| Original language | English |
|---|---|
| Publication status | Unpublished - 2014 |
| Event | AAAL, 2014 - Portland, Oregon, United States Duration: 2014 Mar 22 → 2014 Mar 25 |
Conference
| Conference | AAAL, 2014 |
|---|---|
| Country/Territory | United States |
| City | Portland, Oregon |
| Period | 2014/03/22 → 2014/03/25 |
Bibliographical note
AbstractTraditionally, most studies investigating L2 lexical processing have either focused on single words or idioms, but there is now a growing body of research that focuses on less idiomatic word combinations such as collocations (Wolter & Gyllstad, 2011, 2013; Yamashita & Jiang, 2010). One theoretical approach to word combinations (Howarth 1996, 1998) assumes a continuum model of semantic transparency and restrictedness, from the most transparent category – free combinations – through collocations, to the least transparent category – idioms. Of these three types, collocations, although typically argued to be unproblematic from a comprehension perspective, are often seen as a major hurdle for L2 learners (for a review, see Henriksen 2013), whereas free combinations are largely considered unproblematic. Furthermore, Howarth (1996) has argued that the degree of restrictedness is psychologically related to processing and storage.
An experiment was designed to determine if the division into the two categories of collocations and free combinations has psychological validity. Specifically, a lexical decision task (LDT) was used to assess advanced L2 speakers’ reaction times to English free combination items versus collocational items, with L1 speakers as controls. In the study, collocations were seen as word combinations where one of the constituent words is either used in a delexical, technical or figurative sense (e.g. serve a purpose), whereas free combinations where seen to consist of words used in their literal sense (e.g. serve a drink). The underlying assumption was that if collocations are indeed more difficult to process, due to a slightly lower degree of semantic transparency, then they should be recognised more slowly and less accurately than a matched set of free combinations. The results of the study are discussed in the light of the continuum model in particular and word combination typologies in general, and implications for word combination storage and retrieval are presented.
Subject classification (UKÄ)
- Languages and Literature