What do we remember from the vast quantity of events happening to us, involving us, and relevant to our life-story? For example, the fact that you went on a trip with your classmates in the 5th grade and the details of the trip might be relevant to your life-story if many of your current friends were still some of the children that back then were your classmates. People’s memories for their experiences are not a veridical recording of such experiences, however (e.g., Schacter, 2001), and can be influenced by many factors. An important and influential theory focusing on the emergence and content of autobiographical memory is the social cultural developmental theory (Nelson & Fivush, 2004), which emphasizes the role of society and culture in shaping people’s memories of their autobiographical past.
In essence, autobiographical memory is about defining the self in time and in relation to others, so that individuals gain a sense of who they are by relating to others within a culture and creating a shared past. This theory views autobiographical memory as a function of various socio-cultural factors that interact with basic memory systems, such as the acquisition of language, talk with parents, the style of parental talk, and psychological understanding. Autobiographical memory emerges gradually and is influenced by cognitive developments and social interactions, thus becoming a social-cultural-cognitive system.
One area that has received attention from researchers is the influence of cultural differences in collectivism/individualism on autobiographical memory. This introduction presents the definitions for collectivism and individualism and some of the studies that illustrate the particularities of collectivistic societies and individuals, as well as those of individualistic societies and individuals. The influence of the socio-cultural background on people’s autobiographical memory is then depicted in studies that show the influence of the linguistic and cultural environment on children’s autobiographical memories in Western (e.g., American) and Eastern (e.g., Chinese) cultural backgrounds.
Some studies look at the conversational style between mother-children dyads of different ethnicities, and others are more specifically focused on autobiographical remembering and look at the content and characteristics of the children’s and adult’s autobiographical memories. Studies focusing on bicultural individuals will be examined to show that these individuals integrate the norms of both cultures and apply them accordingly. Finally, the phenomenon of imagination inflation is discussed and evidence towards it is brought through studies that show its effect on past events and future expectations, for both children and adults. Moreover, it is noted that imagination can also create false memories, particularly if the events imagined are highly plausible or if the participants are children.
“Open up to imagination” by Ryan Hickox
Differences Between Individualistic and Collectivistic Cultures
One important dimension of culture is the extent of individualism or collectivism exhibited (Desai, 2007). Collectivism puts an emphasis on distinguishing between in-groups and out-groups, engaging in cooperative tasks, and focusing on what people have in common. Conversely, individualism is characterized by engagement in competitive tasks, by public situations, and by an emphasis on what makes the individual distinct. In general, in societies in which agreeing on social norms is important and jobs are interdependent, collectivism is preponderant, whereas in complex, stratified societies, where affluence, independence, and differences are emphasized, individualism is preponderant.
In particular, individualism is mostly seen in the cultures of Western Europe and North America, whereas collectivism is mostly seen in the cultures of Asia, Africa, and parts of Europe and Latin America (Triandis, 1993; Nelson & Fuvish, 2004). The emphasis on one or another starts in the family, even with the very structure of the family: a large, multigenerational one emphasizes collectivism, whereas a smaller, nuclear family emphasizes individualism (Triandis, 1993). Studies examining differences in collectivistic and individualistic cultures often use either Asian Americans or people from Asian cultures, such as Vietnamese or Filipino and compare them to Caucasians or Americans (Skillman, 2000; Desai, 2007). These studies on families and intergenerational conflict show how individualistic societies value self-reliance, independence, autonomy, personal achievement (Skillman, 2000), and a definition of self apart from the group and personal goals (Desai, 2007).
Collectivistic societies value family cohesion, cooperation, solidarity, and conformity (Skillman, 2000), and thus people is these societies tend to make more references to others, emphasize group goals, and follow the expectations and regulations of the group (Desai, 2007).
Such cultural differences mean that people in different cultures have fundamentally different construals of the self and others. For more collectivistic societies, interdependent construals are the norm: The self is a part of a community, defined relative to others, concerned with belongingness, dependency, empathy, reciprocity and focused on small, selective in-groups at the expense of out-groups. The interdependent self exercises control to the interior, so that cognition and representation involve attentiveness to others, and personal attributes and actions are situationally bound. Autonomy becomes secondary, whereas relationships with others are emphasized, being ends in themselves.
Thus, it is crucial to be aware of other people’s desires, needs, and goals and to work towards them to help the other, even read their minds (Mark & Kitayama, 1991). For more individualistic societies, independent construals are the norm: The distinctiveness of people, the uniqueness of a person, autonomy, and independence are emphasized. This requires construing oneself as an individual and speaking one’s mind. Social responsiveness is determined by the need to assert and express the self, and thus the independent self exercises control to the exterior. The consequence is that larger, more inclusive but superficial in-groups are the norm, as opposed to the small, selective in-groups of the interdependent self construals (Mark & Kitayama, 1991).
Socio-Cultural Influences on Memory
According to the socio-cultural developmental theory, socio-cultural influences can be seen both in the formation and content of autobiographical memories (Nelson & Fivush, 2004). Research has examined this in several ways. One line of research has looked at cultural differences in autobiographical memory by comparing Caucasian Americans with various Asian ethnicities (e.g., Korean, Chinese, and Japanese). An analysis of conversations about reminiscing about one’s experiencess in Caucasian mother-child dyads and Korean mother-child dyads (the children’s age ranged between 3-4 years of age) revealed that Caucasian dyads talked on average as much as three times more than the Koreans dyads (Mullen & Soonhyung, 1995).
In addition, Caucasian mothers talked more during their turns and were more likely to portray the child as the protagonist in the talk, and to emphasize the child’s and others’ feelings and thoughts, whereas Korean mothers focused on norms, social roles, and emphasized behavioral expectations. This suggests that children’s linguistic experiences are related to the development of autobiographical memories, and that the latter are culturally modeled (Mullen & Soonhyung, 1995).
Another study that suggests the influence of the socio-linguistic environment on autobiographical memory at the early stages of childhood looked at conversations about children’s past experiences between American mother-child dyads and Chinese mother-child dyads (the children were 3 years old). The analysis of these conversations revealed that American mother-child dyads had an elaborative, independently oriented conversational style in which the focus was on the child’s predilections and opinions, whereas Chinese mother-child dyads had a low-elaborative, interdependently oriented conversational style in which the mother repeated factual questions and emphasized moral rules and behavioral expectations (Wang et al., 2000).
These results show that parent-child talk focuses on what types of events are considered memorable, on what aspects of those events are more important, on how to organize events in a temporal fashion, and on how to make inferences about people and causality. All these differ according to the values of a specific culture (Mullen & Soonhyung, 1995; Wang et al., 2000). Child talk appears to be more valued in Western societies, where children are encouraged to talk more about their experiences and talk more about themselves (Mullen & Soonhyung, 1995).
Other research also showing how the socio-linguistic environment in which children grow shapes their autobiographical memories has focused on memory specificity and the amount of detail found in young children’s memory reports. When American children (4 and 6 year old) were interviewed about a story presented to them a day before, they gave more voluminous and elaborate accounts for both their own experiences and for the story than did Korean children (Han, Leichtman & Wang, 1998). In addition, American children were more specific and descriptive about specific past events than both Korean and Chinese children, and the American children referred to emotions more and categorized negative emotions, whereas Asian children tried to emphasize the positive aspect of an event and talked more about other people than American children did.
This suggests that the content of memory differs with the cultural background of the individual (Han et al., 1998). Research involving preschoolers describing autobiographical events shows that American children’s memories are generally focused on specific events, individual experiences, and feelings (Wang, 2004). In addition, American children’s memories tend to be expressive, detailed and lengthy, and they focus on the child as being the protagonist in the narrative and present the child in a positive light. In contrast, Chinese children’s memories were found to be general, skeletal, less emotional, more neutral in their expression, and focused on routine events, on collective activities, on social interactions, on others or relations with others.
These patterns are seen because Western cultures promote autonomy and put an emphasis on the individual’s qualities, and children in these cultures are encouraged to stand out and talk about themselves, whereas Eastern cultures promote cohesiveness and put an emphasis on the group, and children in these cultures are discouraged to talk about themselves and the past and focus more on those around them (Han et al., 1998).
Cultural influences on memory persist into adulthood. In one study, American and Chinese college students were asked to recollect early childhood experiences, and they showed the same biases as American and Chinese preschoolers (Wang, 2001). In the study, the American and Chinese college students were asked to recollect their earliest childhood memories and provided self-descriptions. Results indicated that the American participant’s earliest childhood memories were from around the age of 3.5 years, whereas the Chinese participant’s earliest childhood memories were dating from approximately 4.1 years of age.
In addition, American college students’ memories were discrete, focused on specific events, and the individual’s feelings, whereas Chinese college students’ memories were more general, about routine activities, and focusing on family and in-groups. Americans also stressed personal preferences and autonomy in lengthier narratives than the ones reported by the Chinese.
When considering the influence of culture on autobiographical memory, it is important to realize that people can internalize more than one culture, in equal measure, so as to form a bicultural identity (Devos, 2006). For example, young adults in the United States in a Chinese family might be competitive and expose their achievements in the society at large, but inside their community and/or family, they will be respectful to their elders and try to blend in. Studies have primed bicultural individuals with one cultural identity or another, in order to see how that influences their behavior and cognition (Hong, Ip, Chiu, Morris & Menon, 2001; Wang, 2008). In one such study, Chinese Americans were primed with their Chinese cultural identity, by being given collective, Chinese-related statements. Results showed that the participants became more aware of their duties (e.g. “I ought to understand Chinese history,” “We have to pay taxes”).
However, Chinese Americans primed with their American cultural identity, by being given individualistic, American-related statements, they tended to become slightly more aware of their rights (e.g. “I can vote when I’m 18”) (Hong et al., 2001). Moreover, when Asian Americans were primed with their American self before recalling important autobiographical events, they were more likely to recall personal experiences in which they were the protagonists and they tended to emphasize their own perspective (e.g., “I got the acceptance letter for Cornell. I did not like my high school at the time and most of the people in, so this was very good news for me. […] I was getting out of town”) (Wang, 2008). However, when Asian Americans were primed with their Asian self before recalling important autobiographical events, their recollections were more likely to focus on social interactions, and persons from in-groups (e.g., “The day I got my letter of acceptance to Cornell gave me a sense of relief. […] So it’s not the fact of accomplishing that makes my parents happy […]. It’s the ability to plan”). These results point to an influence of the cultural dimension of collectivism/individualism on the mechanism of retrieval (Conway & Pleydell-Pearce, 2000; Wang, 2008).
Thus, previous research provides strong evidence that the cultural dimension of collectivism/individualism influences autobiographical memories, starting with their formation in early childhood and continuing its influence into adulthood, when it influences the retrieval of autobiographical memories.
In autobiographical memories, connections between self and the past are made. The self, as a personality, is considered as having three levels: traits, characteristic adaptations (cognitive, motivational, developmental components), and a life story (McLean & Fournier, 2008). But how do we know that these memories that compose our life story are actually events that we experienced? Was it really me that had a costume party for my 10th birthday? Did I really go the museum on a class trip as I think I went to? How do people differentiate between events that really happened and events that they only thought about, inferred, or imagined?
Research shows that memories of experienced events generally have more sensory and perceptual details than memories for events that did not really occur but were products of the imagination (Sporer & Sharman, 2006). Such qualitative details allow people to differentiate between memories of events that actually happened versus those that they only imagined happening (Johnson, Hashtroudi & Lindsay, 1993), and apparently people use the same criterion to judge the life-story narratives of other people (Sporer & Sharman, 2006).
Nonetheless, imagining events that never happened can have consequences, as the phenomenon of imagination inflation shows. Imagination inflation refers to an increase in confidence that a fictional event that was imagined actually happened (Garry, Manning, Loftus & Sherman, 1996; Garry, Sharman, Wade, Hunt & Smith, 2001). The classic way of testing imagination inflation is with the Life Events Inventory (LEI), and the methodology consists of three steps. First, participants complete an LEI consisting of a long list of possible childhood events that they rate their confidence that each event has or has not happened.
Second, participants are asked to imagine a series of events that were selected from those previously rated as unlikely to have happened to them in childhood. Lastly, participants complete another LEI that included the critical events they had earlier imagined (Garry et al., 1996; Garry et al., 2001; Henkel & McDonald, in press). When participants were asked to imagine target events that were rated on a previous LEI by the participants as unlikely to occur, a positive change in the ratings of confidence that these events actually happened was observed. Specifically, 34% of the events imagined had increased confidence ratings, as compared to only 25% of events not imagined that had increase confidence rating on the second LEI (Garry et al., 1996). It another study, some of the participants imagined adulthood events, whereas other imagined childhood events and both groups showed imagination inflation (Sharman & Barnier, 2008).
Imagination not only shapes people’s memory of the past but can also influence their expectations about the future. In one study, participants had to imagine for 1 minute various events, of which some were past events (e.g, “I fell asleep during a religious service”) and some were future events (e.g., “I will win a prize for a piece of art”).These events were ones that the participants had previously rated low in confidence that they had or will happen (Henkel & McDonald, in press). Some of the participants not only imagined but also described aloud what they had imagined. Afterwards all participants were asked to rate both the vividness and amount of detail for each imagined event.
On a subsequent LEI, participants gave higher confidence ratings for both past and future events that were imagined, compared to events that were not imagined, and imagined past events showed a greater change in confidence ratings than imagined future events. Interestingly, events that were only imagined showed a marginally greater change in confidence ratings than events that were both imagined and described aloud, and effects were stronger for the participants that imagined these events more vividly and detailed.
A natural question would be why does the phenomenon of imagination inflation happen? The source monitoring framework argues that people’s memory for an event (whether perceived or imagined) contains perceptual, emotional, and semantic information, and spatial and temporal details (Johnson, 1988; Mitchell & Johnson, 2009), and that people sometimes make misattributions based on those features. When imagining an event, especially if it is imagined vividly, the information about the details of the event may be more accessible for retrieval but the source can be confused, especially when the event has happened way back in the past and thus access to retrieval of the details and the source is harder to achieve. These processes can cause the phenomenon of imagination inflation (Garry, Manning, Loftus & Sherman, 1996).
For example, remembering you broke a window with your hand, although you did not, can be a consequence of imagining the event and thus making it more accessible and more vivid in your mind. The fact that the event is situated in a period of time when you were 6 or 7 years old and thus the source of the memory is less accessible than the details of the event aids to the appearance of this phenomenon. Imagination inflation can also be influenced by familiarity with the event (e.g., if the participant has ever seen anyone else doing the action or involved in the event he or she has to imagine) and the plausibility of the event (e.g., how likely is the participant to have seen an animal give birth if he or she has grown up in a metropolis) (Garry et al., 1996). Especially prone to this phenomenon are recent, positive events, probably because they support the sense of self and they make us feel as if we have improved over time (Sharman & Barnier, 2008).
Not only can imagination increase confidence levels that an event has actually happened but it can also create false autobiographical memories. One study asked participants to complete an LEI that contained a frequently occurring event and an event that does not occur (Mazzoni & Memon, 2003). Subsequently, participants were asked to either imagine the frequently occurring event and a read a passage about the nonoccuring event or to imagine the nonoccuring event and read about the frequently occurring event. Results showed that imagination alone increased the confidence that the nonoccuring event actually occurred and even produced false memories of the event. However, just increasing the familiarity with the event (i.e., reading about it) did not produce false beliefs and memories. Another study reflects similar results, but is more culturally relevant (Pezdek, Finger & Hodge, 1997). Catholic and Jewish high school students were read three true events and two false events (one Catholic and one Jewish), from the time they were 8 years old.
After a week, students were asked if they had any additional memories about the five events. Catholics tended to show recall for the Catholic false event, whereas Jews tended to show memories for the Jewish false event. These results suggest that false memories can be created for more plausible, culturally relevant events. In yet another study, participants were given an LEI and after a week were presented with four target events, out of which 2 were presented as plausible and 2 as implausible (Pezdek, Blandon-Gitlin & Gabbay, 2006). Participants had to imagine one of each category of target events. A week after this, the participants were asked to fill another LEI. Results showed that imagining events supposedly high in plausibility determined an increase in ratings from the first to the second LEI, whereas imagining events supposedly low in plausibility did not affect ratings.
In a laboratory study where careful control over participants’ experiences was possible, similar findings were obtained. Participants heard an action and then either performed or imagined it (Goff & Roediger, 1998). Later on, participants had to imagine once or several times either new actions or some of the actions that they had previously performed or imagined. The results showed that the more times the participants imagined the actions, the more likely they were to later state that they had initially performed those actions, although they did not do so the first session and even when they had never heard those actions mentioned before. This suggests that not only the plausibility of the event had a role in creating false memories, like the previous studies showed, but that the number of times the events are imagined influences this process, as well. Specifically, the more times the events are imagined, the more likely people are to now claim that the events really occurred in their childhood.
The good news is that if the memory of an imagined event has more than one character, aside from oneself, the person can confront these other characters (e.g., a sibling) and recant their previous “recollection” (French, Sutherland & Garry, 2006). The bad news is that, especially in the case of children, some of these memories take wildly unrealistic forms but are nonetheless experienced as something realistic that has actually happened to them. In one study, children were told about unlikely or impossible events (e.g., going on a trip to the moon) and some of them were let to draw at their leisure afterwards (Strange, Garry, & Sutherland, 2003). Finally, children were asked if those unlikely or impossible events previously presented had ever happened to them. Results showed that when recall is done after drawing something, even if the drawing is not at all related to the events previously presented, children had a strong tendency to “remember” those events happening to them.
The Present Study
Culture has a strong effect on how a person experiences the world (Desai, 2007; Ozawa et al., 1996; Skillman, 2000; Trafarodi & Smith, 2001; Triandis, 1993), and it also influences how a person remembers the world and what he or she will remember (Han, Leichtman & Wang, 1998; Mullen & Soonhyung, 1995; Nelson & Fivush, 2004; Wang, Leichtman & Davies, 2000; Wang, 2001, 2004, 2008). The present study investigates the influence of the cultural dimension of collectivism/individualism on autobiographical memories through the paradigm of imagination inflation.
Is imagination inflation more prone to occur for culturally relevant events? Specifically, will people from Eastern cultures show greater imagination inflation for autobiographical events that emphasize the dimension of collectivism (e.g., “My family, my friends, and I drove across country”), and people from Western cultures show greater imagination inflation for autobiographical events that emphasize the dimension of individualism (e.g., “I won a gold medal or first place in a competition”)? The present study sought to answer these questions, drawing on previous research on imagination inflation and the influence of the cultural dimension of collectivism/individualism on autobiographical memory.
Participants completed a 36-item LEI, containing both collectivistic (e.g., “My grandparents used to help me with my homework”) and individualistic items (e.g., “I cheated on a test at school”). Twelve target, low-probability events (six collectivistic and six individualistic) were selected. Half of the participants were asked to imagine six of them (probed events), and their confidence levels on these were compared to their confidence levels on the other six events that they did not imagine (not probed). One week later, they completed another LEI containing the 12 target events but also new items not presented on the first LEI. The key variable of interest is the extent of changes in the participants’ confidence that the events occurred from the first session to the second session.
Imagination inflation is shown when people are more confident that events that were imagined (probed events) had happened when they were children, compared to the participants’ confidence levels on the not probed events, and we looked at this difference across the two sessions, as well as at the absolute levels of confidence in Session 2. It was also expected that participants coming from a Western background will show increased confidence levels for the individualistic events imagined than for the three collectivistic events.
For the participants coming form an Eastern background, the opposite trend is expected. Such findings would indicate that people are more susceptible to the impact of imagination on memory for culturally relevant events. To this end, the INDCOL scale was used, so as to decide the individualistic or collectivistic background of each participant and how this matched with their Western or Eastern background.
Forty participants (31 coming from a Western cultural background, and 9 coming from an Eastern cultural background) participated in the study, 29 females and 11 males. Participants were college students from Fairfield University, with an age range of 18-22 years (M= 19.18; SD= 1.11). They participated either to fulfill a class requirement or to receive extra credit.
Two separate Life Events Inventories (LEI) were created, each presenting a list of various childhood events and experiences. Some of the items selected were used in previous research (e.g., “I won a gold medal or first place in a competition”)Henkel & McDonald, in press), and some were generated specifically for this study (e.g., “I used to be afraid of spiders”). See Appendix A for a complete list of all events. Each inventory included both positive events (e.g., “I had my picture taken for a newspaper”) and negative events (e.g., “I broke or fractured a bone”), and ones that were more collectivistic (e.g., “My family, friends and I drove across country”) or more individualistic events (e.g., “I won a gold medal or first place in a competition”).
Collectivistic events were defined as general activities or related to general, on-a-daily-basis activities (e.g., “My grandparents used to help me with my homework”), involved the participant in a social context with the accent being put on the context not on the self (e.g., “A team I played on won a big game”), or involved only people of the ingroup (family, kin, friends) (e.g., “My sibling or cousin got in trouble for calling 911”). Individualistic events were defined as events involving only the participant and focus heavily on the self as an autonomous being. This included positive events relating accomplishments (e.g., “I won a stuffed animal at a carnival”), extraordinary events (e.g., “I saw a total solar eclipse”), and fears of the self (e.g., “I used to be afraid of spiders”).
The first LEI included a total of 36 events, out of which 12 target events were preselected as having a low probability of occurring in the lives of the participants before the age of 10. Six of these twelve target events were used as probed events for the imagination task and 6 as not probed events. The specific items were counterbalanced in two sets across participants. Two additional filler events (one collectivistic and one individualistic) were also used as probed events. Both were high probability events (e.g., “A team I played on won a big game,” “I cried the first time I went to the dentist”).
The second LEI consisted of 36 events. Sixteen of these events were new, not included on the first LEI. These events included positive and negative events, as well as more collectivistic or more individualistic events. Randomly intermixed were the 12 target events, and the 2 fillers.
The Social Orientation Inventory was presented to all participants for rating (see Apppendix B). It consisted in 28 statements (e.g., I would enjoy functioning in a high-level, decision making capacity”), that had to be rated on a scale from 1 (strongly disagree) to 7 (strongly agree). Another scale presented to all participants was the Vividness of Visual and Auditory Imagery Questionnaire (see Appendix C). It consisted of 16 items (e.g., “The different colors worn in some familiar clothes”) that had to visualized and then rated on a scale from 1 (no image at all) to 5 (perfectly clear and vivid as normal vision). It also consisted of 7 items (e.g., “Imagine the sound of water dripping”) that had to be rated on a scale from 1 (no sound at all) to 7 (perfectly clear and vivid as normal hearing).
The shortened Individualism-Collectivism (INDCOL) Scale (36 items) was used for all participants, to assess their level of collectivism/individualism (Hui, 1994), and is included in Appendix D. The INDCOL is used to measure feelings, beliefs and behaviors at the individual level that are in line with collectivism or individualism. It consists of subscales addressing relations with one’s spouse (e.g., The decision of where one is to work should be jointly made with one’s spouse, if one is married”), parents (e.g., “I would not share my ideas and newly acquired knowledge with my parents”), kin (e.g., “When deciding what kind of education to have, I would definitely pay attention to the views of relatives of my generation”), friends (e.g., “To go on a trip with friends makes one less free and mobile. As a result, there is less fun”), neighbors (e.g., “I don’t really know how to befriend my neighbors”), and co-workers (e.g., “If a colleague lends a helping hand, one needs to return the favor”).
Participants had to read each of the 36 statements and rated it as T (True) or F (False), with the exception of the last item, which had 6 options to choose from. For the reliability coefficients and the validity values of the shortened form of the INDCOL refer to Hui and Yee (1994). The reliability coefficients of the test-retest, split-half and Cronbach of the original INDCOL were in the .60s, and were considered “a satisfactory value considering the complexity and multifacetedness of the construct” (Hui, 1988). For the validity values, refer to Hui (1988).
Design and Procedure
This study used a 2 x 2 x 2 factorial design where the participants’ cultural background (Western or Eastern) was a between-subjects variable, while the focus of the event (individualistic vs. collectivistic) and the imagery condition (imagined events vs. not imagined events) were within-subjects variables. The main dependent variable was the change in the overall confidence of the event’s occurrence from the first session to the second session. Many other studies using LEIs also examine the percentage of participants who showed increased confidence across the two sessions for probed vs. not probed events, and the absolute levels of confidence in session 2, so both of these dependent variables were examined, as well.
Participants were tested individually in a study advertised to them as being about people’s lifetime experiences and their personalities. All participants were asked to complete the first LEI. They were asked to evaluate how confident they are that each event listed has actually occurred in their life before the age of 10, by checking one of the options: Occurred or Did not occur. Then, they rated their option in terms of their confidence level, on a scale from 0 % (absolutely no confidence that the event occurred/did not occur) to 100 % (absolutely confident that the event occurred/did not occur). There was not time limit for rating all the events, but most people finished in about 5 minutes.
Following the LEI, participants were told that the next task would require them to imagine several events as clearly and vividly as possible as if they had actually happened to them in childhood. Twelve events from the initial Life Events Inventory were preselected for the imagination task. Three collectivistic events and three individualistic events were given, in randomized order for participants on the imagination task. Participants were asked to imagine each event for 60 seconds, as vividly and detailed as possible, and to picture themselves in the situation, thinking about the surroundings, about other people that might be involved, and the emotions the participants felt in relation to these events.
Half of the participants received 6 of the 12 target events as probed events and the other half received the other 6 events as probed events. In addition, there were two more events (one collectivistic and one individualistic) that were imagined by all participants (“fillers”), as the first and last events to be imagined, with the order counterbalanced between participants. All participants rated on a questionnaire how vivid and detailed they imagined each event, on a scale from 1 (not at all vivid/not at all detailed) to 7 (extremely vivid/extremely detailed). At the end of Session 1, participants also received the Social Orientation Inventory. Overall, the first session did not extend beyond 20 minutes.
Participants returned for a second session one week later and were told that past events were again examined in another Life Events Inventory. Participants were told that this inventory was different from the first one, including a number of new items. Subsequently, participants were asked to complete a Vividness of Visual and Auditory Imagery Questionnaire, by imagining each statement and rating it on a scale of clarity and vividness of imagine or sound and the INDCOL scale, by reading each statement and rating it as either T (True) or F (False).
This study sought to determine whether the participant’s cultural background might influence how confident the participant became that an event happened to them as a child after imagining that event. Participants’ confidence was calculated by multiplying their confidence ratings from 0-100% for events rated as “did not occur” by -1. This scalar confidence score therefore associates scores of-100 with high confidence ratings that the event did not occur. Moreover, scores of 0 correspond to a total lack of confidence in that the event has or has not happened (participant is guessing), and 100corresponds to a total confidence that the event did happen.
Changes in Confidence of Event’s Occurrence
Two different measures of confidence were used to examine the extent of imagination inflation exhibited. One measure was based on the levels of confidence from the second session for the targeted events, and the other measure was based on the change in the overall confidence of the event’s occurrence from the first session to the second session for various events. A 2 x 2 x 2 mixed factorial analysis of variance (ANOVA) was conducted on scaled confidence scores from the second session, with the participants’ cultural background (Western or Eastern) as a between-subjects variable, and the focus of the event (individualistic vs. collectivistic) and the imagery condition (being imagine vs. not being imagined) as within-subjects variables (see Tables 1 and 2). The main effect of the imagery condition on confidence scores was significant, F (1, 38) = 7.22, p = .01, with greater confidence levels for probed events (M = -16.14) than for not probed events (M = -38.46). Moreover, the main effect of the event type on the confidence scores was marginally significant, F (1, 38) = 3.10, p = .09, with greater confidence scores for individualistic events (M = -20.78) than for collectivistic events (M = -34.10). The main effect of cultural background on confidence scores was not significant, F (1, 38) = 0.05, p = .83, with no overall difference in confidence scores between participants coming from a Western cultural background (M = -26.26) and participants coming from an Eastern cultural background (M = -28.61).
However, there was a significant interaction between the participants’ cultural background and the focus of the event, F (1, 38) = 7.40, p = .01. When the focus of the event was individualistic, people from an Eastern cultural background were slightly more confident (M = -11.67, SE = 11.28) than people from a Western cultural background (M = -29.89, SE = 6.08). However, when the focus of the event was collectivistic, people from an Eastern cultural background (M = -45.56, SE = 12.27) were much less confident than the people from a Western cultural background (M = -22.63, SE = 6.61).
Means and Standard Errors for Main Effects of 2x2x2 Mixed Factorial ANOVA Conducted on Confidence Levels from The Second Session
|Variable||Marginal Means||Standard Errors|
|Type of event||Individualistic||-20.78||6.41|
Means and Standard Errors for Ethnicity x Type of Event Interaction of 2x2x2 Mixed Factorial ANOVA Conducted on Confidence Levels from The Second Session
|Ethnicity||Type of event||Means||Standard Errors|
A separate 2 x 2 x 2 mixed factorial ANOVA was conducted on changes in confidence from Session 1 to Session 2 by taking into consideration the difference between the scores form the second LEI and the scores form the first LEI, for the targeted events. A positive number represents an increase in confidence, whereas a negative number represents a decrease in confidence from the first to the second session that the event actually occurred in the participant’s childhood. A score of 0corresponds to a lack of change in scores from the first to the second session. Participants’ cultural background (Western or Eastern) was a between-subjects variable, and the focus of the event (individualistic vs. collectivistic) and the imagery condition (being imagine vs. not being imagined) were within-subjects variables (see Tables 3 and 4).
The main effect of cultural background on change in confidence ratings was significant, F (1, 38) = 6.08, p = .02, with greater positive change in confidence ratings for participants coming from an Eastern cultural background (M =18.15) than for participants coming from a Western cultural background (M = 2.37). Moreover, the main effect of the imagery condition on change in confidence ratings was significant, F (1, 38) = 6.59, p < .05, with greater positive change in confidence ratings for probed events (M = 17.05) than for non-probed events (M = 3.46). Finally, the main effect of the event type on the change of confidence ratings was marginally significant, F (1, 38) = 2.99, p = .09, with greater positive change in confidence ratings for individualistic events (M = 15.68) than for collectivistic events (M = 4.84).
There was also a significant interaction between the participants’ cultural background and the focus of the event, F (1, 38) = 5.50, p < .05. When the focus of the event was individualistic, people from an Eastern cultural background showed a greater change in confidence ratings (M = 30.93, SE = 8.11) than people from a Western cultural background (M = 0.43, SE = 4.36). However, when the focus of the event was collectivistic, people from an Eastern cultural background (M = 5.37, SE = 7.66) did not differ in their change in confidence ratings much from the people from a Western cultural background (M = 4.30, SE = 4.13).
Means and Standard Errors for Main Effects of 2x2x2 Mixed Factorial ANOVA Conducted on Changes in Confidence Levels from The First to The Second Session
Means and Standard Errors for Ethnicity x Type of Event Interaction of 2x2x2 Mixed Factorial ANOVA Conducted on Change in Confidence Levels from The First to The Second Session
Correlational analyses were also run to assess whether the change in confidence ratings was related to the participants’ overall imagery abilities, as shown by their score on the VVIQ scale. A marginally significant positive correlation was found between the mean vividness of visual imagery and the change in confidence ratings for individualistic, probed events, r = .29, p = .07.
No significant correlation was found between the mean VVIQ and the change in confidence ratings for collectivistic, probed events, r = 0.00, p = 1.00. Also, no significant correlation was found between the mean VVIQ and the change in confidence ratings for individualistic, not probed events, r = 0.12, p= .45. Finally, no significant correlation was found between the mean VVI and the change in confidence ratings for collectivistic, not probed events, r = -0.12, p = .46.
Qualitative Characteristics of Imagery
Ratings of overall vividness and amount of detail generated during the imagery task were examined to determine whether participants were indeed able to create mental images of past events and whether cultural differences existed. No significant difference in vividness and amount of detail were seen between people coming from a Western cultural background (M = 4.24, SD = 0.94) and people coming from an Eastern cultural background (M = 4.48, SD = 0.92), t (38) = -0.66, p = .51. Ratings of4 corresponded to “somewhat vivid and detailed,” while ratings of 7 corresponded to “extremely vivid and detailed.” Hence, overall, people did create relatively vivid and detailed images.
We also examined overall imagery abilities based on the standardized VVIQ scale. The mean VVIQ score was not significantly different for people coming from a Western cultural background (M = 3.86,SD = 0.46) than for people coming from an Eastern cultural background (M = 3.90, SD = 0.65), t (38) = -0.20, p = .84. Average ratings correspond to “moderately clear and vivid” to “clear and reasonably vivid” images.
The INDCOL Scale was used to assess the degree of collectivism/individualism exhibited by each participant. The scoring was done by giving one point to each answer that was collectivistic in nature (see Appendix D for information on items that were reverse coded). Thus, the higher the score, the more collectivistic the person is. Unexpectedly, the degree of collectivism exhibited was significantly higher for people coming from a Western cultural background (M = 23.68, SD = 3.72) than for people coming from and Eastern cultural background (M = 19.67, SD = 2.74), t (38) = 3.00, p = .01.
This study sought to determine whether the participant’s cultural background might influence the magnitude of the change in confidence levels for the dimension of collectivism/individualism that is emphasized by that cultural background. That is, the two significant main effects of the imagery condition on confidence levels from the second session and on the change of confidence ratings from the first to the second session demonstrate the imagination inflation effect.
Overall, people had higher confidence that the events happened from the probed events, compared to the not probed events. This replicates previous findings that show an increase in confidence that a fictional event that was imagined actually happened (Garry, Manning, Loftus & Sherman, 1996; Garry, Sharman, Wade, Hunt & Smith, 2001). Imagining the event makes it more accessible and more vivid in your mind. The fact that the event is situated in a period of time when you were 10 years old and thus the source of the memory is less accessible than the details of the event aids to the appearance of this phenomenon.
Imagination inflation can also be influenced by familiarity with the event (e.g., if the participant has ever seen anyone else doing the action or involved in the event he or she has to imagine) and the plausibility of the event (e.g., how likely is the participant to have seen an animal give birth if he or she has grown up in a metropolis) (Garry et al., 1996).
The two marginally significant main effects of the event type on the confidence levels from the second session and on the change in confidence ratings from the first to the second session illustrates that the type of event had somewhat of an effect but not in the direction predicted, as the interaction between the participants’ cultural background and the focus of the event will show. Previous studies reflect similar results (Pezdek, Finger & Hodge, 1997), such that Catholics tended to show recall for the Catholic false event presented to them at a previous session, whereas Jews tended to show memories for the Jewish false event.
These results suggest that false memories can be created for more plausible, culturally relevant events. Thus, collectivistic and individualistic events had different effects on the participants, such that participants were more confident that the individualistic events happened in their childhood than the collectivistic events. This may be due to the fact that many of the individualistic events were more straightforward and probably more relevant than the more general collectivistic events.
The main question that this study set to answer was if a person’s cultural background does have an effect on imagination inflation. The contradictory results on the main effect of the participants’ cultural background on the confidence levels from the second session and the change in confidence ratings from the first to the second session, such that the former was not significant and the latter was significant, is probably due to the small sample of people from an Eastern cultural background.
Nonetheless, it was expected that an interaction will be seen, so that participants would show an increase in confidence ratings for the collectivistic events when the participants who rated them were from an Eastern background. Conversely, participants coming from a Western cultural background were expected to show an increase in confidence ratings for individualistic events. This interaction would indicate that the phenomenon of imagination inflation was influenced by the cultural background of the participants, and specifically by the collectivism/individualism dimension.
This could be due to the fact that a culture-specific way of encoding autobiographical information is learned since childhood (Mullen & Soonhyung, 1995; Wang et al., 2000). Thus, because the cultural dimension is so pervasive, it was to be expected that participants would show a tendency to be more confident that an event really happened after imagining it, provided that the event corresponds with the individual’s background. Of course, multiple processes can be involved, aside from the processes already considered by the socio-cultural developmental theory (Nelson & Fivush, 2004), but influenced by the latter, such as the vividness, plausibility of event and amount of detail, all which can be related back to the source monitoring framework. This shows that not only cultural processes are at work, but cognitive ones, as well.
The interaction was, indeed, significant, but it showed the opposite trend from the one expected, such that when the focus of the event was individualistic, people coming from an Eastern cultural background were slightly more confident when considering their confidence levels form the second session and showed a greater change in confidence ratings form the first to the second session. However, when the event was collectivistic, the people coming from an Eastern cultural background were less confident when considering their confidence levels from the second session and did not differ in the change of confidence ratings from the first to the second session.
This is the opposite of what was hypothesized, namely that participants coming from a Western background will show increased confidence levels for the individualistic events imagined than for the three collectivistic events. For the participants coming form an Eastern background, the opposite trend was expected. However, even these surprising results might actually find their answer in the culturally specific childhood memories encouraged in a specific society.
Previous research had shown the influence that culture has on autobiographical memory. Parents, and especially mothers, are the ones that in the first years of life provide the structure and content of discussion with the children (Han et al., 1998; Mullen & Soonhyung, 1995; Nelson & Fivush, 2004; Wang et al., 2000; Wang, 2004). Some mothers are more elaborate in their style, talk frequently, and give rich information with each question they put to their children, even when the latter do not answer, and consequently their children incorporate more information in their narratives about the past and overall recall more, whereas mothers with a less elaborate style pose more redundant questions and consequently their children talk less and incorporate little information in their narratives.
Eastern cultures use personal narratives to express moral and social standards, while Western cultures use them for the entertainment value, and focus more on the emotions of the protagonist(s) (Han et al., 1998; Mullen & Soonhyung, 1995; Nelson & Fivush, 2004; Wang et al., 2000; Wang, 2004). This trend continues into adulthood, and is observable among people with bicultural identity (Devos, 2006; Hong et al., 2001; Nelson & Fivush, 2004; Wang, 2000, 2008).
This suggests that people coming from an Eastern cultural background tend to have few, more general memories, whereas people coming form a Western cultural background tend to have more clear, cohesive, highly specific memories. For this reason, it may be that the people from an Eastern cultural background are especially susceptible to highly specific and clear childhood memories, of which they do not have any recollections, but that they can easily incorporate in their more general memories. However, the people coming from a Western cultural background, because of their much clearer and specific memories, are not as susceptible to show imagination inflation on individualistic type of events.
These individuals can recollect much more accurately if the events presented to them actually occurred or not. On the other hand, if the event presented was a collectivistic one, there seems to be no difference between people coming from these two different cultural backgrounds, and when there is a difference, the people who actually show more confidence are the ones coming from a Western cultural background, compared to those coming from an Eastern cultural background. This phenomenon is probably due to the fact that people coming form an Eastern cultural background are not so susceptible to these types of events, since their childhood memories are more general, and since their main focus falls on the individualistic events, of which they probably are attracted and tend to pay more attention to.
Conversely, people coming from a Western cultural background might be more susceptible to show imagination inflation on such events, since their memories are highly specific and these general, collectivistic events might seem probable.
Finally, it should be noted that the ratings of overall vividness were moderately high, indicating that the participants were able to create real and vivid imagery. Another surprising result was rendered by the INDCOL scale. According to this scale, people from a Western cultural background were more collectivistic, when compared to people from an Eastern cultural background. This might have occurred because of the small number of people from an Eastern cultural background, but also because the scale was not really appropriate (e.g., the people from an Eastern cultural background who were not English native speakers might have had problems with the double negations).
Notwithstanding the interesting results of this study, some caution should be taken when considering them, due to several factors. One possibility would be that the methodology was flawed, such that the items should have been more specific, or that additional personality variables should have been controlled for before administering the LEIs. Another possibility is that the sample might have been insufficiently large or biased in the sense that individuals that are bicultural, for example, might adapt to both sets of target items, “switching” between their two cultural identities, as need be.
Many of the people coming from an Eastern cultural background had actually been born in an Asian country but had lived for many years in the United States. Also, this adaptation to both cultures is a documented phenomenon, since it is necessary for bicultural individuals to react in this way in society, according to which community they find themselves in (Devos, 2006; Hong et al., 2001). Previous research has shown that when Asian Americans were primed with their American self before recalling important autobiographical events, they were more likely to recall personal experiences in which they were the protagonists and emphasized their own perspective. Conversely, when Asian Americans were primed with their Asian self before recalling important autobiographical events, their recollections were more likely to focus on social interactions, and persons from in-groups (Wang, 2008). Accordingly, bicultural individuals might have adapted to the events they had to imagine, and ultimately showed an increase in confidence levels for both types of events.
Thus, future research should employ a larger overall sample, but also a specifically larger sample of people coming from an Eastern cultural background. Also, considering the adaptations that bicultural individuals can employ, it might also be important to use a more distinct sample, comparing, for example, American individuals with Chinese individuals (who did not live in a Western country). Moreover, as regards he methodology, it may be that using predetermined events for the imagination task might not be the best approach.
Thus, considering each participant’s answers on the first LEI, events that received lower confidence ratings can be chosen, and each set of events will be unique to each individual. The INDCOL scale might be another issue, because of its double negations. Therefore, a more culturally and linguistically sensitive scale might be needed in order to correctly assess the individualism/collectivism dimension of culture for each individual.
Ambady, N. & Bharucha, J. (2009). Culture and the Brain. Current Directions in Psychological Science, 18, 342-345.
Conway, M.A. & Pleydell-Pearce C.W. (2000). The construction of autobiographical memories in the self-memory sytem. Psychological Review, 107, 261-288.
Desai, J. (2007). Intergenerational conflict within Asian American families: The role of acculturation, ethnic identity, individualism, and collectivism. Dissertation Abstracts International, 67, 7369.
Devos, T. (2006). Implicit bicultural identity among Mexican American and Asian American college students. Cultural Diversity and Ethnic Minority Psychology, 12, 381-402.
French, L., Sutherland, R. & Garry, M. (2006). Discussion affects memory for true and false childhood events. Applied Cognitive Psychology, 20, 671-680.
Garry, M., Manning, C.G., Loftus, E.F. & Sherman, S.J. (1996). Imagination inflation: Imagining a childhood event inflates confidence that it occurred. Psychonomic Bulletin & Review, 3, 208-214.
Garry, M., Sharman, S.J., Wade, K.A., Hunt, M.J. & Smith, P.J. (2001). Imagination inflation is a fact, not an artifact: A reply to Pezdek and Eddy. Memory and Cognition, 29, 719-729.
Goff, L.M. & Roediger, H.L. (1998). Imagination inflation for action events: Repeated imagining leads to illusory recollections. Memory and Cognition, 26, 20-33.
Han, J.J., Leichtman, M.D. & Wang, Q. (1998). Autobiographical memory in Korean, Chinese, and American children. Developmental Psychology, 34, 701-713.
Henkel, L.A. & McDonald, C. (in press). Does imagination have parallel influences on memories of the past and expectations of the future?
Hong, Y., Ip, G, Chiu, C, Morris, M.W. & Menon, T. (2001). Cultural identity and dynamic construction of the self: Collective duties and individual rights in Chinese and American cultures. Social Cognition, 19, 251-268.
Johnson, M. (1988). Reality monitoring: An experimental Phenomenological approach. Journal of Experimental Psychology, 117, 390-394.
Johnson, M., Hashtroudi, S. & Lindsay, D.S. (1993). Source monitoring. Psychological Bulletin, 114, 3-28.
Mark, H.R. & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological review, 98, 224-253.
Mazzoni, G. & Memon, A. (2003). Imagination can create false autobiographical memories.Psychological Science, 14, 186-188.
McLean, K., & Fournier, M. (2008). The content and processes of autobiographical reasoning in narrative identity. Journal of Research in Personality, 42, 527-545.
Mitchell, K.J. & Johnson, M.K. (2009). Source Monitoring 15 years later: What have we learned from fMRI about the neural mechanisms of source memory?. Psychological Bulletin, 135, 638-677.
Mullen, Mary K., and Soonhyung Y. (1995). The cultural context of talk about the past: Implications for the development of autobiographical memory. Cognitive Development, 10, 407-419.
Nelson, K. & Fivush, R. (2004). The emergence of autobiographical memory: A social cultural developmental theory. Psychological review, 111, 486-511.
Pezdek, K., Finger, K., & Hodge, D. (1997). .Planting false memories: The role of ebent plausibility.Psychological Science, 8, 437-441.
Pezdek, K., Blandon-Gitlin, I., & Gabbay, P. (2006). Imagination and memory: Does imagining implausible events lead to false autobiographical memories?. Psychonomic Bulletin & Review, 13, 764-769.
Sharman, S.J. & Barnier A. J. (2008). Imagining nice and nasty events in childhood or adulthood: Recent positive events show the most imagination inflation. Acta Psychologica, 129, 228-233.
Skillman, G. (2000). Intergenerational conflict within the family context: A comparative analysis of collectivism and individualism within Vietnamese, Filipino, and Caucasian families. Dissertation Abstracts International, 60, 4910.
Sporer, S.L. &Sharman, S.J. (2006). Should I believe this? Reality monitoring of accounts of self-experienced and invented recent and distant autobiographical events. Applied Cognitive Psychology, 20, 837- 854.
Strange, D., Garry, M. & Sutherland, R. (2003). Drawing out children’s false memories. AppliedCognitive Psychology, 17, 607-619.
Triandis, H. (1993). Collectivism and individualism as cultural syndromes. Cross-Cultural Research: The Journal of Comparative Social Science, 27, 155-180.
Wang, Q., Leichtman, M.D. & Davies, K.I. (2000). Sharing memories and telling stories: American and Chinese mothers and their 3-year-olds. Memory, 8, 159-177.
Wang, Q. (2001). Culture effects on adult’s earliest childhood recollection and self-description: Implications for the relation between memory and self. Journal of Personality and Social Psychology, 81, 220-233.
Wang, Q. (2004). The emergence of cultural self-constructs: Autobiographical memory and self- description in European American and Chinese children. Developmental Psychology, 40, 3-15.
Wang, Q. (2008). Being American, being Asian: The bicultural self and autobiographical memory in Asian Americans. Cognition, 107, 743-751.