Belief-related memories: autobiographical memories of the religious self

ABSTRACT

Autobiographical memories play important roles in the development of the self and identity by grounding the self in rich, meaningful experiences. In the present study, we explored whether there is a specific type of autobiographical memories tied to religious belief; namely, belief-related memories. We compared belief-related memories with important and word-cued memories across five religions. We found belief-related memories were characterised as important, positive, intense, vivid, and frequently retrieved. The characteristics of belief-related memories were markedly different from word-cued memories, and they also differed from important memories on several variables. Compared with important memories, belief-related memories were rated as less prevalent, less important, less scripted, and showed a different distribution across the life span. The temporal distributions of belief-related memories varied across religions and showed no reminiscence bump as opposed to important memories and word-cued memories. Our findings suggest belief-related memories form a distinct category of autobiographical memories, consistent with the self being multidimensional and with different types of memories supporting distinct aspects of the self.

KEYWORDS:

• Autobiographical memory
• belief-related memory
• religion
• religious identity

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The terms ‘self-concept’ and ‘identity’ are closely related. They usually refer to the same definition, “who I am.” Thus, we use these terms interchangeably in this article.

2 The number of participants in the present study differs from Tungjitcharoen and Berntsen (2020) because the data were initially spilt into two worksheets. Data cleaning was done separately for each data set after the split. Therefore, some responses may not have qualified for both parts.

3 These results are consistent with the norms from Huber and Huber (2012), which demonstrated that norm values of the CRSi7 from the USA (where the population is predominantly Christian), Islamic countries (e.g., Morocco, Indonesia, and Turkey) and India (Hindu) were higher than those from Thailand (Buddhist) and Israel (Jewish).

4 Because the numbers of memories retrieved in different tasks were unequal (seven for life story memory task versus three for belief-related memory and word-cued memory tasks), we ran a series of 2 (Memory type: Life story and belief-related memories) × 5 (Religion) mixed factorial ANOVAs for every memory characteristic using the first three life story memories, rather than all seven memories, to ensure that the results would hold if participants had been asked to recall only three life story memories rather than seven. We found that important memories were still rated as more prevalent and important than belief-related memories, and no difference was seen on valence ratings. However, in this new analysis, previously significant differences of the ratings of intensity and third-person perspective became non-significant. Additionally, some new effects emerged. Belief related memories were now rated as more vivid, more rehearsed and having more first-person perspective. See Supplementary tables S3 and S4 for details on these analyses. In short, even when limiting the analyses to the first three important memories, we observed clear differences between the two memory types.

5 The procedure can be described as follows. First, extract a number of participants who were between 31 and 35 years of age. These participants could not produce memories that were estimated to occur above their ages. Then, calculate a mean age of the participants. Add 0.5 years to the mean age because on average they had lived 6 months through their currant ages. Subtract 31 from the mean age to get a mean year these participants had lived through 31 years old, which was the lower bound of this age interval. After that, count the numbers of memories these participants produced that were dated between the ages of 31 and 35. Based on the assumption that if these participants had lived past this age interval, they would have generated more memories in proportion to their mean age, we divided the number of memories in the age interval 31-35 by the mean years these participants had lived through 31 years old to get an average number of memories produced per year. Finally, multiply the average number of memories procured per year by 5, which resulted in an estimated number of memories they could have produced had they all lived through the age interval. To illustrate, there were 14 Buddhist participants aged between 31 and 35 years old. Their mean age was 32.71 years. Therefore, on the average they had lived past through their 31^st birthday for (32.71 + 0.5) - 31 = 2.21 years. These participants generated 6 belief-related memories in the age interval of 31 to 35 years. Per year, they generated 6 ÷ 2.21 = 2.71 memories. If they had lived past this age interval, they would have generated 2.71 × 5 = 13.55 memories. Thus, we added 7.55 memories to the number of memories the participants produced in this age interval.

Additional information

Funding

This research was supported by Grant DNRF89 from the Danish National Research Foundation (Dansmark Grundforskningfond). We thank Adam Redman Congleton for comments on the manuscript.