<scp>Mild cognitive impairment</scp> decreases the accuracy of own memory monitoring

Do individuals with mild cognitive impairment (MCI) with impaired memory function also have impaired judgments about their own memories? The monitoring of one's memory is called metamemory and has been a popular topic in psychology for half a century.1 Memory monitoring is measured by a memory monitoring task.1 In the memory monitoring task, participants are first asked to learn items (either words or pictures), and immediately after learning each item, they are asked to provide their judgments of learning (JOLs; namely, judgments of how well a remembered item can be recalled later) ranging from 0% (I will never recall the item on a later test) to 100% (I will definitely recall the item on a later test). Finally, participants are asked to recall the items. The findings from these paradigms have shown that JOLs do not necessarily match actual memory performance.2 Age has also been examined as a factor in this discrepancy between memory performance and memory monitoring,3, 4 and, surprisingly, no differences in the correlation between memory performance and JOLs (i.e., accuracy of memory monitoring) were found between older and younger adults.3, 4 However, the accuracy of memory monitoring in older adults remains controversial. Many studies have examined age factors in memory monitoring and are limited to comparing younger and healthy older adults, whereas very few studies have examined differences in memory monitoring in relation to cognitive function in older adults. In particular, older adults with MCI have not only impaired memory function but also subjective complaints of forgetfulness, which suggests that memory monitoring is also impaired. Therefore, in this study, we aimed to determine the relationship between manipulated MCI by cognitive function test score and the accuracy of memory monitoring. We hypothesize that older adults with MCI show reduced accuracy in memory monitoring compared with healthy older adults. Participants were 113 older Japanese individuals (mean age = 71.56 years, age range = 65–85 years, women = 91%) aged 65 years or older who applied for health programs offered by Japanese local governments in 2021 and 2022. Cognitive function was measured by the Japanese version of the Montreal Cognitive Assessment (MoCA-J), with 25/26 points used as the cutoff point for MCI.5, 6 Metamemory monitoring was measured based on previous studies. The learning phase consisted of 60 pictures, which were presented for 1 s each, and participants were required to learn the pictures. Immediately, after each picture presentation, participants were asked to provide a judgment of learning with a key response (0–100%). After the learning phase, a recognition test was conducted with 120 pictures. In the recognition test, participants were asked to make a yes/no judgment by pressing a key, and participants responded “yes” to pictures presented in the learning phase. A total of 52 participants were assigned to the MCI group (mean age = 72.31, age range = 65–85 years) and 75 to the control group (mean age = 70.93, age range = 65–82 years) as a result of the MoCA. Recognition performance was evaluated as the percentage of correct hits (i.e., the percentage of “yes” responses to items presented in the learning phase). We conducted a t-test between the MCI and control groups for the correct hit rate. The findings showed that the MCI group had a significantly lower percentage of correct hits than the control group [t (107) = 2.36, padj = 0.02, Cohen's d = 0.46]. Goodman–Kruskal's gamma was used to measure the accuracy of memory monitoring.7 The Goodman-Kruskal gamma is one of the indicators of rank correlation.8 In this study, the correlation between JOLs and memory performance ranges from −1 (complete mismatch between JOLs and memory performance) to 1 (complete match between JOLs and memory performance). The analysis showed that the MCI group (gamma = 0.04, SD = 0.16) was less accurate in memory monitoring than the control group (gamma = 0.10, SD = 0.16) [t (107) = 2.12, padj = 0.04, Cohen's d = 0.42]. The results of the analysis of memory performance and memory monitoring are presented in Table 1. The present study suggests that participants with MCI have impaired memory performance and memory monitoring. Despite this significant result, some issues still need to be solved. First, most participants in the health program were female, so future studies should include greater sex diversity. Second, MCI was manipulatively defined in this study using the MoCA. A comprehensive understanding of MCI may be obtained by examining the relationship between MCI selected by various criteria and memory monitoring. This study was a correlational study. Therefore, We cannot determine causality. Further studies are needed to evaluate whether a causal relationship exists between MCI and impaired memory monitoring. This would contribute to providing evidence for the prevention of MCI. This work was supported by JSPS KAKENHI Grant Number JP 22H01098. The authors have no conflicts of interest to disclose. Data supporting the findings of this study are available from the corresponding author upon reasonable request.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.