Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Janet McDonald


The present research explored the processes and mechanisms of first language forgetting from linguistic and psychological perspectives. Two studies, one---a case study, another---a controlled experimental study, yielded evidence on L1 loss. The case study involved a 9 year old Russian girl who was adopted by an American family, brought to the US and completely taken out of a Russian-speaking environment. Over the course of one year, changes in L1 morphology, syntax, and vocabulary were investigated using various tasks, e.g. picture description, picture naming, story telling. Data on L1 retention, L2 acquisition, and reaction time were gathered. The results of the study suggested that L2 transfer may cause certain morphological and syntactic errors. However, the major findings in this study involved vocabulary, where three groups of words showed high vulnerability to loss, i.e. cognates, non-distinguished categories (pairs of words lexically distinguished in L2 and non-distinguished in L1), and high-frequency words. Fast loss of these lexical items was related to the acquisition of their equivalents in L2. Thus, this semantic overlap between L1 and L2 labels may cause L1 forgetting. The experimental study further investigated semantic overlap by comparing the performance of the experimental group who learned lexical labels in two languages for the same concepts and the control group who learned lexical labels for non-overlapping concepts. The number of L1 learning trials remained the same, while the number of L2 trials varied across groups. Reaction time was precisely measured in this study. The results of this study confirmed the hypothesis of semantic overlap as a cause of L1 loss: that is, experimental participants showed more forgetting than control participants, particularly with a high number of L2 trials. Thus, both studies converged on the conclusions that L1 loss is determined by L2 interference, and that semantic overlap is a mechanism of L1 loss and is noticeable with high amounts of L2 learning. The findings of this research may have implications for studies on L1 loss, L2 acquisition, and psychological studies on retroactive interference.