ENGINEERING RESEARCH INSTITUTE THE UNIVERSITY OF MICHIGAN ANN ARBOR Final Report DETECTION THRESHOLDS FOR POINT SOURCES IN THE NEAR PERIPHERY H. Richard Bllackwell Ann B. Mo6dauer Vision Research Laboratories ERI Project 2455 BUREAU OF SHIPS, DEPARTMENT OF THE NAVY CONTRACT NO. Nobs-72038 WASHINGTON, Do C. June 1958

~/z.;,~-,.*.... -_,, Ci.

The University of Michigan ~ Engineering Research Institute 2455-14-F TABLE OF CONTENTS Page List of Figures itl List of Tables iv Sumrnmary v I. Tntroductiion II, Apparatus and Procedures 2 IIIo Results 8 References 11

The University of Michigan * Engineeriin Research Institute 2455 14 F LIST OF FIGURES Number Title 1 Arti sts con eption of the basic psychophysics test facility 2Relative threshold contrast values for various azimuths of the visual field 3'$~tt~~Threshold contrast as a function of fixation eccentricity (high background luminances) 4 Threshold contrast as a function of fixation eccentricity (middle background luminances) 5 Threshold contrast as a function of fixation eccentricity (low background luminances ) 6 Threshold contrast as a function of background luminance (central fixation) 7 Threshold contrast as a function of background luminance (1 degree eccentricity) 8 Thres-hold contrast as a function of background luminance (2 degrees eccentricity) 9 Threshold contrast as a function of background luminance (4 degree' eccentricity) 10 Threshold contrast as a function of background luminance (8 degrees eccentricity) 11 Threshold contrast as a function of background luminance (12 degree eccentricity) -__ _ __ _ _ _.__ __ _ _ _*__ _ tii

The University of Michigan -- Engineering Research Institute 2455- 14-F LIST OF TABLES Number Title Page I: AAverage Threshold Contrast Values 13 _~~~~~g

The University of Michigan ~ Engineering Research Institute 2455 14=F SMMRY Two observers have obtained visual detection thresholds for the foveal center and for thirtytwo. locations in the peripheral retina within a radius of 12 degrees from the fovea. The locations have fallen along eight equally spaced meridians of the visual field, at distances of 1, 2, 4, 8 and 12 degrees firom th ateiona centero These measurements have been made at each of nine levels of background luinance, ranging from zero to 75 foot-lamberts A total of 368,250 observations have been made, utilizing the temporal forced-chiofe variant of the method of constant stimulio The target was a citrcle whose diameter subtended 1 minute of arc; the exposure duration was.01 second. The data were analyzed separately initially for each of the meridians studied, There were significant differences in $ensitivity among the different meridians, the pattern of sensitivity differences differing at different levels of background luminance, Although these differences are of considerable theoretical interest, they were rarely of sufficient size to be of practical significanceo Accordingly, the data for different retinal locations equidistant from the fixational center have been averaged. The primary data concern the effect of the extenLt to which a target falls eccentric to the fixational center upon the threshold contrasto As expected, the threshold contrast was considerably higher for off-axis locations than for the fixational center at high levels of background luminance. On the other hand, at low levels of background luminance the threshold contrast was somewhat lower at off-axis locations than at the fixational center, At a background luminance level of about 10'3 footlamberts, the sensitivity of the visual field within a 12 degree radius of the fixational center was e$sentially uniform. It is possible to plot the relation between threshold contrast and background luminance for the foveal location and for the various peripheral locations0 The data for each peripheral location exhibited a discontinuity which was more marked the farther was the location from the fixational center. These discontinuities presumably represent the well-known rod cone "break"".

The University of Michigan ~ Engineering Research Institute 2455-14-F I, INTRODUCTION Previous reports from these laboratories (for example Ref0 1, 2, 3, 41 5) have been concerned with the detection capability of the eye for targets located on the line of-sight. These studies have been concerned with theinfluence of such variables of the physical environment as the luminance of the background, the size and shape of the target, the time the target is visible, the spectral composition of the target, and the non-uniformity of lumitnance of the backgrournd tthe vicinity of the target. Data on the limneof-sight capabilities of the eye are sufficient for use in only a limited numiber of practical visibility situations. At high luminance levels where target detectability is greatest on the lineof$sight, these data enable us to compute maximum visibility ranges in various practical situations e At low luminance levels, it is generally accepted that targets are more detectable off the line-of-sight0 In these cases, data for the line-of-sight obviously do not enable us to compute maximum visibility ranges, In addition, many practical visibil ty problems are concerned with the visibility ranges of targets under conditions in which the range is less than maximum because the observers are ntaware of the position of the target and hence must search and scan the visual field0 Under these conditions, many -targets never fall on the line-of-sighto Others may fall on the line-of-sight only very briefly in comparison with the time they fall at locations off the line-of-sight. In order to assess visibility distances under these conditions, we require comprehensive data on the visual detection sensitivity of the eye for all regions of the visual field- The influences of such variables as background luminance, target size and shape, and target duration need to be studied for various regions of the visual field~ The present study represents the first step in a program intended to provide these data. The sensitivity of the near periphery of the visual field has been studied out to 12 degrees from the line-of-sight, along each of eight meridians0 Measurements have been made at nine levels of backm ground luminance varying from zero to 75 foot-lamberts. A circular target subtending 1 minute of arc has been presented for 0Q01 second0 In these experiments, the observers had complete knowledge of the off-axis position to be occupied by the target0 Thus, these studies do not take account of the fact that in practical visibility situations, the observers not only are often looking away from the target, but also have no prior information as to where the target will appear. The influence of this informational variable has been studied to some extent is an earlier publication (Ref, 6) and will not be considered here.

The University of Michigan ~ Engineering Research Institute 24'55 14-F II0 APPARATUS AND PROCEDURES The apparatus used to: produce the background and target luminances is illustrated schematically in Figure 1 The observers viewed the far wall of a large cube which served as the background luminance, through a large opening in the near wall"o The far wall was lighted by a series of luminaires located on the near wall of the cube around the opening,', which were shielded from.the observers. eyes The target appeared from time to time in the center of the far wall~ as a luminance incr$ementv The target was produced by a projection syste located behind the far wallo. The larger portion of the far wall was covered with a translucent plastic screen. The target was produced by transilluminating.. the- screen over a restricted area.: The screen was located at distances from 8082 to 12.33 feet from the eyes of: the.observers.in different experiments0 High background luminances were produced by four arrays of tung' sten incandescent lamps, one array on each of the sides of the opening in the near wall. A level of 75 footlamberts was produced by lOOw100att lamps, whereas a level of 1 foot-lambert was produced by 40-Owatt lamps,'In each case, direct illumination of the plastic screen by the lamps was prevented by disc deflectors placed on the lamps, For all background luminances less than 1 fto-L., the luminance of the plastic screen Was provided by two small integrating light boxes containing 6-volt, 32- candlepower incandescent lamps. Each-light-box had an opal glass screen. These screens were not used to illuminate the plastic screen directly, but were used to illuminate the near wall of the cube directly, the plastic screen being illuminated entirely by reflection from the near wall, These two boxes were mounated against the sides of the cube walls at a point whiche prevented the observers from seeing the opal glass surfaces, The luminance of the plastic screen was adjusted by use of Wratten neutral filters placed over the opal screens of the light-boxes. The target projector was based upon a special tungsten ribbonfilament lamp developed by the General Electric Company' the output of which was equivalent to a 1000-watt monoplane projection lampo The ribbonfilament lamp was operated at 6-volts AC. Two sets of Wratten neutral filters were used to adjust target luminanceo "Fixed" filters reduced the target luminance to the threshold rangeQ These filters were fixed during any given session, Additional adjustment of the target luminance.was obtained by the "psychophysical" filters.o These filters adjusted the target luminance to the values desired within the psy'chophysical range. Five psychophysical filters were used, mounted on an electrically-driven filter-selector which could position one or another of them in the projection beam0 The presentation of the target was controlled by the operation of a flag-type shutter. Whenever a solenoid was activated, the shutter was removed from the projection beam0 The target projection system consisted primarily of condensing lenses wlhich iaged the ribbon filament in the plane of a rotating wsedctor 0: disc used to time the target pulse. A collimator was used us$t beyond. _ _ _ _~ s; *4 @.i _ ___________ 2

The University of Michigan * Engineering Research Institute 2455 14F the sector disc Then9m a -very short foca-llength'magnifiert' was used to form a small image of the filament on a metal aperture mounted flush against the rear wall of the translucent plastic screeno The aperture limited the size of the tranillruminated target to approximately o035 inch~ The side of the pla tic screen facing te obs ervers was covered with an extremely thin layer of white sphere paint, designed to eliminate the specularity of the plastic Screen without introducing spectral select-'ivity or blurring of the transilluminated target0 The color temperatures of the target and of the background luminance were set for 2850~ KO:The timing of the target presentation was arranged and controlled by a timer comprising two discs mounted on a single shaft9 one rotating at seven times the speed of the other~ and designed to juxtapose two adjustable slots in the perimeters of the discs and the projection beam0 The timer wheels intercepted th projetor beam in the plane of a filament image0 Variation in the chord length of the slots could be utilized to produce continuous variation in the duration of the pulse of light from about 0 001 to 0~03 second The time required for the target to come to full luminanee was 0,0001 second (Thea ribbon-filament was used in order to minimize the onset timeC) In the present experiment, an exposure duration of 001 second was used t-hroughouto This duration was selected to be shorter than the critical duration (Refo 7) at at l values of Bo The temporal forced-choice variant of the method of constant stimuli as described in detail by Blackwell (Ref, 8) was used in this study. Essentially9 the temporal force4dchoice method presents the observer with four successive9 aurally delineated, 2-second time intervals0 In one of these onlyS as determined randomly by the presentation sequence, there appears a target of randomly selected luminanceo After a cycle of four temporal intervals i.completed the observer is allowed eight seconds to press one of four buttons located on his arm rest indicating in which of the four intervals he believes the target' to have appeared~ These responses are automatically tallied in a dis-tant rom on electrical counters and punched on record cards for permanent reference o Although ten targets of the ame luminance succeed each other9. each block of ten is randomly arranged with respect to all other blocks of ten in terms of luminanceo Fiv such block ranging in difficulty from a target visible nearly 100% of the time to a value virtu~ ally never visible are presented in one experimental session~ After the presentation of fifty targets9 observers are permitted:a five minute break while the equipment is reS eto A fifteen minute break:for refreshment customarily follows e he third block of stimulit Automatic presentation and recording equipment was used, designaed for use with this method and described by Blackwell9 Pritchard, and OImart (Ref0 9)> The amassing of the great amounts of data necessary to t1hi study would have been impossible without thi$ equlipment, The basic experimmental data were percentages of correct choice -for each o ffyve target tauniance incrementsl$0 Analysis of the data begins

The University of Michigan ~ Engineering Research Institute 2455-14=F by eliminating the effect. o chane successe by means of the relationo where p' R corrected proportions and p raw proportiono The corrected proportions were analyzed by a variant of the probit analysis developed by Kineaid and Blackwell (Refo 10), based on the general probit method of Finney (Refo l1)o Baaically, the probit method fits a theoretical curve to the data to satisfy the maximum likelihood criterion. In this case, the theoretical curve was the normal ogiveo Analysis of the data in this manner yields the value of the threshold, the standard error of the threshold9 the slope of the ogive the standard error of the slope% and an estimate of the goodness of fit determined by the Chi-square test. The basis of all photometry was the Macbeth Illuminometer, calibrated against standard lamps and standard reflectance surfaces certified by the Electrical Testing Laboratories New York0 The calibration of the illuminometer and filters was cheke d several times during the experiments O High luminance$ of the screen were measured directly with the ilTluminometer9 fitted with a lens which imaged the screen in the photometric cube of the deviceo Low luminances, produced by the light boxes described above9 were photometered indirectly0 The ratio between the luminances of the opal screens of these light boxes and the resulting screen luminance was measured at maximtaum output of the light-boxe When the output of the light boxes was reduced by filters to produce low luminances of the screen the luminances of the opal screens of the light boxes were measured and the ratio used to compute the screen luminanceo This procedure i$ entirely adequate since the optics of the light sources were not altered by the use of the filters The luminanee of the small target was difficult to photometer0 The basic measurement involved what may be called a "candle-poiwer box"0 A closed metal box was made with an opal disc at one end and a small aperture at the other. The aperture was fitted precisely in place so that the transillumina&ted target lay entirely within it. The target thus became a source for 0lumination of the opal disc at the other end of the boxo Baffles were placed within the box to eliminate interreflectionso'The luminance of the opal disc was measured with the illuminometero From the transmittane of the opal disc and the inverse square law9 the intensity of the small target could be determined0 The luminance of the small target was computed from the measured size of the target0 The measuremsen t with the candlepower box is sotewhat tedious0 -Accordi'ngly9 occasional measurements were made in this way and more frequent measurements vre mEade with a photoelectric telephotometer9 calibrated in terms of the candlepower boo r x mea unrement0o The t elephotomJeter

The University of Michigan * Engineering Research Institute 2455 14-F consisted of a telescope which'imaged the small target on the cathode of a 931 photomultiplier tube.o The relative luminance of the target was determined by the emf requred t compensat the photoeleca tric current produced by the targ et. Te target increment provided by the projector i$ designated B. When there ri a finite value of the background lumin ance B, itt is customary to specify detection ensitivity in ter$ Of target contrast, Cp defined as follows"o c B (2) B The target projection apparatus was fixed in position with respect to the screen and could present targets only in the precise centerE of the screen~ Thus to obtain off-axis target presetations it was necessary to have the observers fixate various points which were not at the center of the screen, For the off-axis atudies which make up the' principal bulk of the data a small bright fixation point was provided the obiserverso The fixation point wa projected onto the side of the screen viewed by the observers by a projector mounted inaide the light ~cube out of the view of the observerso This projector utilized the image of the filament of a 6-volt 32o candlepower tungsten lampo The intensity: of the fixation point was always maintained at a value approximately ten times the threshold intensity siance extended experience has indicated that this intensity is the dime$t which can be comfortably used for the control of fixation and accommodation during extended periods. For the background luminanr e of 75 foot-lamberts it was difficult to obtain sufficient intensity from the projector to meet this criterion~ Accordingly, four such projectors were naed, and the images aligned on top of one another in order to nreae ta h fixation point intensty to the required level, For those comparison experim ents involving foveal presentations multiple fixation points were used to form a pattern around the point to be occupied by the target. Four points were normally used, each of one minute of arc diameteer surrounding the target and equidistant from its edges at a constant distance of eighteen minutes of arc, These four points were arranged in the form of the terminal points of the arms of a cross, with the point of' target appearance ocated at the intersection of an imaginary line through each of the vertical and horizontal pairs$ For a few foveal experiments only the horizontal pair of fixation points was utilizedo Earlier experimentation had revealed that equivalent results are obtained with two or four fixation points arranged in this mannero For experiments involving zern background luminanc e an appreciable error can be introduced by the reillumination of the screen from light produced by the fixation points which is reflected onto the floor and walls of th white cube and back onto th scrteen To ainsure that'the screen was perfect y dark~ a special stifL lack v elvt Covea was constructed which was mounted ain front of the plastic sCr een~ which covered the screen entirely0o A small hole wa~ cu t o for hte tangeta1 and a n.-umber of small holes were cut for the various positions to be occupied: 1~~~~~~~~~

The University of Michigan ~ Engineering Research Institute 2455 14-F by the fixation poinats Plugs were inserted in all the fixation point holes that were not actually in use in a given, experimento Normal binocular viewing was used throug out with natural pupils $so that the data would be directly useful for application to practical visibility problemso The refractive condition of the eye of each observer was deter mined under the conditions of the experiment by a method outlined by Ogle (Ref o 1,2) An "oculometer" was used to measure dynamic refraction with the stigmatoscopic techniqueo Masursueents were made with this apparatus under conditions ide ntical with toee of tohe experilent proper with the exceptibns that a stigma or point of light appeared in the apparent location usually occupaed by the target, The stigima was seen by reflection fr a half-s1lvered mirror located in the oculometer0 The basic oculo.metric procedure may be described as followg- the observe.r fxates the fixation poi nt stuated perhaps 10 feet from his eye, and observes changes in appearance of the bright stigma centered among themo The assembly which houses a tuetaen lampp, $tigmaa9 and reduction filters in the oculometer is movable along the optic axis parallel to a scale~ The distance from. the eye to the field lens is made equal to the focal length of the latter, making possible a linear scale calibrated in diopters whose modulus ui determined by the dioptric strength of the field lenos, (The zero point of the scale corresponds to the point at which the sti gmato-lens distance is equal to the focal length of the field lens and the image is at infinity~) Uisng the pycehophysical method of limits the observer adjusts the position of the lamp housing until the stigma appears to have min imn sizeo A mean of many such settings translated into diopters by the affixed scale, reveals the refractive state of the eye. Acco.modaton on the stigima would give a spurious result, This possibility i$ precluded by causing the lamp to flash intermittently, the "off" period being sustained conliderabl7 longer than the "on"o The discrepancy between the meaa of these oculometric settings and the theoretical "normal" refractive state of the eye at the actual viewing distance repre aents the refractive "error" of each observero From refractive error ist i possible to compute refractive correctionso Two observers were utilized for all experimental measurements, both females of age 30 One. of the osersve r (ABM) was also the experimenter; the other observer (LP) was a laboratory technician. The measurerment$ extended over a period of about 30 monthsh the motivation of the observer in the tedious taak of observing was truly exceptionaL, The observave were fitted wi th f d ophthalmic corrections for the purposes of the experiment on the basia of onleo.etric ~easurement~ as followso Observe r ABM OODB 10E1 $ 00 Sb l

The University of Michigan ~ Engineering Research Institute 2455 14 F Observer LP,ODDo os 3S75 -,25 1650 A regime of prior dark adaptation was adopted9 on the basis of well1known relations between pre-exposure luminance and sensitivity in the dark. At zero background luminWance Nay7 red adaptation goggles (Polaroid type) were used for 20 minutes before entering the experimental roomn, followed by a 10 minute period in the darko Alternativelyy9 a total period of 20 minutes in the dark could be u.ed at the discretion of the observers, The control of pre-exposure lum.inan ce was less stringent for the studies involving higher levels of background luminance At i0o3 footlamberts background luminance. the red goggles were used for 1O minutes before entering the experimental room followed by 3 minutes in the experimental cube o The two observer$ differed in their relative sensitivity under different experimental conditions0 In order to equate them sufficiently so that they ould observe at the e th e ime advantage was taken of the fact that the distances from the observer8s eyes to the screen differed among the four chairs a ailable for the ohbservrss (see Fig ure ), The observers were ahlited freely from. hair to O~iaMr to equate their thhre holds for each experimental condition -the range of viewing distance vary ing from 8082 to 1233 feet. All data obtained at different distances from tthe rcreen were corrected to th standard ditsancs at which the -target subtended 1 minute of arc, on the basis of the inverse square law o Experimental checks substantiated the validity of this manipulation of the data o Nne difaerent lvel of be kground pluomnawce were atudiedi n all. The order in which the different levels were studied waa haphazard. with all the data for a g.iven background generally collected during one conseentive period of time, The observer were extensively practiced before the collection of the data reported.ere having been used for more than 100 hours off observing during preLiminary experinments The standard routine of study at each background lumLinance included several sessions with ifoeal presl entations for compaaisoon purposes, interspersed during the peripheral studiesa An off-axis locationg or eccentricity, of about I degree was studied along No Sp E. and W meridianso Similarly9, a 2 degree eccentricity was stdied along tdhese fouor major meridianso Measurements were made along eight meridians for eccentricities of about 4, 89 and 12 degreess Thus9 there were 32 basic peripheral locations to be stud 0ied~ In general9 two separate sessions weres conductned for each peripheral locationo If the data fromn the two sessions did not show good agreemente an additional session or two were conducted under these same conditions0

The University of Michigan ~ Engineering Research Institute 2455r14iF III, RESULTS The raw data of all the studie$ except those conducted at zero background luminanee represented values of threshold contrast Q for each of the two observers under each of the numerous experimental conditions. These values were expressed in logarithmic terms from the outset and the averages which were obtained represent logarithmic averageso As ha been pointed out in an earlier report (Refo 5) it is possible to render threshold values of A6B, obtained at zero background luminance comparable with threshold contrast values obtained at finite values of background luminance by assigning an arbitrary w background lum9nane~ B, and computing "contrast" from equation (2)o When this was done in the earlier report B was set equal to IO03 foot lambert o This procedure was admissible in the earlier study because only foveal presentation was involved and the cone photoreceptors operate in an equivalent manner at 10o3 or zero luminance9 as shown in an earlier study (Refo 2)o In the present case, rod photoreceptors are presumably involved in at leat maost of the peripheral locationsr In this event, as was reported in 1946 (Ref0 13) a background luminance of 106 ts required to be equivalent to zero0 In this report " contrast" values for the zero luminance data were specified from equation (2) with B - 0-o Data for the two observers were analyzed separately at first, to evaluate the extent of simtlarity observedo The initial data graphs involved plotting the values of log 6 obtained at various peripheral locations in comparison with data for the foveal target0 For these graphs. the data obtained along the different meridians were averagedo It was found that the relations between log C and the extent to which the target was presented eccentric to fxation were very similar for the two observers at all the luminance levels studiedo Accordingly' the data for the two observers were averaged and all data to be presented here represents average results of this type. Concern was felt for the fact that eight meridians were included in the averages for the 4B 8, and 12 degree eccentricities whereas only the four major meridans were included in the averages fotr and 2 degree eccentricitie s Accordingly9 an analysis was made of the similarity between averagea of the 4W 8. and 12 degree data based upon four or eight meridians 0Only haphazard differences were found, so that it was felt admissible to use all available data in the averages at each eccentricity, Analyses were made of the extent to whichi the vales of log C were independent of the aximuth of te aeridian aloER whi ch the targets were presented. It was found that maximal azimuthal diff erenes were found at the 12 degree eccentricity~ Sample data are presented in Figure 2 illustrat$ing the variat ions in log C as ua fuction of aziut E r four of thle leve1 of background luminanee $tud ied0 There appears to be a regulat bange n send ita vit a a bhr au onn of azinounh for fe en75foo lanbent bacekground luninane, but th ne are 1es paronoun ed df~ rence$

The University of Michigan ~ Engineering Research Institute 24551l4~F at the other luminance leve1ls If it is indeed the case that azimuthal differences depend upon background luminance this may lmply that~ as has been suggested elsewhere (Ref. 14). the visual neural systet alters its networko of connections at different luminance levels At least for practical purposes, we are primarily concerned with average data from all aximutoh? The main results of the present experiments are presented in Tdable These data represent the staggering total of 368 250 observations, In averaging the data from the two observers~ account was taken of the fact that as they changed seats in the experimentta room, the eccentricity angle changed. Thus~, in Table I, each average value of log has a value of Ea which represents the eccentricity or extent to which the target was presented off-axis. The initial data graphs are presented in Figures 3~ i, and 5 Each graph contains the results obtained at three of the background luminance level, (These data have been aeparated into three graphs so that sufficiently large seales can be used in ve~a graph to reveal th precisae shape of the functional relationship between log threshold contrast and eccentricity.) It is apparent that there are large differenes in senasi tivity within a 12 degree radius of the foveal center, As expected on the basis of common experieicenf considerably more contrast is needed to detect a point target offaxis tha on ax at the higher luminances whereas lesa contrast is needed off-axs at low background uaminanaes. The fact that the visual field is relatively homogeneouas n sensitivity at a background luminance of about Q103 foot4lamberts is an interesting result. The existence of even less sensitivity I degree offaxis than at the foveal center for the afloweR baSc kgrounRd lumnance levels is a new1 finding of some theoretical intereSt o It is perhaps apparent that the smooth curves drawn through the data points in Figures 3 g~ and 5 do not fit the points as well as they might, The curvea used to fit these data were derived from a method of curve fitting which may be best described by reference to Figures 6 l I. These figures present the relations Between log threshold contrast and log background luminance for a foveal taret location and for each of the following ecg entric lo ationsg 1, 2; 4p 8, and 12 degrees off-axis. Now values for the foieal graph (i-igure 6) are taken directly from Table I when E w On O However all the other graphs have to be produced by interpolation from the $mooth curves i Fin gure 3 5 since the values of E in Table I are unequal and do not represent integer values of the eccentricity angle. The process of data-smoothing proceeded in the following stepas Smoothe a eeaircal w8r~es.rse fitted to xa~ 0m~f8raen X1] data in Fguresa 3 5 and value~ were interpolated for Figure 7 =llo The curves in Figuresa 7 11 were not mso1oth there appea rig to ae congpdera ble hap hazard irregularity tn tie varihn~ graphao It was ausued that the curves from Figures 7 1 hould ot be haphazar bu should be ginerally samooth although a di$acontinuity corresponding to to e We1llknown cOna rod "break" was expected0 Curve fits were alternaated Back and forth until smooth curves were gsenate d in Fig ure 7 t. wdith c urvlwe wwhic * 9

The University of Michigan ~ Engineering Research Institute 2455-14 F fit the experimental data ina Figure 3 5 as well as po sible. It was not required tha the tcrves in Figures 3 - 5 be 1moothP snce the different retinal locations have diff erent rceptor populations different peripheral blood supply etc Thus, there need not necessarily be a smooth relationship between eccentricity and te-hrsold cotrasto The curve for foveal Viewing presented in Figure is very similar to other curves reported from these laboratories both with respect to shape and absolute value. The curves for the peripheral locations always give some evidence of a discontinuity, even for the I degree eccentricity~ The discontinuity is of course more marked the more eccen tric is the location. In the case of a 12 degree eccentricity, presumably the egments at Ithe h igaher 1Luminance levels r eprtesent cone activity whereas the segment at the lower luminane lvaels represents p tpure rod activity, AL less extreme eccentricities mixed cone ad rod activity probably occurs in the segment obtained at the lower levels of background luminance The curves presented in Figures 7 11 allow us to assess the effect of target eccentricity, by comparison with the curve presented in Figure 6o These data should be useful in computations of practical visibility situations in which the target fails to fall upon the observer's line of sighto It should be pointed out however tat the target used in these studies subtended only I minute of are, It is not safe to assume that the effect of target eccentricity will be the same as we have found in thi$ study for targets of considerably larger sizeo

The University of Michigan ~ Engineering Research Institute 2455 14F REFERENCES -1 lIdHamilton, C. Eo and BlackwellS HO Ro "The Effect of an Horizonc line Luminanee Gradient Upon Target Detectability in its Vicinity"o UniverAity of Michigaj i Engineering Research Institute Report 2455~-83F, 30 p (957 ) 2 BdBlackwell H. R. and Law O o To "A Study of Possible 8Photosensitizati on' of the Human Eye by White Light". University of Michigan, Engineering Research Institute Report 2455,9-F, June 1958o 3 Taylor, Jo Ho and Blackwell, Ho Ro "Variations in Spectral Sensitivity Within the Human Fovea'0 University of Michigan, Engineering Research Institute Report 2455410-F, June 1958o 4 Kristofferson~ Ao B0 and Blackwe l, Ho Ro "The Effects of Size and Shape on Visual Detection for Continuous Foveal Targets at Moderate Background Luaminance" University of Michigan, Engineering Research Institute Report 2455-11-F June 1958o 5 Biackwel., Ho Ro and McCready, Do Wo0 Jr o "Foveal Contrast Thresholds for Various Durations of Single Pulsess" University of Michigan, Enginearing Research Institute Report 2455 3:'F, June 1958 o 6 Blackwell, Ho Ro "The Effects of Certain Paychological Variables Upon Target Detectability'o University of Michigan, Engineering Research Institute Report 2455=12F,, June 19580 %7 eGraham, Co Ho and Kemp, Eo Ho "Brightne$s Discrimination as a Function of the Duration of the Increment in Intensity",o J. Gen. P ihol e 6352650 (1938)o 8:3Blackwell, Ho R o PIyds ho i aa Thre otd~ Ex erimentat Studies of Methods of Measuremento University of Michigan, En3gng Re, Bul, No. 36, 227 P g(953)0 9 Blackwell, H o R,, Pritchard, B, So, and Ohmart s Jo Go "Automatic Apparatus for Sti$alu0 Presentation and Recording in Visual Threshotd Experiments " J o So $e9,o44 322 326 (1954 ) 10 Kincaid, Wo Mo and Blackwell9 Ho Ra'"Application of Probit Analysis to Psychophysical Data I, Techniquea for Desk Computation" Un iversity of Nic igan, Engi$eering Research Institute, Report 21l4=283-T (in pre )o 11 Finney P0 Jo Pr bt Ana yo y 0 Cembridge~ Cambridge Univre$ity Prees, 256 p (1947)~

The University of Michigan ~ Engineering Research Institute 245.5L14-F 12 Ogle~ Ko No Private Communication. 13 Blackwellp Ho Ro "Contrast Threshold of the Human Eye"o J t Soc0 Amer,, 6216246g43 (1946)e 14 Blackwell~ H.- R. "'Postulations of Neural Mechanisms Involved in Visual Detection from Experimental Studies of Target Size and Shape". (Abstract) J0'Opt Soo0 Amer._ _ 1054 (1957)

TABLE I Average Threshold Contrast Values Log Background Luminance (foot-lamberts) -60oo0 -5.10 -4.10 =2,75 -~20 -1,50 =1000 E Log C E Log E LogO E LogOC E LogOC E Logo E LogO C E LogC o 0 6, ~54 0 5 o63 0 4.63 0 3 o29 0 2,54 0 2,17 0 1,78 0 1,18 0.38.9 6,o6.9 5.71 1 a 4,63.9 3.48 1l1 2.33.9 2,01 10 1.38.9 057 1.8 6,43 1.8 5 6Q 1 9 4.64 1.8 3,53 2,0 2,76 2,0 2,14 2,0 1,55 1.8.91 m 3.5 6.os5 3.7 5. 22 3,6 4,34 3.6 3,D55 40 3Q03 740 2. 1.0 2.39.0 1.87 3,9 1,41 Ca Repo 5rt2 7 4 4,98 780 4,27 72 35054 7F8 3,20 8.0 2.66 8 252, 19 Igoo 5,,78 II.8 5,o8 12,0 4, j&&- 10,8 3 56 11.4h 3 oag 12 o 3 -o4 12.0 2.81 11.8 2,42 f~2Q:Report 2455-14w-F~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~e

Projectrr Light Cube Target Presentation System and Recording __ _ _ _ __ _ _ _ __ _ _ _ _ __ _ _ _ __ _ _ _ _ __ _ _ _ __ _ _ _ Apparatus/ Response Buttons ~ 4 _ -1.1S~~~cheduling Device * I_ ~~~~~Respaonse Recorder Recording Apparatus Observing Booth and Light Cube Fig. 1. Artist's conception of the basic psychophysics test facility. 12' ECCENTRICITY 9 (FT-L) 44 4.3t / t75 +.2 +I 0 > 0 ~ +.2 I I +. +.2 4.'' SW W NW N NE E SE' S SW AZIMUTH Fig. 2. The effect of target azimuth.

ok U40 3 -1.00 0.00 1.87 50% FORCED CHOICE DETEOTIONS o' 2 4 6 10 12 FIXATION ECCENTRICITY (DEGREES) Fig. 3. Eccentricity data: three highest luminances. as 1D ~~~4-~~~~~~~~~~~4. -2.00 ~, 50% FORCED CHOICE DETECTIONS FIXATION ECCENTRICITY (DEGREES) Fig. 4. Eccentricity data: three intermediate luminances. 15

7F I — 1 8. -4.10 O 4 50% FORCED CHOICE DETECTIONS 0 2 4 6 8 10 12 FIXATION ECCENTRICITY (DEGREES) Fig. 5. Eccentricity data: three lowest luminances.?- CA 1134 CENTRAL FIXATION C) 450% FORCED CHOICE DETECTIONS -6 -5 -4 -3 -2 -I 0 I 2 LOG BACKGROUND LUMINANCE (FOOT-LAMBERTS) FlIg. 6, Background luminance effect. 16

1 tr~~~~~~~~~~~~~~~~~~~~~~~oCA 1135 6 45. ECCENTRIC FIXATION 2 X 50% FORCED CHOICE DETECTIONS -6 -5 -4 3 -2; O I 2 LOG BACKGROUND LUMINANCE (FOOT-LAMBERTS) Fig. 7. Background luminance effect., GA 1136 6 5\ 2' ECGENTRIC FIXATION'3 X 2 50% FORGEO CHOICE DETECTIONS -6 -5 -4 -3 - -; t i LOG BACKGROUND LUMINANCE FOOT-LAMBERTS) Fig. 8. Background luminance effect, 17

7, GA 1137 7A 11340 ECCENTRIC FIXATION 5 8 ECCENTRIC FIXATION Iz4 0 0 0 o 3 ~~~~~~~~~~~~~~~~~1~ ~ ~ ~ ~ ~~~~~~~~~( 2 2 50% FORCED CI4OicE DETECTIONS 50% FORCED CHOIlE DETECTIONS ___ __I __I___ ___ ___I__ ___ __ __j 0.I -6 -5 -4 - -2 -I C I 2 -6 -5 -4 -3 -2 -I C I 2 LOG BACKGROUND LUMINANCE (FOOT-LAMBERTS) LOG BACKGROUND LUMINANCE (FOOT-BEFAIIER Fig. 9. Background luminance effect. Fig. 10. Background luminance effect. 7- GA 1139 5- 120 ECCENTRIC FIXATION 43. 50% FORCED CHOICE DETECTIONS 0 -6 -5 t 3 2 A1 O r LOG ACKGROUND LUMINgAlCE (FOOT-LAMBERTS) Fig. 11.:Back round luminance effect. U~~~~~

UNIVERSITY OF MICHIGAN 3 9015 02514 8225