READING RATES 



509 



35 



30 



CO 25 

 o 

 oc 

 o 



$ 



o 

 cc 



UJ 

 CL 



in 

 a 

 z 

 o 

 u 



UJ 



in 



20 



10 



12 3 4 



SYLLABLES PER WORD 



Fig. 7 — Patterns governing reading time. 



necessarily monosyllables in Chinese and happened to be monosylla- 

 bles in English. 



In one case a word is made up of a sequence of letters, each standing 

 for a sound, and in the other it is made up of a number of strokes which 

 are meaningless individually, yet in each case a word is taken as a unit 

 or pattern reciuiring nearly the same time for reading. 



We found that the rate for reading arable numerals is substantially 

 the same as for reading familiar words. Each numeral is an individual 

 pattern to be recognized. • 



In a fii'st effort to find the effect of syllable length on reading rate, 

 a subject read se^'eral lists made up respectively' from vocabularies of 

 l(i single-syllable, 16 two-syllable, 16 three-syllable, and 16 four-syllable 

 words. None of the words was very unfamiliar to start with, and all were 

 presumably very familiar after the subject had read several randomized 

 lists composed of the same words. 



The outcome of the experiment is shown in Fig. 7. For comparison, 

 the points associated with the lower straight line are time for 60 words 

 for repeating, as rapidly as possible, a one-, a two-, a three- and a four- 

 syllable word. The points on the upper curve are reading time for 60 

 words for the randomized lists of the familiar one-, two-, three- and four- 

 syllable words. 



In dealing with such groups of highly and uniformly familiar words, 

 it appears that, roughly, a certain time is required to recognize the word 

 regardless of length, and this time governs the reading rate up to the 



