I’ve written a few times over the past year about the numerous biases that counter our abilities to behave in a rational and sensible manner in the workplace.  The notable thing about many of these is that they are largely implicit.  In other words, they’re something that we do, often without thinking about it.  This unconscious level of behaviour makes these biases both fascinating but also incredibly dangerous.

Central to the field is the implicit association test (IAT), which is a simple test that can be completed online via the Project Implicit website.  It aims to record the speed of your responses when sorting various targets into different categories, such as good or bad.  These targets could, for instance, be a black or white face.  The test has proved reliable in highlighting prejudices even in those who firmly believe that they have none.

A new study has set out to delve deeper into the various forms of implicit attitudes that are common throughout the workplace, and perhaps more importantly, how one can go about changing them.  The study, undertaken online via the IAT, saw between 3-400 people randomly assigned to various interventions that hoped to change these implicit biases.  There were 17 interventions in total, of which 9 appeared to have some positive results.

Interestingly, the interventions that fared worst attempted to change the underlying attitude of participants.  Interventions such as ‘instilling a sense of common humanity’, ‘training empathetic responding’, encouraging taking the perspective of the outgroup or imagining positive interracial contact all seemed not to work.

So, which interventions did appear to work in regards to shifting behaviours?  The paper suggests that the best methods involved a form of priming or training.  For instance, interventions such as ‘faking the IAT’ achieved good results.    Whilst this is rather sad on one hand in that it suggests the test can be gamed in some way, it also provides a glimmer of hope by suggesting that positive interventions all required an in depth knowledge of the IAT.

Unfortunately, the paper didn’t run for long enough to determine whether the changes in behaviour were sustained over a long period of time, nor in fact whether the impact of the interventions would have changed with a longer exposure to them.  After all, the whole point is to shift those underlying attitudes that create the biases in the first place rather than just our response strategies.  It suggests that more work is needed to really get to the bottom of this prickly issue.