Context-specific Information Processing: Investigating Circumstances that Improve the Retention of Message Content

We investigate the relationship between imagery and discursive processing of information content in contexts where one or the other type of processing is prevalent. We compare the retention of the content in episodic memory obtained in an experiment (a recognition task) where participants studied pictures or words either in an imagery-enhancing (a 3D virtual world) or in a discursive-enhancing context (a traditional web browser). The results show a contrast effect when the exposure time is sufficiently high, that is, measured recognition accuracy increases when content and context mismatch. We do not find that the effect persists at very low exposure time. This result fits our theoretical framework which informs that assimilation/contrast and congruence/incongruence theories make different predictions about episodic memory retention. The former relates to a more shallow form of encoding (lower elaboration) predicting an improvement in mismatching conditions because of a contrast effect between the processing style used within the context and the processing style adopted during the task. The second relates to a deeper encoding (higher elaboration) predicting an improvement in matching conditions, because of congruence between the context and the task. With exposure time sufficiently above a certain threshold and under shallow encoding we find higher accuracy with mismatching content and context, thus providing opening support to the former hypothesis and to the theoretical framework. We discuss the implications for advertising and media selection.