0 - 521 - 15727 - 8 - Components of L 2 Reading : Linguistic and Processing Factors in the Reading Test Performances of Japanese EFL Learners

The practice of developing and testing second language (L2) reading must follow as clear an understanding of this skill as possible, and such an understanding must be a result of rigorous research, both theoretical and empirical. Comparisons have been made between applied linguistic research on L2 reading and cognitive psychological research on L1 reading, both of which directly or indirectly contribute to our understanding of the nature and development of reading. Compared to cognitive psychology’s extensive and cumulative research base resulting from successions of empirical data collection and theory and model generation, not only for description but for explanation and prediction of reading behaviour, L2 reading research eff ort within the applied linguistics community has been said to suff er from lack of a solid empirical research base (Bernhardt 1991a) and a resultant relative weakness in generalizability and predictability of L2 reading behaviour. It is therefore imperative that more principled eff ort be made to accumulate empirical evidence on the nature of L2 reading. While L2 reading research has at times drawn insights from the L1 reading literature, the issues it has addressed and emphasised have tended to be distinct from the ones researched through typical cognitive psychological approaches. Two questions which characterise L2 reading research have been whether L2 reading diffi culty arises from incomplete L2 knowledge or insuffi cient L1 literacy (Alderson 1984) and whether there is a linguistic threshold level which the L2 reader must achieve in order for his or her L1 literacy skills to be positively transferred to the task of L2 reading comprehension (Clarke 1978). These questions obviously relate to the facts that most individuals learning to read in L2 do not have the kind of sophistication in linguistic competence shared by the majority of the native speakers of the target language and that these L2 readers already have varying levels of literacy skills in their own L1. While these features specifi c to L2 reading must be suffi ciently highlighted and refl ected in L2 reading research designs (Alderson 1984, Koda 1994), eff ort also seems to be justifi ed which examines the factors considered to be signifi cant for L1 reading comprehension. The advantages of such an eff ort would be the availability of methodology established in L1 research and its extensive body of research evidence, which enables us to compare L2 reading 1