—-— v~-~_,_-00'0-‘_- 0- ‘ ===!'—vv-'0— — - »----‘"‘_.q- .-v -—-._'.s‘_ -- “M- ’0'.» 0..-mu v' “ww— -w-v-0—vw——v u‘v—u—u — -VV—vvv—V. "V 0. .... O . 0‘.-. .'.o m ...wa V. G. m m 0 SIMULATED JG‘HT AND DAY D ism AND MEAS . URED "BY THE TTTMUS VISION T . ..nn mun m C IL ..L. M 3 ”MIT 0.0.1' . o 00-..: 0 A0 I r - 0. 00 A 0 .0...000.J0.00 000.564.000.001 ...0.0:00.0000.._ 1x10 00’... ... .000. .0. ..00. ...0 -.0...000 00.,0...0.000.0.....000 (.0.. 000’ 00:. o . .00. _...00.. 00 . 0 ...-t 00.00000...00-0000. 000. .000! "O.‘00 .. 0 . . . . .. ........0o. 00-... .00. .00. 0...00.. 0.00... .0.—0:0): 0. 0. 0 . . ._ .. . _ _ ...0..0....000 0... 00.. .0A.0.0. . n0 ..n0‘uld ..0 .y/ 0.00 .00 40000 .. - .u . . . . . .. . . ... ...0..o0 .....0... . ... ...0..00..000... 0.0.0.... 0. 0 10 ...0 .. 0 : . . .. . . .0. . ....0....00... ,.-.-.00..00....0-_0....,.00..;0- ... ... .04 . . . . . .4 ........0.. ... 00.0.0000000.... 00.0.00. 0..000 _ Vfr,‘./000_. 0 0 . 2 . . . 0 _. 0. .0.00 .0 0 ..0000.-.. 0.0. a 00.1.2.0.a- . 0.00 .10 .0.. n . . . . . .. ..0.F..... 0. ...... .0..0.......0.0 000..0 0005 0000.00”: . .- . . 0 0 . . 0 . . ... .4...0.0 0 ...0. 0 0,... 0...... 00 .0000 00.0.00 0.0.0.000 0 . 0 ... . . T 0 . .....00.—0..00_ 0.... 0.0.000..... . .000.0.0 000 30-70-1000 0 1 0 .. 0 0 0 .. ..0. .00. 0.0 0.00. .0 000. .0.....;.... .0000 .0.-00.0400‘0. . . . .... . ........0....0. . 0..0.0...0. 0.!0 ”0.0.0.. 0‘04...’000000..IO . 0. .0....0 .. 0. ....000 .0000 0000.0 00. .00 0 .03000.’ . .0.!“.. 0’ 0,000. '0 1 . 0 .I _ 0 t r.V . . . .. . 0..; .0. .0 .0. ..0. .00 0. 0 . 00.00..0 ’804 .0.. 000040.01" no. .0 - ._ .0 .0. pl. ...—...00 0.......0.00 .0..... .0/00,0 0.0... I . 0 . 0 .0 . 0.. . ..... 0 0. ...0000 .000 0.000. .00 Y.0.0.000¢ ..00 0 0 .0 .- ..I . .0....- .. ...0. 0.0...00 . A 0 - 0.0.0 0 00000’0r300'00t0 F . . 0 . 0 ._ .. .0.00 . .... 00.. ...... .Q. 0.. 00 0.0.10.0 t0 . . . O - . . 0... 0.00 0.0-0.0-., .00.. .000...00.00'.00.’0 .v...040 00 0'0 £0.00’l’ r .. . . . o 0... 0. .0.0. .0... o 0...000.0 0.0.0... . ..00042000000: ..0 O . .. . 0 . ...... .. . 0 .0. .0 0 0.... 0 0000000 0..-V000 00 l0000!0.000’ .r 0 0 .. ._ .0 ... .0000...0 n . 0.... 40050000000. 0 v. U .0.. . . .0 0 .. 0.....V .0.. 0 ..0 0.. .000..0.0. 0.0.00.000 .0 V 0 _ .. .00. .0 .0 .0. . 0.. ..00.0000 ..., - 00 0 _ . 0 .0 .0.... . 0.. 0 . . 000..0 000 . . .. .......0.. . ........-..A.0... _ . . . 0 V, . .0....0... s. 0. 0. 0.0.000 .. . . 0 .00 ..00. 0 ... .0 ...000.0 . 0 0 0. .0 0.. . .. Q-IOIOT v.104 0 ._ 0A 0 .0....0. .0. 000.00..v 0 0 0 .... . .00 .0 .0. . . 0 . . 0. . 0 . ,.0.0.000.. (00.00.00/0 .. 0 0 00.. . .0 .0, 0 0100.10.00. . . 0 0 . .... . .0.-0 . 0 O . I . 0 0 0 .00... 0 .0 0 0 0 .0 .....0 . - 2 f 40 . - .. . . . .,. ...-0.000 .I 0 . 0 .0 .0.- 0 0 .0 .0. o. 0 2 . _. . . 000 . 0 .. . . . .0000. . . .000 0 0.000 00.0. 000.0000.J0. 0..0.H..0....0w0. 0.. 1.0!. 0.)”. 00. 0000.0 .-.'f¢00- . 0 0“ . , 0 1.. 0 .0 ... n 0 0.000 4. ~000..0..‘l.. 7 0'0... 10‘ 0"». J'.‘00€ I. .045. 0 0 A .. a . 0.0 . . A. .. 0 .0v.0-.0A00.. . 0 .0. .0 0. 0. 0. .000 .... .... AA .0; 900.00 0 0.0!”. I.. 0. n ”/ omr’ . . . . n. ._ 0 0.0.00.00. .00000000V. .‘.. 00V.n~. 0P0000o . _ . .. . . . . ... .I...._ .. . 002...... . n . 0 0.. 0 010‘. . 00. 0.00 0 V 0 . a 0. 0.V. . 0. 00.0lav.0r 0 . . c . 0 .0 ..0.I0O . 0 .. 01‘000 . . .. . . .0 0 .0... . .0.... ..0000... . . . . . -... . .. 0.0. .00.000..-. .0. 0.— . . 0. _ .. .. . 0.... ..-0,. .00.-. _ .0 0 ..00 . .... . v .00’. .4 . 0 . 0. _ .... - 0 . . l .. 0 .0 0 0 .0. . ...-.0 0 0. 0 I n 00 I 00. .0 ¢ 0 0- : O 0 . 0 0 _ 0 0 0. '. .00 0 . s . . . .. 0 . ... . 0 a. #00 0. . 1 . . 0 . ...... ...00 ..0. (0000000._0/I.¢\.. .mr09.0#0,0rlrtm.fo 0. 0. .0 0 . . 0.~a. 00.00005P0.o-000.0..'0 :.0$.JI’('-vu' 40. . . .. . . .. 1 . 0 .0.- .000 e .0000 0.00,. (O. ...10 '_ . . . . . . 0 . a. .. a .0.. .0 . 0....00..0.. .... f. 0 0'5“? 0r/3 . 004?. JP; . . . 0 .0. .I .00.. .0 . 0000 0000.0. #0,.IHI’0‘10010. w."0'“ . . . 0 0 . A . . .. 0.0 0. r 40. I] I. 0’0—01fllu 0’0...‘ 0 _ 0 0000 J 0.. .00] )0; .0 0 _ 0 . . 00...;. .000 [000.01.00 0200 . 0 .0 0 00000.0. . 0O _0 .0010 10:0} .glfi‘l. . . . . . 0 . 0. V . . . . .00 4.0..0000.0.-0.. 010:..00.r..0.0.0 \000H0..0.HC0’4r(0Pr0 . . . . ..0 00000000... .0.. ..000 02’ 05...! I 03.; 0 . . 0 ... . . . I .. A0100..’_0vo00 ’r Vr‘lro.’- '0'. . . . . . 00,000.]. .. ”0.0.0.0r...000.000'.0.0 '00 r . .. . . . ..0 .. 0.. ..a.:...00.. 0.0r0.000000.0lr'10 ’0’ .. n . .0 000.. 00 0'.0.01rr ' Cr \0’0.'0 . . . . .00 :0 0 I . . .000 . .0! 0 . ”I v. 30“ . . . . f 0 .0 0‘04000 00: 031’0 .V 0.0000.0.000.0_ 00000.00...A".I0 4 I .0 . . . 0 0 . 0 00, 0 00 .000010..000d 0100' 0.000.200. “000“ . 0. . . . 00.0 ‘0 1 0 5.000007”? 0700’. 0 - . . 0 . .-0.0. . . ..00‘ .7000..00 . Jv'.'400 . 000 .....00....0:.0..- .‘.I0V 300.0. - l 01.“! . .0. . - 0 . 0. 0. ... ...... 0..1. 00.0.0..f‘tr0fi10 0 . 0 . . , v) 00.. I . . _ . . . . 0. . -- .. .. If... . .. . ..............or.v.. 514...} O 0 . 0 (00-.. 0 l -A. . - . ........0 0 0. .0 . 0..-00... 0....0000..000....4A.0.0¥0H'..r.rv \A. 0 00.00 . : . . .0 . . 1 . 0 A 0.0. 0 0. . 0. . - .0 .00 0 . 0. . 0 . .. 0 '0000‘ . 1 . .. .. _ 1. . - . . 0..; . 0 00? f . . 0. _ . . . . .. . 0 ... .0.0 0 . $0.. 011020.05! 04“.”... 0.0.0.0. l.- . . . .. 10. ,. 00.0 0 00.0 . 40.00.07,...0 . . y . . . . .. . . .0 0.0....u0...o-.'.00_0.1.0.’ 0.010000 . I. 0.r.1..~?l¢l./..00.n(/0M00M94$. 0.0 0 ..fl 0 _ O. R... . _ . . 1 . . . . . .. . . 000..-..0rY00vl0- pl. 00 e . .. _ . . 1 . V .- J00... 0:0..0-for ”(is 'H V - .0 . 0.00 a '0‘0'“ 00 3/}. 0 000 . - .. .. . . .0 . 0T. .1 0. I. .00 20-31.. .- . . .00.000 0. t. ’50.. Pffl . 000 1.. 1 0 .0.0000 'DV.0 [0 A. tr 1" , . . . . 0 I.l.r..... :. .... v .00T‘0 ... "\10 0 - 0 . . . . 0.0.00.7-UJI0I'1I. 00 ...N' 0 - 0. 0 v 0 .00.000" '10 .rhlz-V.0. . . . . 100.00.03l004 7-0Ff .7.) . . . _ . ‘00.. ..".r.ocr.\o.'vlo.’0 . 7 . I... .0 .00 v0.00.00)00\0l1.. . ..0 ..140. l.. J? .l0 .l.’0.0r -0. Iflr0tvt00 .. ...-,lslv 0 '0 . 0 0' 0,0000 0.00. ..0 P! 0,}.0082 x . 0 0.0(...0I'0...00l.0l 00"0'1 0.7!: . -. . 0 . . ....’I. 10.0.! - 0 1 0. . - . 0 0.0-0(""‘(‘$0 . 0. . . 0. .huiol!...t-0¢.I . 0.0!! h - 0 00 . _ . \0 10002 . w... ....1r.£-..00o{.0.'.0 H50” .. 0 . A ..0. I . . 0 (I0 0. (0.0400 .10.}. 0.I.T0 J0000.I00 0 .. 0' .0000 V 0 V 0 . 00-: 4 -. V0 00 - 0.0 r .0wn.f0' .fl “main... 1 .0r1100Apln t 0!. . . ... . .0 . . 0.| ..2..o00. .7103 00.0 r0)..~030...r.110400010 l 0. . . 0. .0? '.)\.o}0 r . . ..00 . O V... 30.1 0 30" 0 . . . .0 '0» . . J. . r 00: . a I I 0 a 0 0.00.0.0I0 00 .a . .. .010; 0fr...~ ... 0"- . - K . . . 0..!0..f...0.riuvu000..p.. f. G 3 . 135-? 0 0 .0 0'. .0 I A O .- . . .0 r u a c . I. ... 0 1 . ...00h 1ft),0.' .. 2 cu“. . .. Aff rolf'tlfof .0 ... . '0! . 1 0 ~'.I00n!u...p.-.0O0.0. 0. no .. 0000.. .u. .... . . 0‘.In ... '0' j . . . - T1. . .0u-- .0.. ..hqu... 0 0 I, 00‘ It .000 A. 0. .,0¢0.I e I. . .l00-.00 :0 (I . A 0 .‘I 4.. . .I .100w00 c . {0.0 0'01 0 ’10 . 0 I 0. 0 - . 002.00H 0’. l..0000 . . . 210...- |.-. .0 0 I 04- I ll! .0 0 . .[Va0n-.fl ‘l\.' . . IV0\.0 .1171: . . 0 u.“aro 1’. 0's, ’llvO-IO) . 00 0c.( 0 0. . 0'0. ....0. 0'1. '4‘ 2 00. C 'n .n .4 . ..-.T’ (1 n O . o r . . t.r. . .U}. 0 0 0 . o .- I .'.'2_ . 7 . .0 . . I 0 v n 0. .0 Ir 0 .2! O . 0 0 0 .0 O . . _. 00 .. . J: . I .f 0010 .1020 la ‘ I ’ 0 0 00A I - "“ ..- ... u .r. . o . O . 0 cl ..lO00000. 00V 0 . 00 0000 0 00 I l 0v .00 0 . ~ I 0 I I ' 0. . l 0 O 0 '15: 0 0T - l b v! ‘0 l ‘ .0 0 Ir 0 0 0 0 O 0 l .. l ' I '0. I l 0 'I ODI- l A.4 . . ..I . - ' O .I 0 r I . 0 U ' I 00!! . . . Onoz'L . -..-00.....l0ozfno‘v0lf- .70 00 - . 1.00.0‘00000030000000.’ . . . 0...... 0.0.... 0......0......0000I.0V00:.00000000. . .0.. 0.00.0... . 0.. . . 0 . Q I. 0 . 0 .0 l . iTS‘ii‘zDOQ‘EQ“‘\‘Sé .0 ...- .. 1.. .0 .00“. 0.00 .00.200.913.0000;L900..- r0 . v. 0.. .0 . x . .0 0 I. . .0 0.. .1 . 70 . .0 .0 $0.8...0,\.0\. . .0 0 0|A. 0. 0 _il‘. . . .. 0'. ... ' 0. .000.0..0‘0.0.¢‘ .00(: 0. Ir 1", ... ||||Fu 1 11111111111111 11111111111111 313006310670 LIBRARY I Michigan State University SIMULATED NIGHT AND DAY DRIVING VISUAL ACUITY AS MEASURED BY THE TITMUS VISION TESTER AND THE SNELLEN ACUITY CHART By Jack Vincent Ferrara AN ABSTRACT OF A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1970 G’s- -- L359 75' /-ar7—7/ ABSTRACT SIMULATED NIGHT AND DAY DRIVING VISUAL ACUITY AS MEASURED BY THE TITMUS VISION TESTER AND THE SNELLEN ACUITY CHART BY Jack Vincent Ferrara Night visual acuity represents a special problem to the automobile driver. Not only do lower lighting levels cut down on viewing distances, but lowered contrast makes detail detection all the more difficult. In addition, research in the area has suggested that specific eye disorders of the individual driver may produce even greater reduced acuity. Dis- orders such as night myopia and night hyperopia get progressively worse as lighting levels lower. A night myopic individual will get more and more nearsighted as lighting levels lower. A night hyperopic individual is analogous with respect to farsightedness. Detection of such individ- uals would be of value since many of these individuals' abnormal night acuity losses can be rectified with a lens correction specifically for night driving. Through the use of the Titmus Vision Tester near and far acuity tests at mesopic and photopic illumination, detection of unusual cases of mesopic acuity loss (at simulated night driving illumination levels) was attempted. Scores on the Titmus far acuity test, which simulated a 20 foot testing distance, were compared with scores of the Snellen eye chart at an actual 20 foot distance. Performance on a variable low con- trast test as opposed to the higher contrast acuity tests was also under- taken to look for possible relationship to night myopia. “ 7" ‘7—1-—Ii—._ 1)...- 1‘ ul‘ .u “—T. :“TAA‘y—Tv .- ‘.-.—?'T T "—7 37. -— 1‘";- a u A; no. A... vauLilifi . .- _ .‘Pg 3" _' SIJZJ'; k :53! 1‘30 31 fl .7} 1.. 1. . 'l' , Jack Vincent Ferrara The experimental group was given the Titmus near acuity test, the Titmus far acuity test, the Snellen far acuity test, and the Titmus low contrast test, in mesopic, and then again in photopic, lighting. Thus each subject set his own norms in the photopic condition. A con- trol group was given the same tests in mesopic light both times. The purpose of the control group was to detect improvements in performance due to either improved dark adaptation or increasing familiarization with the tests. Subjects were drawn from the Michigan State University (MSU) student population and ranged in age from 18 to 25. The results showed the Titmus near and far acuity tests were useful for detecting unusual cases of mesopic acuity loss when poor performance on one test was considered in relation to the subjects' performance on the others. Three cases of mesopic acuity loss great enough to cause potential concern for night driving were discovered, but further testing on more sophisticated apparatus is needed to classify the loss as night myopia, hyperopia, etc. The fact that the Titmus did detect such losses indicated its usefulness towards this aim. The Titmus far acuity scores correlated well (.70) with the Snellen scores, but there were significant differences between the means. It was concluded that inaccurate Titmus simulation of visual conditions for 20 feet and differences in test targets accounted for the differ- ence. The experimental sample performed well on the low contrast test in both mesopic and photopic conditions with no potential problems for the night driving situation occurring. The control group showed relatively high reliability (.55-.86) between mesopic treatments of the same tests indicating no significant Jack Vincent Ferrara learning or increase of adaptation. Only the Snellen test showed a small, but significant difference between the means in this group. Approved: T. W. Forbes, Chairman John H. Wakeley S. Howard Bartley Dated: October 21, 1970 SIMULATED NIGHT AND DAY DRIVING VISUAL ACUITY AS MEASURED BY THE TITMUS VISION TESTER AND THE SNELLEN ACUITY CHART By Jack Vincent Ferrara A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1970 -l DEDICATION AND ACKNOWLEDGEMENTS During the 26 months necessary to develop this thesis from a germ of an idea to a completed volume, nine individuals especially aided in its production in their own way. This thesis is respectfully dedi- cated to them: Dr. 8. Howard Bartley James V. Ferrara F. Joan Ferrara Dr. Theodore W. Forbes Ann M. McElhinny Debbie A. Schultz M. Peter Scontrino Dr. John H. Wakeley Carolyn S. Wares Special thanks are offered to the Michigan State University High- way Traffic Safety Center and the Michigan State University Department of Psychology for the use of their facilities. ii TABLE OF CONTENTS LIST OF TABLES.O.......OOOOOOOOOOOOOOOOO......OOOOOOOOOOOO...0...... LIST OF FIGURES.......OOOOOOOOOOOO......OOOOOOOOOOOOOOO... INTRODUCTIONAND LITERATURE REVIEWOOOOOO......OOOOOOOOO.0.0.0...O... THE PROBLEM AND EXPERIMENTAL DESIGN....................... The Problem.................................... ..... The Experimental Design............................. METHOD................................................... Test Room and Equipment............................. Subjects............................................ Procedure........................................... Method of Analysis.................................. RESULTS.................................................. Experimental Group.................................. Control GrOUp....................................... DISCUSSION......................... ..... ................. SUWARYAND CONCLUSIONS.........COOCOOOOOOO0.00.... ..... O BIBLIOGRAPHYOOOOOOOOOOOOO......00.0.0000...OOOOOOOOOOOOOOO iii iv 10 IO 11 13 13 16 16 19 21 21 30 37 43 45 LIST OF TABLES Table l. Brightness and Contrast Levels of Targets................. 17 Table 2. Age and sex Distributions...oooooooooooooooooooooooooooooo 18 Table 3. A Comparison of Titmus and Snellen Means (Experimental 33 Group).........OOOOOOOOOOOOOOOO......OOOOOOOOOO.0.0.0.0... Table 4. Intercorrelations: Titmus, Snellen, and Derived Scores... 34 Table 5. A Comparison of Titmus and Snellen Means (Control Group).. 35 Table 6. Extreme Cases of Titmus Near and Far Acuity............... 38 iv Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 10. 11. 12. LIST OF FIGURES The Relation Between Visual Acuity and Illumination.... The Testing ROWOOOOOOOOOOOOOOOOOO......OOOOOOOOOOOOOOO Titmus Mesopic Near ACUitYOOOOOOOOOOOOOOO0.0..000...... Titmus Photopic Near Acuity.......... Titmus Mesopic Far AcuttYOOCOOOOOOOOOOO0.00.00.00.00... Titmus Photopic Far ACUitYooooooooooooooooooooooooooooo Snellen NeSOPiC ACU1ty0000000.0.0.0.0000...0.0.0.000... smellen Photopic ACUitYOOOOOOOO......OOOOOOOOOOOCOOOOOO Mesopic Low Contrast Vision............................ Photopic Low Contrast Vision.............. Titmus Far Decrement.....OOOOO............OOOOOOOOOOOCO Titmus Near Decrement.. 3 14 24 25 26 27 28 31 32 INTRODUCTION AND LITERATURE REVIEW Good visual acuity is obviously an essential component of safe driving. This thesis seeks to investigate acuity with special emphasis on the night driving situation, through the use of standard visual tests (Titmus and Snellen acuity tests). Near acuity, far acuity, and low contrast vision will be measured among the population sampled. By com- paring acuity in daytime luminance against that in the nighttime condi- tion, one can gauge the loss of acuity, both near and far, which the night driver faces. It is hoped that this investigation will help lead to an increased understanding of the visual handicaps the night driver encounters, and ultimately to the removal of these handicaps. The concept of acuity is central to the subject of this paper, and there are several definitions of it. Lorrin Riggs defines visual acuity as the capacity to discriminate fine details of an object in the field of view (Graham, 1965). Fine, in the previous definition, is a relative adjective, but detail can be defined more concretely. Riggs specifies detail in three ways: 1. width of an object 2. angle subtended by the object at the eye 3. computed width of the retinal image (Graham, 1965, 321). Angle subtended is the convention of most vision researchers, and it is expressed in units of minutes and seconds of arc. Riggs additionally outlines four basic types of tasks to demonstrate acuity: 1. detection 2. recognition 3. resolution 4. localization (Graham, 1965, 322-326). Detection simply involves a response from the subject as to whether or not the test object is present. In the recognition task the subject must name a critical aspect of the stimulus (e.g., the direction an arrow in pointing). In a resolution task the subject must respond to elements in a pattern. Localization involves displacements with respect to two test objects (Graham, 1965, 323-326). Acuity must always be considered at a given level of illumination, since a person's acuity will change as illumination levels rise or fall. Basically there are three levels of illumination: scotopic, mesopic, and photOpic. ScotOpic occurs at just near the threshold illumination, and at this level only the rods are functioning -- thus, no color per- ception is possible. The photOpic level is representative of daytime illumination, and at this level the cones are fully functioning so that color vision is possible. Intermediate to these two levels is the nighttime level, or meSOpic condition. At this level the eye is dark adapting, but the cones are also functioning to some extent. The basic relationship between acuity and illumination is as illustrated in Figure l. Acuity increases markedly as a function of illumination. Low values of acuity under dim illumination are due to the fact that only rods func- tion at this level. As light levels increase and more cones come into play acuity increases. Uniformity of illumination is also important to acuity. During a 1937 experiment, S. Hecht discovered that unless the background of his testing device was of uniform brightness, acuity scores would be sig- nificantly lowered (Hecht, 1937, 593). 1.6 Cone function 1.4 1.2 1.0 0.8 Visual Acuity 0.6 0.4 0.2 Rod function -5 -4 -3 -2 -1 0 1 2 3 Log I - millilamberts Figure 1. The Relation Between Visual Acuity and Illumination Hecht replotted original data from Konig in millilamberts. (The two curves represent probability ogives fitted to rod and cone portions of the data.) (Woodworth, 1965, 385) z, Luminance is generally described in footlamberts (ft1.). One foot- lambert is equal to 1.076 millilamberts, also a widely used standard. "A perfectly reflecting, perfectly diffusing surface, illuminated with a luminance of one footcandle has a luminance of one footlambert, re- gardless of the direction of View" (Judd, in Stevens, 1951, 957). A footcandle is the "illumination produced by a uniform point source of one candle on a surface every point of which is one foot away from the source" (Judd, 22.'gig.). In addition to the illumination of the viewed object, the illumi- nation level the S had adapted to immediately prior to his viewing the object is important. It generally takes 20 minutes for relatively com- plete dark adaptation to take place, although the majority of the adap- tation will take place in approximately five minutes (about 90 percent). Adaptation is much quicker from meSOpic to photOpic than vice versa, being virtually complete in little more than a minute. Changes in the eye during this period include chemical changes in the retina as well as a change in pupil diameter. Generally as one ages, the speed and extent of these changes diminish. Angle subtended (or the size of the "fine" detail) and illumination level (both of the viewed object and adaptation level) have been presented as important factors determining visual acuity. Contrast is another im- portant factor in the determination of acuity, and is subsequently dis- cussed in relation to night driving. A study by Forbes, Vanosdall, Pain, and Bloomquist investigated the question of whether high contrast vision and low contrast vision (at simulated night driving levels) were part of the same function, or whether separate skills were involved in two types of vision. They 5 tested 396 drivers on the Night Vision Performance Test (designed by Dr. Merrill Allen). This device consists of a low contrast target in a dark housing. The test was given in mesopic (.2 ftl.) and photOpic (10 ftl.) lighting. The Titmus vision tester was similarly used at these levels of illumination for far acuity and low contrast tests. The re- sults showed that low illumination scores correlated .50 to .65 with normal (photOpic) acuity. Older age groups had peOple who did poorly on both tests. Results did not indicate that there were any age groups specifically handicapped with respect to low contrast vision. Work is continuing in this area with even lower contrast targets (Forbes, 35 31., 92. g_i_t_.). O.W. Richards makes the point that a common brick subtends 12 mi- nutes at 200 feet, with a contrast of 25 to 50 percent on a moderately clean asphalt road at night. Richards thus concludes that five to 10 mi- nutes of are at this contrast level are representative of potential caus- es of concern to the night driver. Better vision may be required to read signs or see objects on poorly lighted streets. Those on modern high- ways are designed to be readable at legal speeds with 20/40 vision (Richards, 1966). It is not a surprising fact that objects of high contrast may be perceived where a similar sized object of low contrast may not be seen under the same conditions. Richards concluded that de- creasing contrast did not affect the acuity function to any great ex- tent above 30 to 40 percent contrast. With less contrast, vision decreases rapidly. Sixty year olds required 25 percent or greater contrast to see a 20/200 letter at 0.01 ftl. at the standard 20 foot testing distance. At 10 percent contrast objects smaller than three and one half minutes are not likely to be seen. Richards adds that if a 100 percent safety 6 factor is allowed for field conditions, a half brick might not be seen at 100 feet during average road illumination, while a full brick most probably would be seen by these 60 year olds. By the end of the seventh decade, then, some individuals may not see well enough in low contrast conditions to safely drive at night. Besides external factors of target size, illumination, and contrast level, individuals may have "internal" visual problems that affect acui- ty. With reference to the night driving situation night myopia is of particular importance. Night myOpia is a phenomenon in which an indi- vidual tends to get nearsighted as illumination is lowered. Night hyper- 0pia involves the individual getting farsighted as illumination is low- ered. These two trends may cause a change in an individual's acuity under mesopic illumination, and thus a lens prescription based on pho- topic acuity tests may be inadequate for use in the mesopic condition. These phenomena are especially dangerous since individuals may attribute myopia or hyperOpia to the normal acuity decrement everyone suffers at night. H.A. Knoll has attributed a reduction in the amplitude of accommo- dation as one of the causes of night myopia-hyperOpia. Most researchers (according to Knoll) agree on three factors as the main cause of night myopia-hyperopia: 1. spherical aberration 2. chrOmatic aberration 3. accommodation (Knoll, 1952). In 1945 Cabello investigated the rapidity of the occurrence of night myopia. It was found that the first five minutes of dark adapta- tion produced relatively rapid night myopia. The next five minutes 7 showed the myopia at a relatively constant level, requiring a minus 1.5 diOpter correction in order to obtain normal vision. (A diopter (d) is a unit of measurement of the refractive power of a lens, or the de- gree to which it bends light; a minus correction indicates myOpia, while a plus correction indicates hyperopia.) Cabello found that during the last five minutes of a 20 minute adaptation period myOpia approached a limit of minus 2.00 d correction (Knoll, 1952). It is interesting to note that pupil dilation takes little over three minutes, on the average, and it is within this period that Cabello found most of the myOpia de- velOping. Older SS took longer periods of time to deve10p the myopia, and likewise their pupils did not respond to the lowered intensity as quickly as younger SS. This is evidence for pupillary aberration as a probable cause of night myopia (Knoll, 1952). O.W. Richards has studied night myOpia with respect to age, parti- cularly in relation to highway driving. He mentions most night driving is done in an average luminance range of from 0.1 to 0.3 ftl. He tested 65 individuals in luminances of 10 to 0.01 ftl. Testing was done with free binocular vision. Richards' results were consistent with the usual acuity-illumination relationship, except at 95 percent contrast acuity was inferior to acuity at 85 to 91 percent contrast. Due to Richards' small subject sample size at this age, however, this may not be a sig- nificant difference (Richards, 1966). Richards performed another experiment by changing lens corrections on a given individual, and noting the effect on acuity. By checking the lens that would produce best acuity in the lower illumination against the lens used in the daytime one could gauge the presence of night myOpia- hyperopia. He found some individuals would be better off with a different 8 lens prescription for meSOpic and photopic driving illumination levels (Richards, 1966). In a later article (1967) Richards mentions that night myOpia may be inhibited through the use of atropine and homatropine. Some eyes de- velop presbyopia under the influence of drugs. Richards suggests several more reasons for myOpia developing in certain individuals (compare to Knoll's list, above): 1. change in chromatic aberration with the Perkinje shift 2. increased spherical aberration from larger pupils 3. increasing convergence as lights dim 4. retinal curvature 5. forward movement of the lens due to the choroid vessels becom- ing partially empty in dim light 6. greater curvature of the eye from increased tonus in the ciliary process during dark adaptation (Richards, 1967). Spectacle corrections for night myopia range from minus 1.50 d to plus 1.50 d for the SS run in Richards' experiment (1967) in meSOpic il- lumination. Richards' corrections were made for specific luminance lev- els. Richards found that a minus 0.50 d correction would provide the maximum benefit for the most people. Army tests have shown that by im- posing a minus 1.00 d correction, about one third of its personnel would be handicapped with poorer acuity (Richards, 1967). The fullest myopia will develOp in scotOpic levels of light, a condition that would be encountered by the average motorist during head- light failure, but myopia can and will develop at mesopic levels of light. Richards found that a few people improved with a plus 0.50 to 0.75 d correction. While approximately five percent of the total gained in 9 acuity with the plus correction, 20 percent of the sample gained with minus d corrections. This finding implies that night myopia is many times more prevalent than night hyperOpia. Again, Richards found that minus 0.50 to 0.75 d corrections gave the most improvement to the most peOple. He did note that no one with 20/10 or better vision found any improvement with lenses at mesopic levels, and approximately 80 percent of the p0pulation did best with no corrections at all (Richards, 1967). Richards also found that night myopia was rare in the very young and very old. The greatest concentration of this phenomenon was in the 203 and 405. He added that considering the small numbers in his sample, there was no marked indication of any age group significantly demonstra- tive of night myopia (Richards, 1967). Richards' final conclusions were that 10 to 20 percent of the p0pulation would see better at night with a minus 0.50 to 0.75 d correc- tion. More than 20 percent would see less well, though, eliminating the possibility of adding a minus correction to everyone in the pOpulation (Richards, 1967). Thus it is possible that an individual may see well during the day, but not see well enough to drive safely at night. This points to a need to measure night driving visual acuity. Attempts to measure meSOpic driv- ing vision should focus on near and far acuity at meSOpic lighting levels and a representative contrast level. By controlling illumination and contrast, the third determinant of acuity, target size, can be validly measured, and recorded in minutes of arc subtended. This, in turn, will tell just how "fine" a detail the night driver can accurately perceive. THE PROBLEM AND EXPERIMENTAL DESIGN A. The Problem From the previous discussion of the literature it can be seen that the phenomenon of meSOpic acuity loss may be a significant factor in nighttime driving safety. The first aim of the thesis was to attempt to detect persons who encounter a typical meSOpic acuity loss, such as night myopes or hyperOpes, through the use of such readily available acuity tests as the Titmus acuity tests (near and far acuity). The de- finition of nearsightedness (myopia) indicates that the nearsighted in- dividual would see near objects relatively well and far objects relativ- ely poorly. Thus a person deve10ping night myopia would have a tendency to see near objects well as compared to far objects at low levels of illumination. It follows that the night myopic individual would show a greater drop in his far acuity score than his near acuity score. These individuals could not be predicted from correlation coefficients since they deviate from the group trend of a relatively even drop for both near and far acuity. An analogous argument was made for night hyper- Opes. These individuals would show a greater loss in near vision than for far vision as illumination levels drop. As a second aim of the experiment, the Titmus far acuity test was correlated with the Snellen far acuity test in both meSOpic and photopic conditions. By correlating the Titmus far acuity test to a widely used standard (the Snellen) an estimate of its validity was made. Finally the low contrast Titmus test was included to check perfor- mance of experimental group extreme near and far acuity cases with the experimental sample as a whole on the low contrast function. 10 11 Spgcific Objectiveg 1. To determine the usefulness of the Titmus acuity tests in de- tecting unusual cases of mesopic acuity loss which might be detrimental to night driving. 2. To validate the Titmus far acuity test in meSOpic and photopic conditions against the Snellen far acuity test in the same conditions. 3. To investigate the performance of the experimental subject group on the Titmus low contrast test. B. The Experimental Desigg To meet the experiment's three objectives four scores were obtained from each S: l. the Titmus far acuity test 2. the Titmus near acuity test 3. the Snellen (far) acuity test 4. the Titmus low contrast test. The experimental group was given the above four tests in mesopic light, after an adequate dark adaptation period, and then the same four tests in photOpic light, after an adequate light adaptation period. MeSOpic and photopic light levels were chosen to match realistic driving illumination levels previous investigators used. A control group received the same treatment as the experimental group, except that the tests were given in mesopic light on both occa- sions. Use of the control group permitted investigation of two pos- sible problems: 12 1 1. that the dark adaptation period was inadequate, and 2. that there was a learning effect from the first giving of the test to the second. The acuity test scores were in minutes of visual angle subtended by target detail discriminated. The Titmus low contrast test was scored in percent contrast of the target against the field. The lower the visual angle, the finer detail discriminations the individual can make; and similarly, the lower the percent contrast score, the finer contrast discriminations the S can make. METHOD A. Test Room and Equipment The testing room was a narrow rectangle with matte black walls. Seven (40 watt) cool white fluorescent tubes were required to bring the room to photOpic illumination, along with two 60 watt goosenecks, as shown in Figure 2. Mesopic lighting was achieved through the use of the goosenecks only. A drape was hung from the first set of fluorescent tubes in order to eliminate direct glare from the lighting, without blocking the Ss' view of the Snellen chart. SS sat 20 feet from the chart in the room, with the chart hung at eye level (sitting). The binocular Titmus tester was placed on a table and S was required to look directly into it in order to take the three tests in it. Each S took a near acuity test, a far acuity test, and the variable contrast test in the Titmus tester, in addition to the Snellen far acui- ty chart. The means and standard deviations of the luminance levels of these tests are presented in Tables 2 and 3. A lighting level of .2 ftl. was used for the mesopic condition, and 10 ftl. for the photOpic condition. These are levels that the driver would encounter on the highway during the nighttime and daytime, respec- tively (Richards, 1966, 1967). This is also the level that Forbes used t al., 22. cit.). in his study of low contrast vision (Forbes, The Titmus presents two slides (one per ocular) to the S. The slides are mounted on a drum so that different slides depicting different vision tests may easily be switched. The Titmus test is internally illuminated and mesopic and photopic illumination were produced by a rheostat attached to a light source, this being four GE "night light" 13 14 .fiflmfi 0H03 uflOfld. one? on =o>oo 0:» mafia exam: nouoow osu coauao naoo camouosm cu moon: onus mxooa noaoow 03» ecu mane :oauwocoo canouoa.au “ouoz Boom wawumoa 058 .N ouswam 30H> ouosuo>o Anwvwm Haw no magma Jonas ouuwav 1 uoom on ma ooaaumwo coaaoam cu m. _ a a A now a .I 33. 88 53 on n. :6 a E on» scum macaw mamas me ouoauaaao e um omaa rd 3 ocean ...uom magnum l- (J - \ axuoaomoow I I All; domwmnm sums co ”\D I , IA? . Emcee: uuaz as N— u —mnooa nun? ooiwa son> moan 2% T .mumo coauuooH .o’oa/aa—-s\ /$’..—.~.\ mace: \“ ouoaw unmeaEaHo uuuno mosuo>o on cacao saws :oHHoam once: vmosuo>o 15 bulbs of seven watts each. Each side of the drum is lit with a pair of the white coated bulbs. The near and far acuity Titmus tests used the same type of test figures. A cluster of three Landolt rings and one complete ring were provided for each discrimination level. The S was required to name the location of the complete ring from the four alternatives. The variable contrast test required the S to give the location of a break in a single Landolt ring. The far Titmus tests simulated a 20 foot distance from the S through the use of a mirror and a reduced image slide. A 14 inch test distance for the near test was simulated by reducing the size of the slide image to half that necessary from an actual 14 inches with placement seven inches from the S's eyes. The far and near acuity tests measured acuity from 10 to seven tenths minutes of arc. The Snellen ranged from 10 to five tenths. The contrast on these three tests all were approximately 90 percent; the variable contrast test ranged from 69 percent contrast to one percent contrast. The variable contrast test figures all required discrimina- tion of a two minute visual angle target. Illumination measurements for mesopic and photopic conditions were taken from the background of the Snellen chart with a Pritchard photo- meter at a 20 foot distance from the chart. Internal Titmus lighting was measured off the background of the far acuity test, again with the Pritchard photometer. This instrument was also used to check contrast on the variable contrast test. The Snellen proved remarkably easy to "tune" to meSOpic and photo- pic lighting levels, mainly due to the fact that its lighting was 16 obtained through the use of reflected light, with no internal illumina- tion as in the Titmus. As can be seen in Table 1, actual mesopic and photopic illumination were almost identical for the Snellen with the desired .2 and 10 ftl. reflectance, respectively, from the background of the test. The Titmus far acuity test also displayed similar readings, but its greater standard deviations than the Snellen for the same series of readings indicated the lighting was not as uniform. The contrast levels of both tests at both mesopic and photopic illumination were close to 90 percent. B. Subjects The Ss were exclusively college students who were taking the MSU introductory psychology course. Ss were obtained through the use of a sign-up sheet, and the students volunteered for the experiment in order to partially fulfill a requirement for experimental credits. A total of 96 Se were used: 25 in the control group, and 71 in the experimental group. Table 2 shows the age and sex distributions of both groups. C. Procedure Each.S was run individually and the procedure took from 15 to 40 minutes per‘S. ‘Ss were ushered into the experimental room and seated while experimentor read a brief description of the aim of the experiment. Ss were told that the aim of the experiment was to gauge loss of visual acuity at low levels of illumination, with emphasis on potential appli- cations towards highway driving. Then the Ss were asked four questions relating to night driving. After the introductory remarks and the four questions, experimentor would wait until five minutes had passed from the Ss' first having en- tered the experimental room. Table 1. Snellen Mesopic Snellen Photopic Titmus Far Acuity Mesopic Titmus Far Acuity Photopic 17 Brightness and Contrast Levels of Targets Measurements Taken with Pritchard Photometer Figure (N=4) Background (N=7) Figure (N=4) Background (N=7) Figure (N=4) Background (N=7) Figure (N=4) Background (N=7) Percent Contrast .021 ftl. .003 ftl. ‘——T‘“ 90 1.250 0.250 89 10.550 0.410 .003 0.020 :1 96 .472 0.520 12.710 9.820 T‘— NOTE: Circled background readings were used to determine .2 ftl. mesopic condition and 10 ftl. photopic condition. Percent Contrast = BBkgr.“'BFig. BBkgr. 18 Table 2. Age and Sex Distributions Experimental Group Age Male Female 18 10 l4 19 16 8 20 14 4 21 1 1 24 0 1 25 l l N = 42 N = 29 Modal Age = 19 Modal Age = 18 Control Group Age Male Female 18 2 S 19 9 5 20 2 21 2 N = 15 N = 10 Modal Age = 19 Modal Ages = 18, 19 19 During this entire five minute period the room was at mesopic il- lumination, with the Snellen chart being covered with a white paper, and the Titmus being internally illuminated to the mesopic illumina- tion. The order of the four tests was balanced so that each of the 24 different combinations of the four tests was given an equal number of times (three times each in the experimental group, and one each in the control group). Upon completion of the set of four tests, the room was lighted to photOpic illumination, and S was given 30 seconds to adapt to the new lighting level. The S was then given the same four tests as in the mesopic condition, and in the same order as in the mesopic condition. The control group took the tests in meSOpic lighting both times. They were given the two adaptation periods, but the lights remained at mesopic intensity throughout these periods. D. Method of Analysis Results of the three acuity tests given (Snellen, Titmus far, Tit- mus near) were measures of visual angle subtended from a low of .50 to over 3.00 minutes of arc. The Titmus low contrast scores ranged from one to 25 percent contrast. For convenience in graphing the acuity scores were lumped in step intervals of .25. Statistical analysis was performed on original data. In addition to recording the original visual angle data in minutes of arc, derivative scores were calculated in accordance with the attempt to detect night myOpes-hyperOpes. Decrements were calculated for Titmus near and far acuity by subtracting for each S his photOpic acuity score from his meSOpic acuity score. 20 For all scores (including derived scores) means, standard devia- tions, and Pearson correlation coefficients were calculated. In addi- tion, for certain pairs of scores two tailed Ts for correlated means were calculated. RESULTS Note that acuity results are presented in minutes of arc, but the Titmus low contrast test is presented in percent contrast of target against background. For reference, an acuity score of 2.00 minutes of arc is equivalent to the more well—known Snellen score of 20/40. A score of 20/20 is equal to 1.00 minute of arc. A. Experimental Group The Titmus near acuity test in the mesopic treatment resulted in a mean of 1.02 and a standard deviation of .26. The photopic treatment mean was .84, and the standard deviation .17. The normal acuity loss expected under lower illumination accounts for the difference here. The Titmus mesopic far acuity mean was 1.32, with a standard deviation of .61. The Titmus photopic far acuity mean was .94, and the standard de- viation .29. Again the expected acuity loss occurred. The standard deviations rose as the means did, indicating greater spread of the scores around the mean. (See Figures 3-6) The Snellen mesopic acuity mean was 1.52 and the standard deviation .47. The Snellen photopic acuity mean was .86, and the standard devia- .tion .25. Again, the normal loss of acuity expected in mesopic condi- tions occurred along with a greater spread of scores in the meSOpic con- dition. (See Figures 7 and 8) The Titmus low contrast test mean was .074 in the meSOpic condition and .028 in the photOpic condition. The mean is again lower in the pho- t0pic condition, and the standard deviation again bears the same rela- tionship. (See Figures 9 and 10) The "derived" score used in this study was a decrement score, re- presenting photOpic acuity subtracted from meSOpic acuity, for each 21 %u«=o< uwoz oedemoz maauae .m ouswfim ou< Hmsmfl> «0 mouse“: :q suu=o< 22 .dN.Nnoo.NLrom.HnmNHMLm«N.H|om.a m¢.HImN.H «N.Hloo.a mm.lmn. «n.1om. o¢.|nm. .\ o~.no a _ N a 111 N m OH ma ON on an ac on an I z I cm. I m mm m0 No.H I m. honaaz 23 muaso< awoz caucuonm usauwH ou< Hosmw> mo nouns“: ca huuso< .¢ unawam «N.Huom.fi m¢.~1mN.H #N.~uoo.a mm.umn. «n.10m. aq.nmN. Lu 6N.uo w _ ~ . H NH NH no as n 2 NH. I m em. a x OH ON Om oe om mm.wo Honasz moaned Ham aquamoz assuae .m shaman ou< Nanmw> no mouscwz cu howso4 «N.m-om.m m: m¢.NumN.N eN.Nnoo.N ma.aunn.~ ¢N.Huom.H m¢.HunN.H «N.Huoo.H mm.-mn. «n.10m. :qu.ao ml! . VW“ o n N m «”1 0H OH NH «a ma 0N 24 on ow on AN Ho. Nm.H mw.mo wooasz a I n Ix m:z muwao< new oNQOuonm msEuHH .o ouswwm ou< Hmama> mo mouscwz aw muwno< -eN.Nnoo.N mm.HumN.H «N.Hnom.fi m¢.HnmN.H «N.H|oo.N o¢.|mm. «5.10m. m¢.nmN. \ ¢N.Io T L c o N OH NH «H ON 5 2 on om oq on an n z mN. I e am No em. I x wonanz 26 N.Nnom.N mq.NumN.N qN.Nuoo.N mm.HumN.H «N.Hnon.H a¢.HumN.H «N.Huoo.H huH=o< UHQOmoz coHHmcm .N ouame ou< Hman> mo mouscHz :H huHao< m :nz 3. as $4 um NH «H Nm «wk «NFC oH oN on ow on mm mo nonaaz zuHso< OHQOuonm :oHHocm .w ouamHm ou< Hast> mo mouacHz :H muH=o< «N.Nnoo.N mm.HumN.Hih¢N.Hnom.H m¢.HumN.H ¢N.Huoo.H ma.umn. «n.10m. m¢.umN. H «N.uo 0 Hi L 4. L H m OH «H ON 7 2 3. mN on OS on HN n z I mN. I m mm mo em. I M monasz 28 aonH> ammuucoo 30H UHQOmoz .m ouszm umwuucoo unmouom nu. ma. . no. NH. so. so. . . , so 8 N m s N N . 3 Z R S u z 30. n 6 SE. n m CH CN Om on on mm_mo Monasz conH> ammuucoo 304 UHQOuonm .oH oustm u meUCOU ufiwohwm fl. 2. a oo. oo. oo. 8.. \ FIIAIIL oH oN 9 2 Z on oq on HN MNo. mNo. mm «0 nonssz I l>< m:z 30 individual. These decrement scores were calculated for both near and far Titmus acuity. The larger the decrement, the greater the Ss' loss of acuity from the photOpic to the meSOpic condition. The Titmus far acuity decrement was .38 with a standard deviation of .46. The Titmus near acu- ity decrement was .17, with a standard deviation of .16. Far acuity suf- fered the greater drop, and in addition showed a more widespread distri- bution of these decrements, although both scores have high standard devia- tions. A few SS had negative decrements. Strictly interpreted, these individuals actually had slightly better scores in the mesopic condition than in the photopic. The difference was so slight, however, that it was well within test-retest reliability error. (See Figures 11 and 12) Table 3 shows the results of T tests and correlations between cer- tain pairs of the tests. When the meSOpic treatments were compared with the photopic treatments, across all tests, the Ts between all the means were significant at the .01 level. The correlations between meSOpic and photOpic scores of the same test ranged from .71 to .79. This same table shows that the Snellen phot0pic and Titmus far acu- ity photOpic tests correlated .70, but the difference between the means was significant. This same relationship is true for the mesopic treat- ment of the two tests, but in the meSOpic case the Snellen test had the larger mean, while in the photOpic case the Titmus had the larger mean. Table 4 shows intercorrelations of all the tests given, and the derivative scores. B. Control Group Table 5 reflects means, T tests, and correlations between control group data. (In this group conditions were at meSOpic light both times the test was given.) The Titmus near meSOpic correlated with itself 31 Am 50mm you zuHso< oHoOuosm macHz hquo< OHQOmozv uaoaouoon new maauHH .HH ouanm ou< we mouacHz :H onc< HmzmH> mo omuouocH oo.~-ma.~ oa.o-om.fl oo.o-n~.a o~.H-oo.H oo.-ma. o~.-on. oo.-mm. o~.-o Ho.- on m~.- H q _ IEJ H H m m N CH ON om Hm I z 0%. I m om. u m oH ON ON on on mm wo nonasz 32 Am comm now huH=o< UHQOuocm macHz huHso< oHoomozv ucosouuoo umoz mneuHH .NH ousmHm ou< mo mouacHz aH onn< HmsuH> mo omwouocH «n.10m. m¢.nmN. «N.uo Ho.u cu mN. «H Hm I 2 Nn OH. I m S. «m oH ON om as on am we monasz wwwwwwww‘w 33 Table 3. A Comparison of Titmus and Snellen Means (Experimental Group) ‘R T'(Two Tailed) s r Titmus Photopic Near Acuity 0.84 9.28++ .17 1L79 Titmus Mesopic Near Acuity 1.02 i .26 fl Titmus Photopic Far Acuity .94 i 7.08++ .29 .71 Titmus Mesopic Far Acuity 1.32 1‘ .61 “ Snellen Photopic Acuity 0.86 i 17.63++ .25 .77 Snellen Mesopic Acuity 1.52 i} .47 “ Photopic Low Contrast Vision 0.03 ‘i 8.78it .02 ".72 Mesopic Low Contrast Vision 0.07 i .05 i Snellen Photopic Acuity 0.85 1' 3.57++ .25 .70 Titmus Photopic Far Acuity 0.94 h .29 1 Snellen Mesopic Acuity 1.52 {' 3.84++ .4? I .70 Titmus Mesopic Far Acuity 1.32_ ‘1 .61 “ Titmus Mesopic Near Acuity 1.02 1’ 4.64+t .26 1i.45 Titmus Mesopic Far Acuity 1.32 f} .61 “ Titmus Photopic Near Acuity 0.84 i 4.27++ .17 i.75 Titmus Photopic Far Acuity 0.94 ii .29 ” Titmus Far Decrement 0.38 1’ 3.82++ .46 1i.15 Titmus Near Decrement 0.17 Ti .16 “ .LEEEEQ‘ 4+ Significant at .01 __ Experimental N = 71 X - Mean t - T Test 3 - Standard Deviation r - Correlation Coeficient I... 3 as u z Houaoaouooxm Hosanna 639v Ho. no uaououoawom 1+ no. no uaouamoomom + oHnouosm I UHQOooz I unoaouoon I uoouuaou 30H naauHH I new I unoz I huHsom aoHHoam I zuwao< ooauHH I "odouoq 429-: O H E-IVJZIII Una UH. :— mm 2m Gnu. mama «HEB 42H. 29H. 2 Na. i+ao. l+oo. l+mm. i+oo.‘i+om. oo. +AN. am. #2. #Z. :9. in. too. 8. #2. +8.. #2. +7.3. tom. #3. No... #3. +3. too. #3. ooh. Ho. +5. 2. in. too. a. 2. 2. ask. oH. sea. Ego. om. too. +13. HN. .xoa. .xoa. "oaouunHouuooh 35 ..V oHoeH mouoom vo>Huoo was .aoHHoam .osauHH ...“ mm 2m awe hmH mZB so“ zmH 35 Table 5. A Comparison of Snellen and Titmus Means (Control Group) ,_ t (Two X Tailed) s r Titmus Mesopic1 Near Acuity .89 2.04 .17 .74 Titmus Mesopic2 Near Acuity .94 .17 Titmus Mesopic1 Far Acuity 1.13 1.15 .36 .55 Titmus Mesopicz Far Acuity 1.22 .44 Snellen Mesopic1 Acuity 1.42 2.304F .43 .71 Snellen Mesopic2 Acuity 1.28 .30 Mesopic1 Low Contrast Vision .06 0.00 .04 .86 Mesopic2 Low Contrast Vision .06 .03 Legend: ++ Significant at .01 Control N - 25 - Mean - T Test Standard Deviation Correlation Coefficient H m n:fi| 36 .74, and there was no significant difference between the means; the Tit- mus far meSOpic correlated .55 with itself, again with no significant difference between the means. The Titmus low contrast test-retest cor- relation was .86, again with no significant difference between the means. The Snellen test showed a high correlation .71, but there was a significant difference between the means at the .01 level. DISCUSSION The primary aim of the experiment was to compare acuity scores ob- tained from the Titmus near and far acuity tests and determine if un- usual cases of acuity loss, such as night myopes and hyperOpes, could be detected from the results. For this analysis experimentor derived Titmus far and Titmus near decrement scores as a potential short cut to such detection (Figures 11 and 12). These scores gave the loss in acuity from photopic to meSOpic conditions for near and far vision, and were calculated by simple subtraction of the near (or far) photOpic score from the near (or far) mesopic score for each S. A correlation of .90 was found between the Titmus far decrement score and the Titmus far acuity meSOpic scores (Figure 5). The photOpic correlation was only .32. Thus a large prOportion of the variance of the derivative score could be traced back to one of the two "parent" scores (the meSOpic score). In addition a subtractive disadvantage could occur with the decrement scores, i.e., a poor visual acuity score being masked by its subtractive derivation. Therefore the original acu- ity scores were the best data base to form realistic conclusions about unusual acuity losses. Similarly the Titmus near decrement score correlated .76 with the Titmus near mesopic scores and .21 with the Titmus near photopic scores. Again there was little point in dealing with the derivative score since the majority of its variance came from a single score. Since unusual mesopic acuity loss was of interest, experimentor examined the extreme cases in Titmus near and far mesopic acuity (Table 6). The three extreme cases in Titmus far mesopic distribution stood out quite apart from the rest of the distribution (Figure 5). A look at 37 Hoooo>aoaH «ago a Hou< mo mounaHz aH noncomv so. so. om.o ma.o oN.o oo.~ mooo6< oases»: umoz maauHH oo. mo. oo.~ om.m so.o soo.~ "ouooo «souuxm w oo. 2. oo.~ oem so; ..oo.~ huH=o< 0Haomoz so. mH. mN.o om.m Hn.o SH.H nah oDEuHH ”oomoo oaouuxm mH. mN. oo.N om.m om.o ¢H.H huH=o< you muH=o< Mom 0Haouonm uuoz canoooz umoz UHQOUOHE OHQOQOZ huw50< man—awn. IAUH§O< man—UHH. huw50< QEUHH hufiflofi mgufiH. (2 :0HQH> UQQHUgU 30A huH=o< Hum one uwoz anuHH mo moooo oaouuxm .m oHnoH 39 these cases shows that the first case showed a greater than average loss of acuity in the far acuity condition, but did not show any great loss in the near acuity condition. The second case showed the same re- lationship, but to an even greater extreme, this S ranging from 0.75 in the photopic far acuity condition to 3.50 in the meSOpic far acuity con- dition. The third case showed a different relationship; his Titmus near acuity scores (both meSOpic and photOpic) were poorer than the others. This last extreme case in the Titmus far acuity situation was also an extreme case in the Titmus near acuity extreme cases. An examination of the low contrast scores in this extreme group showed them to be poorer than normal, but not so poor as to cause con- cern in the night driving situation, this level thought to be 40 per- cent. The low contrast meSOpic scores were .25,.l8, and .15 with the experimental group mean at .074 (Figure 9). This correlation between relatively poor low contrast meSOpic scores and relatively poor far acu- ity mesopic scores may be due to a common visual function; but the fact that the low contrast target is a simulated 20 feet from the subject's eye and subtended a visual angle of two minutes of are complicated the situation. It is suggested that the low contrast test's targets be in- creased to four minutes of arc in order to better eliminate covariance of far acuity and contrast in the test. There were two extreme cases in the near acuity mesOpic scores that stood apart from the distribution, both with scores of 2.00 minutes of arc. One of these Ss also had an extreme far acuity score in mesopic and photopic conditions. The other individual had a 1.75 minute Titmus meSOpic far acuity score -- not especially bad. To sum, the Titmus was a useful instrument for detecting extreme ‘1.- 40 cases using base data. Using an arbitrary cutoff of 3.00 minutes of arc, experimentor found three cases of relatively poor far mesopic visual acuity -- all 3.50 minutes. All three showed a similar record with rela- tively poor meSOpic low contrast scores (but still not sufficient for concern in night driving) and a relatively small decrement in near acuity in meSOpic light. This relationship between a large far acuity decrement and a small near acuity decrement was indicative of night myopia -- a more exact diagnosis on a direct lens testing device (such as Richards') would be needed to confirm this. The Titmus was useful as a meSOpic acu- ity detection instrument though additional diagnosis and possible remedy should be undertaken after such detection. The second aim of the experiment was to correlate the Titmus far acuity against the Snellen far acuity test in order to establish validity against a known standard. Table 3 shows the Snellen photopic acuity test had a mean of .85, and a standard deviation of .25, while the Titmus photopic far acuity test had a mean of .94 and a standard deviation of .29. These two tests correlate .70, but the difference between the means was not significant at the .01 level. In mesopic conditions the Snellen test gave a mean of 1.52, while the Titmus far acuity test gave a mean of 1.32 with standard deviations of .47 and .61, respectively. Again the correlation between the means was .70, and again the difference between the means was significant. The Snellen was thus the more conservative test in the mesopic condition, while the Titmus was more conservative in the photopic condition. Differences between testing conditions could account for the differ- ences found. Snellen targets were letters of the alphabet, while the Titmus used Landolt rings. S would be more likely to guess correctly 41 with the letters of the alphabet since he has had experience with them before. In addition the Titmus simulated the 20 foot distance through the use of slides and a viewer; the Snellen was an actual 20 feet from the S. Some few SS had trouble getting the images the Titmus presented binocularly to merge, and others complained of glare from the viewfinder type oculars. The Titmus therefore appeared less valid than the Snellen, Minor but annoying problems prevented the Titmus from achieving a com- pletely realistic simulation of the 20 foot distance. Correlations be- tween the two tests were still impressive -- indicating they shared 49 percent of their variance. Although the difference between the means was significantly different, the difference was not of such a magnitude to discourage its use for visual acuity testing. If we accept Richards' hypothesis that contrast does not become a critical factor in visual acuity until it gets below 30 or 40 percent, there were no Ss exhibiting such poor low contrast vision that they would be of concern for night driving. The worst cases in the meSOpic condition were .25 percent contrast; under photopic conditions the worst case was .15 (Figures 9 and 10). Again we do note a lower mean (and a lower stand- ard deviation) in the photOpic condition, but there was no indication of any serious visual handicap in the low contrast condition. The young age of the sample may have been a reason for this. Observing the inter- correlation matrix, experimentor noted no pattern of low contrast correla- tions, except for a general higher series of correlations with the Snellen acuity and Titmus far acuity. Since the low contrast test tests vision at a simulated distance of 20 feet, this was probably due to the far vision function being common to all three tests to some extent. Table 5 represented the last aim of the experiment. If there were 42 no significant learning effects, and if the dark adaptation period were adequate there would be no significant difference between the first show- ing of the tests in mesopic light and the second. The results show rela- tively high correlations between the two treatments (.55 to .86), indi- cating high test-retest reliability, but there was one difference between the means significant at the .01 level, that of the Snellen acuity tests' first and second mesopic treatments. The Titmus tests behaved consistently in one direction while the Snellen again disagreed with the Titmus somewhat, although it too did have a high test-retest relia- bility. It was apparent again that the Snellen was somewhat at odds with the Titmus simulation of distance. Again the complications of in- ternal lighting, simulated distance, and hard image fusion appear to have caused minor shortcomings in the test validity of the Titmus tests. Since the correlations were consistently high between the first and se- cond meSOpic treatments of each test in the control group, and the sig- nificant Snellen difference was not so great as to cause concern in this experiment (amounting to .20 of a minute of arc), the control group was concluded to have demonstrated no significant dark adaptation or learn- ing effect. SUMMARY AND CONCLUSIONS Through the use of the Titmus Vision Tester near and far acuity tests at meSOpic and phot0pic illumination, detection of unusual cases of mesopic acuity loss (at simulated night driving illumination levels) was attempted. Scores on the Titmus far acuity test, which simulated a 20 foot testing distance, were compared with scores of the Snellen eye chart at an actual 20 foot distance. Performance on a variable low con- trast test as opposed to the higher contrast acuity tests was also under- taken to look for possible relationship to night myopia. An experimental group (N=71) and a control group (N=25) were tested. The experimental group was given the Titmus near acuity test, the Tit- mus far acuity test, the Snellen far acuity test, and the Titmus low contrast test, in mesopic, and then again in phot0pic, lighting. Thus each subject set his own norms in the photopic condition. A control group was given the same tests in meSOpic light both times. The pur- pose of the control group was to detect improvements in performance due to either improved dark adaptation or increasing familiarization with the tests. Subjects were drawn from the MSU student p0pulation and ranged in age from 18 to 25. The results showed that: l. The Titmus far and near acuity tests, when used in conjunction on Ss in meSOpic and photOpic lighting conditions were of value in detecting unusual cases of meSOpic acuity loss, both near and far. Four extreme cases were found, and one S was an ex- treme case in both near and far mesopic acuity. Three of the extreme cases had far acuity scores of 3.50 minutes of arc, and as such represent potential causes of concern in the night 43 44 driving situation. To confirm the presence of night myopia or hyperOpia further testing was suggested on more SOphisti- cated apparatus such as that used by Richards (Richards, 1966). The Titmus far acuity test correlated well with the Snellen, but did show significant differences between the means. This was attributed to a number of problems the Titmus far acuity test presents with its less than ideal simulation of the 20 foot testing distance. There were no cases with Titmus mesopic low contrast greater than .25. This was not of a magnitude to cause concern in night driving. The fact that the subject ages were exclusively in their teens and twenties may have been a reason for this. The Titmus low contrast showed some correlation with other far acuity tests -- namely the Titmus far acuity and the Snellen. This was attributed to the fact that the Titmus low contrast test uses 20/40 targets at a simulated distance of 20 feet. Finally, the control group scores showed high correlations be- tween the first administration of the tests in mesOpic conditions and the second. The Snellen test, however, did Show a signifi- cant difference between the means while the Titmus tests did not. Differences between the two tests could have caused the differ- ence, and the difference, although significant, was not of a magnitude to cause concern. B IBLIOGRAPHY BIBLIOGRAPHY Adams, L. A. (Ed.) Webster's Dictionary of the English Lagguagg. New York: Educational Book Guild, 1957. Barret, N. R. Dark Adaptation and Light Adaptation. In C. H. Graham (Ed.), chp. 8, pp. 185-206. Bartley, S. H. Handbook of Experimental Psychology. New York: Wiley, 1951. Berger, C. The Dependency of Visual Acuity on Illumination and Its Relation to the Size and Function of the Retinal Units. American Journal of Psychology, 1941, SS, 336-352. Brown, J. L. The Structure of the Visual Stimulus. In C. H. Graham (Ed.), chp. 2, pp. 39-59. Bruning, J. L. & Kintz, B. L. Computational Handbook of Statistics. Glenview: Scott, Foresman, & Co., 1968. Davson, Hugh (Ed.) The Eye: Volume II. New York: Academic Press, 1962. Davson, Hugh (Ed.) The Eye: Volume IV. New York: Academic Press, 1962. Forbes, T. W., Vanosdall, F., Bieler, R., Schuller, J., & Ferrara, J. Very Low Contrast Vision at Night Driving Luminance Levels, Michi- gan State University, 1969. Unpublished. Forbes, T. W., Vanosdall, F., Pain, R. F., & Bloomquist, D. W. Low Contrast and Standard Visual Acuity under MeSOpic and PhotOpic Illumination. Journal of Safety Research, 1969, 1(1), 5-12. Geldard, Frank A. The Human Senses. New York: J. Wiley & Sons, 1953. Graham, C. H. Fundamental Data. In C. H. Graham (Ed.), chp. 4, pp. 68-80 0 Graham, C. H. (Ed.) Vision and Visual Perception. New York: Wiley, 1965. Graham, C. H. Vision III: Some Neural Correlates. In C. H. Murchison (Ed.), chp. 15, pp. 829-879. 45 46 Hecht, S. The Relation Between Visual Acuity and Illumination. Jour- nal of General Psychology, 1928, 11, 255-281. Hecht, 8. Vision II: Nature of the Photo-Receptor Process. In C. Murchison (Ed.), chp. 14, pp. 653-703. Hecht, S. Visibility of Single Lines at Various Illuminations and the Retinal Basis of Visual Resolution. Journal of General Physi- ology, 1939, SS, 593-612. Kimble, G. A. & Garmezy, N. Principles of General Psychology. New York: Ronald Press, 1963. Knoll, H. A. A Brief History of 'Nocturnal Myopia' and Related Phe- nomenon. American Journal of Optometgy, 1952, 22, 69-81. McCormick, E. J. Human Factors Engineering, New York: McGraw-Hill, 1964. Murchison, C. (Ed.) Handbook of General Experimental PSychology. London: Oxford University Press, 1934. Richards, 0. W. Vision at Levels of Night Road Illumination. Ameri- can Journal of Optometgy, 1966, SS, 313-319. Richards, 0. W. Night Myopia at Night Automobile Driving Luminances. American Journal of Optometry, 1967, 33, 517-523. Riggs, L. A. Visual Acuity. In C. H. Graham (Ed.), chp. 11, pp. 321- 349. Stevens, S. 8. (Ed.) Handbook of Experimental Psychology. New York: Wiley, 1951. Troland, L. T. Vision 1: Visual Phenomena and Their Stimulus Corre- lations. In C. Murchison (Ed.), chp. 13, pp. 653-703. Woodworth, Robert S. & Schlosberg, Harold (Eds.) Experimental Psycho- logy. New York: Holt, Rinehart, & Winston, 1965. 125V 3 1977‘ T ,.jflflflfllflmim‘njjml'llu1111114: R155 10570