Jerwen Jou, Ph.D.
TITLE: Professor of Psychology
COMPANY: University of Texas- Pan America
Advisor: Dr. Richard J. Harris
The automatic processing of some grammatical rules
The nature of the automatic processing of some English inflectional rules was investigated in this study. In Experiment 1, it was found that native-English-speaking college students showed a lengthened reading time for a sentence that contained an occasional inflectional error. Thus, these native speakers of English were sensitive to violations of verb inflections even in fast reading, suggesting automatic processing of the verb inflections.
Experiment 2 found that subjects could successfully ignore the violations of inflectional rules when every sentence they read contained such a violation. This result indicated that the processing of inflectional rules could be inhibited when the frequency of such violation was very high, suggesting that the automatic processing of the inflections was not uncontrollable.
Experiment 3 showed that the successful inhibition of inflection processing could be developed within a dozen or so trials, when the errors consistently occurred in each sentence, suggesting that suppressing the processing of the inflections did not require too much practice. Throughout the first three experiments, the redundant nature of the English verb inflections was also experimentally demonstrated.
Experiment 4 showed that the correct use of the inflectional rules in speech production was not affected by dividing the attention and overloading the memory, strongly suggesting that the process of applying these rules was automatic.
Experiment 5 demonstrated that oral reading of separate sentences and connected discourse containing inflectional errors was more difficult than reading these texts not containing the errors. Subjects showed strong response competition or conflict effects when they read the error texts, suggesting the strong automatic activation of the correct inflections. The experiment further showed that consistent occurrence of errors made reading less difficult compared with inconsistent occurrence of the errors.
The combined findings from the first set of three experiments showed that some basic low-level syntactic processing in reading was not independent of higher-level information in the text. The combined results from four of the five experiments indicated a consistent pattern of the importance of a consistent task requirement for the changing of an automatic mental process.
Ph.D., Psychology, Kansas State University, 1990