Remedying the Metamemory Expectancy Illusion in Source Monitoring: Are there Effects on Restudy Choices and Source Memory?

Abstract

Metamemory monitoring, study behavior, and memory are presumably causally connected. When people misjudge their memory, their study behavior should be biased accordingly. Remedying metamemory illusions should debias study behavior and improve memory. One metamemory illusion concerns source memory, a critical aspect of episodic memory. People predict better source memory for items that originated from an expected source (e.g., toothbrush in a bathroom) rather than an unexpected source (e.g., shampoo in a kitchen), whereas actual source memory shows the opposite: an inconsistency effect. This expectancy illusion biases restudy choices: Participants restudy more unexpected than expected source–item pairs. The authors tested the causal relationships between metamemory and source memory with a delay and a source-retrieval attempt between study and metamemory judgment to remedy the expectancy illusion and debias restudy choices. Debiased restudy choices should enhance source memory for expected items, thereby reducing the inconsistency effect. Two groups studied expected and unexpected source–item pairs. They made metamemory judgments and restudy choices immediately at study or after delay, restudied the selected pairs, and completed a source-monitoring test. After immediate judgments, participants predicted better source memory for expected pairs and selected more unexpected pairs for restudy. After delayed judgments, participants predicted a null effect of expectancy on source memory and selected equal numbers of expected and unexpected pairs. Thus, the expectancy illusion was partially remedied and restudy choices were debiased. Nevertheless, source memory was only weakly affected. The results challenge the presumed causal relationships between metamemory monitoring, study behavior, and source memory.

Metacognitive differentiation of item memory and source memory in schema-based source monitoring.

“…Participants rendered item-wise JOLs and JOSs. Replicating prior research (Mieth et al, 2021; Schaper et al, 2019a, 2019b), participants predicted expectancy effects on item memory (via JOLs, see Konopka & Benjamin, 2009; Shi et al, 2012) and source memory (via JOSs, see Schaper & Bayen, 2021; Schaper et al, 2022a, 2022b). The expectancy effect was stronger on JOSs than JOLs.…”

“…Such manipulations of expected and unexpected source–item pairs have been implemented in a myriad of studies with a variety of item and source materials. For example, researchers used scenes with expected and unexpected objects (Küppers & Bayen, 2014; Lew & Howe, 2017; Schaper & Bayen, 2021; Schaper et al, 2019a, 2019b, 2022a, 2022b), people with different professions making expected and unexpected statements (Arnold et al, 2013; Bayen & Kuhlmann, 2011; Dodson et al, 2008; Hicks & Cockman, 2003; Konopka & Benjamin, 2009; Kuhlmann et al, 2012; Shi et al, 2012; Spaniol & Bayen, 2002; Wulff & Kuhlmann, 2020), people associated with certain stereotypes showing expected and unexpected behavior (e.g., concerning prosocial behavior: Ehrenberg & Klauer, 2005; Sherman et al, 1998; Sherman & Bessenoff, 1999; gender: Marsh et al, 2006; or partisanship: Mather et al, 1999), and trustworthy and untrustworthy people who expectedly or unexpectedly cheated or cooperated (Bell et al, 2012; Kroneisen & Bell, 2013; Kroneisen et al, 2015; Mieth et al, 2021). These studies show that schema-based expectations affect item memory and source memory differentially.…”

“…On the one hand, people predict that item memory and source memory are similarly affected by schema-based expectations. Across multiple studies, participants predicted that both were better for expected than unexpected trials (Mieth et al, 2021; Schaper et al, 2019a, 2019b; for studies assessing JOLs only, see Konopka & Benjamin, 2009; Shi et al, 2012; for studies assessing JOSs only, see Schaper & Bayen, 2021; Schaper et al, 2022a, 2022b). That is, participants predicted expectancy effects on item memory and source memory.…”

“…That is, study choices are based on convictions about memory and learning (see review by Kornell & Finn, 2016). In line, in schema-based source monitoring, participants restudied unexpected source-item pairs more often than expected pairs, because they predicted worse source memory for unexpected pairs (Schaper & Bayen, 2021;Schaper et al, 2022b). One interesting direction for future research would be to test whether differential monitoring of different aspects of memory translates into differences in metamemory control.…”

Metacognitive Illusions: A Positivity Effect in Judgments of Learning for Older but Not Younger Adults

“…Improving the accuracy of metacognitive processes can improve older adults’ quality of life ( Tauber and Witherby 2016 ), and effective use of metacognitive monitoring can help older adults compensate for memory decline through selective, compensatory learning efforts ( Hertzog and Dunlosky 2011 ). Because of the causal relationship between metamemory monitoring, control, and memory, correcting metamemory illusions holds promise for de-biasing learning behaviors and ultimately improving memory ( Schaper et al 2022 ). Future research could explore in greater depth the effect of emotion on metacognitive monitoring accuracy in different age groups, which is crucial for understanding the factors that limit monitoring accuracy and how to improve individual monitoring accuracy.…”