The study assessed how stimuli that contradict pre-experimental histories affect the formation of new relations. The study also assessed whether social variables such as race would influence college students’ perceptions of people of different races and job types. Twenty-six college students at a Midwestern University participated in the study. During the pre-test, participants rated the degree of “Good” or “Bad” of various pictures of African American males, police officers, and random objects on a Likert-type scale. Next, based on their pre-test results, participants completed a match to sample task that paired pictures of African American males and police officers opposite of their initial perceptions. Afterward, all participants again completed the Likert-scale rating task. Pre-test results revealed that some participants demonstrated strong negative pre-experimental biases toward police officers and that the race of the participants influenced their pre-test ratings. Individual data showed that 22 out of 24 participants changed their perceptions for at least one stimulus. Match to sample and post-test results revealed that participants with strong pre-experimental biases took more trials to complete the task, scored less accurately when conditions included socially loaded stimuli, and were less likely to change mean ratings for police officers during the post-test rating scale.

Made available in DSpace on 2016-06-02T20:30:59Z (GMT). No. of bitstreams: 1 6757.pdf: 1103934 bytes, checksum: 1e243bc6fc800e7f55079679a98ba983 (MD5) Previous issue date: 2015-03-09 / Financiadora de Estudos e Projetos / Research upon attitude s formation and change using the stimulus equivalence paradigm has shown, at the same time, a potential for the formation of new classes, and a difficulty with respect to teaching new relationships when the stimuli used are familiar and socially loaded. As racial prejudice is a problem worldwide, and stimulus equivalence paradigm has proven to be useful when studying attitude s formation and change, the aim of this study was to verify, from a systematic replication, if teaching new relations to children who showed a negative racial bias towards black people could revert the pre-existing classes. The level of 54 children s bias was assessed by the Self-Assessment Manikin (SAM), an instrument that measures the affective experience of individuals on pictures or events. The 22 participants who had attributed to black people s pictures lower levels of pleasure compared to white people s pictures also performed a second test aimed to confirm the bias shown. The 13 participants whose negative bias towards black people was confirmed were trained to match indirectly black people s pictures with positive attributes in a matching-to-sample task. Two experimenters, a black and a white one conducted the research with different children, to evaluate possible differences in the participants performance. The performances of these children were compared in two conditions: using simultaneous or delayed matching-to-sample, analyzing 1) How many children formed the expected equivalence classes; 2) in which of the two conditions the results were more robust; and 3) if the presence of white faces as a third comparison-stimuli in a modified equivalence test caused changes in the previous responses given by the participants who responded in accord with stimulus equivalence. Two instruments, a self-report and a implicit measure were used as complementary measures of transfer of functions, to evaluate the meaning of the stimuli. Results showed no differences between delayed and simultaneous matching-to-sample, or between the experimenters in the formation of the equivalence classes. All 13 participants who demonstrated a negative racial bias showed formation of the equivalence classes experimentally planned. Of those 13 children, nine have maintained their responses on the modified equivalence test, and the group data showed transfer of function, evidenced by SAM. Although there was a statistically significant difference between the pleasure levels of white and black faces before the class formation, post-test data revealed no statistically significant differences; this was also confirmed by the results of another instrument, called Implicit Relational Assessment Procedure (IRAP). Future research should recruit a larger number of participants and systematically change variables such as the use of mixed training of trained relations and the review of baseline relations, in order to identify which of them may be responsible for the positive results of equivalence. / Pesquisas sobre formação e mudança de atitudes utilizando o paradigma de equivalência de estímulos têm mostrado, ao mesmo tempo, um potencial para a formação de novas classes, e a dificuldade com relação ao ensino de novas relações, quando os estímulos utilizados são familiares e socialmente carregados. Como o preconceito racial é um problema presente no mundo todo, e o paradigma de equivalência de estímulos tem se mostrado útil no estudo da formação e mudança de atitudes, o objetivo dessa pesquisa foi verificar, a partir de uma replicação sistemática, se o ensino de novas relações de equivalência a crianças que demonstrassem ter viés negativo para faces negras reverteria as classes pré-existentes. O nível de viés racial de 54 crianças foi avaliado pela Auto-Avaliação Manikin (SAM), instrumento que mede a experiência afetiva dos indivíduos diante de imagens ou eventos. As 22 crianças que avaliaram fotos de negros com níveis de prazer inferiores aos atribuídos nas fotos de brancos realizaram um segundo teste com o intuito de confirmar o viés apresentado. Os 13 participantes cujo viés racial negativo foi confirmado foram treinados a emparelhar, de maneira indireta, fotos de negros com um símbolo positivo, a partir de tarefas de emparelhamento de acordo com o modelo. Duas experimentadoras, uma negra e outra branca conduziram a pesquisa com diferentes crianças, para avaliar possíveis diferenças no desempenho dos participantes. O desempenho dos participantes foi comparado em duas condições: com o uso de emparelhamento ao modelo simultâneo versus com atraso, analisando 1) quantas crianças formariam as classes de equivalência esperadas; 2) em qual das duas condições os resultados obtidos foram mais robustos; e 3) se a presença das faces brancas como 3º. estímulocomparação em um teste de equivalência modificado ocasionaria mudanças nas respostas prévias dadas pelos participantes que demonstraram a formação das classes de equivalência. Dois instrumentos, um de autorrelato, e outro de medidas implícitas foram utilizados como medidas complementares de transferência de função, para verificar o significado dos estímulos. Os resultados mostraram não haver diferenças entre o uso de emparelhamento de acordo com o modelo simultâneo ou atrasado, ou entre as duas experimentadoras na formação das classes de equivalência. Todos os 13 participantes com viés negativo para faces negras demonstraram a formação das classes de equivalência planejadas experimentalmente. Desses 13, nove mantiveram suas respostas no teste de equivalência modificado, e os dados de grupo evidenciaram transferência de função obtida pelo SAM. Apesar de haver diferença estatisticamente significativa entre os níveis de prazer das faces brancas e negras antes da formação das classes, os dados de pós-teste com o SAM revelaram não haver diferenças estatisticamente significativas, dado confirmado pelos resultados de outro instrumento, chamado Procedimento de Avaliação Relacional Implícita (IRAP). Pesquisas futuras devem recrutar um número maior de participantes, além de alternar sistematicamente variáveis como o uso de treino misto das relações treinadas e a revisão das relações de linha de base, de modo a identificar quais delas podem ser responsáveis pelos resultados positivos de equivalência.