Abstract:

Collective memory often appears in everyday conversations. Communicating with others about what happened to the past inevitably affects our collective memory by social context. In recent years, several studies have believed that the formation of collective memory depends on a dynamic system of communication, but few research has started from dialogue and interaction backgrounds and used empirical methods to explore factors that may affect collective memory at the group level. On the basis of existing studies, this research thus explores factors that may influence collective memory in an interactive way (i.e., memory conversation). It investigates how memory outcomes under this framework are affected by the interaction social context, which includes memory conversation, information matching, and group identity.

First, the effect of conversation on collective memory was investigated. Specifically, Experiment 1 explored the effect of activating information content through memory conversation and adopted a single factor (activation mode: conversation vs. individual) pretest-posttest experimental design (Figure 1). Second, the effect of information matching on collective memory was determined. Experiment 2 adopted a single factor (information matching: matching vs. mismatching) pretest-posttest experimental design. Third, the effect of group identity on collective memory in memory conversation was investigated. Experiment 3 adopted a single factor (group identity: in-group vs. out-group) pretest-posttest experimental design (Figure 2).

The results of the three experiments are as follows: (1) Conversation improved participants’ free recall scores at the individual level, whereas the nominal group did not. At the collective level, the conversation did not affect the group’s collective memory convergence. However, the conversation reduced the number of items forgotten by participants in common, F(1, 19) = 6.55, p = 0.019 (Table 1). (2) Given matching information on memory conversation, individuals can perform better in free recall. Information matching does not affect collective memory convergence, but information matching can increase the number of items remembered in common among participants, F(1, 13) = 13.60, p = 0.003 (Table 2). (3) When inconsistent information comes from the outgroup, individuals’ free recall performance is better. Only if the inconsistent information comes from the outgroup can make the collective memory among participants have more parts to remember in common, F(1, 13) = 25.37, p < 0.001. Meanwhile, inconsistent information from the ingroup and outgroup did not affect the number of items forgotten by participants in common (Table 3 & Figure 3).

In conclusion, these findings have important implications for understanding the mechanism underlying the effects of memory conversation and group identity on collective memory. Moreover, the function of communication is not to make the memory content close to facts, but to abstract our memories and remember things effectively. Our findings confirm that collective memory is not only recognized as an individual psychological phenomenon but also likely involves a kind of social property. Therefore, given the significant role of communication with collective remembering, placing people in a conversational background is a direct way to investigate collective memory.

Key words: collective memory, memory conversation, group identity, item remembered in common, item forgotten in common