Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.637218
Title: The development of automatic semantic and syntactic context effects in reading
Author: Hapeshi, K.
Awarding Body: University College of Swansea
Current Institution: Swansea University
Date of Award: 1987
Availability of Full Text:
Access from EThOS:
Abstract:
Empirical evidence is presented which shows that readers can utilise semantic and syntactic context in order to aid word recognition, either through a conscious strategy-related process, or through an automatic spread'ng excitation process. The experimental results described show that, when average readers are used as subjects, semantic context effects may be evident in readers as young as 6 or 7 years of age, and in readers aged 11 years and over. There is little evidence for the use of context in 8 and 9 year old readers. This age-related pattern of results is consistent with the theory that readers at different developmental stages utilise semantic context primarily through different processes; young readers through a conscious process, and older, more skilled readers, through an automatic process. Readers between the ages of 8 and 10 years of age are at some transition stage between the two types of process, which results in the absence of context effects. Experimental evidence was also presented, showing that syntactic context can be utilised through either a conscious or an automatic process, but only in fluent adult readers. The automatic effects of syntactic context are similar to automatic semantic context effects, which suggests that they involve the same type of mechanism of spreading excitation through an associative network. In order to show how such a mechanism could work, a model for the Lexicon is described, which argues that both semantic and syntactic information can be encoded through associative learning providing directional links between nodes in a neural network. Spreading excitation through the resulting structural network could then account for automatic context effects.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.637218  DOI: Not available
Share: