Understanding the Illusory Truth Effect Through Processing Perspectives

Abstract

Assessing what is true from what is not, is a critical task in increasingly overwhelming information environments. A range of factors influence people's perceptions of truth, including the source that provides the statement or how well it converges with prior knowledge. But a more insidious factor can shape people's perceptions of what seems real: the mere repetition of a claim. Decades of research show that people are more inclined to believe a claim that has been repeated, the Illusory Truth Effect (ITE) (Hasher et al., 1977). In a typical ITE study, people see a series of claims (half are true and half are false) at encoding and then, after a long or short delay interval, people see another series of claims and make truth judgments about them. Half the claims are repeated from the previous exposure and half are new. Repetition increases belief across different types of claims, such as people's opinions, consumer testimonials, political statements, and even implausible claims. One prominent theoretical account for the ITE is that repeated claims are processed more fluently than new claims which, in turn, are evaluated as more valid. Extending this processing fluency account, the Referential Theory of the ITE suggests that processing semantic elements of a claim (conceptual processing) at encoding is important in producing a fluent processing experience at re-exposure, as semantic references are reactivated in the memory network at the time of judgement. In the thesis, we draw on classic cognitive paradigms to explore the impact of shifts in processing demands on the ITE, considering the impact of overlap in cognitive operations, addition of secondary tasks, and stimulus features that may impact conceptual processing. In the first study (five experiments), we applied the Transfer-Appropriate Processing framework to examine to what extent matching/mismatching cognitive operations influenced the magnitude of the ITE. We found the ITE persists regardless of the match or mismatch of processing at encoding and testing stages. In the second study (three experiments), we examined whether introducing a secondary task, compared to a control condition (classic ITE paradigm), influences the size of the ITE. We found that the presence of a secondary task (e.g., counting vowels) reduces the ITE compared to control, but there was individual variation in this effect, suggesting a possible role for dual-task focus in moderating the ITE. In the third study (two experiments), we introduced photographs to examine the extent to which adding semantic context may have an additive impact on the ITE across long (48 hours) and short (10 minutes) delays. We found that photos had an additive effect on repeated claims in the 10-minute, but not the 48-hour, delay experiment. Considered together these studies extend understanding regarding processing demands and the ITE. The ITE literature has focused largely on features of a claim in understanding cognitive mechanisms contributing to the impact of repetition. This research considers how processing demands may alter the magnitude of the ITE through shaping the way people engage with content. We find mixed evidence for processing demands shaping the ITE. Our findings highlight the need to further understand the ITE by considering features of people's information environments and the manner in which they engage with content they encounter. Show more