WHEN DO MEMORIES GO AWAY? L1 ATTRITION EFFECT ON BILINGUAL AUTOBIOGRAPHICAL MEMORY

Abstract

Language plays a crucial role in remembering, storing, maintaining, accessing, and sharing of memories. This evidence comes from the disciplines of psychoanalysis (Javier et al., 1993; Schwaneberg, 2010), developmental psychology (Fivush, 2011; Nelson, 2004), and bilingual memory (Larsen et al., 2002; Marian & Neisser, 2000). Some of the most telling examples come from bilingual psychotherapy where therapist switch to the patient’s native language in order to access childhood memories (e.g., Aragno & Schlacher, 1996). The loss of language, therefore, may have a detrimental effect on memory storage and recall. Until now, however, this possibility has not been tested. The purpose of the present study is to address this gap by investigating autobiographical memory in speakers undergoing L1 attrition. The study is grounded in the dual-coding theoretical framework (Paivio, 1971; 2014). According to the dual-coding theory, memories that are encoded with multiple memory traces (audio, visual, etc.) are remembered better and a weakening of a memory trace due to its inactivation might lead to forgetting. Therefore, memories that were encoded in a language that is no longer available might show signs of deterioration. Twelve non-attriters ages 18-28 (M=22.08; SD=3.73), 13 moderate attriters ages 18-33 (M=24.29; SD=5.43), and 10 advanced attriters ages 18-30 (M=23.1; SD=3.7) shared autobiographical memories with the help of free recall and cued-recall procedures (Marian & Neisser, 2000; Schrauf & Rubin, 2000). The pool of 420 free recall memories were analyzed for amount of detail (Levine et al., 2002) and the set of 1,988 cued recall memories were analyzed for phenomenological properties of vividness, significance, emotionality, and confidence in the event (Schrauf, 2009; Schrauf & Rubin, 2004). The results revealed that moderate attriters who had vague L1 Russian memories recalled memories with lower confidence rating in comparison to non-attriters, which suggests a negative L1 attrition effect on bilingual autobiographical memory. However, the advanced attriters were able to recall vivid and detailed L1 Russian memories. The aforementioned finding did not support the study’s hypothesis that memories might be forgotten if the language of encoding is no longer available. This can be explained by the following observation. Advanced attriters recalled their L1 Russian memories very frequently which insured a preservation of these memories. This rehearsing of L1 memories in their dominant L2 English also caused re-encoding of these memories into the dominant language. This was observed qualitatively (participants commenting on how L1 Russian memories were coming to them with L2 English words) and quantitatively (L1 Russian memories were reported to be accompanied by L2 English words). These findings have several theoretical implications for the discipline of bilingualism. L1 attrition may have a negative effect on bilingual autobiographical memory, at least as far as the phenomenological properties are concerned. However, this negative effect can be reversed by frequent rehearsing of memories that were encoded in an attrited language. Rehearsing may lead to re-encoding of L1 memories into the L2 which suggests that memories may be malleable. This demonstrates flexibility of bilingual mind and how it can adjust to L1 attrition.