M. (1992). Anxiety: The cognitive perspective. Hove, UK: Lawrence Erlbaum. Eysenck, M. (1997). Anxiety and cognition: A unified theory. Hove, UK: Psychology Press. Eysenck, M. (2000). A cognitive approach to trait anxiety. European Journal of Personality, 14, 463–476.
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99 Eysenck, M., Mogg, K., May, J., Richards, A., & Mathews, A. (1991). Bias in interpretation of ambiguous sentences related to threat in anxiety. Journal of Abnormal Psychology, 100, 144–150. Eysenck, M., Payne, S., & Santos, R. (2006). Anxiety and depression: Past, present, and future events. Cognition and Emotion, 20, 274–294. Fraley, R., & Waller, N. (1998). Adult attachment patterns: A test of the typological model. In J. Simpson & W. Rholes (Eds.), Attachment theory and close relationships (pp. 77–114). New York: Guilford Press. Fraley, R., Waller, N., & Brennan, K. (2000). An item-response theory analysis of selfreport measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365. Grossman, P., Niemann, L., Schmidt, S., & Walach, H. (2004). Mindfulness-based stress reduction and health benefits: A meta-analysis. Journal of Psychosomatic Research, 57, 35–43. Hazan, C., & Shaver, P. (1987). Romantic love conceptualised as an attachment process. Journal of personality and Social Psychology, 52, 511–524. Kabat Zinn, J. (1982). An out-patient program in behavioural medicine for chronic pain patients based on the practice of mindfulness meditation: Theoretical considerations and preliminary results. General Hospital Psychiatry, 4, 33–47. MacLeod, C. (1990). Mood disorders and cognition. In M. Eysenck (Ed.), Cognitive psychology: An international review (pp. 9–56). Chichester, UK: Wiley.
99
Mikulincer, M., & Arad, D. (1999). Attachment working models and cognitive openness in close relationships: A test of chronic and temporary accessibility effects. Journal of Personality and Social Psychology, 77, 710–725. Mikulincer, M., & Florian, V. (1998). The relationship between adult attachment styles and emotional and cognitive reactions to stressful events. In J. Simpson & W. Rholes (Eds.), Attachment theory and close relationships (pp. 143–165). New York: Guilford Press. Mikulincer, M., & Shaver, P. (2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. Advances in Experimental Social Psychology, 35, 53–152. Schachner, D., Shaver, P., & Mikulincer, M. (2005). Patterns of nonverbal behaviour and sensitivity in the context of attachment relationships. Journal of Nonverbal Behavior, 29, 141–169. Spielberger, C. (1983). State-trait anxiety inventory for adults. Redwood City, California: Mind Garden. Teasdale, J., Segal, Z., Williams, J. M. G., Ridgeway, V., Soulsby, J., & Lau, M. (2000). Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. Journal of Consulting and Clinical Psychology, 68, 615–625. Walach, H., Buchheld, N., Buttenmüller, V., Kleinknecht, N., & Schmidt, S. (2006). Measuring mindfulness – the Freiburg mindfulness inventory (FMI). Personality and Individual Differences, 40, 1543–1555.
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
Predicting individual differences in mindfulness: The role of trait anxiety, attachment anxiety and attentional control James J. Walsh *, Marc G. Balint, David R. Smolira SJ, Line Kamstrup Fredericksen, Stine Madsen School of Psychology, University of East London, Romford Road, London E15 4LZ, UK
a r t i c l e
i n f o
Article history: Received 8 February 2008 Received in revised form 4 September 2008 Accepted 10 September 2008 Available online 23 October 2008 Keywords: Mindfulness Trait anxiety Attachment anxiety Attentional control
a b s t r a c t Two correlational studies sought to identify possible predictors of individual differences in naturally occurring mindfulness. In study one, trait anxiety and attachment anxiety, but not attachment avoidance, were negatively predictive of mindfulness. In study two, trait anxiety ( ) and attentional control (+), but not openness or parental nurturance, predicted mindfulness. In addition, there was evidence of a partial mediation effect of attentional control on the association between trait anxiety and mindfulness. Key features of trait anxiety such as attentional and interpretative processing biases, as well as those of attachment anxiety such as rumination and hypersensitivity, are at odds with mindfulness characteristics such as attention to what is present coupled with an attitude of openness and acceptance. Thus, whether generalised or specific, anxiety appears to be antagonistic to mindfulness; control over one’s attentional resources may form part of the underlying explanation. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Originating in contemplative traditions such as Buddhism, mindfulness is defined as a state of enhanced attention to, and awareness of, what is taking place in the present (Brown & Ryan, 2003). Such awareness is characterised as open and receptive, but not judgemental (Bishop et al., 2004; Deikman, 1982). Mindfulness appears to be absent when attention is captured by rumination and fantasy (Brown & Ryan, 2003). Building on the original work of Kabat Zinn (1982), mindfulness-based interventions have proliferated over the past twenty years, mostly with very positive outcomes (Baer, 2003; Grossman, Niemann, Schmidt, & Walach, 2004). Underpinning most of these studies is an assumption that mindfulness training increases levels of mindfulness and such increases mediate the observed positive outcomes. However, attempts to validate this assumption have rarely been undertaken, the most likely reason being the absence of appropriate measures. It is only in the last few years that psychometrically valid measures of mindfulness have appeared in the literature (Baer, Smith, & Allen, 2004; Brown & Ryan, 2003; Walach, Buchheld, Buttenmüller, Kleinknecht, & Schmidt, 2006). In developing their measure, Brown and Ryan (2003) argued that mindfulness could be considered a ‘naturally occurring characteristic’ (p. 822) with both inter- and intra-personal variation. Similarly, Walach et al. (2006) argued that mindfulness could be expressed dispositionally and in state form, depending on the
* Corresponding author. E-mail address: [email protected] (J.J. Walsh). 0191-8869/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.paid.2008.09.008
time-frame in question. Given the importance of the construct in terms of physical and psychological well-being, Brown and Ryan (2003) emphasised the need to explore its antecedents. As an initial step in this direction, the two studies reported here were designed to investigate possible predictors of mindfulness. 2. Study 1 Whilst it is likely that individual differences in mindfulness will eventually prove to be multiply determined, a decision was made in the first instance to explore both the developmental and personality domains for factors that might possess predictive utility. Attachment and trait anxiety emerged as strong candidates from their respective fields. 2.1. Attachment Attachment processes developed in early childhood are believed to remain active throughout the lifespan (Bowlby, 1988). Current conceptualisations of attachment suggest a two-dimensional system, namely anxiety and avoidance (Brennan, Clark, & Shaver, 1998; Fraley & Waller, 1998). Low levels of anxiety and avoidance reflect secure attachment (Schachner, Shaver, & Mikulincer, 2005). Insecure attachment, however, can take one of two forms. If proximity-seeking is highly desired, then a ‘hyper-activating’ attachment strategy is adopted. This is characterised by intensive efforts to seek proximity and protection, hypersensitivity to signs of rejection and abandonment, and excessive rumination upon
95
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99
one’s personal deficiencies and threats to one’s relationships (Mikulincer & Florian, 1998). Collectively, these features constitute attachment anxiety (Mikulincer & Shaver, 2003). In contrast, where it is felt that proximity-seeking will not serve to reduce threat, a ‘deactivating’ attachment strategy is adopted. Here, one removes oneself from stimuli that are likely to activate the attachment system. This results in the avoidance of proximity-seeking, the denial of attachment needs and the suppression of signs of vulnerability. This strategy is known as attachment avoidance (Mikulincer & Shaver, 2003). The current study seeks to determine whether attachment anxiety and attachment avoidance predict mindfulness. According to Bishop et al. (2004), mindfulness involves attending to immediate experience and adopting an attitude of curiosity, openness and acceptance. It also includes a ‘decentred’ approach to one’s thoughts which serves to reduce cognitive elaboration and prevents rumination (Teasdale et al., 2000). Finally, instead of observing one’s experiences through various filters of beliefs and expectations, mindfulness involves direct observation of them. Thus, given the hypersensitivity to rejection and rumination on personal deficiencies associated with attachment anxiety, together with the thought suppression, relationship avoidance and person perception biases associated with attachment avoidance, it seems that these dimensions are essentially antithetical to many of the key features of mindfulness. Therefore, it is hypothesised that both attachment anxiety (H1) and attachment avoidance (H2) will be negatively predictive of mindfulness. Furthermore, since low scores on both attachment dimensions are associated with secure forms of attachment, the analysis will examine the interaction of the two main predictors to see if it accounts for additional amounts of criterion variance. 2.2. Trait anxiety Trait anxiety is closely related to one of the ‘big five’ personality factors, namely neuroticism; however, the underlying cognitive architecture of the former is probably better understood (Eysenck, 1992,1997). People high in trait anxiety are thought to possess both attentional and interpretative processing biases (MacLeod, 1990). Specifically, they are more likely than low anxious counterparts to detect threat, regardless of its source (Eysenck, 2000). Similarly, they are more likely to make threatening interpretations of ambiguous stimuli whereas low anxious counterparts are inclined to make neutral interpretations (Eysenck, Mogg, May, Richards, & Mathews, 1991). This bias towards detecting, interpreting and elaborating threat may be contrasted with mindfulness which advocates an open acceptance of what is present. Accordingly, it may be hypothesised that trait anxiety will be negatively predictive of mindfulness (H3).
3. Method 3.1. Design and measures A cross-sectional, correlational design was employed. Attachment-related anxiety, attachment-related avoidance and trait anxiety constituted the predictor variables. The criterion variable was mindfulness. Attachment-related anxiety and attachment-related avoidance were measured by the Experiences in Close Relationships-Revised (ECR-R; Fraley, Waller, & Brennan, 2000) questionnaire. The measure comprises 36-items and employs a six-point Likert scale. Higher scores signified more attachment-related anxiety (a = 0.92) and attachment-related avoidance (a = 0.92). Secure attachment patterns are reflected in low scores on both scales.
Trait anxiety, a ‘facet’ of neuroticism, was measured using an 8item scale from the revised NEO personality inventory (NEO-PI-R; Costa and McCrae (1992)). Respondents were asked to read the item statements and to rate their level of agreement with each using a five-point Likert scale. Higher scores indicated more trait anxiety (a = 0.76). Mindfulness was measured by the Mindful Attention Awareness Scale (MAAS) (Brown & Ryan, 2003). Although each of the 15 items signified the absence of mindfulness, the six-point Likert scale is arranged such that higher scores reflected more mindfulness (a = 0.85). 3.2. Participants Of the 334 questionnaires distributed to psychology students and staff, 127 usable replies were obtained (38%). The sample comprised 100 females and 23 males (four respondents failed to provide demographic details). Ages ranged from 17–60 years, with just over half (53%) of the participants being 30 years or younger. 3.3. Procedure Students and teaching staff were approached on campus and invited to participate in the study. Participants were asked to complete the questionnaire anonymously and return it either to a designated box on campus or post it to the researcher in an accompanying stamped self-addressed envelope. 4. Results The means and standard deviations for the variables, together with their inter-correlations, are presented in Table 1. As expected, mindfulness correlated negatively and significantly with all three hypothesised predictors. Next, mindfulness was regressed onto attachment anxiety, attachment avoidance, their interaction, and trait anxiety (see Table 2). Overall, the model accounted for 18% of the variance in mindfulness and proved to be reliable [F(4,122) = 6.57, p < 0.001]. Individually, attachment anxiety and trait anxiety were significant predictors, the latter emerging as slightly stronger. Neither attachment avoidance, nor its interaction with attachment anxiety, emerged as significant. 5. Discussion In line with the first hypothesis, adult attachment anxiety emerged as a significant independent predictor of individual differences in mindfulness. Examination of the attachment literature offers a solid conceptual basis for such a finding. Firstly, anxiously attached individuals are hypervigilant to threat-related cues, especially cues of rejection and abandonment, and ruminate extensively on distress-related material (Mikulincer & Florian, 1998). In contrast, mindful people do not deliberately monitor the environment for threat signals and, given their tendency not to elaborate incoming sensations and thoughts, they avoid rumination Table 1 Mean scores, standard deviations, and inter-correlations between variables in study 1 (N = 127)
1. 2. 3. 4.
Mindfulness Attachment anxiety Attachment avoidance Trait anxiety
*p 6 0.01; **p 6 0.001.
Mean
SD
1
3.87 2.71 2.68 3.09
0.81 0.96 0.95 0.72
–
0.32** 0.25* 0.33**
2
3
4
– 0.36** 0.31**
– 0.22*
–
96
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99
Table 2 Regression of mindfulness onto attachment anxiety, attachment avoidance, their interaction, and trait anxiety
Attachment anxiety Attachment avoidance Interaction Trait anxiety
B
SE(B)
Beta
0.18 0.11 0.05 0.27
0.08 0.08 0.06 0.10
0.21* 0.13 0.08 0.24**
*p < 0.05; **p < 0.01.
(Brown & Ryan, 2003). Secondly, while those who are anxiously attached resist information that disconfirms their expectations and fail to update existing schemas in light of contradictory evidence (Mikulincer & Arad, 1999), mindful people are much more open to new information and seek to avoid schema-congruent processing (Brown & Ryan, 2003). Finally, attachment anxiety is known to be associated with intolerance of ambiguity (Hazan & Shaver, 1987). In contrast, Brown and Ryan (2003) have shown that mindfulness is associated with greater cognitive flexibility and openness. In sum, the most notable features of mindfulness – attending to one’s immediate experience with an open and accepting attitude, avoiding the tendency to elaborate one’s thoughts and sensations, decentring from self-derogatory thoughts and not ruminating – contrast substantially with attachment anxiety. Attachment avoidance failed to predict mindfulness even though their zero-order correlation was significant. Whilst some features of attachment avoidance (thought suppression, defensive projection and cognitive closure) point towards a negative association with mindfulness, other features (inhibited processing of threat and the non-elaboration of a broad range of cognitions) may be positively related (Mikulincer & Shaver, 2003). This combination of attachment avoidance features appears to have differential, and potentially conflicting, predictive impact on mindfulness and may underlie the observed null result. Developing a multifaceted measure of attachment avoidance that contains independent sub-scales would facilitate a more precise testing of the hypothesised association with mindfulness. An alternative reason why attachment avoidance failed to emerge as significantly predictive may be due to its conceptual overlap with both attachment anxiety and trait anxiety. Trait anxiety emerged as the strongest predictor of mindfulness. According to Costa and McCrae (1992), trait anxiety represents a generalised tendency to be fearful, worried, and apprehensive about the future. Other conceptualisations of trait anxiety suggest the presence of danger- and threat-related schemas in long-term memory which guide information processing (Beck & Emery, 1985). The tendency to cognitively elaborate threat-related stimuli, coupled with schema-driven processing, conflicts with the key elements of mindfulness, namely, the allocation of attention to immediate stimuli and their non-elaboration (Bishop et al., 2004). An attempt will be made in Section 6 to explore more thoroughly this association between trait anxiety and mindfulness.
6. Study 2 Study 2 was designed partly to replicate the main finding from Section 2 concerning the key predictive role of trait anxiety. On this occasion, however, an alternative measure of mindfulness was employed, namely the Freiburg mindfulness inventory (FMI – Walach et al. (2006)). Brown and Ryan’s (2003) measure of mindfulness (MAAS) is known to be unidimensional whereas other measures of mindfulness are predicated upon a multidimensional conceptualisation (Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006). The FMI depicts mindfulness as a general construct, which incorporates
inter-related facets such as non-judgemental acceptance, mindful presence and insight. In a study examining the associations between five measures of mindfulness, the correlation between the MAAS and the FMI (r = .31) was the weakest (Baer et al., 2006). If trait anxiety once again proves to be predictive of mindfulness when an alternative, multidimensional, measure is employed, then this would confirm the findings from Section 2 and consolidate the underlying process-based link. As in Section 2, a negative association between trait anxiety and mindfulness is hypothesised (H1). It should be noted that as well as changing the measure of mindfulness, a different measure of trait anxiety was also used, namely Spielberger’s (1983) STAI-T. This is probably the most widely used measure of trait anxiety in the literature. Three more possible predictors of mindfulness were also examined – openness to experience, parental nurturance and attentional control. Openness to experience is believed to be a key dimension of individual differences and constitutes one of the big five personality factors (Costa & McCrae, 1992). People high in openness are attentive to inner feelings, intellectually curious, and independent in judgement. Low scorers, in contrast, are much more conservative in outlook and prefer the familiar to the novel. Since mindfulness is conceptualised as an open form of attention to, and awareness of, the present (Brown & Ryan, 2003), it seems reasonable to argue that openness would positively predict it (H2). Brown and Ryan (2003) measured this association in two student samples and found surprisingly low coefficients. However, as noted above, it may be that their measure of mindfulness was not broad enough to capture in full the hypothesised association between the two constructs. The use of Walach et al. (2006) multidimensional measure will provide a more comprehensive test of the hypothesis. Parental nurturance (Buri, 1989; Buri, Kirschner, & Walsh, 1987) is thought to be one of the key mechanisms through which individuals develop their attachment styles. In contrast to adult measures of attachment, it relates more explicitly to one’s formative years. It was theorised that experiencing a secure and supportive childhood via nurturing parents would facilitate a tendency to employ an open, explorative and accepting attitude towards the environment (mindfulness) in later life. Therefore, the third hypothesis predicts a positive association between parental nurturance and mindfulness (H3). While most theoretical conceptualisations of mindfulness highlight ‘attention’ as a core feature (Bishop et al., 2004; Brown & Ryan, 2003; Walach et al., 2006), they rarely specify which aspect of attention is implicated. Derryberry and Reed (2002) identified individual differences in attentional control. People high in attentional control are thought to be better at focusing attention, shifting attention and controlling thoughts in a flexible manner than low control counterparts. Such characteristics seem to overlap considerably with those of mindfulness. If so, one would expect to find a significant positive association between attentional control and mindfulness and this constitutes the fourth hypothesis (H4). When validating their measure, Derryberry and Reed (2002) found a correlation of r = 0.55 between trait anxiety and attentional control. The tendency for trait anxious participants to have poorer attentional control may underpin the negative association between trait anxiety and mindfulness observed in Section 2. This suggestion will be explored empirically in the current study. Specifically, it is hypothesised that as well as having a direct effect on mindfulness, attentional control also will mediate the association between trait anxiety and mindfulness (H5). 7. Method 7.1. Participants Undergraduate and graduate students (N = 153) participated in the study. Nurturance from just one parent was reported by 17 stu-
97
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99
dents and four others failed to complete the mindfulness measure; as a result, their data were removed from further analysis. Of the remaining 132 participants, 26 (20%) were male and 105 (79%) female with one respondent failing to indicate their sex. The mean age of participants was 25.9 years (SD = 6.7 years). 7.2. Design and measures
Table 4 Regression of mindfulness onto trait anxiety, openness, attentional control and parental nurturance (mother)
Trait anxiety Openness Attentional control Parental nurturance (Mo)
B
SE(B)
Beta
0.37 0.06 0.20 <0.01
0.07 0.06 0.10 0.04
0.46** 0.07 0.17* 0.001
A cross-sectional, correlational design was employed. Individual differences in mindfulness constituted the criterion variable whilst the predictors included openness, trait anxiety, parental nurturance (maternal and paternal) and attentional control. Each variable was measured as follows.
*p < 0.05; **p < 0.001. R2 = 0.33; Model F(4,127) = 15.39, p < 0.001.
7.3. Mindfulness
The means and standard deviations for the variables, together with their inter-correlations, are presented in Table 3. As predicted, mindfulness correlated positively and significantly with openness, attentional control and parental nurturance (mother), and negatively with trait anxiety. A multiple regression analysis was carried out to determine the predictive power of each of the significant correlates of mindfulness. Examination of the standardised regression coefficients indicates that trait anxiety was the strongest predictor of mindfulness although attentional control was also significant. Openness and parental nurturance (mother), on the other hand, failed to predict mindfulness independently (see Table 4). To test the mediating effect of attentional control on the relationship between trait anxiety and mindfulness, three regressions were carried out in line with Baron and Kenny (1986). In the final model, the mediator proved to be significant and the effects of trait anxiety decreased (beta was reduced from 0.54 to 0.46, a drop of almost 15%) but still remained significant. Thus, a partial mediation effect was noted. The amount of variance accounted for was R2 = 0.32, p < 0.001 (see Table 5).
Walach et al. (2006) Freiburg mindfulness inventory was used. This 14-item measure uses a four-point Likert scale with higher scores indicate greater levels of mindfulness (a = 0.77). 7.4. Trait anxiety The trait component of Spielberger’s (1983) state-trait anxiety inventory, comprising 20 items, was used. A four-point response scale was employed with higher scores indicating more trait anxiety (a = 0.90). 7.5. Openness Openness to experience was assessed using Costa and McCrae’s (1992) 12-item scale from the NEO-FFI (neuroticism, extraversion, openness – five factor inventory) – form S. A five-point Likert scale was used with larger scores indicating greater openness (a = 0.73).
8. Results
7.6. Parental nurturance Parental nurturance was assessed using Buri et al.’s (1987) 24item scale. Items describing positive and negative aspects of parental nurturance were rated on a five-point Likert scale on two occasions, once for mother/maternal carer and once for father/paternal carer. Higher scores indicated greater levels of parental nurturance (a = 0.96 (maternal) and 0.95 (paternal)). 7.7. Attentional control Derryberry and Reed’s (2002) 20 item measure was employed. The four-point Likert scale was scored such that larger scores signified greater attentional control (a = 0.83). 7.8. Procedure Questionnaires were completed anonymously during a research methods class in the following order: attentional control, openness, mindfulness, trait anxiety and parental nurturance.
9. Discussion Once again, trait anxiety emerged as the best predictor of mindfulness, thereby confirming the findings from study one as well as those of Baer et al. (2006) and Brown and Ryan (2003). People high in trait anxiety are believed to possess ‘danger’ schemata which influence information processing in a top–down manner (Beck & Emery, 1985). Being predisposed to attend to threat, and preferentially interpreting threat from ambiguous stimuli, requires the utilization of limited processing resources such that less is available for being openly attentive to, and aware of, current experience. Indeed, there is good evidence that anxiety seems to direct thinking towards future events (Eysenck, Payne, & Santos, 2006). The combination of a relative inability to maintain focus on the present, whilst simultaneously over-interpreting incoming stimuli, may explain why more anxious people are less mindful. It would be fascinating to see if mindfulness-based interventions could reduce trait anxiety and its associated processing biases over time.
Table 3 Mean scores, standard deviations, and inter-correlations between variables in study 2 (N = 132)
1. 2. 3. 4. 5. 6.
Mindfulness Trait anxiety Openness Attentional control Parental nurturance (Mo) Parental nurturance (Fa)
*p < 0.05; **p < 0.01; ***p 6 0.001.
Mean
SD
1
2.69 2.19 3.61 2.58 3.85 3.54
0.45 0.56 0.55 0.40 0.91 0.91
–
0.54*** 0.16* 0.39*** 0.16* 0.07
2
3
4
5
6
– 0.20** 0.07 0.09
– 0.19* 0.16*
– 48***
–
– 0.12 0.46*** 0.27*** 0.30***
98
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99
Table 5 Regression of mindfulness onto trait anxiety with and without the mediator (attentional control) Without mediator B Trait anxiety Attentional control
SE 0.44
–
0.06 –
With mediator Beta
t
0.54 –
7.36 –
Attentional control, incorporating the abilities to focus attention, avoid distraction and switch attentional focus, also predicted mindfulness. This is not surprising given that mindfulness requires a strong attentional focus on what is present together with a facility to switch easily to incoming, replacement, stimuli (Bishop et al., 2004). Using a direct measure of attentional control, the current results endorse those of Baer et al. (2006) who found that mindfulness was negatively associated with poor attentional control. In Derryberry and Reed’s (2002) initial research, as well as in the current study, trait anxiety was negatively associated with attentional control. Exploration of the mediating role of attentional control in the association between trait anxiety and mindfulness revealed a small but interesting effect. The combination of a predisposition to attend to threat on the one hand (high trait anxiety), and an inability to switch attention away from threat on the other (low attentional control), may prevent high anxious people from being mindful. It should be noted, of course, that trait anxiety still has a direct predictive association with mindfulness and so this indirect route via attentional control is only part of the explanation. However, attentional control does constitute a promising predictor of mindfulness alongside personality and developmental factors. Contrary to the second hypothesis, openness failed to predict mindfulness. The low zero-order correlation between the two constructs confirmed the findings of Brown and Ryan (2003). Thus, the facets of openness which the measure taps into were reexamined (Costa & McCrae, 1992). It was felt that openness to fantasy and openness to aesthetics did not depict the kind of openness that was conceived to be similar to mindfulness with its emphasis on current experience. Accordingly, the relevant items were removed from the overall openness score which was recalculated and correlated with mindfulness. The resultant coefficient reduced to 0.10 and became non-significant. It may be concluded that openness to experience and mindfulness are essentially different theoretical constructs, the conceptual breadth of the two rendering them incompatible. While openness constitutes a fairly broad dimension of personality, mindfulness is much narrower in scope and focuses mainly on attention to one’s immediate experience. Such differences in conceptual latitude may account for the very modest association detected. Parental nurturance failed to predict mindfulness. By encouraging participants to think about each statement as it applied to them ‘. . . during your years of growing up’, it may be that the instructions lacked temporal specificity in terms of early, middle or late childhood. Anecdotal reports from the participants confirmed this. Alternatively, it is likely that the temporal gap between childhood and adulthood would have provided participants with ample opportunity to influence the hypothesised association between variables. In essence, exposure to non-nurturing parents may serve to motivate an individual to search for change and to actively pursue psychological repair. The fact that participants had all chosen to study psychology suggests that they were interested in gaining insight into human (and by inference their own) behaviour and experience. It should also be noted that parental nurturance correlated significantly and negatively with trait anxiety and positively with attentional control; these latter two constructs accounted for significant amounts of variance in mindfulness in the regression model.
p
B
SE
Beta
t
p
<0.001 –
0.37 0.21
0.07 0.09
0.46 0.19
5.61 2.28
<0.001 =0.024
10. Conclusions and limitations In the two studies reported here, trait anxiety emerged as a reliable predictor of mindfulness irrespective of how either variable was measured. Key features of trait anxiety such as attentional and interpretative processing biases (Eysenck, 1992) are incompatible with what are believed to be the central components of mindfulness, namely, attention to one’s immediate experiences and an attitude of openness and acceptance (Bishop et al., 2004; Brown & Ryan, 2003; Walach et al., 2006). In addition, attachment anxiety is also negatively predictive of mindfulness. Again, examination of the mechanisms and processes underpinning hyper-activation and attachment anxiety such as effortful proximity-seeking, hypersensitivity and excessive rumination (Mikulincer & Shaver, 2003) provides a clear rationale for the observed association. Thus, regardless of the specificity with which it is assessed, anxiety appears to militate against the expression of mindfulness. While the lack of attentional control associated with anxiety (Derryberry & Reed, 2002) constitutes part of the explanation of its association with mindfulness, other factors are clearly implicated. However, the reliance of the current studies on self-report methods, crosssectional and correlational designs, limits the extent to which conclusions can be drawn about possible pathways between anxiety, attentional control and mindfulness.
References Baer, R. (2003). Mindfulness training as a clinical intervention: A conceptual and empirical review. Clinical Psychology: Science and Practice, 10, 125–143. Baer, R., Smith, G., & Allen, K. (2004). Assessment of mindfulness by self-report: The Kentucky inventory of mindfulness skills. Assessment, 11, 191–206. Baer, R., Smith, G., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using selfreport assessment methods to explore facets of mindfulness. Assessment, 13, 27–45. Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182. Beck, A., & Emery, G. (1985). Anxiety disorders and phobias: A cognitive perspective. New York: Basic Books. Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. D., Carmody, J., et al. (2004). Mindfulness: A proposed operational definition. Clinical Psychology: Science and Practice, 11, 230–241. Bowlby, J. (1988). A secure base: Clinical applications of attachment theory. London: Routledge. Brennan, K., Clark, C., & Shaver, P. (1998). Self-report measurement of adult attachment: An integrative overview. In J. Simpson & W. Rholes (Eds.), Attachment theory and close relationships (pp. 46–76). New York: Guilford Press. Brown, K., & Ryan, R. (2003). The benefits of being present: Mindfulness and its role in psychological well-being. Journal of Personality and Social Psychology, 84, 822–848. Buri, J. (1989). Self-esteem and appraisals of parental behaviour. Journal of Adolescent Behavior, 4, 33–49. Buri, J. R., Kirschner, P. A., & Walsh, J. M. (1987). Familial correlates of self-esteem in American young adults. Journal of Social Psychology, 127, 583–588. Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-R) and NEO five factor inventory (NEO-FFI): Professional manual. Odessa, FL: Psychological Assessment Resources. Deikman, A. J. (1982). The observing self. Boston: Beacon Press. Derryberry, D., & Reed, M. (2002). Anxiety-related attentional biases and their regulation by attentional control. Journal of Abnormal Psychology, 111, 225–236. Eysenck, M. (1992). Anxiety: The cognitive perspective. Hove, UK: Lawrence Erlbaum. Eysenck, M. (1997). Anxiety and cognition: A unified theory. Hove, UK: Psychology Press. Eysenck, M. (2000). A cognitive approach to trait anxiety. European Journal of Personality, 14, 463–476.
J.J. Walsh et al. / Personality and Individual Differences 46 (2009) 94–99 Eysenck, M., Mogg, K., May, J., Richards, A., & Mathews, A. (1991). Bias in interpretation of ambiguous sentences related to threat in anxiety. Journal of Abnormal Psychology, 100, 144–150. Eysenck, M., Payne, S., & Santos, R. (2006). Anxiety and depression: Past, present, and future events. Cognition and Emotion, 20, 274–294. Fraley, R., & Waller, N. (1998). Adult attachment patterns: A test of the typological model. In J. Simpson & W. Rholes (Eds.), Attachment theory and close relationships (pp. 77–114). New York: Guilford Press. Fraley, R., Waller, N., & Brennan, K. (2000). An item-response theory analysis of selfreport measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365. Grossman, P., Niemann, L., Schmidt, S., & Walach, H. (2004). Mindfulness-based stress reduction and health benefits: A meta-analysis. Journal of Psychosomatic Research, 57, 35–43. Hazan, C., & Shaver, P. (1987). Romantic love conceptualised as an attachment process. Journal of personality and Social Psychology, 52, 511–524. Kabat Zinn, J. (1982). An out-patient program in behavioural medicine for chronic pain patients based on the practice of mindfulness meditation: Theoretical considerations and preliminary results. General Hospital Psychiatry, 4, 33–47. MacLeod, C. (1990). Mood disorders and cognition. In M. Eysenck (Ed.), Cognitive psychology: An international review (pp. 9–56). Chichester, UK: Wiley.
99
Mikulincer, M., & Arad, D. (1999). Attachment working models and cognitive openness in close relationships: A test of chronic and temporary accessibility effects. Journal of Personality and Social Psychology, 77, 710–725. Mikulincer, M., & Florian, V. (1998). The relationship between adult attachment styles and emotional and cognitive reactions to stressful events. In J. Simpson & W. Rholes (Eds.), Attachment theory and close relationships (pp. 143–165). New York: Guilford Press. Mikulincer, M., & Shaver, P. (2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. Advances in Experimental Social Psychology, 35, 53–152. Schachner, D., Shaver, P., & Mikulincer, M. (2005). Patterns of nonverbal behaviour and sensitivity in the context of attachment relationships. Journal of Nonverbal Behavior, 29, 141–169. Spielberger, C. (1983). State-trait anxiety inventory for adults. Redwood City, California: Mind Garden. Teasdale, J., Segal, Z., Williams, J. M. G., Ridgeway, V., Soulsby, J., & Lau, M. (2000). Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. Journal of Consulting and Clinical Psychology, 68, 615–625. Walach, H., Buchheld, N., Buttenmüller, V., Kleinknecht, N., & Schmidt, S. (2006). Measuring mindfulness – the Freiburg mindfulness inventory (FMI). Personality and Individual Differences, 40, 1543–1555.