Tell me what you remember and I will know who you are: The link between collective memory and social categorization

Abstract

The present article aims to show that collective memories could serve as a criterion in social categorization. We predicted that a target person who shares common collective memories will be perceived as similar (to the self), relatively more favorably and categorized as an ingroup member. We conducted four studies using memories of historical events or childhood objects. These studies consistently showed that a target who shares common memories is more likely to be perceived as an ingroup member than someone who does not. This effect is mediated by perceived similarity to the self. Finally, individuals who share common memories are perceived more favorably than when they do not. However, according to the type of collective memories (historical events vs. childhood memories) sharing memories impacts either perceived competence or perceived warmth. The current sets of studies support the idea that collective memory influences social categorization processes, exemplifying its group identity-defining function.

Keywords

collective memory social categorization social identity social perception

The link between memory and identity has been frequently suggested in the literature. Since Locke (1869/2009), an individual’s memories have been shown to be related to his or her personal identity (Greish, 2001). That is, autobiographical memory system contains memories that define an individual’s identity (Conway, 2005). Along the same line, Halbwachs (1950/1997) suggested that collective memory is linked to social identity: collective memory is a defining feature of the society and groups to which people belong. Group membership thus defines identity within which a current of collective thoughts (e.g., representations) is shared. In this perspective, collective memory can be seen as a set of representations (Jedlowski, 2001; Licata & Klein, 2005; Licata, Klein, & Gely, 2007) and beliefs on the past (Dudai, 2002; B. Schwartz, 2008) related to a common identity of group members. Collective memory is thus shared (Dudai, 2002; B. Schwartz, 2008; Wang, 2008; Wertsch & Roediger, 2008; Zerubavel, 1996) and emerges from social interactions between group members (Jedlowski, 2001; Wang, 2008). Therefore, as social representations, collective memory can cover a large set of objects from the past and is shared and collectively produced within a culture or a specific group (Moscovici, 2008; Rateau, Moliner, Guimelli, & Abric, 2011).

In this sense, collective memory could act (partly) as markers that define the (in)group and its identity. In the present paper, we aimed to explore the link between collective memory and social identity. Particularly, we examined how collective memory influences social categorization and thus may represent one of the criteria to determine membership.

Collective Memory and Social Identity

As stated by Licata and Klein (Licata & Klein, 2005), collective memory is tightly linked to social identity. This relationship is bidirectional (Apfelbaum, 2000; Liu, 1999; S. H. Schwartz, Struch, & Bilsky, 1990; Wertsch & Roediger, 2008): on the one hand, social identity influences the (re)construction of collective memory (Baumeister & Hastings, 1997; Dresler-Hawke, 2005) for identity protection or enhancement purposes. On the other, collective memory influences social identity. The latter refers to collective memory’s function in defining social identity or “who we are” as a group.

Previous studies have mainly focused on the enhancement and protection functions of collective memory (Licata et al., 2007). Drawing a parallel between rumors (Allport & Postman, 1945) and collective memory, Paez, Marques, Valencia, and Vincze (2006) suggested that collective memory protects social identity of group members who remember. To this end, collective memories could be distorted or reconstructed through a set of processes. The elements of a memory that could harm social identity would be omitted, whereas elements that enhance the group’s identity would be highlighted (Baumeister & Hastings, 1997; Rateau, 2002). For example, although historical studies showed that German population largely supported the Nazi regime, German students reported that their grandparents were instead victims of the Nazi regime (Dresler-Hawke, 2005). Likewise, Sahdra and Ross (2007) showed the influence of group membership on the recall of memories that was biased to protect group’s positive social identity: high identifiers recalled fewer memories during which their group showed hostility towards an outgroup than did low identifiers. The role of collective memory on identity enhancement seems thus established: social identity shapes memories through reconstruction, selection, and maintenance in order to achieve or maintain a positive social identity (Tajfel & Turner, 1986).

If the influence of social identity on memory recall has been established, little has been studied concerning the role of memories in defining social identity. Beyond the influence of social identity on memories or the recall of events (aimed at maintaining a positive identity), it is necessary to examine the role of collective memory in defining group social identity (Licata et al., 2007). Indeed, as Hirst and Echterhoff (2008) stated, a shared memory shapes collective identity. That is, shared representation of the past may allow group members to answer the question “who are we?” (Bar-Tal, 2000; Licata et al., 2007) and define group/social identity (Liu & Hilton, 2005). As per Bar-Tal (1990, 2000), collective memories appear to share the properties of group beliefs (see also Hirst & Echterhoff, 2008). That is, collective memory grounds the existence of one’s group by defining its essence.

Moreover, shared beliefs are intimately linked to social communication and everyday conversation (Bar-Tal, 1990). As conversations or social interaction lead to a consensus (about values or judgment), group beliefs maintain cohesion between interlocutors and reinforce social cohesion by highlighting similarities between individuals who share those beliefs (Hogg, Hardie, & Reynolds, 1995). In this line, collective memory could be a source of group cohesion and, as proposed by Licata and colleagues (Licata & Klein, 2005; Licata et al., 2007), creates a “horizontal solidarity” between group members and a “vertical solidarity” with past or future members. As group beliefs, collective memories play a role in the definition of (in)group boundaries (Chryssochoou, 1996) and thus complete the mere self-categorization process by adding sense to social identity (Bar-Tal, 2000). Thus, common history or memories shared by a group define and organize group identity (Jetten & Wohl, 2012; Liu & Hilton, 2005). In conclusion, we suggest that shared memories may provide a criterion for social categorization of others as ingroup or outgroup members.

Collective Memory as a Criterion for Social Categorization

Social categorization is central in social identity approach. In order to decide if a person is a member of the ingroup, individuals would assess his/her proximity or similarity to other members of the group and on group characteristics. We suggest that shared social/group memories may be part of these characteristics and influence social categorization.

Social categories may be represented as prototypes or schematic representations of the characteristics that define group membership (Hogg & Reid, 2006). Considered as a set of elements that include behaviors, attitudes, and beliefs, group prototype provides a frame to categorize others (Hogg, 2002; Hogg & Reid, 2006). Indeed, by making salient the characteristics that will be relevant to the ingroup–outgroup comparison (Hogg et al., 1995), prototype represents the position that maximizes, on the one hand, the differences with outgroups and, on the other, the similarities within the ingroup. This is what the authors called the “meta-contrast principle” (Hogg & Reid, 2006). Social identity theorists have already suggested that group history, and by extension collective memory, emphasizes how the ingroup differs from outgroups. Thus, salient collective memories may be included in the group prototype (as group beliefs) and allow the categorization of a target regarding group membership.

Moreover, when categorizing oneself as member of a group, individuals expect to find people who share their point of view in the group (Tindale, Meisenhelder, Dykema-Engblade, & Hogg, 2002). Indeed, when people discover that they share similar cognitions (e.g., shared representation of the past), they tend to infer a common group membership (Swaab, Postmes, van Beest, & Spears, 2007). Similarity, particularly on knowledge, is thus a central aspect in the categorization of a target person as a group member. Shared knowledge increases the prototypical similarity between the self and ingroup members (Hogg et al., 1995). This increased perception of similarity subsequently reinforces social attraction and attraction (Boyanowsky & Allen, 1973; Hogg & Hardie, 1992). In brief, according to the context, collective memories (i.e., shared representation of the past) may be part of those attributes that encapsulate similarities among ingroup members and differences from outgroup members. Liu, Wilson, McClure, and Higgins (1999) also showed that although historical or past events might be important for many groups, their hierarchy and thus their importance may differ across groups.

In conclusion, if collective memory contributes to the group definition and is a criterion in determining group membership, a person who shares common memories with members of a particular group should be more likely seen as an ingroup member than someone who does not. According to social identity theorists, this could be explained by perceived similarity between the self and a person who shares memories compared to a person who does not (Hogg & Hardie, 1992).

Overview

In four studies, the present article examined the influence of shared memories on social categorization. More specifically, we studied the impact of sharing memories (of either events or objects from the past) important for a group on the social categorization of a target person as an ingroup member. In the first two studies, participants were asked to evaluate a target who did versus did not share common important historical events with the participants. Participants evaluated the perceived similarity of the target to themselves, his/her perceived competence and warmth (Study 2) and the extent to which he/she was seen as belonging to the participants’ ingroup(s). In the last two studies, we replicated our findings using past objects (rather than historical events) such as toys and figures from childhood (Study 3) and cartoons (Study 4).

We expected participants to perceive the target as more similar to themselves and evaluate the target more favorably (in terms of competence and/or warmth) when the target shares common memories with the participants, than when he or she does not. Moreover, we hypothesized that in the shared collective memory condition, participants would be more likely to categorize the target as an ingroup member than in the nonshared condition. Finally, if sharing a common memory is a social criterion for ingroup membership due to increased perceived similarity, we can expect perceived similarity to mediate the effect of shared common memory on the categorization of the target as a group member.

Study 1

In this study, we aimed to provide empirical evidence for the influence of shared memories on social categorization. We presented participants with events allegedly chosen by a target person from a list of historical events. In the shared memory condition, the target chose the most salient historical events for our population (that is, the same choices likely to be made by the participants themselves), whereas in the nonshared memory condition, the target chose less salient events. Next, participants indicated to what extent they thought the target belonged to their group. We predicted that when the target chose the same salient events as they did, participants would perceive the target as similar to the self and would be more inclined to perceive the target as an ingroup member than when the target chose less salient events. Moreover, we predicted perceived similarity to mediate the link between shared collective memory and the categorization of the target as ingroup member.

Method

Participants

Thirty-six students (33 women, M_age= 21.3, SD = 3.47) at Paris Descartes University participated voluntarily in the study. All were French and native speakers of French. No interactions with participants’ gender or age were found and these will not be discussed further.

Materials

In order to identify the most important events throughout history in our population, 54 students (42 women, M_age = 24.35, SD = 7.01) at Paris Descartes University participated voluntarily in a preliminary study. All were French and French native speakers. Participants completed a free association task (Liu et al., 2005): they were asked to spontaneously indicate three historical events that came to mind. In line with previous studies (Pennebaker, Paez, & Deschamps, 2006), the five most cited events were: World War II (44.4%), the attack of September 11 (29.6%), the first men on the moon (20.4%), World War I (18.5%), and the Shoah (16.7%). For our main study, we selected the two World Wars and the September 11 attack to operationalize the choice made by the fictitious target in the shared memory condition. To operationalize the nonshared memory condition, we chose three events that did not appear among participants’ responses, but belonged to the same category of events: The first Iraq war in 1991, the 2006 Israel–Lebanon War (two wars in which France was a member of the coalition and/or of the U.N. resolution) and the Madrid train bombing of March 2004. Two fillers were added: the election of Nicolas Sarkozy and of Charles de Gaulle as French presidents.

Procedure

Participants were presented with a list of eight historical events and were told that another person (the fictitious target) had been asked to choose the three events he/she considered as the most important from the list. In the shared memory condition, the fictitious target chose the three most important historical events: World War I, World War II, and the September 11 attack. In the nonshared memory condition, the choices included: the Israeli–Lebanon conflict, the first war in Iraq in 1991, and the bombing attack in Madrid on March 11 2004.

Next, participants evaluated the target on nine items using a 7-point scale (1 = not at all to 7 = extremely). Four were critical items related to perceived similarity of the target with the self (e.g., close to your values, similar to you, α = .84) and the remaining five were fillers (e.g., loyal). Using the same 7-point scale, participants also indicated to what extent they thought it likely that the target belonged to the following groups: European, French, students, and students in psychology. Although (supra)national ingroups seem the most relevant for the collective memories at stake, these groups represent ingroups for all participants: either citizenship/national or generational (i.e., students) ingroups. Finally, participants specified the country they believed the target was from.

In order to ensure that our results could not be attributed to differences in participants’ identification with the ingroups, we asked participants to evaluate to what extent they would define themselves as European, French, student, student in psychology, using a 7-point scale from 1 = not at all to 7 = extremely.

Results and Discussion

Ingroup Identification

First we checked the extent to which participants in our study identified with the ingroups used. The analyses showed no significant difference between the shared memory and the nonshared memory conditions for all groups: European (M = 5.23, SD = 1.66 and M = 5.57, SD = 1.51, respectively, t < 1), French (M = 5.41, SD = 1.71 and M = 5.07, SD = 1.82, respectively, t < 1), student (M = 6.23, SD = 1.34 and M = 5.71, SD = 2.09, respectively, t < 1), and psychology student (M = 6.05, SD = 1.53 and M = 5.64, SD = 2.06, respectively, t < 1).

Perceived Similarity Between the Target and Participants

As expected (see Table 1), the analysis showed that participants perceived the target as more similar to them in the shared than in the nonshared memory condition, t(35) = 2.46, p = .017, η²_p= .15.

Table 1.

Target perceived similarity and ingroup membership according to shared memory condition (Studies 1 to 3).

		Similarity	European	French	Student	Psychology student
Study 1
Shared CM	M	4.75	5.50	4.86	4.14	3.55
Shared CM	SD	0.72	1.41	1.36	1.21	1.16
Nonshared CM	M	3.91	4.29	3.36	3.79	2.64
Nonshared CM	SD	1.34	1.38	1.55	1.76	1.15
Study 2
Shared CM	M	3.37	—	4.24	4.15	—
Shared CM	SD	0.83	—	1.58	1.20	—
Nonshared CM	M	2.61	—	4.37	3.58	—
Nonshared CM	SD	0.92	—	1.50	1.22	—
Study 3
Shared CM	M	4.38	—	5.73	4.58	3.52
Shared CM	SD	1.27	—	1.37	1.51	1.22
Nonshared CM	M	2.69	—	4.78	3.49	3.03
Nonshared CM	SD	1.30	—	1.52	1.83	1.22

Note. CM = Collective memories.

Perceived Target’s Group Membership

We expected that sharing collective memories would lead to a greater likelihood of categorizing the target as an ingroup member. Ingroup membership was assessed in two ways. First, participants indicated to what extent the target belonged to the mentioned groups on a 7-point scale. As predicted (see Table 1), in the shared memory condition, participants were more likely to perceive the target as European, t(35) = 2.54, p = .016, η² = .14; French t(35) = 3.08 p = .004, η² = .23; a psychology student, t(35) = 2.16 p = .038, η² = .12 than in the nonshared memory condition. No significant difference was observed for the group student, t(35) = 0.67, p = .51, η² = .01.

Second, participants were asked to indicate the likely country of origin of the target. We coded mentioned countries as follows: France (ingroup), European countries, and non-European countries. The analysis showed a consistent pattern of results χ²(2) = 12.17, p = .002. In the shared memory condition, the target was perceived as more likely to have come from France (81%) than in the nonshared memory condition (21%). In contrast, in the nonshared memory condition, the target was more likely to be perceived as coming from another European country (43%) or a non-European country (36%), compared to the shared memory condition (European Countries: 9.5% and non-European countries: 9.5%).

Mediation Analysis

For each membership in which a significant difference across conditions was found, we conducted a mediation analysis with shared versus nonshared collective memory as our independent variable, the target’s perceived ingroup membership (European, French, Psychology Student) as our dependent variable, and perceived similarity, as our mediator. All presented mediation analyses were conducted using PROCESS macro for SPSS (Model 4; Hayes, 2013). PROCESS allows examining the size of indirect effect using bootstrapped confidence interval (5,000 samples). Indirect effect is significant when the bias corrected confidence interval does not include zero.

As expected (see Table 2), the total effect of conditions on each of the tested group memberships was: European (B = 0.61, 95% CI [0.12, 1.09]), French (B = 0.75, 95% CI [0.26, 1.25]) and Psychology Student (B = 0.45, 95% CI [0.03, 0.88]). The indirect effect through similarity was significant for each group membership: European (B = 0.27, 95% CI [0.02, 0.69]), French (B = 0.20, 95% IC [0.01, 0.62]), and Psychology Student (B = 0.23, 95% CI [0.03, 0.60]). The direct effect of our condition, while controlling for similarity, was no longer significant for European and Psychology Student memberships (B = 0.34, 95% CI [−0.14, 0.82] and B = 0.22, 95% CI [−0.20, 0.64], respectively). The direct effect of conditions remained significant for the French membership (B = 0.55, 95% CI [0.03, 1.07]), indicating partial mediation. Sobel tests were only marginal for European and Psychology Student memberships and nonsignificant for French membership (see Table 2).

Table 2.

Mediation analysis including perceived similarity as mediator, shared collective memory as independent variable, and ingroup membership as dependent variable for Studies 1 to 3.

	Perceived similarity			European			French			Student			Student in psychology
	B	SE	t	B	SE	t	B	SE	t	B	SE	t	B	SE	t
Study 1
Shared Collective memory	0.42	0.17	2.46*	0.61	0.24	2.54*	0.75	0.24	3.08**	0.17	0.26	0.67	0.45	0.21	2.16*
Shared Collective memory				0.34	0.23	1.45	0.55	0.25	2.16*				0.22	0.21	1.09
Perceived similarity				0.27	0.17	2.92**	0.33		2.07*				0.23	0.14	2.84**
Sobel				Z = 1.82^†			Z = 1.53						Z = 1.80^†
Study 2
Shared Collective memory	0.90	0.21	3.73**	—	—	—	−0.13	0.36	−0.36	0.57	0.28	2.03*	—	—	—
Shared Collective memory										0.05	0.28	0.321
Perceived similarity										0.52	0.21	4.15**
Sobel										Z = 2.91*
Study 3
Shared Collective memory	1.69	0.23	7.26**	—	—	—	0.95	0.26	3.60**	1.09	0.30	3.58**	0.48	0.23	3.06*
Shared Collective memory							1.05	0.32	3.33**	0.28	0.34	0.820	−0.14	26	−0.532
Perceived similarity							−0.11	0.17	0.618	0.81	0.22	4.27**	0.62	0.15	4.27**
Sobel							—			Z = 3.65**			Z = 3.65**

Consistent with our hypotheses, when an individual shared common memories with the participants, she/he would be more likely to be perceived as belonging to the participants’ groups compared to when he/she did not. Moreover, this effect is (partially) determined by perceived similarity with the self. These first results are consistent with Hogg et al. (1995) results. However, although we tried to choose the same kinds of events in both experimental conditions (bombing attack, wars), the material used in this study presented some limitations. The most salient events were not always equivalent in other aspects with the less salient events. Indeed, in the nonshared collective memory condition, the events took place outside France and/or involved those in which France was not or relatively less directly involved. In contrast, in the shared collective memory condition, the events directly impacted France, such as the two World Wars. This bias may account for why participants did not categorize the target as an ingroup member in the nonshared memory condition.

Study 2

The purpose of the second study was to replicate and extend the results of Study 1, using relatively more equivalent (salient and nonsalient) events. To this end, all the events concerned bombing attacks that took place in Paris. Participants evaluated a target person based on the importance the target gave to the bombing attacks. Next, participants evaluated similarity between the target and themselves and the target’s group membership. As ingroup members tend to be perceived more positively than outgroup members (e.g., Brewer, 1979), participants also completed an impression formation task where they rated the target on traits related to competence and warmth, considered as fundamental dimensions in person perception (Rosenberg, Nelson, & Vivekananthan, 1968). As in Study 1, we predicted that when the target ranked as important the bombing attacks most salient in our participants’ collective memory, they would perceive the target as more similar to them and as more likely to be an ingroup member than when the target chose the less salient bombing attacks. Nonshared collective memory could be either attributed to membership in another national group (having no knowledge of these events) or another generation (i.e., being older or younger). Additionally, we hypothesized that the target would be perceived more favorably in the shared than in the nonshared collective memory condition. Finally, we expected perceived similarity to mediate the link between shared collective memory and perceived group membership.

Method

Participants

Seventy-eight students (67 women; M_age= 21.00, SD = 5.73) at Paris Descartes University participated voluntarily in the study. All were French and were native speakers of French. No interaction with the participants’ gender was found and this will not be discussed further.

Materials

In order to ensure comparable or equivalent events, we chose only bombing attacks in Paris that targeted the French government. These attacks took place either before 1990 (i.e., before our participants were born) or after 1990. Two series of bombing attacks fit these criteria and have already been described as somewhat similar (Hebding, 1995). Indeed, since the 1980s, France has had two major waves of terrorism: one during the mid-1980s (1985–1986) attributed to Hezbollah and one in the mid-1990s (1995–1996), attributed to the Armed Islamic Group or the al-Jama’ah al-Islamiyah al-Musallaha (Gregory, 2003). For each period, we selected two bombings that were highly mediatized at that time. For the 1985–1986 period, we chose the attack of the Printemps Haussmann (i.e., a well-known Paris department store) on December 7, 1985 and the attack in rue de Rennes on September 17, 1986. For the 1995–1996 period, we selected the attack of the (train) station Saint Michel on July 25, 1995 and of the station Port-Royal on December 3, 1996. The bombing attacks of the mid-1980s did not lead to many commemorations after 1990. In contrast, the attacks of the mid-1990s led to commemorations 10 years later in 2005 and 2006 respectively.

Our participants were born in/around 1989. Thus, they were not yet born during the bombing attacks of the mid-1980s, were 6 to 7 years old during the attacks of the mid-1990s and about 16 years old during the commemorations of the latter events. The attacks of the 1990s should be thus the most salient events in the collective memory of this population. We added another event: the attack of the high speed train TGV Paris–Marseille of December 31, 1983. In contrast to the previous attacks, it is less the event than its perpetrator, the terrorist Ilich Ramirez Sanchez—known as Carlos—that was covered by the media.

In the shared collective memory condition, the fictitious target ranked the importance of bombing attacks in the following manner: Station Saint Michel (1995), station Port Royal (1996), TGV Paris–Marseille (1983), rue de Rennes (1986), and Printemps Haussmann (1985). In the nonshared collective memory condition, the ranking order supposedly made by the target was reversed.

Procedure

Participants were presented with a list of five events (dates were not indicated). They were told that another person (fictitious target whose sex was not indicated) had been asked to prioritize the events in the order of importance to him or to her. In the shared memory condition, the target ranked the events from the most to the least salient (a ranking that our participants would have done themselves). In the nonshared memory condition, the order was reversed. Participants next evaluated the target on 11 items using a 7-point scale (1 = not at all, 7 = extremely). Four items were related to perceived similarity with the self (e.g., close to your own values, similar to you; α = .79), four were related to perceived competence (e.g., smart, cultured; α = .80), three items were related to perceived warmth (e.g., nice, warm; α = .77). Participants then indicated to what extent they thought the target belongs to their groups, namely, French and Students. Finally, participants were asked to order the presented bombing attacks from the most to the least important.

Results and Discussion

Manipulation Check

In order to ensure that our participants perceived the bombing attacks of the mid-1990s as the most salient and important attacks in both conditions, we conducted a 2 (collective memory: shared, not shared) x 5 (bombing attack: Saint Michel, Port Royal, Paris–Marseille, Rennes, Haussmann) ANOVA with shared memory as a between and bombing attack as a within factor. This analysis revealed a significant effect of the bombing attack, F(4, 59) = 6.06, p < .0001, η²_p = .22. The main effect of shared memory was not significant, F(1, 62) = 0.83, p = .37, η²_p =.01, nor was the interaction, F(4, 59) = 1.08, p = .37, η²_p = .08. The mean rank of each event was similar to the ranking presented in the shared memory condition: the bombing attack at Saint Michel station had the lowest mean rank (M = 2.45, SD = 1.37), followed by the event of the TGV Paris–Marseille (M = 2.72, SD = 1.46) and the Port-Royal bombing (M = 2.98, SD = 1.23). A movie on Carlos’s life was released in 2010, which may explain why the attack of the TGV was more salient than expected. Finally, as expected, the events with the higher mean rank were the attacks in rue de Rennes (M = 3.66, SD = 1.30) and the Printemps Haussmann (M = 3.17, SD = 1.43).

Target Perception: Similarity, Warmth, and Competence

Consistent with our hypothesis, the analysis showed that in the shared memory condition, participants perceived the target as more similar to themselves than in the nonshared condition, t(76) = 3.73, p < .0001, η² = .16 (see Table 1). Likewise, in the shared memory condition, participants perceived the target as more competent (M = 4.07, SD = 0.86) than in the nonshared condition (M = 3.60, SD = 0.92), t(76) = 2.29, p = .025, η² = .06. No significant effect of conditions was found on perceived warmth, t(76) = 0.20, p = .84, η² = .001.

Perception of Target’s Group Membership

As presented in Table 1, in the shared memory condition, participants were more prone to perceive the target as a student than in the nonshared condition, t(75) = 2.03, p = .046, η² = .05. No significant effect of condition was found for the French attribution, t(75) = −0.36, p = .72, η² = .002.

Mediation Analysis: Ingroup Membership

As presented in Table 2, the indirect effect through similarity was significant (B = 0.52, 95% CI [0.20, 1.04]). Moreover, the significant effect of shared collective memory on student membership was no longer significant once perceived similarity was controlled for (B = 0.05, 95% CI [−0.51, 0.62]; z = 2.91, p = .004), indicating a full mediation. This analysis suggests that the difference in perception of the target as a student across experimental conditions was determined by the perceived similarity between the participant and the target.

The present results replicate those of Study 1. Indeed, when a target shared common memories (i.e., accorded more importance to the most salient historical events), he/she was perceived as more similar with the participants and more likely to be perceived as an ingroup member. Moreover, this study showed that remembering important and salient events led participants to perceive the target more favorably on competence. No effect on warmth was found.

In order to generalize our findings, we conducted further studies using a different type of materials than shared historical events. Instead, we conducted studies using memories such as action/toy figures (Study 3) and cartoons (Study 4) from our population’s childhood.

Study 3

This experiment aimed to investigate whether social categorization as ingroup member could be based on sharing common memories of childhood. Collective memories can refer to historical memories, but also to everyday life memories of popular culture (Lipsitz, 2001). Thus, childhood memories can be objects of representation of the past that are socially produced and shared (cf. supra), for instance through books showcasing childhood objects.

We presented the choice of figures made by a target person on a series of childhood figures. In the shared memory condition, the target chose the most memorable childhood figures for our participants’ generation, whereas in the nonshared memory condition, the target chose the least memorable. Then, as in Study 2, participants evaluated the target on her/his perceived similarity to themselves, her/his perceived competence and warmth, and finally on her/his membership in their groups. We predicted that in the shared memory condition, participants would be more likely to evaluate the target as a member of their group (national or generational) than in the nonshared memory condition. Again, we expected perceived similarity to mediate the predicted difference in perceived group membership. Finally, we predicted that participants would perceive the target more favorably in the shared than in the nonshared memory conditions.

Method

Participants

One hundred and twenty-one students (104 women, M_age = 20.14, SD = 1.75) at Paris Descartes University, participated voluntarily in the study. All were French, were native speakers of French, and had spent their childhood in France. No interaction with the participants’ gender or age was found and these will not be discussed further.

Materials

To operationalize our experimental conditions, 52 students (45 women, M_age = 21.04, SD = 1.93) at Paris Descartes University were presented with a list of 18 childhood figures. They reflected childhood figures of two generations: either people born after 1990 or those born around 1980. Participants were to indicate to what extent they remembered each presented figures using an 8-point scale (1 = not at all to 8 = completely). Among these 18 figures, we selected three that were the best remembered and three the least remembered, that is, those with a ranking significantly above or under the midpoint scale (i.e., 4.5). Selected items were: In the shared collective memory condition, we included: “J’aime lire” (i.e., “I Love Reading,” a monthly magazine for children, M = 7.48, SD = 1.23), the Tamagotchi (M = 7.67, SD = 1.02), and the Polly Pockets (M = 7.33, SD = 1.54). In the nonshared collective memory condition, we included Big Jim action figure toys (M = 1.29, SD = 0.78), the Masters of the Universe (M = 2.31, SD = 1.92), and the world cup soccer mascot of 1990 (M = 1.20, SD = 0.90).

Procedure

Participants were presented with a description of a (fictitious) target person. They were told that the target had chosen, from a list of six figures, the three that evoked the most memories for him/her. In the shared memory condition, the target had chosen the three figures pretested as most memorable. In the nonshared memory condition, the choices included the three other figures. Next, participants evaluated the target on 14 items using a 7-point scale (1 = not at all to 7 = extremely). Four items were related to judged similarity of the target with the self (e.g., shares the same values, similar; α = .88), five were related to competence (e.g., smart, cultured; α = .83), and five were related to warmth (e.g., nice, warm; α = .85). Finally, participants were asked to indicate to what extent they thought likely that the target belonged to participants’ groups, namely French, Students, and Psychology Student.

Results and Discussion

Target Perception: Similarity, Warmth, and Competence

As expected (see Table 1), in the shared memory condition, participants perceived the target as more similar to themselves than in the nonshared memory condition, t(119) = −7.03, p < .0001, η²_p =.31. In addition, in the shared memory condition, participants perceived the target as warmer (M = 4.97, SD = 1.02) than in the nonshared memory condition (M = 4.44, SD = 0.92), t(119) = −2.98, p = .003, η² = .07. No significant difference across conditions on competence was found, t(119) = −0.14, p = .89, η² = .001. In contrast to Study 2, sharing common (positive) memories about childhood led participants to perceive the target as warmer than not sharing (and no effect on perceived competence).

Perception of Target’s Group Membership

As presented in Table 1, participants in the shared (vs. nonshared) memory condition were more inclined to perceive the target as a psychology student, t(119) = −2.06, p = .042, η² = .03; as a student, t(119) = −3.58, p = .001, η²_p = .10; and finally, as French, t(119) = −3.60, p < .001, η² = .10.

Mediation Analysis

As seen in Table 2 and as expected, the total effect of conditions was significant for Student (B = 1.09, 95% CI [0.40, 1.69]) and Psychology Student memberships (B= 0.48, 95% CI [0.02, 0.95]). The indirect effect through similarity was significant for Student (B = 0.81, 95% CI [0.43, 1.33]) and Psychology Student memberships (B = 0.62, 95% CI [0.36, 0.97]). Finally, once perceived similarity was controlled for, the effect of shared memory on ingroup membership was no longer significant: for Student (B = 0.28, 95% CI [−0.40, 0.96]; z = 3.65, p < .001) and for Psychology Student (B = −0.14, 95% CI [−0.66, 0.38]; z = 3.65, p < .001). However, the effect of perceived similarity is not significant when the effect of the experimental conditions on the attribution of French membership was controlled for (B= −0.11, 95% CI [−0.43, 0.25]), which still predicted group membership (B = 1.05, 95% CI [0.43, 1.86]).

As expected, even for memories of childhood, sharing common memories led participants to perceive the target as more similar to themselves, and thus more likely to be an ingroup member. Moreover, sharing memories about childhood led participants to perceive the target as warmer. We thus replicate the pattern found in the two first studies: collective memories influence social categorization as an ingroup member, and this, via perceived similarity. Second, we found an influence of sharing common memories on target perception, particularly on rated warmth. This may be related to the memory objects used in this study that differed from Study 2 (where we found differences in rated competence).

It should be noted, however, that the selected childhood objects might also be gender specific. In the shared memories condition, the figures could have been more specific to girls (e.g., Pollypockets, Tomagotchi). Thus, common memories and gender may have been confounded and could account in part for the presented results, even if gender does not seem to be a relevant criterion to categorize the target as a student or French.

Study 4

To ensure that only shared memories rather than gender influence social categorization, we replicated Study 3 with cartoons that differed on two dimensions: gender-type (girl vs. boy) and age or generation (people born around 1990 vs. around 1980). The methodology was almost the same than the one used in the Study 3 and only female participants were included in this study in order to examine the influence of both (mis)matches in gender and common memories. We expected to replicate previous findings in both gender-type cartoons conditions. Indeed, to categorize a target as a student or as French, sharing common memories should be the relevant criterion compared to gender.

Method

Participants

One hundred and twenty-one students (M_age = 19.47, SD = 1.40) at Paris Descartes University participated voluntarily in the study. All participants were females, native French speakers, and have continuously lived in France before university.

Materials

To construct our experimental conditions, 64 students (55 women, M_age = 19.00, SD = 1.13) at Paris Descartes University were presented with a list of 22 cartoons for children, represented by their main characters. Participants indicated to what extent they remembered each cartoon using a 7-point scale (1 = not at all to 7 = completely). Additionally, participants indicated to what extent they thought each cartoon was more “girl-type” versus “boy-type” using a 7-point scale (1 = very boy-type to 7 = very girl-type). We selected four cartoons (two girl-types and two boy-types) that were best remembered and another four least remembered (two girl-types and two boy-types). They were significantly above or under the midpoint scale (i.e., 4). Selected items were (see Table 3): In the shared memories condition, two girl-type cartoons: Totally Spies! (broadcasted in France in 2002), The Powerpuff Girls (broadcasted in 2000); and two boy-type cartoons: Rugrats (broadcasted in 1998) and Pokémon (broadcasted in 2000). In the nonshared memories condition, we chose two girl-type cartoons: Juliette je t’aime (i.e., French adaptation of Mezon Ikkoku, broadcasted in 1988), Popples (broadcasted in 1986); and two boy-type cartoons: Ulysses 31 (broadcasted in 1981) and Bibifoc (i.e., Seabert, broadcasted in 1985).

Table 3.

Selected cartoons evaluation on the level of memories elicited and on gender type.

Collective memory	Gender	Title	Memory			Gender
			M	SD	t	M	SD	t
Shared memory	Girl-type	Totally spies	6.53	0.96	21.12**	6.47	0.85	23.13**
	Girl-type	Powerpuff girls	6.00	0.81	10.89**	6.41	0.81	23.74**
	Boy-type	Rugrats	6.47	1.01	19.61***	3.75	0.77	−2.51*
	Boy-type	Pokémon	6.64	0.76	27.67*	3.39	1.03	−4.72**
Nonshared memory	Girl-type	Juliette je t’aime	1.78	1.71	−10.36**	4.63	1.08	4.65**
	Girl-type	Popples	2.02	1.77	−8.98**	4.98	1.23	6.41**
	Boy-type	Ulysses 31	1.78	1.59	−11.18**	2.34	1.19	−11.06**
	Boy-type	Seabert	1.69	1.51		3.73	0.93	−2.29*

Note. Higher score on gender type means girl-type cartoons.

p < .05. **p < .0001.

Procedure

The procedure was similar to that in Study 3 with the following modifications: In the shared memory condition, the choice made by the fictitious target included either the two girl-type cartoons (“Totally Spies” and “Powerpuff Girls”) or the two boy-type cartoons (“Rugrats” and “Pokémon”) pretested as most remembered (sex of target, years of broadcast, and titles were never indicated). In the nonshared memory condition, the choice included the least remembered cartoons, that were either two girl-type cartoons (“Juliette, je t’aime” and “Popples”) or two boy-type cartoons (“Ulysses 31” and “Seabert”). Next, participants evaluated the target on 20 items using a 7-point scale (1 = not at all to 7 = extremely). Four items were related to perceived similarity of the target with themselves (e.g., sharing the same values, similar to you; α = .82), four were related to competence (e.g., smart, cultured; α = .71), four were related to warmth (e.g., nice, warm; α = .81), and the remaining were fillers. Participants also indicated to what extent they thought the target was a student and French (i.e., the ingroups). Finally, participants chose, among the eight cartoons, the two that evoked the most memories for them.

Results and Discussion

Preliminary Check

We first ensured that participants chose the same cartoons identified as evoking childhood memories in the pretest. In the main study, 91% of participants chose two of the cartons identified previously as best remembered (82% to 100% across conditions) and 100% chose at least one.

Perceived Similarity With the Target

As expected, the analysis showed that participants perceived the target as more similar to themselves (M = 16.00, SD = 5.04) in the shared than in the nonshared memory condition (M = 11.71, SD = 4.67), F(1, 117) = 23.83, p < .0001, η²_p = .17. The analysis revealed a marginal effect of gender-type, F(1, 117) = 3.64, p =. 06, η²_p = .03. Choosing “boy-type” cartoons led participants to perceive the target as marginally less similar (M = 13.18, SD = 5.17) than choosing “girl-type” cartoons (M = 14.81, SD = 5.40). Finally, the interaction was not significant, F(1, 117) = 0.09, p = .76, η²_p = .001.

Perceived Warmth and Competence

Consistent with our hypothesis, the analysis showed that in the shared memory condition, participants perceived the target as warmer (M = 15.91, SD = 2.41) than in the nonshared condition (M = 14.50, SD = 3.12), F(1, 117) = 8.31, p = .005, η²_p = .07. The analysis revealed a significant effect of gender type, F(1, 117) = 4.09, p = .045, η²_p = .03. Choosing “girl-type” cartoons led participants to perceive the target as warmer (M = 15.72, SD = 2.73) than choosing “boy-type” cartoons (M = 14.79, SD = 2.88). Finally, the interaction was not significant, F(1, 117) = 1.97, p = .16, η²_p = .02.

For the competence score, the analysis did not reveal any significant effect of shared memory, F(1, 117) = 0.27, p = .61, η²_p = .002), nor of gender-type objects, F(1, 117) = 0.79, p = .38, η²_p = .007). The interaction was not significant, F(1, 117) = 0.10, p = .75, η²_p = .001.

Target Ingroup Perception

As expected, in the shared memory condition, participants were more inclined to perceive the target as a student (M = 4.82, SD = 1.57) than the in the nonshared condition (M = 4.02, SD = 1.58), F(1, 118) = 7.98, p = .006, η²_p = .06. The analysis did not reveal any significant effect of gender type, F(1, 118) = 1.12, p = .29, η²_p = .009. Choosing “girl-type” cartoons did not lead participants to perceive the target as more likely to be an ingroup member (M = 4.59, SD = 1.59) than choosing “boy-type” cartoons (M = 4.31, SD = 1.64). The interaction was not significant, F(1, 118) = 0.66, p = .42, η²_p = .006.

Moreover, in the shared memory condition, participants were more inclined to perceive the target as French (M = 4.97, SD = 1.47) than the in the nonshared memory condition (M = 4.42, SD = 1.37), F(1, 117) = 4.45, p = .04, η²_p = .04. The analysis did not reveal any significant effect of gender type, F(1, 117) = 0.10, p = .75, η²_p = .001, nor interaction, F(1, 117) = 0.10, p = .75, η²_p = .001.

Mediation Analysis

As mentioned earlier, common memories significantly affect perceived membership in the Student group (B = 0.39, SE = 0.14, 95% CI [0.11, 0.68]) and perceived similarity (B = 2.14, SE = 0.44, 95% CI [1.26, 3.02]). The indirect effect through similarity was significant (B = 0.17, SE = 0.08, 95% CI [0.05, 0.36]). Finally, once perceived similarity was controlled for, the effect of shared collective memory on student membership was no longer significant (B = 0.22, SE = 0.15, 95% CI [−0.08, 0.52]; z = 2.35, p < .02), indicating a full mediation. This suggests that target’s perceived similarity explained the difference in perception of the target as a student across conditions. As in Study 3, the attribution of French membership to the target was not mediated by perceived similarity. Indeed, the effect of perceived similarity on French membership attribution was not significant when the effect of conditions was controlled for (B = 0.04, SE = 0.06, 95% CI [−0.07, 0.17]). That is, experimental conditions did not predict French membership attribution (B = 0.24, SE = 0.14, 95% CI [−0.04, 0.52]).

Thus, we replicated the pattern of Study 3’s results and found consistent findings with Studies 1 and 2. Perceived similarity with the self and perceived ingroup membership were significantly predicted by sharing memories, but not by gender-type memories. Moreover, the effect of sharing memories on similarity, warmth and ingroup membership was not moderated by gender type. Finally, as in Study 3, sharing memories influences the target perception on perceived warmth that may be more pertinent in the context of childhood memories than on competence.

General Discussion

The present article extended previous research on the role of collective memory related to its group identity-defining function. We suggested that if collective memories define group identity and group boundaries, they should also be a relevant criterion for the social categorization of others and for the assessment of similarity. In this optic, a target who shares the same collective memories would lead members of a group to perceive him/her as more similar to themselves and to categorize the target as an ingroup member.

We conducted four studies to test our ideas, using memories comprising shared historical events (e.g., WWI and WWII or bombing attacks, for the first two studies) or childhood objects (e.g., books showcasing toys of a generation; cartoons, for the last two studies). They concerned effectively memories that are shared by all group members (Zerubavel, 1996), here national or generational. Results support the idea that collective memory influences social categorization as an ingroup member. The four studies consistently showed that someone who shares memories is more likely to be perceived as an ingroup member than one who does not. Moreover, this effect is mediated by perceived similarity between the self and the target. Inasmuch as the representation of the self overlaps with that of the salient ingroup (Smith, Coats, & Walling, 1999; Smith & Henry, 1996), rated similarity is (partly) based on a salient social identity. As an individual who shares memories with members of a group, he/she will be perceived as more similar to ingroup members than an individual who does not. This determines the perception that he/she is likely to be an ingroup member.

Past studies have focused on the function of collective memory in the maintenance of a positive identity through the reconstruction or the selection of memories (e.g., Sahdra & Ross, 2007). In the present research, we explored a complementary facet of collective memory functions: how collective memories are “weighing on the group’s present” (Klein, Licata, van der Linden, Mercy, & Luminet, 2011). In our studies, collective memories served as a criterion to categorize a target as an ingroup member and in this manner, participate in the satisfaction of needs such as defining a group’s identity.

Moreover, the results are consistent with the assertion that collective memories, as shared representations of the past, act as shared cognitions leading to the inference of a common group membership (Swaab et al., 2007). The results of the present studies support this idea. These results are consistent with those of Conway (1997): individuals sharing common knowledge, such as public events (Holmes & Conway, 1999) or everyday objects (movies; Sehulster, 1996) form a social group or a generational unit with its identity. In the same vein, collective memories should be part of a group prototype or shared representations of elements that represent a social category. In our studies, collective memories are the salient element to evaluate differences and similarities with the ingroup. In Study 4, although gender could have been a criterion, it was collective memories that served as criterion to assess similarity. Collective memory thus participates in the so-called metacontrast processes.

Finally, our studies showed that individuals who share memories are perceived more favorably than those who do not. The type of collective memories (historical events vs. childhood objects) impacts either perceived competence or perceived warmth, two important dimensions in intergroup perception (Fiske, Cuddy, Glick, & Xu, 2002).

For historical memories, one explanation could be that acknowledging important historical memories leads participants to think that the target has the appropriate knowledge concerning their ingroup collective history. In this context, the competence dimension is the salient one. In contrast, the last two studies involved choice of items related to childhood, for which the warmth dimension is salient. These memories are positive and might have elicited nostalgia. Indeed, nostalgia is a “positive, social, and past-oriented emotion” (Zhou, Wildschut, Sedikides, Chen, & Vingerhoets, 2012) that reinforces connections to others and increases perception of a common identity (Turner, Wildschut, & Sedikides, 2012).

Another possible explanation exists: the different types of memories may influence the importance, relevance, or centrality of the warmth and competence dimensions for ingroup members. As suggested by Oldmeadow and Fiske (2010), the dimension on which we observed an influence of shared collective memory could be the one that allowed participants to differentiate positively from potential outgroups. Ingroups, in this case students, are usually perceived as both competent and warm (Fiske et al., 2002). The collective memories highlighted could make one dimension more relevant for intergroup distinction than the other. Thus, different types of collective memories can lead participants to focus on particular characteristics of their ingroup identity, in the same manner that autobiographical memories allow individuals to justify their desired self-concept (Sanitioso, Kunda, & Fong, 1990).

The four studies presented in this paper contribute to research linking collective memory and social identity. Sharing collective memories is an important criterion to assess the likelihood that a target belongs to the ingroup, as mediated by perceived similarity to the self. However, this was true for the Student ingroup, but not for the French. An explanation could be, especially for Studies 2 to 4, generational identity as Students was the most salient one in our experimental setting. Consequently, perceived similarity might have been based on the representation of the self as a student and not as French. This self-categorization as, and identification to, the ingroup Students could explain why perceived similarity mediated Student but not French membership attribution. Future studies should therefore take into account self-categorization as a group member as well as social identification to the group.

Additionally, perceived similarity between the target and the ingroup prototype (and not the self) might account for the influence of shared memories on membership in national ingroup. This may explain why he/she was less likely to be perceived as an ingroup member (Abrams, Marques, Bown, & Henson, 2000; cf. Castano, Paladino, Coull, & Yzerbyt, 2002). Future studies should examine the similarity of a target to ingroup prototype and, to allow further intergroup differentiation (Abrams et al., 2000), his/her similarity to the prototype of a specific outgroup. Finally, norms of remembering may also be pertinent to the present results. Norms can be defined as a description or a prescription of a behavior, an attitude that is valued in a given social group or society (Reno, Cialdini, & Kallgren, 1993). Historical events and their acknowledgement may be related prescriptive norms that specify what is socially valued or not. As these events are commemorated and associated with a duty of remembrance through commemorations (e.g., bank holidays) and/or laws (e.g., the law related to Holocaust deniers), remembering is thus a norm or duty. Consequently, not remembering constitutes a deviance. In contrast, some memories, such as childhood memories, come under descriptive norms describing what most members of a group do (descriptive or denotative characteristics; see Abrams, Randsley de Moura, Hutchison, & Viki, 2005), leading to categorization as an ingroup member. Future research should explicitly look at the type of norms associated with collective memories to better understand the identity function of collective memories. Though the present research shows pervasive influences of shared collective memories on social categorization, it is mute on how shared memories differ from other shared cognitions. Future studies may thus examine if collective memories have specific proprieties or functions (e.g., when social identity is threatened), compared to other shared cognitions.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Abrams

Marques

J. M.

Bown

Henson

(2000). Pro-norm and anti-norm deviance within and between groups. Journal of Personality and Social Psychology, 78, 906–912. doi:10.1037/0022–3514.78.5.906

Abrams

Randsley de Moura

Hutchison

Viki

G. T.

(2005). When bad becomes good (and vice versa): Why social exclusion is not based on difference. In Abrams

Hogg

M. A.

Marques

J. M.

(Eds.), Social exclusion and inclusion (pp. 161–190). Philadelphia, PA: Psychology Press.

Allport

G. W.

Postman

(1945). The basic psychology of rumor. Transactions of the New York Academy of Sciences, 8, 61–81.

Apfelbaum

(2000). And now what, after such tribulations? Memory and dislocation in the era of uprooting. American Psychologist, 55, 1008–1013. doi:10.1037/0003-066X.55.9.1008

Bar-Tal

(1990). Group beliefs. A conception for analyzing group structure, processes, and behavior. New York, NY: Springer-Verlag.

Bar-Tal

(2000). Shared beliefs in a society: Social psychological analysis. Thousand Oaks, CA: Sage.

Baumeister

R. F.

Hastings

(1997). Distortions of collective memory : How groups flatter and deceive themselves. In Pennebaker

J. W.

Paez

Rimé

(Eds.), Collective memory of political events: Social psychological perspectives (pp. 277–293). Mahwah, NJ: Lawrence Erlbaum Associates.

Boyanowsky

E. O.

Allen

V. L.

(1973). Ingroup norms and self-identity as determinants of discriminatory behavior. Journal of Personality and Social Psychology, 25, 408–148. doi:10.1037/h0034212

Brewer

M. B.

(1979). In-group bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307–324. doi:10.1037//0033–2909.86.2.307

10.

Castano

Paladino

Coull

Yzerbyt

V. Y.

(2002). Protecting the ingroup stereotype: Ingroup identification and the management of deviant ingroup members. British Journal of Social Psychology, 41, 365–385. doi:10.1348/014466602760344269

11.

Chryssochoou

(1996). How group membership is formed: Self categorisation or group beliefs? The construction of a European identity in France and Greece. In Breakwell

Lyons

(Eds.), Changing European identities: Social psychological analysis of social change (pp. 181–194). Oxford, UK: Butterworth.

12.

Conway

M. A.

(1997). The inventory of experience: Memory and identity. In Pennebaker

J. W.

Paez

Rimé

(Eds.), Collective memory of political events: Social psychological perspectives (pp. 21–56). Mahwah, NJ: Lawrence Erlbaum Associates.

13.

Conway

M. A.

(2005). Memory and the self. Journal of Memory and Language, 53, 594–628. doi:10.1016/j.jml.2005.08.005

14.

Dresler-Hawke

(2005). Reconstructing the past and attributing the responsibility for the Holocaust. Social Behavior and Personality, 33, 133–148. doi:10.2224/sbp.2005.33.2.133

15.

Dudai

(2002). Memory from A to Z: Keywords, concepts and beyond. Oxford, UK: Oxford University Press.

16.

Fiske

S. T.

Cuddy

A. J. C.

Glick

(2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878–902. doi:10.1037//0022–3514.82.6.878

17.

Gregory

(2003). France and the war on terrorism. Terrorism and Political Violence, 15, 124–147. doi:10.1080/09546550312331292987

18.

Greish

(2001). Paul Ricoeur. L’itinérance du sens. Grenoble, France: Editions Jérome Millon.

19.

Halbwachs

(1997). La mémoire collective. Paris, France: Albin Michel. (Original work published 1950)

20.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis. A regression-based approach. New York, NY: The Guilford Press.

21.

Hebding

(1995). L’explosion terroriste. Autres Temps. Les Cahiers Du Christianisme Social, 47, 63–65. doi:10.3406/chris.1995.1797

22.

Hirst

Echterhoff

(2008). Creating shared memories in conversation: Toward a psychology of collective memory. Social Research, 75, 183–216.

23.

Hogg

M. A.

(2002). Social categorization, depersonalization, and group behavior. In Hogg

M. A.

Tindale

R. S.

(Eds.), Blackwell handbook of social psychology: Group process (pp. 56–85). Malden, MA: Blackwell.

24.

Hogg

M. A.

Hardie

E. A.

(1992). Prototypicality, conformity and depersonalized attraction: A self-categorization analysis of group cohesiveness. British Journal of Social Psychology, 31, 41–56. doi:10.1111/j.2044-8309.1992.tb00954.x

25.

Hogg

M. A.

Hardie

E. A.

Reynolds

K. J.

(1995). Prototypical similarity, self-categorization, and depersonalized attraction: A perspective on group cohesiveness. European Journal of Social Psychology, 25, 159–177. doi:10.1002/ejsp.2420250204

26.

Hogg

M. A.

Reid

S. A.

(2006). Social identity, self-categorization, and the communication of group norms. Communication Theory, 16, 7–30. doi:10.1111/j.1468–2885.2006.00003.x

27.

Holmes

Conway

M. A.

(1999). Generation identity and the reminiscence bump: Memory for public and private events. Journal of Adult Development, 6, 21–34. doi:10.1023/A:1021620224085

28.

Jedlowski

(2001). Memory and sociology: Themes and issues. Time & Society, 10, 29–44. doi:10.1177/0961463X01010001002

29.

Jetten

Wohl

M. J. A.

(2012). The past as a determinant of the present: Historical continuity, collective angst, and opposition to immigration. European Journal of Social Psychology, 42, 442–450. doi:10.1002/ejsp.865

30.

Klein

Licata

van der Linden

Mercy

Luminet

(2011). A waffle-shaped model for how realistic dimensions of the Belgian conflict structure collective memories and stereotypes. Memory Studies, 5, 16–31. doi:10.1177/1750698011424028

31.

Licata

Klein

(2005). Regards croisés sur un passé commun: Anciens colonisés et anciens coloniaux face à l’action belge au Congo. In Sanchez-Mazas

Licata

(Eds.), L’Autre: Regards psychosociaux (pp. 241–278). Grenoble, France: Presses Universitaires de Grenoble.

32.

Licata

Klein

Gely

(2007). Mémoire des conflits, conflits de mémoires : Une approche psychosociale et philosophique du rôle de la mémoire collective dans les processus de réconciliation intergroupe. Social Science Information, 46, 563–589. doi:10.1177/0539018407082593

33.

Lipsitz

(2001). Time passages. Collective memory and American popular culture. Minneapolis: University of Minnesota Press.

34.

Liu

J. H.

(1999). Social representations of history: Preliminary notes on content and consequences around the Pacific Rim. International Journal of Intercultural Relations, 23, 215–236. doi:10.1016/S0147-1767(98)00036-4

35.

Liu

J. H.

Goldstein-Hawes

Hilton

D. J.

Huang

L.-L.

Gastardo-Conaco

Dresler-Hawke

Hidaka

(2005). Social representations of events and people in world history across 12 cultures. Journal of Cross-Cultural Psychology, 36, 171–191. doi:10.1177/0022022104272900

36.

Liu

J. H.

Hilton

D. J.

(2005). How the past weighs on the present: Social representations of history and their role in identity politics. The British Journal of Social Psychology, 44, 537–556. doi:10.1348/014466605X27162

37.

Liu

J. H.

Wilson

M. S.

McClure

Higgins

T. R.

(1999). Social identity and the perception of history: Cultural representations of Aotearoa/New Zealand. European Journal of Social Psychology, 29, 1021–1047. doi:10.1002/(SICI)1099-0992(199912)29:8<1021::AID-EJSP975>3.0.CO;2-4

38.

Locke

(2009). Essai sur l’entendement humain. Paris, France: Le Livre de Poche. (Original work published 1869)

39.

Moscovici

(2008). Psychoanalysis: Its image and its public. Cambridge, UK: Polity Press.

40.

Oldmeadow

J. A.

Fiske

S. T.

(2010). Social status and the pursuit of positive social identity: Systematic domains of intergroup differentiation and discrimination for high- and low-status groups. Group Processes & Intergroup Relations, 13, 425–444. doi:10.1177/1368430209355650

41.

Paez

Marques

J. M.

Valencia

J.-F.

Vincze

(2006). Dealing with collective shame and guilt. Psicología Política, 32, 59–78.

42.

Pennebaker

J. W.

Paez

Deschamps

J.-C.

(2006). The social psychology of history: Defining the most important events of the last 10, 100, and 1000 years. Psicología Política, 32, 15–32. Retrieved from http://www.uv.es/garzon/psicologia%20politica/N32.htm

43.

Rateau

(2002). Pensée sociale, mémoire collective et saillance d’un événement : L’affaire de Carpentras. Psychologie & Société, 4, 105–130.

44.

Rateau

Moliner

Guimelli

Abric

J.-C.

(2011). Social representation theory. In van Lange

P. A. M.

Kruglanski

A. W.

Higgins

E. T.

(Eds.), Handbook of theories of social psychology (pp. 477–487). Thousand Oaks, CA: Sage.

45.

Reno

R. R.

Cialdini

R. B.

Kallgren

C. A.

(1993). The transsituational influence of social norms. Journal of Personality and Social Psychology, 64, 104–112. doi:10.1037//0022–3514.64.1.104

46.

Rosenberg

Nelson

Vivekananthan

P. S.

(1968). A multidimensional approach to the structure of personality impressions. Journal of Personality and Social Psychology, 9, 283–294. doi:10.1037/h0026086

47.

Sahdra

Ross

(2007). Group identification and historical memory. Personality & Social Psychology Bulletin, 33, 384–95. doi:10.1177/0146167206296103

48.

Sanitioso

R. B.

Kunda

Fong

G. T.

(1990). Motivated recruitment of autobiographical memories. Journal of Personality and Social Psychology, 59, 229–241. doi:10.1037/0022-3514.59.2.229

49.

Schwartz

(2008). Collective memory and abortive commemoration: Presidents’ day and the American holiday calendar. Social Research, 75, 75–110.

50.

Schwartz

S. H.

Struch

Bilsky

(1990). Values and intergroup social motives: A study of Israeli and German students. Social Psychology Quarterly, 53, 185–198. doi:10.2307/2786958

51.

Sehulster

J. R.

(1996). In my era: Evidence for the perception of a special period of the past. Memory, 4, 145–158. doi:10.1080/096582196389013

52.

Smith

E. R.

Coats

Walling

(1999). Overlapping mental representations of self, in-group, and partner: Further response time evidence and a connectionist model. Personality and Social Psychology Bulletin, 25, 873–882. doi:10.1177/0146167299025007009

53.

Smith

E. R.

Henry

(1996). An in-group becomes part of the self: Response time evidence. Personality and Social Psychology Bulletin, 22, 635–642. doi:10.1177/0146167296226008

54.

Swaab

Postmes

van Beest

Spears

(2007). Shared cognition as a product of, and precursor to, shared identity in negotiations. Personality & Social Psychology Bulletin, 33, 187–199. doi:10.1177/0146167206294788

55.

Tajfel

Turner

J. C.

(1986). The social identity theory of inter-group behavior. In Worchel

Austin

L. W.

(Eds.), Psychology of intergroup relations (pp. 7–24). Chicago, IL: Nelson-Hall.

56.

Tindale

R. S.

Meisenhelder

Dykema-Engblade

A. A.

Hogg

M. A.

(2002). Shared cognition in small groups. In Hogg

M. A.

Tindale

R. S.

(Eds.), Blackwell handbook of social psychology: Group process (pp. 1–30). Oxford, UK: Blackwell.

57.

Turner

R. N.

Wildschut

Sedikides

(2012). Dropping the weight stigma: Nostalgia improves attitudes toward persons who are overweight. Journal of Experimental Social Psychology, 48, 130–137. doi:10.1016/j.jesp.2011.09.007

58.

Wang

(2008). On the cultural constitution of collective memory. Memory, 16, 305–318. doi:10.1080/09658210701801467

59.

Wertsch

J. V

Roediger

H. L.

(2008). Collective memory : Conceptual foundations and theoretical approaches. Memory, 16, 318–326. doi:10.1080/09658210701801434

60.

Zerubavel

(1996). Social memories: Steps to a sociology of the past. Qualitative Sociology, 19, 283–299. doi:10.1007/BF02393273

61.

Zhou

Wildschut

Sedikides

Chen

Vingerhoets

A. J. J. M.

(2012). Heartwarming memories: Nostalgia maintains physiological comfort. Emotion, 12, 678–684. doi:10.1037/a0027236