READING PRINTED TEXT 



63 



letter height, but in height- width ratio and 

 in thickness of stroke (boldness) . In a given 

 type face, a smaller point size may not be 

 a photographic reduction of a larger, since 

 certain letters may differ in details of design, 

 and the reductions in letter height and in 

 thickness of stroke may not be in proportion 

 to each other nor to the reduction in point 

 size. Type designers have resorted to these 

 dimensional disparities to preserve certain 

 perceptual uniformities as pleasing appear- 

 ance, legibihty, and style. 



A comprehensive investigation of typo- 

 graphical factors carried out by Paterson, 

 Tinker, and associates has been brought to- 

 gether in their book, How to Make Type 

 Readable (33). Considerable work by Luc- 

 kiesh and Moss is reported in The Science 

 of Seeing (29) and Reading as a Visual Task 

 (30). 



Paterson and Tinker gave standard speed 

 tests, taking If minutes, under classroom 

 conditions, to school and college students, 

 in groups of 80 or 90 to several hundred, 

 totaling 33,000. Luckiesh and Moss re- 

 ported results usually in terms of bhnk rate, 

 sometimes in terms of reading speed when 

 subjects were reading at their normal rate, 

 and occasionally in terms of reading speed 

 when subjects were reading at their maxi- 

 mum rate. Experimental groups contained 

 typically 30 or 40 subjects, lighting was 

 controlled, and sometimes retests were made. 

 The test period was usually five minutes. 



Kinds of Type 



With other variables held constant, style 

 of type face appears to be a minor factor in 

 ordinary reading. Paterson and Tinker (33, 

 p. 16) found a difference of less than three 

 percent in reading speed among eight type 

 faces in common use, but American Type- 

 writer reduced speed about five percent and 

 Old English about 14 percent. Luckiesh 

 and Moss (30, p. 163) in testing three type 

 faces found insignificant differences in read- 

 ing speed, but up to 14 percent differences in 

 blink rate. The relative blink rates, how- 



ever, were in the order of relative boldness 

 for the three type faces. 



Boldness 



In another study, Luckiesh and Moss (30, 

 p. 171) found some differentiation be ween 

 degrees of boldness, in a particular type face, 

 in terms of reading speed and blink rate, the 

 optimum boldness being somewhat heavier 

 than that of the average type face in common 

 use. Paterson and Tinker (33, p. 27) found 

 boldness not to be a significant variable. 



Italics and Capitals 



In comparison with ordinary lower case, 

 Paterson and Tinker (33, pp. 21-23) found 

 that itahcized material was read about three 

 percent more slowly and material set in all 

 capitals about 12 percent more slowly. 



Type Size, Line Width, and Leading 



On the question of type size, the several 

 investigators emphasize the interaction with 

 line width and leading. When one factor 

 alone is varied experimentally, the results 

 tend to be specific for the conditions of the 

 experiment. Therefore, more helpful con- 

 clusions can be drawn from experiments 

 which include combinations of variables. 

 Type sizes from 9- to 12-point were found 

 by Paterson and Tinker (33, p. 81) to be 

 read with about equal speed when each was 

 set in its optimum line width at 2-point 

 leading. In a somewhat analogous experi- 

 ment with the bHnk-rate criterion, Luckiesh 

 and Moss (30, p. 150) found generally similar 

 results. A further experiment by Paterson 

 and Tinker (33, p. 80) enabled them to 

 define, for each type size, certain safety 

 zones of Hne width and leading within which 

 reading speed was not impaired. For 6- or 

 8-point type set solid (0 leading), line width 

 could be varied from 14 to something under 

 28 picas. Increasing leading through two 

 points tended to extend the line width range. 

 Increasing type size tended to extend the 

 range and raise the lower limit slightly, so 

 that for 12-point type leaded two points, the 



