EASP Seedcorn Grant Report by Prof Iris Žeželj and Dr Goran Knežević
10.10.2017, by Sibylle Classen in grant report
University of Belgrade, Serbia; Project: Prejudice Against the Roma in Serbia: An Analysis of Basic Personality Traits and Socio-Political Attitudes at the Implicit Level of Prejudice
The EASP seedcorn grant enabled us to investigate the relation of personality to the following socio-political attitudes: prejudice towards ethnic minorities (PTT), right wing authoritarianism (RWA), and social dominance orientation (SDO). But first of all, we would like to thank the EASP for this valuable support that encouraged us to conduct the research that will be described in the continuation of the text. The research was done in collaboration with Professor Johannes Keller, University of Ulm, Germany.
In this study we built on the work of Ekehammar, Akrami, Gylje, and Zakrisson (2004), but also on our previous findings concerning the relations between basic personality traits and socio-political attitudes (Keller & Knežević, 2016). The specificity and novelty of our approach is that the basic personality space is complemented by a trait - psychosis-proneness - with a potential to further contribute to the understanding of socio-political attitudes, above and beyond the well-established Big Five traits. Indeed, it was found recently that psychosis-proneness operationalized as personality trait (termed Disintegration) is significantly related to RWA and ethnic prejudice, even when the Big Five personality traits are taken into account (Keller & Knežević, 2016). It was shown that the relation is partly direct and partly mediated via RWA (but not via SDO) The second specificity of our approach is the ethnic group – Roma. This is the minority group which is probably most strongly confronted with negative stereotypes and prejudice in South-Eastern European countries (cf. Urh, 2013). However, systematic research addressing the bases of the particularly fierce and very widely spread negative stereotypes and prejudice against Roma in South-Eastern Europe is largely lacking so far. Accordingly, it is high time to devote more scientific efforts to this pressing social issue.
Our main research question was whether the incremental explanatory value of Disintegration over and above the Big Five personality traits can also be documented at the implicit level of measurement of anti-Roma prejudice. The dual process model of prejudice and its relation to personality was put to a test, with two important additions: (a) Disintegration was added to the Big Five traits (as it was the case in our previous work; Keller & Knežević, 2016) and (b) the explicit measure of anti-Roma prejudice was complemented by an implicit measure (as in, for example, Žeželj, Jakšić & Jošić, 2015). The underlying rationale for our research was that the predictive validity of Disintegration incremental to the Big Five traits - when an implicit measure of prejudice is the dependent variable - would testify to the importance of Disintegration not only concerning those aspects of prejudice of which individuals are well aware, but also to the aspects which are less conscious and less controlled. Another important question that our study was designed to answer was whether the model relating personality traits. RWA SDO and prejudice (that was established with Latino people and African Americans in the USA. as well as Turkish immigrants in Germany as targets of prejudice) also holds for prejudice against Roma people in Serbia.
Thinking about the Disintegration – prejudice relations we came to the conclusion that an additional type of evidence would be very meaningful to obtain in order to come to a more definite conclusion concerning the causal nature of these relations: An experimental approach based on manipulating a basic component of Disintegration (perceptual distortion). To address this issue, we decided to make use of the so-called out-of-body illusion paradigm to elicit a perceptual distortion experience and to test the influence of this experience on prejudice. Thus, we invested parts of the seedcorn grant money in the equipment necessary to realize this crucial part of the evidence on Disintegration – prejudice relations.
Study 1: Relations between personality and an implicit measure of prejudice
Data were collected in Serbia, comprising 154 participants. Eighty-four students from the Faculty of Philosophy and 70 students from the Faculty of Sport. University of Belgrade, (81 of them women); Mage (SD) = 20.06 (2.23). NEO-FFI, a 60 item measure (Costa & McCrae, 1992) of FFM traits was used to assess Big Five traits (N, E, O, A, C). DELTA (Knezevic et al., 2017), a short measure (20-items) was used to assess Disintegration. RWA (Altemeyer, 1996; Rangel & Keller, 2011), a short. 12-item measure was used to assess Right-Wing Authoritarianism. PTT (Pettigrew & Meertens, 1995), a 20-item measure was used to capture prejudices towards minorities. SDO (Pratto et al., 1994; Rangel & Keller, 2011), a short, 8-item questionnaire was administered to assess Social Dominance Orientation. Implicit Anti-Roma prejudice measure: Implicit stereotypes were measured with the Sorting Paired Features Task (SPF; Bar-Anan, Nosek, & Vianello, 2009) This measure builds on the well-established finding that the processing of a stimulus activates associated concepts (Higgins, 1996) and that an available association facilitates performance (i.e., speed of processing) resulting in reduced reaction times.
Disintegration proves to be related not only to an explicit measure of prejudice, but to the implicit one also. Correlational analysis singles out Disintegration as the most important correlate of the implicit measure (SPF) of prejudice. In case of the explicit measure of prejudice and other socio-political attitudes Disintegration turned out to be almost as important as Openness.
The path model containing the explicit measure of prejudice tested here is similar to the one tested on German and USA citizens (Keller & Knežević, 2016). This model had excellent fit to the data. The structure of the personality effects on prejudice mediated via RWA was basically the same as in case of the aforementioned model that was based on German and USA samples. One difference is that the direct effect of Disintegration on prejudice was not significant in this sample. There are also important differences from what would be expected from the Ekehammar et al. (2004) dual process model: the path from A to SDO was not found to be significant in this sample, but the path from O to SDO was significant, instead.
When the implicit measure of prejudice was entered in the analyses (instead of the explicit one), the relations of personality traits to prejudice via RWA remained highly similar to what was obtained with the explicit prejudice measure. The model had excellent fit to the data. The direct effect of Disintegration on prejudice when the latter was implicitly measured was marginally significant. However, the branch of personality effects on prejudice via SDO entirely disappeared when prejudice was measured with an implicit measure: now, even the path from SDO to prejudice was not significant.
Overall, these findings persuasively document the role of Disintegration concerning RWA and prejudice. This conclusion holds no matter whether an explicit or an implicit measure of prejudice has been employed. In line with previous conclusions by Ekehammar et al. (2004) our findings point to the robustness of the relations of personality to prejudice mediated via RWA. The fact that the direct effect of Disintegration on prejudice was not found in the path analysis in this sample (except marginally statistical effect of Disintegration on the implicit measure of prejudice). can be ascribed to the fact that our study was slightly underpowered for the detection of a coefficient of that size within the path model.
Personality relations to prejudice mediated via SDO proved to be less robust than those via RWA. The reason for this might partially lie in the slightly lower reliability of the implicit measure of prejudice. However, the fact that A -SDO- implicitly measured prejudice effects were absent, while those mediated via RWA held. suggests another possibility: RWA mediated personality - prejudice relations might not be based completely on conscious and intentional processes. as SDO mediated personality - prejudice relations seem to be. but also on more automatic and less controlled processes. This conclusion is already announced by the robust finding in all path analyses presented so far, showing that RWA predicts SDO rather than the reverse (Ekehammar et al., 2004) as well as the finding that RWA was more closely associated with personality traits than SDO. These findings were interpreted by Ekehammar et al. (2004) as reflecting the fact that RWA “appears to be more closely associated with the Big Five personality sphere than SDO”, while “...SDO appears to be closer to the prejudice sphere than RWA” (p. 478). Our results suggests that this stronger rootedness of RWA in personality reflects the fact that RWA captures automatic. less conscious processes of prejudice better than SDO.
Because rooted in more automatic and less conscious processes, the personality - prejudice relations mediated via RWA are, at least to some extent, less sensitive to numerous contextual factors that might influence explicit measures. The role of Disintegration in these implicit prejudice processes seems to be even more important than in those of more consciously controlled. It can be concluded that Openness and Disintegration seem to play the central role in the relation between personality and RWA as well as prejudice, especially those operating in a more spontaneous, less conscious and less controlled manner.
Study 2: Experimental manipulation of psychotic-like experience and moderator effect of Disintegration
Data were collected in Germany, the sample consisting of 74 students from the University of Ulm, (65 women); Mage (SD) = 24.19 (5.14). The same DELTA and PTT-E measures were used as in Study 1. Additionally, LIKE EE (Keller, 2005), a measure of likability ratings of Eastern European citizens was administered to the participants as another measure of prejudice. As part of the questionnaire introduced as an opinion poll about the European integration and EU enlargement, participants were asked to indicate their feelings towards citizens of several Eastern European countries (Bulgaria, Poland, Rumania, Russia, Turkey) using a 9-point scale ranging from not at all likeable to very likeable. Ratings for the citizens of Eastern European countries formed the critical measure of prejudice in the present study.
After being informed of the procedure. participants filled-in PTT-E and DELTA questionnaires. Then, they were exposed to the procedure intended to elicit an out-of-body-illusion. After the stimulation, participants were asked to complete a questionnaire by which the presence of the illusion was checked. Finally. they filled in a questionnaire assessing likeability ratings for Eastern Europeans citizens.
We tried to follow the original procedure described by Ehrsson (2007) as closely as possible. Participants sat on a chair, wearing a pair of head-mounted displays that were connected to two video cameras. In our experiment, the cameras were placed 1 m behind the participant’s back. The images from the left video camera were presented on the left display and the images from the right camera on the right display. The person could see his or her back from the point of view of a person sitting behind her/him. The experimenter stood beside the participant. She was holding two plastic rods with which she touched the person’s chest (being out of participant’s view). and the chest of the “illusory body” at the same time. The stimulation lasted for two minutes. The participants were randomly assigned to two experimental conditions: synchronous and asynchronous. In the synchronous condition (experimental condition) the touches of a person's actual chest and the chest of the “illusory body” were synchronized, while in the asynchronous condition (control condition) they were not. The presence of a strong out-of-body illusion in the synchronous condition and the absence of the illusion in the asynchronous condition were reported by the participants in a post-experimental questionnaire.
The crucial dependent variable was the likeability ratings for Eastern European citizens. The pre-experimental measure of prejudice PTT-E served to control for initial differences in prejudice. Crucial predictors in the regression analysis were condition (experimental vs control), score on the Disintegration scale and their interaction. Zero-order correlations were calculated between predictors and dependent variable. In order to test Condition x Disintegration interaction effect and to control for the initial level of prejudice, the data were analyzed using hierarchical OLS multiple regression. Prejudice as covariate was introduced in the first block in the multiple regression analysis followed by Disintegration. condition and the interaction term (Disintegration x condition) in the second block. To avoid problems with the collinearity between interaction term and its components all variables were standardized.
Prejudice towards Turkish minority explained 18 % of the variance in the likeability ratings for Eastern European citizens. Disintegration, experimental condition, and their interaction explained another 10 % of the variance in the dependent variable. As expected, the interaction effect of Disintegration and the experimental condition on the dependent variable was found: out-of-body illusion tends to decrease likability of Eastern Europeans proportionally to their level of Disintegration. In other words, the higher a person’s level of Disintegration. the stronger is the effect of the out-of-body illusion in decreasing her/his likability ratings for out-group members (Eastern Europeans). A significant main effect of the condition did not emerge indicating that it is not the out-of-body illusion per se that influences likeability ratings for out-groups. Interestingly, the effect of the illusion on prejudice seems to be transmitted through those aspects of Disintegration related to cognitive and perceptual rather than emotional processes. Namely, when Disintegration score was split into cognitive/perceptual and emotional disintegration. it was only the former showing a moderator effect. It could be that the engagement of former or latter process depends on the type of stimuli - in case of out-of-body illusion the proneness to cognitive/perceptual disintegration is important. but in case of some other stimuli, such as stress and frustration, it might be that proneness to emotional aspects of disintegration plays the crucial role.
To conclude, the inclusion of Disintegration into the set of basic personality traits turned out to be fruitful when it comes to explaining individual differences in socio-political attitudes. Disintegration together with Openness seem to be the most important personality correlates of prejudice towards minorities. Disintegration relates directly to prejudice, but the largest part of the relation has an indirect character, via RWA (not SDO). Disintegration together with O and C are important and robust predictors of RWA. The only personality factor showing a relation via SDO - the one originating in low A - is also confirmed by our data. However, the relations of personality to prejudice mediated via SDO are found to be weaker and less robust across samples and methods of prejudice assessment (explicit vs. implicit). The effects of N and E are negligible and highly varying across samples.
The robustness of the Disintegration - prejudice correlation has been demonstrated not only across various samples (samples from different cultures, students, general population), but also across methods (explicit and implicit measures of prejudice). Finally, our data demonstrated that the experimental induction of a psychotic-like experience (out-of-body illusion) interacts with Disintegration, making those individuals scoring high on Disintegration react in a more prejudiced manner. This experimental finding demonstrates the causal relevance of Disintegration concerning prejudice.
- Adorno, T.W., Frenkel-Brunswick. E., Levinson. D.J., & Sanford. R.N. (1950). The Authoritarian Personality, NY: Harper-Row.
- Altemeyer. B. (2007), The Authoritarians, University of Manitoba. http://home.cc.umanitoba.ca/~altemey/
- Bar-Anan. Y., Nosek. B. A., & Vianello. M. (2009). The sorting paired features task: A measure of association strengths, Experimental Psychology, 56, 329–343. doi:10.1027/1618-3184.108.40.2069
- Costa. P. T., & McCrae. R. R. (1992). NEO Personality Inventory-Revised: Professional manual. Odessa. FL: Psychological Assessment Resources. Inc.
- Ehrsson. H. H. (2007). The experimental induction of out-of-body experiences, Science, 317, 1048.
- Ekehammar. B. & Akrami. N. Gylje. M., & Zakrisson. I. (2004). What matters most to prejudice: Big Five personality: Social dominance orientation, or right-wing authoritarianism? European Journal of Personality, 18, 463-482.
- Higgins. E. T. (1996). Knowledge activation: Accessibility. applicability. and salience. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 133–168), New York. NY US: Guilford Press.
- Keller. J. (2005). In genes we trust: The biological component of psychological essentialism and its relationship to mechanisms of motivated social cognition, Journal of Personality and Social Psychology, 88, 686–702, doi:10.1037/0022-35220.127.116.116
- Keller. J. & Knežević. G. (2016). Personality roots of right-wing authoritarianism and prejudices. Paper presented at the 18th meeting of the European Association of Personality Psychology. July 2016, Timisoara, Romania.
- Knežević. G., Savić. D.. Kutlešić. V., & Opačić. G. (2017). Disintegration: A Reconceptualization of Psychosis Proneness as a Personality Trait Separate from the Big Five, Journal of Research in Personality, doi:10.1016/j.jrp.2017.06.001.
- Mierke. J., & Klauer. K. C. (2001). Implicit association measurement with the IAT: Evidence for effects of executive control processes. Zeitschrift für Experimentelle Psychologie, 48, 107–122, doi:10.1026//0949-3918.104.22.168
- Pettigrew. T. F.. & Meertens. R. W. (1995). Subtle and blatant prejudice in Western Europe. European Journal of Social Psychology. 25. 57–75.
- Pratto. F.. Sidanius. J.. Stallworth. L. M.. & Malle. B. F. (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology. 67. 741–763. doi:10.1037/0022-3522.214.171.1241
- Rangel. U.. & Keller. J. (2011). Essentialism Goes Social: Belief in Social Determinism as a Component of Psychological Essentialism. Journal of Personality and Social Psychology. 100. 1056-78. doi: 10.1037/a0022401
- Uhr. S. (2014). Anti-Roma racism in Europe: Past and recent perspectives. In M. Lavalette & L. Penketh (Eds.). Race. racism. and social work (pp. 115-130). Bristol. UK: University Press.
- Žeželj. I.. Jakšić. I.. & Jošić. S. (2015). How contact shapes implicit and explicit preferences: Attitudes toward Roma children in inclusive and non-inclusive environment. Journal of Applied Social Psychology. 45. 263-273. doi: 10.1111/jasp.12293