Abstract
Over the past two decades, hundreds of studies have been published on issues related to implicit attitudes. However, there is a strong debate in the literature concerning the nature and properties of implicit attitudes, including how malleable they are. Are they malleable momentary constructs or inflexible stored summary evaluations? One of the key issues in the research on attitudinal malleability is the context-sensitivity of implicit attitudes. Do implicit attitudes towards the same object differ in various contexts? Is such context-sensitivity a property of implicit attitudes or some of its measures? The context-sensitivity of implicit attitudes has attracted considerable research attention, but so far no consensus has been reached. Some models suggest that implicit attitudes are highly insensitive to contextual manipulations, while others propose the opposite. The disagreement can be partly attributed to differences in the understanding of context-sensitivity in relation to implicit attitudes. Certain contexts change the meaning and interpretation of an attitude object (interactive context) while others do not (independent context). For instance, an unfamiliar person in an office may be regarded as an office worker, whereas the same person in a family setting is interpreted as a father (interactive context). By contrast, a cat in a garden or in a room might be considered as a pet in both contexts (independent context). Research on the independent context sensitivity of implicit attitudes has important implications for how these attitudes are represented and stored, yet most of the empirical research has been dedicated to examining the effect of interactive context on implicit attitudes. The present research therefore attempts to investigate whether implicit attitudes are sensitive to independent context, which does not alter the meaning of the attitude object. The independent context sensitivity of implicit attitudes is tested by manipulating the modality (picture, word) of test materials in implicit attitude tests. In four experiments it was investigated whether the modality (picture, word) of test stimuli leads to attitudinal evaluations in line with the valence of within-modality association only. More specifically, we tested whether pictorial associations have more impact than verbal associations in a pictorial attitude test, and vice versa in a verbal attitude test. The results of these experiments reveal mixed findings. Two experiments utilising the Implicit Association Test reveal that implicit attitudinal scores can be sensitive to test modality. However, two further studies that used the evaluative priming task as an implicit attitude measure did not find similar modality-specificity. The implications for the independent context-sensitivity of implicit attitudes and implicit attitude measures of these findings are discussed.