| Popis: |
Recent research has found that biracial categorization accuracy is far below monoracial categorization accuracy. This is troublesome for both internal reliability and ecological validity. As a response to the lack of biracial categorizations and reliability in real biracial stimuli, we have developed a set of altered (“biracialized”) faces. Our recent research has found that perceived “unusualness” or “distinctiveness” of the faces partially explains biracial categorization accuracy. If this is true, then enhanced unusual traits (e.g., eye color, freckles) and the interaction between traits (e.g., blue eyes with darker skin tone) may increase unusualness ratings. Our biracialized faces, altered in photoshop to enhance these unique features, might therefore be a solution to increase biracial categorization accuracy. The purpose of our research is to compare our biracialized stimulus set to existing biracial stimuli that are commonly used in the field. Existing research on the racial categorization of biracial individuals has heavily relied on the use of morphed faces to represent real biracial people (Minear & Park, 2004; Peery & Bodenhausen, 2008; Chen & Hamilton, 2012). Morphed faces are usually created using morphing software (e.g. Fantamorph) and combine two monoracial parent faces to create 50/50 morphs. For example, morphs are commonly made by combining a Black face and a White face to create a Black-White biracial face. In morphed stimuli, hair is usually left out of the final image due to the unsatisfactory product of morphing two different hair styles. Therefore, we will test our biracialized stimuli against these cropped morphed faces. We will also test them against real biracial faces from a new expansion to the Chicago Face Database (Ma, Kantner, & Wittenbrink, submitted for publication). Moreover, we also want to examine the underlying factors resulting in categorization differences across face types. The traits we plan to examine are: attractive, unusual (would stand out in a crowd), feminine, baby-faced, and artificial-looking. These are based on existing research in the field and our prior research. |
