Intra-Familial Similarities and Cross-Generational Differences in the Earliest Childhood Memories of Three Generations of Women

• In Event: 1-164 - Your First Memory may be Earlier than we Thought: Evidence from Retrieval and Cohort Effects

Integrative Statement

Researchers have extensively studied earliest childhood memories. Previous literature reveals systematic differences between individualistic and collectivistic cultures; North American memories are dated earlier and are richer in context and self-emphasis, whereas East Asian memories are dated later, provide only skeletal descriptions and include more emphasis on significant others (Wang, 2001). Another line of research focuses on self-construal differences and reveals that subjects who are both individuated and related provide richer and earlier memories (Sahin & Mebert, 2013).

Psychologists often attribute these differences to variations in early mother-child reminiscing styles, which are theorized to demonstrate a trans-generational pattern. To explore this possibility, we examined the earliest childhood memories of daughters, mothers, and grandmothers from the same families in two investigations using different methodologies. We hypothesized that the age, volume, specificity and inclusion of references to self and other would show similarities among family members. We also predicted that the same set of variables would show differences among the three generations.

In Study 1, 59 daughter-mother-grandmother triads filled out a memory questionnaire that included an open-ended question, asking participants to write about their earliest childhood memories and the ages at which the remembered events occurred. In Study 2, we recruited 105 additional triads and conducted semi-structured interviews with each participant. In these interviews, the age of the memory was a separate question. In both studies, daughters were in their early 20s, mothers were in their mid-40s, and grandmothers, about 70. All daughters were college students, and all mothers and grandmothers were literate.

A second researcher coded 20% of the data in each investigation, with inter-rater reliability coefficients ranging from .86 to 1.00. Findings of Study 1 showed significant intra-class correlations (ICCs) for the volume (total number of words), level of detail, self-related words, and other-related words, but not for other/self ratio, age of the memory, or specificity (Table 1). The results also showed a cross-generational difference for the age of the memory, F (2, 174) = 5.64, p < .01, η2 = 0.06 (Table 2). Post-hoc comparisons revealed that daughters’ earliest childhood memories were dated significantly earlier than grandmothers’ memories. There were no trans-generational differences for the other variables.

The follow-up study replicated the results of Study 1. There were significant ICCs for volume, level of detail, self-related words, and other-related words, but not for other/self ratio, age of the memory, or specificity (Table 1). For the cross-generational differences, the previous results were replicated for the age of the memory, F (2, 236) = 24.96, p = .000, η2 = 0.17 (Table 2). The age of the earliest memory again differed across the generations, with a minor difference in post hoc comparisons: mothers’ reports also differed significantly from grandmothers’ and daughters’ memories.

Both studies showed intra-familial similarities, indicating the trans-generational nature of reminiscing style among female members of families, and indicated that earliest childhood memories date from an earlier age for the younger generations. Further studies should examine male members of the families, and other cultural settings.

Authors

• Basak Sahin-Acar, Middle East Technical University
  Presenting Author

• Tugce Bakir-Demir, University of Otago
  Non-Presenting Author

• Elif Gizem Kus-Tursel, Uludag University
  Non-Presenting Author