Page 158 - Emotions through the eyes of our closest living relatives- Exploring attentional and behavioral mechanisms

Chapter 7
this time the concept-attribute combination was switched (i.e., if participants saw the face of a man of Moroccan descent combined with a negative image in block three, they now saw the face of a man of Moroccan descent combined with a positive image, Figure 2A and 2B).
The IAT is known for its order effects such that when congruent trials are presented first in the critical block, larger IAT effects are found than when incongruent trials are presented first (Nosek et al., 2005). Order effects can be counteracted by counterbalancing the presentation order of the critical blocks, and therefore half of the participants started out with congruent trials (i.e., “Dutch” + positive, “Moroccan” + negative), while the other half started out with incongruent trials (i.e., “Dutch” + negative, “Moroccan” + positive). Furthermore, in the practice blocks, participants either started out with categorizing faces (concepts) or with categorizing positive and negative scenes (attributes). Reaction times on trials in the critical blocks were expected to depend on the congruency of the trials, i.e., following our hypothesis, participants were expected to respond slower to incongruent trials in which faces of men of Dutch descent were linked to negative attributes, and the faces of men of Moroccan descent to positive attributes.
The PIAT was performed on a touchscreen using only one hand. This is different from the typical procedure where participants use their left and right hand to press a left and right key on a keyboard. Previous studies have shown that handedness and the assignments of the left or right response key to a particular IAT category have little to no influence on the IAT effect (Greenwald, 2001; Greenwald et al., 1998), thus we expect no difficulties with using one hand in the PIAT.
Stimuli3
Adult
In non-verbal versions of the IAT, stimuli usually consist of pleasant and unpleasant words (attributes) and words referring to the two categories that are being investigated (concepts). In the PIAT, concepts and attributes were replaced with images. Concepts
3 Given the sensitivity of topics such as ethnic prejudice and discrimination, I wish to provide some more context to our choice for measuring implicit racial attitudes (and not, for instance, implicit attitudes towards insects, established in prior studies (Greenwald et al., 1998)). Originally, when designing the study, we aimed to validate it in humans and subsequently test bonobos and chimpanzees on their implicit associations with familiar group mates (the so called “ingroup”) and unfamiliar others (“outgroups”). By doing so, we wanted to gain more insights into the evolutionary roots of discrimination. We therefore chose to test the implicit association that humans may have with their ingroup (e.g., individuals of the same descent) versus one potential outgroup (e.g., individuals of Moroccan descent).
156