The Parafoveal Processing Influence Selection Of Target During Chinese Reading

The selection of target is a basic problem in eye movement research of reading. As the next fixate area, the preprocessing of parafovea play an important role in target selection during reading. In this study, we investigated the influence of preprocessing in parafovea on selection of target during Chinese reading.Three studies including 7 experiments were carried out. Study 1 focus on the extent of parafoveal processing that would influence the selection of target. In Experiment 1, the preview of the first character(N+l) and second character(N+2) of a 2-character word was manipulated by using the boundary paradigm and the preview was either an identical character or a pseudo-character, to explore whether both the processing of character N+1 and character N+2 in parafovea influence the selection of target or not in Chinese reading. The results showed that the length of outgoing saccade that from the fixated word to parafoveal word was longer under identical preview than pseudo-character preview. The finding suggested that the processing of character N+1 and N+2 in parafovea both affect targeting in Chinese reading.In Experiment 2, the first character (N+1) of a highly-frequent used 2-character target words was either high or low frequency while the preview of the second character (N+2) was either an identical character or a pseudo-character to explore whether the influence of preprocessing of character N+2 was modulated by the processing load of character N+1. The results showed that neither the frequency of character N+1, nor the preview of character N+1 impacted the length of outgoing saccade from the fixated word to parafoveal word. The finding suggested that the preprocessing of character N+2 has nothing effect on selection of target.In Experiment 3, we selected a high-frequency character as character N+l. In the "two-character word" condition, the characters N+1 and N+2 formed a word, but in the "two single character words" condition, the two characters did not form a word. Using the boundary paradigm, we also manipulated the preview of character N+2, such that it was either an identical or a pseudo-character preview. The purpose of this experiment is to investigate whether the length of word which in parafovea modulate the parafoveal processing then affect the selection of target. The results showed that there was no significant difference in the length of the outgoing saccade from the fovea to parafoveal area between the two-character word condition and two single character word condition. In the two-character word condition, the length of the outgoing saccade from the fovea to the parafoveal region was significantly longer when the preview of character N+2 was identical compared to when it was a pseudo-character, however, the preview of character N+2 had no influence on the outgoing saccade in the two single character word condition. The results indicate that parafoveal processing of the upcoming characters influenced saccade target selection only when the characters in parafovea constituted a word. However, since saccade targeting was similar regardless of whether the the two parafoveal characters formed a single word, or two different words, overall, the results indicate that Chinese readers did not target the parafoveal word with their saccade.The study 2 examined whether the strokes and character frequency of the first character in parafovea affect target selection in Chinese reading. In Experiment 4, the preview of the first character(N+l) was manipulated, and the preview of character N+1 was identical character or fewer strokes character or more strokes, to explore whether the strokes of preview character influence selection of target. The results showed that the fewer of the preview of character, the outgoing saccade from fixated word to parafoveal word was longer, and the location of the first fixation on parafoveal word was further away from the word beginning. The results indicate that the strokes of the first character in parafovea influence the selection of target.In Experiment 5, the first character (N+1) of a highly-frequent used 2-character target words was either high or low frequency, of which the preview was either identical to N+1 or a pseudo-character by using the boundary paradigm. The results showed that the length of outgoing saccade from fixated word to parafoveal word was longer under identical preview than pseudo-character preview, but not the frequency of character N+1. The results state clearly that the preview of character N+1 affect the selection of target but not the frequency.The study 3 inspected the relationship between the foveal processing and parafoveal processing on saccade target selection. In Experiment 6, the fixated word in fovea was either high or low frequency, of which the preview of the first character (N+l) of a 2-character word was either identical to N+1 or a pseudo-character in parafovea by using the boundary paradigm. In Experiment 7, the frequency of the fixated word which in fovea was also manipulated like as in experiment 6, and the the preview of the second character (N+2) of a 2-character word was identical to character N+2 or a pseudo-character by using the paradigm. The results showed that the length of outgoing saccade from fixated word to parafoveal word was significantly longer under high frequency word than low frequency, and the length of outgoing saccade from fixated word to parafoveal word was significantly longer under identical preview than pseudo-character preview. These findings suggest that processing of fixated word and processing of parafoveal word independently affect selection of target.The conclusion was made as follow based on the findings of the three studies:(1) The processing of character N+1 and N+2 in the parafoveal affect the subsequent saccade target selection; but the influence of character N+2 play a role on saccade target selection only when the two characters constitute a word.(2) The information of strokes of the first character in parafovea can be obtained from parafoveal processing, and the Chinese reader use the strokes information to decide the next saccade target.(3) The information of character frequency of the first character in parafovea can not be used to decide location of the next fixation.(4) The processing load of fixated word affect the selection of target, the length of outgoing saccade from fixated word was longer under low processing load than high processing load.(5) The processing of fixated word and processing of parafoveal word independently affect selection of target.