Learning new words: effects of meaning, memory consolidation, and sleep

Although encountering novel words in one's own language in adulthood is not an uncommon event, the relevant cognitive processes have become the target of systematic investigation only in recent years. This thesis addressed three main questions regarding word learning. The first was concerned with the role of meaning: to what degree is meaning necessary in integrating new representations in the lexicon? Experiments 1-3 suggested that meaning is indeed important. In the absence of trained meaning novel words may "inherit" the meaning of neighbouring familiar words, possibly explaining some seemingly incompatible reports in the literature (Experiment 1). Experiment 3 showed that such inherited meaning is sufficient to allow integration of novel words in the lexicon. Having established the importance of meaning in lexical integration, the thesis moved to the second question: does knowledge of novel word meanings benefit from offline memory consolidation? Experiments 4-7 suggested that this is the case. Experiment 4 showed that consolidated novel words elicited faster semantic decisions than words learned just before testing, while Experiment 5 showed that cued recall of word forms is also enhanced over time. Experiments 6-7 refined these conclusions by using semantic priming paradigms, showing that novel word primes facilitate processing of semantically associated familiar words after a period of offline consolidation has been allowed to operate over an extended period of time involving several days and/or nights. The third question focused on the role of sleep in the consolidation of novel words: which aspects of sleep architecture are associated with lexical integration? Experiment 8 looked at sleep during the night after word learning and sought to clarify the roles sleep spindles and different sleep stages play in word learning. Spindle activity was associated with the emergence of lexical competition effects, suggesting that sleep has an active role in word learning, and that spindles in particular are associated with lexical integration. These effects were interpreted in light of complementary learning systems theories.