Priming effects in early readers. A quantitative study of children's response times in visual lexical decision-making in their first and second language.

Abstract

This thesis investigates the effects of various conditions of open priming on children in a series of lexical decision tasks in their first and second language. The primes are related through meaning and/or form to the target words in the experiments. The five conditions used are same word priming, meaning-related, shared first syllable, shared last syllable or unrelated. In addition, the experiment consists of 50% pseudo-words. There are two experiments which both consist of two parts. The first experiment is designed to investigate how priming affects children in their first language, Norwegian. The second experiment is designed to investigate how priming affects children in their second language, English. Each experiment consists of two equal parts, one part where the target word is primed and one part where the target word is not primed. The unprimed responses provide information used to calculate a predicted response time. This is used to ensure that the results are not skewed by the fact that children read at different speeds. The results show that children benefit from priming, especially in L2 where they appear to rely on orthographic similarity between the words. In L2 response times were significantly reduced when primed with the same or a form-related word. In L1, the results for priming were weaker, and only one type of form-related priming significantly reduced response times in the model that takes expected response times based on baseline responses into account. This supports mainly Seidenberg’s (2005, 2012) PDP model, especially in L2. There are few indications in the data that meaning-related primes has an excitatory effect on lexical decision making tasks in 12-year-old Norwegian children as suggested by Levelt (1989, 2001).