Lending an ear and giving a hand: the interaction of working memory, long-term memory, and gesture, when adults and children learn new words

The learning of new words occurs over the lifespan and involves visual, auditory and embodied modalities such as gesture. To be successfully learned, or retained in long-term memory (LTM), a word must first be encoded in working memory (WM), the limited-capacity system in which immediate processing takes place. Both short- and longer-term memory of words may be affected by their modality, and the presence of co-speech gestures, as well as by individual differences in WM abilities, prior knowledge and strategy use. The aim of this thesis was to explore the relationship between learning new words and the interaction of these factors in adults and in children.

Research suggests overloading WM inhibits long-term learning, but few previous studies of the WM-to-LTM bottleneck used verbal stimuli. Chapter 2 reports six experiments presenting words in varying modalities to separate samples of adult participants. The likelihood that items were actively held in WM was controlled, and WM and LTM retention were tested. Evidence for an encoding bottleneck for visual, but not auditory, verbal information was found. Also with adults and using a similar paradigm, a second study (Chapter 3) tested the effects of co-presented gestures on novel word retention in WM and LTM. Gesture stimuli which had been rated for iconicity in a separate norming study (Chapter 4) were used. Gestures’ iconicity did not affect memory of words, however, words appeared more likely to be retained in LTM when strategies were reported and in the absence of gestures. Finally, (Chapter 5), the impact of gestures on young children’s learning of real-word words and concepts was examined. Gestures were found to modulate effects of prior vocabulary knowledge and WM abilities. Together, these findings elucidate complex interactions between modality of verbal information, WM capacity limits, the presence of gestures, and learners’ individual differences.