Abstract
The present research examined the effects of sharing different types of memories on perceived relationship closeness and how that is related to psychological well-being in a cross-cultural context. In two studies, European American and Asian participants (total N = 714) reported their feelings of closeness to a conversation partner in hypothetical situations that involved five types of information sharing: specific and general autobiographical memories, specific and general vicarious memories, and non-person information. Sharing memories led to greater feelings of closeness than sharing non-person information in both groups. Asians felt closer to the conversation partner than European Americans across all types of information sharing. Sharing autobiographical and specific memories led to greater closeness than sharing vicarious and general memories in both groups, although sharing specific autobiographical memories appeared most effective for relationship closeness for European Americans. Perceived relationship closeness following conversational exchanges was positively associated with psychological well-being across cultures.
Sharing memories serves important social functions by facilitating relationship closeness between conversation partners (Alea & Bluck, 2007; Bluck et al., 2005; Wang, 2004). During memory conversations, the partners together mentally travel back in time, exchange thoughts and feelings, express empathy toward each other, and form joint perspectives about what happened. These processes further contribute to the development and maintenance of relationship closeness (Bluck, 2003; Pillemer, 2001; Wang, 2013). Indeed, some researchers regard social bonding to be the primary function of human memory (Bruce, 1989; Nelson & Fivush, 2004). Little is known about what factors influence the social functions of memory. In particular, does sharing different types of memory matter for how people feel about their relationship closeness following the conversation and if so, does culture play a role in moderating the process? Moreover, is the feeling of closeness from memory sharing indeed beneficial for individual well-being? To answer these questions, we set out to examine in a cross-cultural context how different types of memories—autobiographical versus vicarious memories and specific versus general memories—influence perceived relationship closeness following memory sharing and how the feeling of closeness is further linked to psychological well-being.
Autobiographical Versus Vicarious Memories
Memories differ in the identity of their protagonists. Autobiographical memory has the rememberer him- or herself as the protagonist, involving the recollection of significant personal experiences from an individual’s life (Brewer, 1986; Pillemer, 2001; Wang, 2013). In everyday life, autobiographical memories play an important role for developing and maintaining relationship closeness (Alea & Bluck, 2007; Bluck et al., 2005; Wang, 2004). Indeed, 44% to 75% of everyday conversations involve sharing personal experiences, at least in the Western cultural context (Beike et al., 2016; Pasupathi & Carstensen, 2003). Such memory conversations often involve not just families and friends but also strangers (Demiray et al., 2019). When a conversation draws information from one’s own experiences, it often appears credible and convincing and can thus engage listeners and induce their empathic responses, especially from listeners who have had similar experiences (Pillemer, 2001). Furthermore, sharing autobiographical memories provides the listeners with self-revealing information, which serves as a natural context for exchanging thoughts, emotions, and opinions with others (Alea & Bluck, 2003; Cohen, 1998; Demiray et al., 2019; Rasmussen & Habermas, 2011). These processes facilitate relationship development and intimacy.
In contrast to autobiographical memories, vicarious memories have others rather than the self as the protagonist, involving the recollection of other people’s experiences, such as events that happened to family members, friends, and even celebrities and the society at large in the form of news events (Pillemer et al., 2015; Reese et al., 2017; Wang, 2018). Although vicarious memories share similar characteristics with autobiographical memories in content focus and emotional reactions (Pillemer et al., 2015; Wang, 2006), they are lower in phenomenological experiences such as emotional intensity, vividness, clarity, and positivity than autobiographical memories (Thomsen & Pillemer, 2017). Like autobiographical memories, vicarious memories can create and strengthen social bonds among individuals (Miller, 2009; Wang, 2013). Still, given the physical absence of the protagonist from the context of the conversation and the low degree of self-disclosure by the memory sharer, sharing vicarious memories may not be as influential as sharing autobiographical memories on interpersonal connection between the conversation partners. Indeed, vicarious memories are viewed as less important than autobiographical memories in affecting relationships (Pillemer et al., 2015). Whether autobiographical and vicarious memories variably influence perceived relationship closeness in the context of memory sharing remains an empirical question.
Specific Versus General Memories
Memories further differ in specificity. No matter whether the protagonist is oneself or another person, memories can be of one-moment-in-time events that occurred at a particular place and time in the past typically lasting no longer than a day—specific memories, or repeated events occurring regularly during a period of time—general memories (Conway & Pleydell-Pearce, 2000; Pillemer, 2001; Rubin, 2005; Tulving, 2002; Williams & Broadbent, 1986). 1 Specific memories involve the recollection of unique and discrete episodes, often accompanied by a sense of mentally re-experiencing the past. In contrast, general memories often involve the provision of semantic or scripted information without evoking details from specific instances. Compared with general memories, specific memories contain more vivid sensory and emotional details and therefore are more likely to captivate and engage listeners (Alea & Bluck, 2003; Baron & Bluck, 2009). Sharing specific memories also allows individuals to effectively convey personal meanings of particular past events (Pillemer, 2001). Furthermore, although both specific and general memories can provide listeners with information about the memory protagonist (e.g., preferences, occupation, relationships), sharing specific memories can evoke a sense of mental time travel for the conversation partners to re-experience a past moment together (Beike et al., 2017; Rubin, 2005; Wang, 2013). Thus, specific memories may better facilitate relationship closeness than general memories in the context of memory sharing.
Very few studies have directly compared specific and general memories as they relate to relationship closeness. In one study, Beike et al. (2016) asked unacquainted pairs of participants to share specific autobiographical memories, general self-information (e.g., a summer job), or nonself-related information (e.g., answering trivia questions). Participants then reported their feelings of closeness with their conversation partners. Although sharing specific autobiographical memories facilitated greater relationship closeness than sharing nonself-related information, it did not differ from sharing general self-related information. Yet, in another study, Beike et al. (2017) found that sharing specific, but not general, memories of co-experienced events facilitated intimacy in close relationships. In contrast, Waters et al. (2014) asked college students to provide personal narratives of single (specific) and recurring (general) events. Based on the coding of the narratives, the researchers found that general memories conveyed a greater sense of relationship closeness than specific memories. However, this finding may not apply to the conversational context given the important influence of contextual factors on memory functions (Kulkofsky et al., 2010). Obviously, additional research is required to examine the effect of memory specificity on relationship closeness following memory sharing.
Culture, Memory Sharing, and Relationship Closeness
Culture influences how people use memories in everyday life and for relationship purposes in particular (for a review, see Alea & Wang, 2015). In Western cultures, relationships tend to be voluntary and individuals have much freedom to choose their friends and partners based on their personal preferences (Yuki & Schug, 2012; Yuki et al., 2007). Given the high relational mobility, actively developing and maintaining relationships is required in this cultural context, and sharing memories—particularly memories for unique personal events—is considered an effective means to achieve the purpose (Chen, 1995; Kito, 2005; Wang, 2013). In comparison, in many Asian cultures, relationships are largely defined by geographical vicinity or circumstances of birth and therefore tend to be more stable and require less maintenance through means such as memory sharing (Brint, 2001; Kito et al., 2017; Wang, 2004). Indeed, European American individuals report sharing memories—specific autobiographical memories in particular—more frequently than do Asians for social bonding purposes (Kulkofsky et al., 2009, 2010; Maki et al., 2015; Schug et al., 2010).
Culture further influences individuals’ sense of self in relation to others, which may, in turn, affect feelings of closeness following memory sharing. Whereas Western cultures greatly value independence and personal uniqueness, many Asian cultures prioritize interdependence and social harmony (Markus & Kitayama, 1991; Wang, 2013). Accordingly, when recalling personal experiences, Asians often focus on social interactions and group activities, whereas European Americans often focus on their own roles and perspectives (Wang, 2001, 2006; Wang & Conway, 2004; Wang & Ross, 2007). When interacting with others, Asians are more sensitive to social cues and focus more on consensus building than European Americans (Setlock et al., 2004; Wang et al., 2009). Furthermore, individuals who are more interdependently oriented perceive their relationships more optimistically than those who are more independently oriented (Cross & Morris, 2003). It is thus possible that social exchanges such as memory sharing can elicit greater feelings of closeness among Asians than European Americans.
Perceived Relationship Closeness and Psychological Well-Being
Relationship quality is essential for psychological well-being (Baumeister & Leary, 1995; Umberson et al., 2010). Whereas supportive relationships are associated with psychosocial adjustments, unsupportive and neglectful relationships lead to disruptions to well-being (e.g., Repetti et al., 2002; Townsend et al., 1988; Umberson et al., 2010). More generally, regardless of culture, social connectedness is a basic human need that must be fulfilled for individuals to maintain psychological functioning (Ryan & Deci, 2000).
Importantly, individuals’ interpretations of social interactions and the associated relationship qualities have implications for their psychosocial adjustments (Song et al., 2018; Warren et al., 2000). For example, children who perceive interpersonal interactions more in terms of caring themes of affiliation, reparation, and prosocial behaviors exhibit more positive long-term trajectories in development, compared with children who perceive more negative themes such as aggression and destruction (Warren et al., 2000; Zahn-Waxler et al., 2008). Young adults who reconstrue peer-victimization experiences with positive resolution show fewer depressive symptoms and greater self-acceptance (Song & Wang, 2014; Song et al., 2021). In the context of memory sharing, individuals who are better able to derive positive meanings from the conversation and perceive greater closeness with the conversation partner may benefit more from the social exchange.
The Present Research
The present research examined in a cross-cultural context the effect of memory protagonist and specificity during memory sharing on developing relationship closeness and the relation of perceived closeness to psychological well-being. We collected data online to obtain samples from demographically diverse populations. In two studies, European American and Asian participants recruited through Amazon Mechanical Turk reported their feelings of closeness to a conversation partner in hypothetical scenarios that involved five types of information sharing: specific and general autobiographical memories, specific and general vicarious memories, and non-person information that did not involve any protagonist (e.g., talking about the weather).
Notably, memory sharing serves both the functions of nurturing existing relationships and forming new relationships (Bluck et al., 2005; Rasmussen & Habermas, 2011). We focused on new relationships here to eliminate the potential effect of pre-existing relationship qualities, which is especially important in a cross-cultural context (e.g., Sparks et al., 2016). Furthermore, the use of hypothetical scenarios allows the control of individual variations in memories to be shared and has been a useful approach to studying memory and other cognitive processes (Goyal et al., 2020; Li & Wang, 2004; Wang & Leichtman, 2000; Zhao & Kushnir, 2019). Research has further shown that guided by their relationship schemas, individuals interpret hypothetical social interactions in line with their actual interpersonal experiences, which reveals their views of the self and others and predicts their psychological well-being (Oppenheim, 2006; Song et al., 2018; Wang & Ross, 2005; Zahn-Waxler et al., 2008).
Given the social functions of memory sharing (Nelson & Fivush, 2004; Pillemer, 2001; Wang, 2013), we expected that in both cultural groups, sharing memories would elicit greater feelings of closeness than sharing non-person information. On the other hand, given their interdependent value orientation and their attentiveness to social cues during interpersonal exchanges (Markus & Kitayama, 1991; Setlock et al., 2004; Wang et al., 2009), Asians would perceive greater relationship closeness in all conversational contexts than European Americans. Furthermore, recognizing the limited and sometimes conflicting empirical findings, we hypothesized based on the theorization of the different types of memories (Alea & Bluck, 2003; Pillemer et al., 2015; Wang, 2013) that in both cultural groups, sharing autobiographical memories and specific memories would lead to greater perceived relationship closeness than sharing vicarious memories and general memories, respectively. In addition, given the special value of specific autobiographical memories for social bonding in Western cultures (Kulkofsky et al., 2010; Maki et al., 2015; Wang, 2004), sharing such memories would be particularly effective for relationship closeness among European Americans.
We first tested these hypotheses in Study 1 and then replicated the findings in Study 2. In addition, we measured in Study 1 cultural value orientations at the individual level and expected that those who exhibited a greater interdependent value orientation would perceive greater relationship closeness following conversations with an imagined partner. In turn, greater perceived relationship closeness, as a reflection of individuals’ relationship schemas (Oppenheim, 2006; Song et al., 2018; Wang & Ross, 2005; Zahn-Waxler et al., 2008), would be related to better psychological well-being across cultures.
Study 1
Method
Participants
A total of 410 participants were recruited through Amazon Mechanical Turk, including 272 self-identified as White or European American (hereinafter referred to as European Americans; 134 males and 138 females) and 138 self-identified as Asian or Asian American (hereinafter referred to as Asians; 102 males and 36 females). Participants’ ages ranged from 19 to 68 years (M = 34.66, SD = 10.79). Among the participants, 74 received high school education or less, 174 had college degrees, and 162 had graduate degrees. A power analysis (G*Power; Faul et al., 2007) showed that a sample size of 266 would be needed to achieve a power of 0.9 to detect effects with a size of f = 0.10 and α = .05. To recruit enough Asian participants, we kept the data collection running until the number of Asian participants reached the planned size. An additional 42 participants did not complete the survey or follow the instructions and another 95 participants self-identified as belonging to other ethnic groups; those participants were excluded from the study.
Procedure and measures
Participants took part in a 30-minute online survey. They were asked to imagine that they newly met a person who shared information with them in a short conversation. The following instruction was presented before each scenario, “Imagine this is the first time you meet someone and you two have a short conversation. In the conversation, this person shares the following information.” The 10 conversational scenarios that included five types of information sharing were presented in a random order (see Table 1). The complete scenarios can be found in Supplemental Materials. The length of the scenarios was similar across the different types of information sharing; it ranged from 51 to 59 words (M = 55.3, SD = 2.4). Furthermore, efforts were made to match the scenarios on features irrelevant to the memory type (i.e., protagonist, specificity) to ensure that effects would be indeed due to memory type and not confounding variables. All memory-sharing scenarios were positive in valence, contained some evaluative elements, and were similar in structure.
Conversation Scenario Excerpts by Information Sharing Type.
Following each conversation description, participants were asked to rate in response to four questions their feelings of closeness to the conversation partner. The first question was adapted from the Inclusion of Other in Self Scale (IOS; Aron et al., 1992), which is a one-item pictorial measure for people’s sense of interpersonal interconnectedness. Participants chose one of the seven Venn-like diagrams, each showing different degrees of overlap between two circles labeled “Self” and “Other.” The IOS has good test-retest reliability and predictive validity and has proved to be a valid measure of closeness with intimate partners as well as strangers (Aron et al., 1992).
Following rating on the IOS Scale, participants responded to the question “How likely would you be to use the term “we” to characterize you and your conversation partner?” by rating on a 7-point scale from 1 (not at all) to 7 (extremely) (Cialdini et al., 1997). They then responded to two additional questions adapted from the Subjective Closeness Index (Gächter et al., 2015), including “Relative to all your other relationships (both same and opposite sex), how would you characterize your relationship with your conversation partner?” and “Relative to what you know about other people’s close relationships, how would you characterize your relationship with your conversation partner?” Participants rated on 7-point scales from 1 (not at all close) to 7 (extremely close).
Cronbach’s alphas for the ratings on the IOS Scale and the three subsequent questions used to assess the participants’ perceived closeness with the conversation partner ranged from .94 to .96 across the 10 conversations. The four ratings were thus averaged for each conversation to index the degree of perceived closeness following the conversation.
In addition, participants completed the Self-Construal Scale and the Flourishing Scale and provided demographic information. The Self-Construal Scale (SCS; Singelis, 1994) consists of two subscales that measure individuals’ value orientation toward independence (e.g., “I enjoy being unique and different from others in many respects”) and interdependence (e.g., “I often have the feeling that my relationships with others are more important than my own accomplishments”), respectively. Each subscale includes 15 items, and participants rated each item from 1 (strongly disagree) to 7 (strongly agree). The SCS has been used widely in cross-cultural research (e.g., Neff et al., 2008; Wang, 2001; Yoon et al., 2000). Cronbach’s alpha in the current sample was .83 for independence and .87 for interdependence. A composite score of independent score minus interdependent score was often used for analysis (e.g., Holland et al., 2004). Given our focus on relationship closeness, we computed a composite score of interdependent score minus independent score to index the degree of an interdependent value orientation and submitted the composite score to relevant analyses.
The Flourishing Scale (Diener et al., 2010) tests individuals’ psychological resources and strengths. It consists of eight items to assess various aspects of human functioning, such as relationships, self-esteem, purpose, and optimism (e.g., “I am competent and capable in the activities that are important to me”). Participants responded to each item on a 7-point scale (1 = strongly disagree; 7 = strongly agree). Cronbach’s alpha for the scale in the current sample was .92. The total score across the eight items (maximum = 56) was used to index the participants’ psychological well-being.
Results
Preliminary analyses showed that European Americans (Mage = 37.16, SD = 11.62) were on average older than Asians (Mage = 29.66, SD = 9.15), t(338) = 7.14, p < .001, d = .72, and there was a greater percentage of females (51%) in the European American sample than in the Asian sample (19%), χ2 (1, N = 410) = 22.77, p < .001, φ = .24. Asians had higher educational levels than did European Americans, χ2 (2, N = 410) = 118.45, p < .001, φ = .54. All subsequent analyses were conducted both with and without the demographic variables as covariates, which yielded identical patterns of results. We report the results of analyses with the covariates. The full statistical results can be found in Supplemental Materials.
Memory Characteristics and Perceived Closeness Across Cultures
In this section, we examined the participants’ perceived closeness with the imagined conversation partner as a function of the type of information sharing and culture. The closeness ratings were averaged between the two scenarios for each type of information sharing and submitted to analyses. The means and standard deviations are presented in Table 2 by culture and information type.
Means and Standard Deviations of Perceived Closeness.
First, to test whether sharing memories would result in feeling closer to the conversation partner than sharing non-person information, a 2 (culture: European American vs. Asian) × 2(conversational content: memory vs. non-person information) repeated- measures ANOVA was conducted on the closeness rating, with culture as a between-subjects factor, conversational content as a within-subjects factor, and an error term accounting for within-individual variance. Age, gender, and education were included in the model. There were main effects of culture, F(1, 404) = 51.07, p < .001, ηp² = .112, and conversational content, F(1, 408) = 163.20, p < .001, ηp² = .286, and a marginally significant Culture × Conversational content interaction, F(1, 408) = 3.09, p = .079, ηp² = .008. Asians perceived greater closeness than did European Americans regardless of the type of information sharing. Both European Americans and Asians felt closer to the imagined conversation partner when sharing memories than when sharing non-person information, with the difference being larger for European Americans, F(1, 271) = 126.90, p < .001, ηp² = .319, than for Asians, F(1, 137) = 37.88, p < .001, ηp² = .217.
Then, to examine the influence of sharing different types of memory on perceived relationship closeness, we focused on the four types of memory sharing in subsequent analyses. We conducted a 2(culture: European American vs. Asian) × 2(specificity: specific vs. general) × 2(protagonist: autobiographical vs. vicarious) repeated-measures ANOVA on the closeness rating, with culture as a between-subjects factor, memory specificity and memory protagonist as within-subjects factors, and an error term accounting for within-individual variance. Age, gender, and education were included in the model. There were main effects of culture, F(1, 404) = 48.99, p < .001, ηp² = .108, memory protagonist, F(1, 408) = 15.81, p < .001, ηp² = .037, and memory specificity, F(1, 408) = 4.63, p = .032, ηp² = .011, qualified by a two-way interaction between specificity and protagonist, F(1, 408) = 23.20, p < .001, ηp² = .054 and a three-way interaction between culture, specificity, and protagonist, F(1, 408) = 11.18, p = .001, ηp² = .027. The adjusted mean ratings from the model, in addition to the distribution of the raw data, are shown in Figure 1.

Closeness ratings as a function of the type of memory sharing and culture.
Asians perceived greater closeness than did European Americans regardless of the type of memory sharing. Further repeated-measures ANOVAs were then conducted within each culture to examine the effects of memory protagonist and memory specificity on closeness ratings. In the Asian group, only the effect of protagonist was marginally significant, F(1, 137) = 3.25, p = .074, ηp² = .023, whereby Asians felt somewhat closer to the imagined conversation partner when sharing autobiographical memories than when sharing vicarious memories. Memory specificity did not affect their perceived closeness. In the European American group, there was a main effect of memory protagonist, F(1, 271) = 12.74, p < .001, ηp² = .045 which was qualified by a Protagonist × Specificity interaction, F(1, 271) = 31.06, p < .001, ηp² = .103. Follow-up Tukey’s honestly significant difference (HSD) tests (p < .05) showed that when sharing specific memories, European Americans felt closer to the imagined partner when the conversation concerned autobiographical than vicarious events, whereas there was no significant difference pertaining to memory protagonist when sharing general memories. In addition, there was greater perceived closeness for sharing specific than general autobiographical memories, whereas there was greater perceived closeness for sharing general than specific vicarious memories. Overall, for European Americans, sharing specific autobiographical memories with an imagined conversation partner led to greater perceived closeness than sharing other types of memories.
Interdependent Value Orientation and Perceived Closeness
A composite score of self-construal was calculated by subtracting independence score from interdependence score for each participant to index the degree of an interdependent value orientation. Consistent with previous studies (e.g., Holland et al., 2004; Wang, 2001), Asians (M = .41, SD = 8.52) scored higher on interdependent value orientation than did European Americans (M = −7.93, SD = 19.29), F(1, 404) = 10.59, p = .001, ηp² = .026, independent of age, gender, and education.
Linear regression analyses were conducted to test the relation of participants’ interdependent value orientation (as the predictor) to their perceived closeness following memory sharing (as the outcome), with culture, age, gender, and education included in the models (as covariates). The patterns of results were consistent across different types of memory sharing. We thus report results of the mean closeness score across the four memory types as the outcome variable. The regression model showed that, independent of culture, age, gender, and education, an interdependent value orientation was positively associated with perceived closeness following memory sharing, β = .13, t(404) = 3.10, p = .002. A similar linear regression on perceived closeness following conversations about non-person information showed a marginally significant effect of value orientation, β = .07, t(404) = 1.66, p = .097. Additional analyses were conducted with Culture × Value orientation interaction included in the above models. The interaction was nonsignificant and the pattern of results remained identical.
Perceived Closeness and Well-Being
Asians (M = 45.14, SD = 6.91) scored higher than European Americans (M = 42.76, SD = 9.04) on the Flourishing Scale at marginal significance, F(1, 404) = 3.18, p = .075, ηp² = .008, independent of age, gender, and education. Linear regression analyses were conducted to test the relation between participants’ perceived closeness following hypothetical memory sharing (as the predictor) and their flourishing score (as the outcome), with culture, age, gender, and education included in the models (as covariates). The patterns of results were consistent across different types of memory sharing and we therefore report results pertaining to the mean closeness score across the four memory types as the predictor. Independent of culture, age, gender, and education, perceived closeness following imagined memory sharing was a positive predictor for well-being, β = .25, t(404) = 4.48, p < .001. A linear regression with perceived closeness following conversations about non-person information as the predictor for well-being also showed a significant effect, β = .18, t(404) = 3.12, p = .002. Additional analyses were conducted with Culture x Perceived closeness interaction included in the above models. The interaction was nonsignificant and the pattern of results remained identical.
Study 2
Given the original method that we developed to study the social functions of memory sharing, we conducted a second study to replicate the main findings. We also made some improvements to the task. Some of the Study 1 scenarios (e.g., about college roommate) were not relevant to older participants in our community sample. The pair of scenarios in some categories (e.g., general vicarious memory) were not as similar as planned, and features (e.g., structure, evaluation) irrelevant to the memory type could be better matched for some scenarios. We addressed these issues in Study 2. In addition, across all Study 1 scenarios, participants were asked to imagine being the listener of conversations. It is unclear whether the pattern of results was driven by this setup. We therefore introduced a within-subjects factor in Study 2 to examine whether being the listener versus being the sharer would play a role in perceived relationship closeness.
Method
Participants
A total of 304 participants were recruited through Amazon Mechanical Turk (MTurk), including 155 self-identified as White or European American (91 males and 64 females) and 149 self-identified as Asian or Asian American (96 males and 53 females). All participants were Master workers on MTurk, with the HIT approval rate set at 95%. Participants’ ages ranged from 18 to 69 years (M = 34.24, SD = 9.82). Among the participants, 19 received high school education or less, 76 had college degrees, and 209 had graduate degrees. An additional 41 participants did not complete the survey or follow the instructions and another 67 participants self-identified as belonging to other ethnic groups; those participants were excluded from the study.
Procedure and measures
The conversation task was the same as the one used in Study 1 except the following changes. 2 For the pair of scenarios in each type of information sharing, participants were asked to imagine being the sharer in one scenario and being the listener in the other scenario. The “sharer” and “listener” scenarios were counterbalanced such as half of the participants in each cultural group were the sharer in one scenario and the other half of the participants were the sharer in the other scenario. The content of each pair of scenarios was closely matched and was appropriate for participants of all ages (see Table 3). The complete scenarios can be found in Supplemental Materials. The length of the scenarios was similar across the different types of information sharing; it ranged from 55 to 62 words (M = 58.7, SD = 2.6). Additional efforts were made to match the scenarios on features (e.g., valence, evaluation, and structure) irrelevant to the memory type (i.e., protagonist, specificity) to eliminate potential confounding variables.
Conversation Scenario Excerpts by Information Sharing Type.
Following each conversation description, participants rated their feelings of closeness to the imagined conversation partner in four questions, same as in Study 1. Cronbach’s alphas for the ratings on the four questions ranged from .92 to .97 across the 10 conversations. The four ratings were thus averaged for each conversation to index the degree of perceived closeness following the conversation.
Results
Preliminary analyses showed that European Americans (Mage = 35.77, SD = 11.49) were on average older than Asians (Mage = 32.64, SD = 7.42), t(265) = 2.84, p = .005, d = .32, and Asians had higher educational levels than European Americans, χ2 (2, N = 304) = 11.36, p = .003, φ = .19. Analyses with and without the demographic variables as covariates yielded identical patterns of results. We report the results of analyses with the covariates. The full statistical results can be found in Supplemental Materials.
As in Study 1, we examined the participants’ perceived closeness with the imagined conversation partner as the function of the type of information sharing and culture. Across the 10 hypothetical conversations, participants’ role as the sharer versus listener did not have any significant effect on closeness ratings, ps > .05, and the ratings did not differ between the two scenarios within each type of information sharing, ps > .05. Thus, participants’ role in the conversations was not considered further, and the ratings were averaged between the two scenarios for each type of information sharing and submitted to analyses. The mean closeness ratings for different types of information sharing are presented in Table 4.
Means and Standard Deviations of Perceived Closeness.
First, to test whether sharing memories would better serve the function of developing relationship closeness than sharing non-person information, a 2 (culture: European American vs. Asian) × 2(conversational content: memory vs. non-person information) repeated-measures ANOVA was conducted on the closeness rating, with culture as a between-subjects factor, conversational content as a within-subjects factor, and an error term accounting for within-individual variance. Participants’ age, gender, and education were included in the model. The analysis revealed main effects of culture, F(1, 298) = 19.00, p < .001, ηp² = .060 and conversational content, F(1, 302) = 89.39, p < .001, ηp² = .228. There was no significant interaction between culture and conversational content. Thus, Asians perceived greater closeness than did European Americans regardless of the type of information sharing, and both European Americans and Asians felt closer to the imagined conversation partner after sharing memories than sharing non-person information.
Next, we examined the influence of specificity and protagonist of shared memories on perceived closeness. We conducted a 2 (culture: European American vs. Asian) × 2(specificity: specific vs. general) × 2(protagonist: autobiographical vs. vicarious) repeated-measures ANOVA on the closeness rating, with culture as a between-subjects factor, memory specificity and memory protagonist as within-subjects factors, and an error term of within-individual variance. Age, gender, and education were included in the model. There were main effects of culture, F(1, 298) = 21.59, p < .001, ηp² = .068 specificity, F(1, 302) = 39.95, p < .001, ηp² = .117 and protagonist, F(1, 302) = 41.31, p < .001, ηp² = .120 qualified by an interaction between specificity and protagonist, F(1, 302) = 6.02, p = .015, ηp² = .020. The adjusted mean ratings from the model, in addition to the distribution of the raw data, are shown in Figure 2. Asian participants perceived greater closeness than did European American participants across all types of memory sharing.

Closeness ratings as a function of the type of memory sharing and culture.
We further examined the effects of memory specificity and protagonist within each cultural group using repeated-measures ANOVAs. In the Asian group, conversation scenarios about specific memories led to greater feelings of closeness than those about general memories, F(1, 148) = 20.08, p < .001, ηp² = .119 and scenarios about autobiographical memories led to greater feelings of closeness than those about vicarious memories, F(1, 148) = 26.71, p < .001, ηp² = .153. Similarly, in the European American group, there were main effects of specificity, F(1, 154) = 20.57, p < .001, ηp² = .118 and protagonist, F(1, 154) = 16.20, p < .001, ηp² = .095 with a marginally significant Specificity x Protagonist interaction, F(1, 154) = 3.35, p = .069, ηp² = .021. Follow-up Tukey’s (HSD) tests (p < .05) showed that European Americans felt closer to the imagined conversation partner after sharing specific than general autobiographical memories, whereas the difference was not significant for vicarious memories. Also, European Americans felt closer to the imagined conversation partner after sharing specific autobiographical than vicarious memories, whereas the difference was not significant for general memories. Overall, for European Americans, sharing specific autobiographical memories with an imagined conversation partner led to greater feelings of closeness than sharing other types of memories.
General Discussion
Research to date has focused on the social functions of specific autobiographical memories in Western populations. Using hypothetical scenarios, the present research is the first to examine in a cross-cultural context the effect of sharing different types of memories on developing relationship closeness and the relation of perceived closeness to psychological well-being. We developed and replicated a novel method in which participants reported their feelings of closeness to a conversation partner in hypothetical scenarios of information sharing. The two studies revealed generally consistent patterns of results. For both European Americans and Asians, sharing memories with an imagined conversation partner led to greater feelings of closeness than sharing non-person information. This is consistent with the notion that memory and memory sharing serve social functions of relationship development and maintenance (e.g., Alea & Bluck, 2007; Bluck, 2003; Pillemer, 2001).
Furthermore, in both studies, Asians perceived greater relationship closeness than did European Americans across all conversational contexts. Given their interdependent value orientation and social sensitivity (Markus & Kitayama, 1991; Setlock et al., 2004; Wang, 2013), Asians may be particularly attuned to interpersonal exchanges that involve any kind of communication and expect relationship development as a result of the exchanges. In contrast, European Americans tend to view themselves as independent and separate from others and may therefore be less sensitive to social exchanges to experience relationship closeness. This finding was further confirmed at the individual level in Study 1, whereby, independent of culture and other demographic variables, participants with a greater interdependent value orientation perceived greater relationship closeness following memory sharing of all types, and even after sharing non-person information at marginal significance.
Also as predicted, memory protagonist and memory specificity influenced perceived relationship closeness with the imagined conversation partner, and the findings are generally consistent across the two cultural groups. For European Americans in both studies, sharing specific autobiographical memories had the strongest impact on perceived relationship closeness. This is in line with extant research showing that memories for distinct, momentary past events happening to oneself play an important role for social bonding, particularly among Western individuals (Kulkofsky et al., 2009, 2010; Maki et al., 2015). Notably, we matched the scenarios in our studies on features irrelevant to the memory type to eliminate confounding variables. In real life, memories for specific autobiographical events are often richer in sensory details and subjective experiences than general or vicarious events, which may further help to engage listeners in a joint mental time travel and facilitate intimacy (Alea & Bluck, 2003; Baron & Bluck, 2009; Pillemer, 2001; Pillemer et al., 2015; Rasmussen & Habermas, 2011).
Similarly, in both studies, Asians felt closer to the imagined partner when the conversation involved autobiographical than vicarious memories, although the effect was only marginally significant in Study 1. The finding is consistent with the notion that autobiographical memories are more effective for facilitating interpersonal connections than vicarious memories (Cohen, 1998; Pillemer et al., 2015; Rasmussen & Habermas, 2011). The effect of memory specificity on perceived closeness was less stable among Asians; it was non-significant in Study 1 and only emerged in Study 2. Wang (2013) contends that general memories often concern rules and regularities that inform individuals of social conventions and thus facilitate interpersonal harmony and group cohesion. Sharing general memories may therefore serve important relationship or group functions in the Asian cultural context. Importantly, culture may further interact with the context of memory usage (private recall vs. social sharing) and the nature of relationships in determining the effect of memory specificity on relationship closeness (Beike et al., 2017; Kulkofsky et al., 2010; Waters et al., 2014).
Finally, independent of culture and other demographic factors, participants who perceived greater relationship closeness following conversations with an imagined partner scored higher on psychological well-being. This relation was significant both following sharing memories and sharing non-person information. The finding is consistent with previous research showing that relationship quality, perceived or actual, is a universal key factor associated with well-being (Baumeister & Leary, 1995; Ryan & Deci, 2000; Umberson et al., 2010; Warren et al., 2000). While the relationship between perceived closeness and well-being is likely bi-directional, the current finding highlights interpersonal communication as an important context for relationship closeness and individual psychological functioning.
Some limitations of the current research warrant acknowledgment. Our participants only imagined having conversations. Although studying memory and other cognitive processes with hypothetical scenarios is an informative approach both within and across cultures (e.g., Goyal et al., 2020; Wang & Ross, 2005; Zhao & Kushnir, 2019), additional research using other methods are called for to fully understand the correlates and outcomes of memory sharing. In particular, methods involving natural conversations coupled with experimental controls are much needed. Also, although the current research included samples with a more diverse and representative demographic profile than in-lab studies with college students, some of the effects were small and the findings need to be corroborated in more controlled settings. Furthermore, this research focused on new relationships of equal partnership, which might explain why the role of listener versus sharer did not affect perceived closeness with the partner. Future research should examine other types of relationship dynamics in affecting memory sharing and intimacy across cultures. Finally, specific and general memories may differ in many dimensions (e.g., mental time travel, sensory and emotional details, personal meaning making) (Pillemer, 2001; Wang, 2013), and we did not systematically vary these dimensions in our scenarios. Future research should tackle these dimensions separately to understand the nuances of influences of memory specificity.
In conclusion, the present findings provide additional insights into the social functions of memory and memory sharing. Both memory characteristics and the participants’ cultural backgrounds influence feelings of closeness following memory conversations. The findings further suggest that sharing memories not only brings people closer but is also associated with psychological well-being, which has important application values in real life.
Supplemental Material
sj-docx-1-jcc-10.1177_00220221211072809 – Supplemental material for Does Sharing Memories Make Us Feel Closer?: The Roles of Memory Type and Culture
Supplemental material, sj-docx-1-jcc-10.1177_00220221211072809 for Does Sharing Memories Make Us Feel Closer?: The Roles of Memory Type and Culture by Li Guan and Qi Wang in Journal of Cross-Cultural Psychology
Supplemental Material
sj-docx-2-jcc-10.1177_00220221211072809 – Supplemental material for Does Sharing Memories Make Us Feel Closer?: The Roles of Memory Type and Culture
Supplemental material, sj-docx-2-jcc-10.1177_00220221211072809 for Does Sharing Memories Make Us Feel Closer?: The Roles of Memory Type and Culture by Li Guan and Qi Wang in Journal of Cross-Cultural Psychology
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by a grant to Qi Wang from the Jeffrey Sean Lehman Fund at Cornell University.
Supplemental Material
Supplemental material for this article is available online.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
