Wednesday, 2 October 2013

Implicit Bias, Moral Assessment and Awareness

Chloë FitzGerald
I am a postdoctoral fellow at the Institute of Biomedical Ethics, University of Geneva, currently working on a project about implicit bias in clinical care.  My background is in philosophy and I work in areas at the intersection of moral psychology, philosophy of mind, ethics and bioethics.


As Natalia and Jules have explained in previous posts, talk of implicit bias in social psychology typically refers to implicit associations between particular social groups, such as minority or oppressed racial groups, the obese, the elderly, women, transgender individuals etc., and negative evaluations or characteristics.  Much remains to be understood about how these associations work.  Psychologists may also distinguish between implicit stereotypes and implicit prejudices/attitudes, where implicit stereotypes refer to sets of belief-like states and implicit prejudice or implicit attitudes are ‘hotter’ states, more like preferences.  This use of ‘attitude’ can be confusing for philosophers, who tend to use attitude to refer to a wider range of mental states, including beliefs.

So much for differences between attitudes, bias, stereotype or prejudice, what about the ‘implicit’ part?  Implicit processes are often equated with unconscious processes or uncontrollable processes. It is easier to give a negative definition of an implicit process than a positive one.   One way is to contrast implicit processes with explicit processes, which are characterised as products of intention, consciously available, controlled and requiring mental resources (Nosek & Riskind, 2012).  Often it is unclear what exactly researchers mean by implicit and this is because ‘implicitness’ is a concept similar to that of  ‘automaticity’, used to cover a range of features, such as the consciousness, controllability, intentionality, efficiency and speed of processes.  These features may admit of degrees and do not necessarily co-occur so should ideally be investigated separately (de Houwer & Moors, 2007).  There is still much disagreement about how these features relate to one another and the exact relation between explicit and implicit processes.
Implicit attitudes fascinate me because they probe us to reflect on what we prioritise in our moral assessments of others.  For a long time, I’ve been bothered by the fact that philosophers suffer from a tendency to concentrate on whether a person has the correct moral principles or theories rather than on how she actually behaves towards others.  Let’s say that we have two people of high socio-economic status (SES):  Devina, who espouses egalitarian moral and political views that I find admirable and correct, yet tends to treat people of lower SES significantly worse than those of a higher status in subtle ways in her personal interactions;  and Delial, who holds reactionary views about the inferiority of those who are of lower SES than her, but nevertheless treats each low SES individual she meets with the same high level of respect, sympathy and sensitivity.  Who should I consider a better person overall, Devina or Delial?  I would be inclined to judge Delial more favourably and would certainly prefer to have her as a friend than Devina.
One way of understanding these examples would be to say that Devina has explicit egalitarian views, but that she has negative implicit attitudes (or bias) towards those of a lower SES that influence her behaviour towards those people.  Delial, on the other hand, could be described as having negative explicit attitudes towards those of lower SES, while holding no negative implicit attitudes towards this group (or even positive implicit attitudes?).  Or we might think that Delial does indeed have implicit biases towards those of a lower SES that correlate with her explicit attitudes, but that they are not manifested in her behaviour.  Perhaps when she encounters someone and interacts with them, she doesn't focus on that person’s SES  and instead treats each person as an individual so her implicit biases are not ‘activated’.  Which of these scenarios is correct matters for our moral assessment of Delial and this highlights how important it is to understand how implicit attitudes work empirically to grasp their normative significance.


Jules and Natalia both mentioned that lack of awareness of implicit biases is sometimes thought to mean that a person cannot be held responsible for having them or manifesting them.  I'm interested in the idea that different individuals might have different levels of awareness of their implicit attitudes and that this is, in part, a matter of training and cultivation.  Thus, it might be the case that we have a duty to become more aware of our implicit attitudes even if we are currently not very attuned to them.  In a forthcoming paper in Bioethics (2014), I argue that having a well-functioning conscience requires that one cultivate awareness of one's implicit attitudes to ensure one is actually living according to one's professed moral values.  

The psychologist Elizabeth Norman employs the concept of 'fringe consciousness' in her work on implicit learning, arguing that we have different gradations of awareness of our mental states.  She has used an Openness to Feelings scale to measure individual differences and to see if these correlate with the intuitive feelings that subjects report in implicit learning situations (Norman et al. 2006, 2007).  This individual variety interests me because it suggests that we could learn to become more aware of states at the 'fringes' of our consciousness, or implicit states.  I hope to work more on this in the future.

No comments:

Post a Comment

Comments are moderated.