Implicit Racial Stereotypes may Temporarily Shift after Reading a Story

Implicit Racial Stereotypes may Temporarily Shift after Reading a Story Maddalena Marini (maddalena.marini@unimore.it) Dipartimento di Comunicazione e Economia, Viale Allegri 9 42100 Reggio Emilia, ITALY Sandro Rubichi (sandro.rubichi@unimore.it) Dipartimento di Comunicazione e Economia, Viale Allegri 9 42100 Reggio Emilia, ITALY Giuseppe Sartori (giuseppe.sartori@unipd.it) Dipartimento di Psicologia, Via Venezia 8 35100 Padova, ITALY Abstract This study demonstrated that implicit preferences for Whites compared to Blacks can be temporarily (i.e., less than a week) reduced or reversed by reading a counter- stereotypical scenario about an assault. Explicit racial associations (i.e., White = Bad, Black = Good) appear to be not a crucial component of the scenario for shifting the bias but seem to play a role in increasing the malleability effect of implicit racial attitudes. When they were not included in the scenario, a reduction of the implicit preferences for Whites compared to Blacks (Experiment 2) was observed, whereas when they were included we found a reverse of the implicit racial bias indicating a preference for Blacks than to Whites (Experiment 1). Keywords: Implicit Association Test; racial bias; malleability; explicit racial associations; persistence over time. Introduction One of the most used implicit measure of social cognition is the Implicit Association Test (IAT; Greenwald, McGhee & Schwartz, 1998; Nosek, Greenwald & Banaji, 2007; Nosek, Smyth, Hansen et al., 2007). In a classic IAT paradigm (Greenwald et al., 1998), participants are asked to classify four types of stimuli by pressing one of two keys. In particular, they are asked to discriminate stimuli belonging to two target categories and stimuli belonging to two evaluative dimensions. Faster responses are made when the interest category shares the same motor response with the stereotypic evaluative dimension (congruent condition) than when it shares the same motor response with the non-stereotypic evaluative dimension (incongruent condition). Originally, implicit measures were assumed to be inflexible and resistant to change and insensitive to external pressures and strategic processes (Bargh, 1999; Dovidio & Fazio, 1992; Fazio, Jackson, Dunton, & William, 1995). In contrast to this assumption, recent research has shown that the implicit measures are highly sensitive to contextual influences (for a review, see Blair, 2002; Dasgupta & Asgari, 2004; Mitchell, Nosek & Banaji, 2003). Multiple interventions were found to elicit shifts of implicit biases as measured by IAT (e.g., Dasgupta & Greenwald, 2001; Blair, Ma, & Lenton, 2001; Foroni & Mayr, 2005). Foroni and Mayr (2005), for example, have demonstrated that reading a counter-stereotypical scenario produces a reduction in implicit preferences for flowers compared to insects. In their scenario, participants were asked to imagine a fictional post- apocalyptic world in which, due to radiation effects, flowers have become highly noxious, whereas insects serve as a crucial, indirect food resource for humans. Moreover, explicit counter-stereotypical associations between the two categories of interest were included. Insects were positive and flowers were negative, contrary to the normal established association flowers are good and insects are bad (Greenwald et al., 1998). Even though it has been demonstrated that reading a scenario influences evaluations in the IAT, the underlying mechanisms of this malleability effect are not sufficiently well understood. Foroni and Mayr (2005) suggested that shift of the implicit preferences can be attributed by embedding the relevant categories (e.g., in their case flowers and insects) within a novel and coherent conceptual network that entails a newly knowledge structure and a revaluation of the categories. We conducted the present research to investigate the underling mechanisms and the persistence over time of the malleability of implicit attitudes as measured by the IAT, focusing on the Foroni and Mayr’s scenario paradigm (2005). Specifically, we investigated whether implicit racial stereotypes could be shifted after reading a counter-stereotypical scenario and whether this change could be detected after one week. Experiment 1 In Experiment 1, according to Foroni and Mayr (2005), we used a counter-stereotypical scenario in which the to- be-evaluated categories (i.e., in our case White and Black) were embedded within a meaningful conceptual network that entails a revaluation of the categories. In addition, explicit associations were included. Specifically, we used a pro-black scenario describing an assault, in which a positive action was assigned to a Black individual (i.e., rescuer role) and a negative action to a White individual (i.e., aggressor role). Method