Content area
An ongoing debate in psycholinguistics concerns the structure of the language processor. One class of theories suggests that the language processor contains subprocessors which are autonomous. Another class of theories suggests that the subprocessors are interactive. Both types of theories have proposed that the initial syntactic structure assigned to linguistic input is a function of the structure of the language processor. This dissertation proposes that syntactic processing is an automatic process that is governed by the general principles of automaticity. In particular, the proposed theory characterizes syntactic processing as an automatic process in the sense intended by Logan (1988). Based on Logan's theory, it is hypothesized that the initial syntactic structure assigned to linguistic input is determined by the retrieval of past syntactic structures from memory. Logan's theory makes a number of predictions and assumptions about both the qualitative and quantitative properties of automatic processes. Automatic processes are fast, effortless, autonomous, and unavailable to conscious awareness. Phenomenal experience and empirical evidence suggest that syntactic processing also possesses these qualities. The experiments conducted represent an attempt to determine whether syntactic processing is also quantitatively similar to automatic processes in general. Logan's theory makes two quantitative predictions: performance will improve as a power function of practice and learning is item specific. Using artificial and natural language stimuli, the current experiments provide evidence that syntactic processing also improves as a power function of practice and that learning of particular syntactic structures does not transfer to processing of other syntactic structures. The results indicate that syntactic processing may not be a function of the structure of the language processor as previous theories have suggested. Rather, syntactic processing may be an automatic process that has developed through experience with a given language. The results also suggest that the role of experience in syntactic processing should not be ignored by any theory which attempts to provide a thorough explanation of language processing.