Although word learning is the foundation of language, little is known about the processing mechanisms that underlie it. Recent evidence has emphasized the relationship between word learning and verbal short-term memory and long-term memory systems. However, here, too, little is known about the basis of such relationships. Investigation of these issues is of key importance in understanding the underpinnings of language, and in improved understanding of language disorders. This project will test and further develop a novel theoretical framework that relates the mechanisms of word learning to verbal short-term memory and long-term memory, and will apply this framework to investigation of the developmental language disorder known as specific language impairment (SLI), which has been estimated to affect up to 7% of children in the US. According to this framework, learning the sound pattern or phonology of a new word is initially supported by verbal short-term memory, but the sound pattern is eventually learned by drawing on the long-term memory mechanism termed procedural learning. Learning the associations between a sound pattern and its meaning or semantics requires support from the long-term memory system termed declarative learning. Word learning is thus seen as the confluence of linguistic and memorial processing. This framework has been implemented as a computational model.
The specific aims of the present project are to (i) test the theoretical framework through behavioral studies, (ii) extend it through computational modeling, (iii) study the performance of a group of children with SLI using a set of tasks that are grounded in the theoretical framework, and (iv) simulate the behavioral performance of these children at the group level and at an individual level. A total of eighteen behavioral and computational experiments will be conducted. The ultimate goal of this program of research is to construct a theory of word learning that integrates across language and memory and that is specified at the functional, computational, and neural levels. This theory has important long-term implications for language disorders. Specifically, the use of experimental tasks that are interpretable in terms of the theory will facilitate identification of specific language functions that are compromised in a given disorder; computational modeling will enable hypothesis testing regarding the precise nature of the impairment.
| Kapnoula, Efthymia C; Packard, Stephanie; Gupta, Prahlad et al. (2015) Immediate lexical integration of novel word forms. Cognition 134:85-99 |
| Gupta, Prahlad; Tisdale, Jamie (2009) Word learning, phonological short-term memory, phonotactic probability and long-term memory: towards an integrated framework. Philos Trans R Soc Lond B Biol Sci 364:3755-71 |
| Gupta, Prahlad (2008) The role of computational models in investigating typical and pathological behaviors. Semin Speech Lang 29:211-25;quiz C 5-6 |
| Gupta, Prahlad; Lipinski, John; Aktunc, Emrah (2005) Reexamining the phonological similarity effect in immediate serial recall: the roles of type of similarity, category cuing, and item recall. Mem Cognit 33:1001-16 |
| Gupta, Prahlad; Lipinski, John; Abbs, Brandon et al. (2004) Space aliens and nonwords: stimuli for investigating the learning of novel word-meaning pairs. Behav Res Methods Instrum Comput 36:599-603 |
