Effect of window size on reading - scrolling window revisited

Main author: Suzuki, Michiko; Tokyo Woman's Christian University; Japan

Co-authors: Kawashima, H;Kobayashi, A;Oda, K

Purpose

Reading performance is known to be degraded with limited field of view and it explains how severe peripheral field loss eventually leads to slower reading. We investigate into how limiting window size makes reading difficult.

Method

Thirty-character easy-to-read sentences from MNREAD-J charts were presented scrolling from right to left in a window of either 1, 2, 4, or 8 letter widths at one of the following rates; 10, 20, 40, 80, 160, 320, 640, 1280, and 2560 letter/minute. Eighteen normally sighted subjects were asked to read the sentence aloud as precisely as possible. Each sentence consists of 8 Kanji and 22 Kana characters and we compared correct recognition rate between character types and among positions in the sentence.

Results

Probability of correct reading was 1 up to the scrolling speed of 320 letter/minute except 1-letter window for which correct probability was 1 up to 160 letter/minute. For faster scrolling speed, correct reading ratio fell down quickly and how it fell interacted with the window size in an orderly fashion. However, there was not any difference between Kanji and Kana letters in psychometric function, nor any position dependence in error rate.

Conclusion

Reading becomes difficult when letters presented very quickly and in a small window. However, our data suggests that reading error occurs stochastically and simply independent of letter types and positions. When we take the probability value for 1-letter window and apply probability summation formula, error ratio data for other window size were well explained.