Explaining Memory Amplification: Is It All About the Test Format?

Abstract

Trauma-exposed people commonly exhibit a "memory amplification" effect, endorsing exposure to more traumatic events over time. Studies reporting this phenomenon have typically relied on checklists, where participants read event descriptions and indicate (yes/no) their exposure. We examined whether that approach is vulnerable to response biases and memory errors. In two experiments, participants viewed negative photos and completed an Old-New recognition test. In Experiment 1, participants completed either a photo recognition test or description test-composed of written descriptions of negative photos. In Experiment 2, we measured analogue PTSD symptoms and participants completed the description test twice, 24 hr apart. Those in the description test condition performed worse on the memory test and were more biased to endorse negative photos compared with the photo test condition. Furthermore, this bias to endorse negative photos increased over time and was related to analogue PTSD symptoms. Overall, our findings suggest that test format plays a role in memory amplification.

Memory consistency for sexual assault events.

“…Therefore, it is possible that increased recall of trauma is a result of a reappraisal process: People who are more distressed may try to make sense of their PTSD symptoms by relabeling and endorsing events that they had not endorsed earlier (see Engelhard et al, 2008). This explanation might be especially likely because we used an exposure scale composed of verbal written descriptions, which can be subjective and therefore particularly open to reinterpretation (see Takarangi, Oulton, & Strange, 2018). However, we do note that participants may have appraised their current distress through the lens of the number of trauma events they had just brought to mind and endorsed (e.g., believing that they should not be distressed because they endorsed only one item) rather than endorsing based on their current distress.…”

“…Also, because yes/no reports are potentially inaccurate (e.g., if participants always select “no” when remotely unsure), future research could use an exposure scale that assesses people’s (un)certainty that an event occurred. Furthermore, items on our exposure scale may have been ambiguous and subjective enough to be vulnerable to reinterpretation (see Engelhard & McNally, 2015; Takarangi et al, 2018). Indeed, Engelhard and McNally (2015) found that when they asked Dutch soldiers how they explained the discrepancy in their memory reports over time, the most common explanation was that they had interpreted items differently.…”