LETTER TO THE EDITOR Reply to Burchett, T.S. and Glenn, L.L. ‘Measurement validity of tests for implicit negative bias’ Sylvia Terbeck & Guy Kahane & Sarah McTavish & Julian Savulescu & Phil Cowen & Miles Hewstone Received: 31 May 2012 / Accepted: 31 May 2012 / Published online: 15 June 2012 # Springer-Verlag 2012 The letter by Burchett and Glenn raises several issues about our study (Terbeck et al. 2012). Their main concern relates to the implicit association test (IAT), the instrument we used to measure implicit racial bias. They also question what they refer to as our ‘claim for clinical treatment due to amygdala suppression’ (p. 2). Burchett and Glenn’ s doubts about the IAT do not, in fact, relate specifically to our study, but to a standard method for measuring implicit bias that has been used by numerous studies for over a decade (Cunningham et al. 2001; Greenwald et al. 2009; Nosek et al. 2007). They assert, with little argu- ment, that the IAT has not been sufficiently validated and that it possesses negligible measurement reliability, citing only one critical paper and ignoring the multiple citations we provide in our study, which themselves summarise and analyse numer- ous further studies. As we noted in our study, there is a considerable literature supporting the reliability and validity of the IAT, including the reviews by Greenwald et al. (2009) and Nosek et al. (2007). Cunningham et al. (2001) had earlier reported consistency and convergent validity specifically for the racial IAT measure. IAT scores have been shown to predict subtle but significant differences in behaviour towards out- group members, such as the frequency of positive versus negative words being used in a conversation (see Nosek et al. 2007). We thus find it hard to understand why Burchett and Glenn claim that the IAT ‘has been used prematurely in research’ (p. 2), let alone why they think that it attempts to ‘measure the immeasurable’. While research and debate on the IAT will continue without a doubt, as is often the case for a widely used measure, a large and growing literature strongly suggests that racial and other forms of implicit bias are a real phenomenon, which may have significant effects on everyday behaviour. Research into implicit bias towards out-group members is not merely of theoretical interest, but is potentially of great practical and ethical significance. Our study’ s methodology and analysis closely follow the established procedure described in Greenwald et al. (2009), including the criteria used for exclusion of data, which are in line with those used for many other psychological measures. In addition, both groups of participants were in the same experimental setting (sitting in the experimental room with- out the experimenter) while completing the IAT. It is thus hard to see what irrelevant situational influences could have played a role in generating the effect we reported. Burchett and Glenn’ s last point of criticism relates to our suggestion that the reported effect of propranolol on IAT response latencies may have been due to inhibition of amygdala activation. We are puzzled by this criticism. As we made clear in our paper, our main finding is that norad- renergic pathways appear to play a causal role in implicit bias. As we explicitly remark, further research is needed to clarify the exact causal mechanisms involved. Based on considerable research showing both that IAT responses are correlated with amygdala activity (e.g. Phelps et al. 2000) and that propranolol reduces such activation (Chamberlain et al. 2006; Hurlemann et al. 2010; van Stegeren et al. 2005), we proposed precisely such inhibition of amygdala activation as one possible explanation of our finding. This proposal is thus largely consistent with available evidence, though, as we emphasised, further research is required to test it directly. In our paper, we point out that some evidence seems in tension with this proposal, such as the study of Phelps et al. (2000) that found an intact IAT effect in one patient with S. Terbeck (*) : G. Kahane : S. McTavish : J. Savulescu : P. Cowen : M. Hewstone University of Oxford, Oxford, UK e-mail: sylvia.terbeck@psy.ox.ac.uk Psychopharmacology (2012) 222:723–724 DOI 10.1007/s00213-012-2768-z