An error in the analysis of "An eye for the I".

Reports an error in "An eye for the I: Preferential attention to the eyes of ingroup members" by Kerry Kawakami, Amanda Williams, David Sidhu, Becky L. Choma, Rosa Rodriguez-Bailón, Elena Cañadas, Derek Chung and Kurt Hugenberg (, 2014[Jul], Vol 107[1], 1-20). Kawakami and colleagues (Kawakami et al., 2014) conducted a series of studies in which participants saw pairs of faces (e.g., one Black and one White, or one ingroup and one outgroup) while eye tracking data were recorded. Because two faces were presented simultaneously, participants were allowed to choose how to allocate attention between the two faces. Often, they attended to one type of face more the other, creating a difference in base rates. The original analysis suggested that (a) preferential attention to the eyes was more pronounced for same-race faces than for cross-race faces, (b) preferential attention to the eyes was more pronounced for a novel ingroup than a novel outgroup, (c) preferential attention to the eyes was responsive to manipulations of motivation, and (d) preferential attention to the eyes was related to subsequent recognition accuracy. Although the first conclusion seems justified, the last three conclusions are either incorrect or unjustified based on the analyses reported in the article. When we correct for base rates, we see that most of the effects disappear. At the date of this writing, Kawakami and colleagues' article has 64 citations in Google scholar, including several that base their arguments on analyses we have shown to be incorrect (e.g., Xiao, Coppin, & Van Bavel, 2016). It is important for the field to acknowledge the error in these analyses and ensure that future theoretical and empirical work rests on a solid foundation. It is also important for researchers to understand the issues that gave rise to this mistake and ensure that, in future work, methods avoid this kind of confound or analyses appropriately adjust for it. (The following abstract of the original article appeared in record 2014-25638-001.) Human faces, and more specifically the eyes, play a crucial role in social and nonverbal communication because they signal valuable information about others. It is therefore surprising that few studies have investigated the impact of intergroup contexts and motivations on attention to the eyes of ingroup and outgroup members. Four experiments investigated differences in eye gaze to racial and novel ingroups using eye tracker technology. Whereas Studies 1 and 3 demonstrated that White participants attended more to the eyes of White compared to Black targets, Study 2 showed a similar pattern of attention to the eyes of novel ingroup and outgroup faces. Studies 3 and 4 also provided new evidence that eye gaze is flexible and can be meaningfully influenced by current motivations. Specifically, instructions to individuate specific social categories increased attention to the eyes of target group members. Furthermore, the latter experiments demonstrated that preferential attention to the eyes of ingroup members predicted important intergroup biases such as recognition of ingroup over outgroup faces (i.e., the own-race bias; Study 3) and willingness to interact with outgroup members (Study 4). The implication of these findings for general theorizing on face perception, individuation processes, and intergroup relations are discussed. (PsycInfo Database Record (c) 2020 APA, all rights reserved).