The misinformation effect: A contemporary replication and extension of Loftus et al. (1978) to investigate its underlying mechanisms

Although the misinformation effect (the tendency for memory of an event to be altered by misleading postevent information) is recognized as robust, its mechanisms remain debated. First, we replicated Loftus et al. (1978) Experiment 3 online. In Session 1, participants watched a car accident video created in Grand Theft Auto; in Session 2, they viewed cowitness statements providing consistent, misleading or no traffic sign information. A binary memory test then assessed memory for the original traffic sign. Misled participants reported the correct sign less often than controls. Second, we investigated whether demand characteristics explain the misinformation effect, indicating strategic reporting rather than true memory distortion. With an identical procedure but a four-option memory test and confidence-based point betting, we still found the effect, though smaller. Hence, demand effects are not the sole driver of the misinformation effect. Third, we reversed the order of original and misleading events to differentiate between memory-based theories: Whether misinformation blocks or overwrites the original memory or is more often reported due to its higher activation. The binary memory test showed no misinformation effect, favoring single-trace or blocking explanations over activation-based theories. However, caution is warranted, as the results may also stem from the order rather than solely supporting single-trace or blocking mechanisms. In conclusion, our studies confirmed memory's susceptibility to misinformation, showing a robust effect partly driven by demand effects and largely, though not definitively, explained by single-trace or blocking mechanisms. This contemporary online paradigm provides a valuable tool for further study. (PsycInfo Database Record (c) 2026 APA, all rights reserved).