Mindfulness meditation and implicit bias

A recent study purports to demonstrate that mindfulness meditation techniques can reduce implicit biases. Affecting all manner of interpersonal interactions, implicit biases are unconscious attitudes or associations that influence our understanding, behavior and decisions. Implicit biases can be revealed using implicit association tests (IATs), which often measure the degree to which a participant associates particular stimuli (e.g. white or black faces) with negative and positive words by recording the speed and accuracy with which words are categorized when presented alongside ‘congruous’ or ‘incongruous’ stimuli. For example, white people are better at categorizing positive words when they presented alongside white, rather than black faces and better at categorizing negative words when they are presented alongside black, rather than white faces. Crucially, these implicit biases often do not correspond to participants’ explicit, reported attitudes to racial or other demographic minorities: even the most fervent (white, young) egalitarian can display implicit bias against black or elderly faces.

In their study, Lueke and Gibson sought to investigate whether mindfulness meditation could reduce implicit associations. Mindfulness, according to them, ‘focuses the individual on the present and encourages practitioners to view thoughts and feelings nonjudgmentally as mental events, rather than as part of the self. This allows the individual to understand and reflect on these events as transient moments that are separate from the self, which inhibits the natural tendency toward reaction and automatic evaluation.’ In particular, Lueke and Gibson wanted to isolate the effect of mindfulness on ‘automatic activation’ – ‘the likelihood that an association or evaluation is automatically activated when a stimulus object is encountered’ as distinct from its effect on ‘overcoming bias’ – ‘the likelihood that an initially activated association can be overcome and replaced by a correct response’ – amongst other parameters. Such a distinction suggests that their hypothesis pertained specifically to the power that mindfulness might have to forestall or remove the automatic association itself, rather than any power it might confer on the agent to exert control over the behavioral influence of automatic associations.

To conduct their study, Lueke and Gibson used a sample of young, white participants, and randomly assigned them to either the experimental condition or the control condition. Those in the experimental condition received a ten-minute mindfulness recording, which instructed the participants to ‘become aware of bodily sensations (heartbeat and breath) and fully accept these sensations and any thoughts without restriction, resistance, or judgment’. Those in the control condition listened to a recording discussing natural history. All participants then completed an IAT for race (white/back) and an IAT for age (young/old). The results showed that those who had listened to the mindfulness recording showed less implicit race and age bias than the control group. Crucially, whilst the analytical model revealed a significant difference between the two groups on ‘automatic activation’, there was no significant difference between the groups on ‘overcoming bias’. This suggests, according to the authors, that mindfulness does something to prevent the automatic associations from being activated rather than helping participants to control activated associations.

Reduction in these sorts of implicit bias is undoubtedly a good thing. However, the precise way in which this is achieved will be of philosophical interest. The authors acknowledge that their study does not make transparent the mechanism underlying the reduced bias effect. They speculate based on their understanding of what mindfulness achieves, particularly in relation to its effects on an agent’s awareness:

… [T]he novice that briefly undergoes meditation is transformed into a state of awareness of sensations and thoughts and nonjudgmental acceptance of those sensations and thoughts. However, this brief meditation is likely to dissipate […]. In contrast, the experienced practitioner not only engages in meditation more often, which allows nonjudgmental awareness to be experienced more often, but through this consistency creates a new default state of being—an awareness that permeates greater aspects of the self and of everyday experience. In this way, a deeper mindful experience can be cultivated, widening the area of awareness that the individual can attend to. This consistent and widened awareness likely has stronger effects on implicit attitudes and accompanying behavior.

However, it is not clear quite what is meant by ‘awareness’ here. Is the supposition that the mindful agent becomes more aware of his implicit attitudes? This reading would be in tension with the authors’ finding that mindfulness prevented automatic associations from being activated, particularly if it is the case that awareness of cognitive feature x requires at least some manifestation of cognitive feature x. Further complicating the picture, the references to ‘nonjudgmental acceptance’ of sensations and thoughts appears to sit uneasily alongside the meditation-mediated suppression of biased associations.

Jules Holroyd has recently written on the relationships between implicit bias, awareness and moral culpability. Her work suggests that much could be riding on the nature of the awareness that mindfulness precipitates (if it is indeed some type of awareness that does the work to reduce implicit bias in this study). She distinguishes between three types of awareness relating to implicit bias: 1) introspective awareness of one’s (biasing) cognitive processes, 2) inferential awareness of one’s (likelihood of manifesting) bias given the empirical data on the phenomenon, and 3) observational awareness of the effects of implicit bias on one’s behavior – e.g. awareness of one’s differential treatment of those belonging to an ‘out-group’. Holroyd examines the extent to which failures of awareness are culpable in relation to implicit bias and, crucially, the discriminatory behaviors they precipitate. She argues that, for different reasons, neither introspective awareness nor inferential awareness of implicit bias is a condition for moral responsibility. Introspective awareness is usually not possible and knowledge of psychological research is not something we can expect most people to acquire. Accordingly, it is not the case that agents should be introspectively or inferentially aware of implicit bias. However, failures of awareness long these lines, she argues, will not serve to exculpate the agent who manifests bias-begotten discrimination. Rather, Holroyd argues that a lack of observational awareness of the effects of implicit bias can sometimes render an agent blameworthy under particular conditions; for example, when an agent fails to pay attention to her differential treatment of others or when she deceives herself into believing she does not manifest discriminatory behavior.

From this (regrettably brief – do go and read Holroyd’s fascinating paper) sketch, we can begin to understand why the kind of awareness that mindfulness yields might be morally relevant as well as empirically and conceptually interesting. In particular, if the awareness that Lueke and Gibson invoke is akin or related to the introspective type of awareness distinguished by Holroyd, this might have implications for one of her assumptions:  ‘implicit processes in general do not appear to be “operationally transparent” to us, such that it is not possible to have introspective awareness of their operation.’ If it were possible to have introspective awareness of implicit associations then our failures of awareness in this regard might become blameworthy under certain circumstances.

I personally do not know much about mindfulness, although I hear enthusiastic reports from its practitioners. My intuition is that Lueke and Gibson are not suggesting that practicing mindfulness meditation makes implicit biases operationally transparent, not least because such an effect would be more likely to reduce scores on the IAT through the ‘overcoming bias’ mechanism, whereby control is exerted over the influence of associations on behavior. On the contrary, their suggestion is that the awareness that mindfulness facilitates reduces scores on the IAT through deactivating the automatic biases in the first place. So, I am left wondering what role awareness is really doing here, if anything. One possible explanation of their finding is that mindfulness serves to calm and compose the practitioner, such that emotions – and hence positive and negative emotional responses – are subdued. Subdued emotions perhaps then lead to a more neutral processing of stimuli. Accordingly, an alternative reading of Lueke and Gibson’s analysis is that the awareness they invoke should be understood not as an awareness of (thoughts or associations) but instead as a calm state of mind, which modifies the agent’s style of cognitive processing. On such a reading, awareness is connected less to knowledge and beliefs and more to the achievement of a particular brain state. Whatever the correct interpretation, this study suggests that psychologists should investigate further the mechanisms at work in reducing implicit bias and, crucially, the extent to which effects transfer to ecologically valid situations.


Holroyd, J. (2014). Implicit bias, awareness and imperfect cognitions. Consciousness and cognition, Online First

Lueke, A., & Gibson, B. (2014). Mindfulness Meditation Reduces Implicit Age and Race Bias The Role of Reduced Automaticity of Responding. Social Psychological and Personality Science, Online First.

  • Facebook
  • Twitter
  • Reddit

Comments are closed.


Subscribe Via Email