This dissertation outlines two experiments aimed at investigating how adults learn words in a second language (L2) that do not have one exact equivalent in the first language (L1). In Experiment 1 participants learned a set of vocabulary words in an artificial L1, followed by a new set of vocabulary in an artificial L2. Results revealed that participants had greater difficulty with differentiation - learning two specific words in the L2 that corresponds to a single, more general equivalent in the L1 (e.g., a native English speaker learning the two forms of to know in Spanish -- saber and conocer), than they did with unitization -- learning a single, general word in the L2 that corresponds to two, more specific L1 words (e.g., a native Spanish speaker leaning the single form of to know in English). In Experiment 2 participants performed a similar task, except when differentiating, the two L2 words had an asymmetrical relationship in terms of how general their meanings were (e.g., saber, which refers to knowing things factually or how to do things, is more general than conocer, which refers to knowing people or places personally). Frequency of these words was manipulated, so that for half of the participants, the more general word occurred more frequently, as is often the case in natural language. For the other half of participants, frequency of these items was held constant. It was predicted, and confirmed that those in the Different frequency group would have greater difficulty learning the specific words that were differentiated in the L2, as opposed to the general words. Results from both Experiments 1 and 2 confirm predictions of the Competition Model (2008) concerning the relative difficulty of differentiation versus unitization, and the effects of word frequency on L2 learning.
展开▼
