A —Fst-. L—'" -—-‘—w—L-'-1‘—v 4". ~—
‘ v r . .

 

 

AN INVESTIGATION OF CHROMATIC
BRIGHTNESS ENHANCEMENT TENDENCIES

Thesis Ior II'ne Degree oI pI1. D.
MICHIGAN STATE UNIVERSITY
Richard James Ball

1963

THESIS

This is to certify that the
thesis entitled

AN INVESTIGATION OF CHROMATIC BRIGHTNESS
ENHANCEMENT TENDENCIES

presented by

Richard James Ball

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in PSYCh010gY

34mm;

Major professor (I

Date 3 / 6

0-169

      
   

  

LIBRARY

Michigan State
University

ABSTRACT

AN INVESTIGATION OF CHROMATIC BRIGHTNESS
ENHANCEMENT TENDENCIES

by Richard James Ball

A modified Fry type prism monochromator with episcotister and
surround field attachments was utilized to quantitatively investigate
chromatic brightness enhancement tendencies. Thirteen narrow band
targets were used ranging from 460mp. tc 680 mu. For most of the
investigation a constant target luminance of 50 foot lamberts was
maintained. Rates of intermittency from 6. 5 to 20. 0 cycles per second
were used and pulse to cycle fractions of 1/16 to 3/4. Target sizes of
1%- and 4 degrees were used. Most of the investigation was done with
dark target surround but target surrounds of varying intensity and
dominant wavelengths of 540mg and 600mg were also utilized. In one
portion of the study target luminance was also systematically varied.

, Three color normal subjects were used. A

A bipartite target was used with one-half steady and one-half
intermittent. The two halves were separated by a narrow black band.
.Data were collected on the brightness, hue, and saturation of the inter-
mittent target in relation to the steady target.

The following results were obtained. Only pulse to cycle fractions
of around 1/4 gave marked brightness enhancement tendencies. Only
rates of intermittency of 12 cycles per second or less gave marked
brightness enhancement tendencies. Only wavelengths close to 500319

gave marked brightness enhancement tendencies. Increased target size

Richard James Ball

or decreased target luminance tended to decrease brightness enhance-
ment tendencies. Illuminating the target surround tended to decrease
the brightness enhancement tendencies except for the condition where
target and surround were matched on both luminance and hue.
Desaturation or "washout" of the intermittent target was obtained
only in the regions of 500m}; and 620mg. At 500ml; with the proper
target rate of intermittency, pulse to cycle fraction, and luminance
level the intermittent target became almost achromatic as well as en-
hanced in brightness. At 620mu the intermittent target became
moderately desaturated but brightness remained at about the Talbot level.
Hue shifts in the intermittent target were often obtained and when
they occurred they always followed a specific pattern. For wavebands
below 500mg the hue of the intermittent target appeared to shift to that
of a lower wavelength; that is it became purplish. For wavebands from
500mg to 560mg the hue appeared to shift to that of a higher wavelength;
that is it became yellowish. For wavebands from 580m}; to 680mg the
hue appeared to shift to that of a lower wavelength; that is it became
orangish. Thus, neutral points in hue shift were obtained at 570mg
where the intermittent target did not shift in hue and at 500mg where the

intermittent target became almost achromatic.

AN INVESTIGATION OF CHROMATIC BRIGHTNESS
ENHANCEMENT TENDENCIES

BY

Richard James Ball

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPH Y

Department of Psychology

1963

ACKNOWLEDGMENTS

It is doubtful that many truly original pieces of work occur.

Most of what we consider original research is the result of digestion,
assimilation, alteration, and regurgitation of various facets of what is
already known. This makes it very difficult to give proper recognition
for assistance on this project. Acknowledgment must be given to the
entire human race, past and present, for providing the necessary
foundation of knowledge which we call civilization.

Special acknowledgment must also be given to certain individuals
who have been particularly instrumental in my work. First of all my
parents whose sacrifices and encouragement made possible my primary,
secondary, and undergraduate education as well as providing me a
wholesome and stimulating environment in which to grdw. I owe an
especially great debt to my mother whose demonstration of complete
lack of selfishness in dealing with others deeply imbued me with the
realization that the only worth-while things in life are obtained by work—
ing with others and not against them.

To Dr. Glenn A. Fry I owe the inspiration which has guided me
through my graduate work. Dr. Fry has for over ten years provided me
a living symbol of the highest scientific endeavor and attainment toward
which I can only strive. Also, without Dr. Fry's generous loan of
equipment I could never even have commenced this investigation.

Dr. S. Howard Bartley is owed a great debt of thanks for his in-
spiring and generous guidance as my graduate advisor. Without his
untiring assitance this project could not have been completed. Certainly

great credit is due the other members of my guidance committee,

ii

Dr. Paul Bakan, Dr. Charles Hanley, and Dr. Byron VanRoekel for
their contributions of time and assistance in my graduate work. Also,
a debt of thanks is owed Dr. Charles Bourassa who served as subject
and to Dr. Thomas M. Nelson who not only served as subject but made
many other valuable suggestions and contributions. An extremely
valuable service was also rendered by Dominic J. Zerbolio who drew
the final copies of the figures and to Richard Clark who photographed
and reproduced the figures.

Of invaluable assistance in all phases of my graduate program
has been my wife. Without her abundance of help and understanding
throughout the trials and rigors of the doctoral program I most certainly
would never have made it. Also, it was her initial idea that started me
on the train of thought which led to this research. Finally, and above
all, I give thanks to the transcending Power from above that has made

possible my work and all of the foregoing assistance.

***************

TABLE. OF CONTENTS

Page

INTRODUCTION . . . . . . . . .

HiStorical Review 0 O O 0 O O O O O O O O O O O O O O O 2
ExPeCtations O O O O O I O O O O O O O O O O O O O O O O 5
METHOD 0 O O O O O O O O O O O O O O O O C O O O O O O O 0 O O 7

:APparatuSoooooooooooooooooooooooo 7
'Procedure....................... 9
RESULTS-0000000000000... 000.0000... 12

Summary of Results . . . . . . . . .

~CONCLUSIONS . . . . . . . .

FUTURE PLANS O O O O O O O O O O O O 0 O O O O O O O O O O 29
REFERENCES 0 O O O O O O O O O O O O O I O O O O O O O O O O 0 30
APPENDIX 0 O O C O O O O O O O O O O O O O O O O O O O O I O O 0 36

iv

LIST OF FIGURES

FIGURE

9.

10.

Schematic illustration of apparatus . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus for various pulse to cycle fractions. . . . . . .

Rate of intermittency plotted against luminance of the
matching stimulus for various wavelengths. . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various rates of
intermittency;PCF.1/1.6. . . . . . . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various rates of
intermittency;PCFl/4. . . . . . . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washing" for various rates of
intermittency; PCF 1/4; target4 degrees . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various rates of
intermittency;PCFl/2. . . . . . . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective"'washout" for various rates of
intermittency;PCF3/4. . . . . . . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various luminance
levels; PCF 1/8 O O O O O O O O O O O O O O O O O O O O O O

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various luminance
levels; PCF 1/4 O O O O O O O O O O O O O O O O O O O O O O

Page

37

38

39

40

41

42

43

44

45

46

LIST 'OF FIGURES - Continued

FIGURE

11.

12..

13.

l4.

15.

16.

17.

18.-

19.

20.

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various luminance
levels; PCF 1/4; target4 degrees. . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various luminance
levels; PCF 1/2 O O O O O I O O O O O O I O O O O O O O O O

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various luminance
levels;PCF 1/4; observer TN . . . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various luminance
levels;PCF 1/4; observer TN; target 4 degrees. . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various luminance
levels; PCF 1/4; observer CB. . . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various luminance
levels; PCF 1/4; observer CB; target 4 degrees. . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various 540mg
surround luminance levels. . . . . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various 600ml;
surround luminance levels. . . . . . . . . . . . . . . .

Wavelength plotted against luminance of the matching
stimulus and subjective "washout" for various day
repetition of the same condition . . . . . . . . . . . .

Rate of intermittency plotted against luminance of the

matching stimulus and subjective "washout" for three
widewavelengthbands. . . . . . . . . . . . . . . . . .

vi

Page

47

48

49

50

51

52

53

54

55

56

LIST OF FIGURES - Continued

FIGURE Page

21. Hue shift for experimental conditions using thirteen
wavebands. O O O O O O O O O O O O O O O O O O O O O O O O 57

22. Hue shift for experimental conditions using five wave-
bands . O O O O O O O O O O

O O O O O O O O O O O O O O O O 58

INTRODUCTION

This study is an investigation of chromatic brightness enhance-
ment. The brightness enhancement phenomenon or the "Bartley effect"
has been extensively investigated by Bartley and others (6, 8, 10, 11, 12,
14, 20, 21, 37, 38, 65). All of this research, however, has utilized
essentially achromatic or white sources. The influences of rate of
intermittency (17, 19), pulse to cycle fraction (PCF) (12, 16, 18, 20), and
stimulus intensity (4, 6) have been studied but not the possible effects
of varying the wavelength composition of the source. Recent work by
Bartley and Nelson (15, 51, 53) has shown the existence of strange color
effects when rate of intermittency, PCF, and intensity are properly
manipulated. Color appearance in an achromatic source has been
demonstrated and strong "washout" effects in certain chromatic stimuli.
The present study has been designed to quantitatively investigate the
effects of intermittent stimulation on brightness, hue, and saturation.

No claim is made that this is more than an introductory investi-
gation into what has turned out to be a potentially fruitful area.
However, there is an attempt to cover a good deal of territory in this
area and also to make the investigation as systematic as possible.
Since a number of variables are influential in their effect on brightness
enhancement, 3. rather complex and extensive investigative procedure
is necessary. The effects of the following variables have been investi-
gated in this present study.

1. Wavelength of the photic stimulation.

2. Pulse-to-cycle fraction of intermittenpy,(PCF). This is defined

as the fraction of the total cycle during which photic energy is

transmitted through the episcoti ster.

1

3..Size of the retinal area stimulated.

4. Rate of intermittency.

5. Intensity of photic source.

6. Effect of dark target surround.

7. Effect of intensity of target surround.

8. Effect of chromaticity of target surround.

All of these variables except the one under investigation during a
particular phase of the study were held constant while this one was
systematically varied. Relative brightness was quantitatively measured
by manipulating a rotary polaroid arrangement over the steady half of
the field to match it to the intermittent half. Saturation was determined
by relating the appearance of the intermittent target to a predetermined
five-step scale of desaturation. Hue was determined by recording the
apparent hue of the intermittent target in relation to the known hue of

the steady: target.

Historical Review

 

A review of the literature in chromatic brightness enhancement
can be extremely short because almost no investigations exist in this
area. No systematic quantitative study with controlled conditions has
been reported. However, a literature review for this study should en—
compass mention of work in a number of very closely related facets of
the problem.

Many mentions have been made of unusual color changes initiated
by proper intermittent stimulation. Some of these concern the creation
of hue in an intermittent achromatic stimulus and others in the "washout"
of a chromatic stimulus. Recently, Bartley and Nelson (15, 51, 53) have

investigated these phenomena but not in a highly quantitative manner.

Many persons in the past have made note of color phenomena in flicker .
Nelson (50) has compiled a bibliography containing seventy-eight
studies where mention is made of a color effect in flicker.

All of these studies could be discussed but mention will be made
of just a few in order to show the progression of thinking and methodology.
Smith (61) in 1881 used a bicycle wheel as an episcotister and the sun as
a source in trying to show that light might be broken up by interruptions
proportional to the wavelength of a particular ray in a composite beam.
In 1918 Baurnann (22) discussed the factors involved in the production of
Fechner-Benham colors. DuBois (26) in 1922 formulated a modulation
theory very similar to Troland's (64) based on research with mollusks.
During this same period Pieron (55, 56, 57, 58) studied Fechner-Benham
phenomena under monochromatic illumination and also concerned himself
with temporal differences between Fechner-Benham colors and after
images. Pauli and Wenze (54) postulated that receptors have natural
temporal periods and response occurs when the frequency of the pulse is
the natural period of the receptor. Marked color changes were obtained
by LeGrand (43) in 1937 during peripheral intermittent stimulation.

In 1940 Segal (60) also put forth a modulation type theory. Gebhard (35)
in 1943 discussed Fechner-Benham colors in relation to color vision
theory including modulation theory. Another explanation of Fechner-
Benham colors on the basis of modulation theory was made by Roelofs
and Zeeman (59) in 1958. An excellent review of the literature in this
area was made by Cohen and Gordon (25).

The first to report a brightness effect of considerable magnitude
in flicker was Briicke (23) in 1884. This work was done with black and
white Disks and Bartley (14) has succinctly differentiated this Briicke
effect from the Bartley effect of brightness enhancement obtained with

episcotisters.

Fry (30, 31, 32, 33, 34) conducted a series of experiments using
chromatic flicker phenomena as a basis for prediction on color vision
theory. He expanded ‘Troland's modulation theory into the Fry modu-
lation theory as the best theory to explain these chromatic flicker
phenomena.

Bartley has formulated his Alternation of Response theory to
explain the brightness enhancement phenomenon and other sensory phe-
nomena. This theory has been presented in a number of publications
(3, 5,10,11,12,14, etc.), but is most comprehensively covered in Bartley's
Freiburg Symposium presentation (14).

Recently a number of animal studies have been done concerning
electrical discharge in the visual system. These studies have the
advantage of allowing direct recording of electrical activity at any level
of the visual system. However, they have the disadvantage of all animal
studies in that there is no assurance that the results are applicable to
human organism behavior. Lennox and co-workers (44, 45, 46, 47, 48, 49)
have conducted studies at the levels of retina, Optic tract, lateral
geniculate, and occipital cortex. They found among other things that
cortical responses were of differing amplitude dependent on spectral
differences in stimulation and spectral differentiation in On and Off fiber
responses. Ingvar (40) in measuring spectral sensitivity in the cerebrum
of cats made the speculation that "in a general way these results support
the hypothesis that differences in conduction velocity may also be a
possible mechanism by which visual centers are informed about color. "

Granit and Wirth (36) in 1953 in an animal study made the observ-
ation that under light adaptation there is a "blue shift" in the sensitivity
curve. This is an interesting finding when compared to the results of
the present investigation.

Halstead and co-workers (37, 38) in working with monkeys found

that they could produce cortical driving at rates of intermittency very

close to Bartley's enhancement rates. A finding that is of particular
interest for this present study is that driving was greatest for wave-
lengths around 500mg.

DeValois and co-workers (27, 28, 29) have isolated layers in the
monkey lateral geniculate which have a selective spectral sensitivity.
Hartridge (39) and Clark (24) have also studied electrical responses in
the lateral geniculate.

Mention must also be made of Landis (41, 42). His annotated

bibliography is a monumental piece of work in the general area of flicker.

Expectations

 

The results do not lend themselves well to statistical evaluation.
Therefore, no attempt has been made to set up null hypotheses and test
them. However, on the basis of previous research I attempted to make
predictions concerning the effects of the individual variables. These
predictions follow as a list of "expectations. "

1. A pulse-to-cycle fraction (PCF) of approximately 1/4 will

give maximal enhancement effect.

2. A rate of intermittency of approximately 10 cycles per

second will give maximal enhancement effect.

3. Greater intensity of stimulation will give maximal enhance-

ment effect.

This next group of "expectations" are on much-less firm footing
because the literature provides no concrete basis for prediction. These
are my "expectations" based solely upon what seems reasonable in the
light of existing knowledge and were made before I began to collect data.

4. Varying wavelength of source, as long as luminosity is kept

constant, should have no significant effect on brightness en-

hancement. Concurrent research by Bartley and Nelson has

shown this to be an erroneous "expectation" but when I started
my research it seemed like a reasonable guess.

Dark target surround should give greater brightness enhance-
ment effect than illuminated surround.

As intensity of surround is increased, enhancement effect
should be decreased.

Chromaticity of surround, as long as luminosity is kept con-
stant, should not give any significant effects with respect to
brightness enhancement.

Larger target size should give increased brightness enhance-
ment effect up to a certain point. This would be expected on
the basis of activating a larger grouping of visual channels.
However, this might be negated by Bartley's finding (7, 14, 21)
that in a bipartite viewing field greater target size causes
greater stray light during the supposed dark phase of inter-

mittency.

METHOD

Apparatus

 

The apparatus utilized is a modified Fry type prism monochrom-
ator with'episcotister and field surround attachments. This is
schematically shown in Figure 1A. A ribbon filament bulb source (1)
is powered through a constant voltage battery eliminator. A collimating
lens (2) causes two parallel beams of light to pass through the double
slit (3). The width of these slits is 1 mm. The bottom beam of this
pair can be controlled in intensity by a rotary polaroid arrangement (5).
This is composed of a fixed polaroid over the bottom slit with a rotating
polaroid in front of this. This rotating polaroid is connected to a shaft
with a knob and calibration dial which allows recording of the percentage
transmission through the polaroid system. A neutral density 0. 2 filter
(4) is placed over the upper slit of pair (3) to reduce intensity of this
beam. Thus, when the brightness of the upper and lower beams are
equated by turning polaroid (5) the scale reading is kept below 100 per
cent.

An episcotister (6) chops only the upper beam. Thus, the upper
beam is intermittent and the lower beam steady. Rate of intermittency
and PCF are varied respectively by changing the disk at (6) and the rate
of rotation of the disk. A direct gear drive through a synchronous motor
is used to drive the episcotister.

A lens (7) focuses the two parallel beams onto the face of the
prism (9). A filter holder (8) allows for introduction of filters to vary
the two target beams without changing the surround illumination. The

spectrum formed by the prism(9) is reflected off the first surface

mirror (10) and through the biprism lens (11). This lens separates the

two original beams from (5) and collimates them. The field stop (12)

is a first surface mirror. The bipartite aperature in this mirror can

be varied in size to provide different sized targets. The mirrored

surface allows for surround illumination to be introduced into the system.

— A filter holder (13) allows the entire target and surround field to be

varied together. A lens (14) focuses the spectrum onto the slit (15)

which is 1 mm. in width. The observer places his eye just behind slit (15).

The prism (9) is mounted on a spring loaded stage which can be
accurately rotated to provide any desired narrow portion of the spectrum
as the target. The settings of the stage for providing various wavelengths“-
were calibrated with a spectrometer.

The surround illumination is provided by a ribbon filament source
(18) and collimated by lens (17). Filter holder (16) allows the surround
to be varied in chromaticity and intensity. The parallel beam of the
surround illumination is reflected off first surface mirror (12) while the
bipartite target field passes through the aperature in mirror (12).

The field seen by the subject is illustrated in Figure 1B. The
bipartite target is composed of a steady top half (1) and an intermittent
lower half (2). The fixation point (4) is at the center of a narrow black
band (3) which separates the halves of the field. The surround (5) can
be dark or illuminated to any desired intensity and chromaticity.

In summary the variables used in this study are controlled in the
following manner. Reference should be made to Figure 1A.

1. Wavelengthis controlled by turning a calibrated knob which
rotates the mounting stage for prism (9). This causes a dif-
ferent portion of the spectrum formed by prism (9) to pass
through the eyepiece slit (15). The width of the wavelength
band varies in increasing magnitude from 5. 3mg at 420mg

to 33. 5m|i at 700mg.

2. The size of the target is controlled by changing the size of the
aperature in mirror (12).

3. Pulse-to-cycle fraction is controlled by the episcotister
disk (6).

4. Rate of intermittency is controlled by the gear ratio in the
direct drive between a synchronous motor and the shaft of
disk (6).

5. Intensity of source is controlled by neutral density filters
at (8).

6. Surround is kept dark by eliminating power to bulb (18).

7. Surround intensity is controlled by a variac on the power
supply to bulb (18) or by neutral density filters at (16).

8. Chromaticity of the surround is controlled by Corning glass

filters at (16) .

Pr oc edur e

 

The first portion of the investigation was done with a dark surround,
maximum target intensity, and a foveal target size (1. 250). Thus, the
variables manipulated were wavelength of source, PCF, and rate of
intermittency.

It was desirable in this part of the study to maintain equal lumi-
nosity for all wavelengths, have the highest possible stimulus intensity,
and still utilize a wide portion of the visible spectrum. A necessary
compromise of these last two requirements was made by utilizing wave-
lengths from 480m}; to 680mg. These two were a match in luminosity
while the necessary neutral density filters were introduced into the
system for all wavelengths between 480m}; and 680m}; to maintain equal
luminosity. The amount of neutral density filter necessary was de-

determined by matching a blue-green surround illumination in brightness

10

to the 480m}; target and then stepwise matching the higher wavelength
targets to the surround. The chromaticity of the surround was varied
several times to reduce the problems of heterochromatic photometry
but each time the surround was changed it was matched in brightness to
a target which'matched the previous surround. This procedure was
repeated starting at 680ml; and working downward. Thus, the neutral
density filters necessary to maintain equal luminosity were determined.
The results of this procedure correlated well with an additional pro-
cedure done where the neutral density filter necessary to reach CFF at
rate 36 cps for each waveband was determined. Thus, agreement was
obtained by two completely different methods that the desired condition
of equal luminosity was a reality in this experiment. The equal lumi-
nosity condition for the 460m“ waveband was obtained by using 2 mm.
rather than 1mm. slits in front of the ribbon filament source and making
the necessary prism dial setting corrections.

An example may illustrate best how the data were collected.
Suppose that PCF was set at 1/4, rate of intermittency at 9. 8 cycles
per second, and wavelength at 480mu. The observer viewed the bipartite
target for one minute prior to making any evaluations. This one minute
adaptation period each time the subject commenced viewing was found to
be sufficient. The observer then noted the state of desaturation of the
intermittent target in relation to the steady target. A response was made
on the basis of a five-step scale of "no washout, " "slight washout,- 'I
"moderate washout, " "extreme washout, " and "total washout. " Then the
observer made an evaluation of the hue of the intermittent target in
relation to the known hue of the steady target. This was recorded as to
whether the intermittent target appeared to be the same hue as the steady,
the hue of a longer wavelength, or the hue of a shorter wavelength.

Then the observer turned the dial controlling the brightness of the

steady target until the steady and intermittent halves of the bipartite

11

field appeared the same brightness. Three ascending and three descend-
ing readings were taken. Then the waveband was changed to 500mg,
proper neutral density filters put into the system to maintain equal
luminosity, and the same procedure repeated. This was done for each
waveband step up through 680mg. Then the rate of intermittency was
changed and the whole procedure repeated. When all the desired rates
had been tested then PCF was varied and the entire procedure of twelve
waveband steps at each rate repeated. Data were also collected for steps
going downward from 680ml]. to 480m}; but this was found to give no dif-
ferent results than going in steps from 480 mp upward to 680mu.

After extensive investigation had been made utilizing the variables
of wavelength, PCF, and rate, the other variables were introduced one
at a time and the whole procedure repeated. Thus, each time a new
variable was introduced, an increasing number of stimulus situations had
to be tested.

An additional procedure was done utilizing 13 mm. slits rather
than 1 mm. slits in front of the ribbon filament source. This allowed a
relatively wide spectral band rather than the previously used narrow
bands. Three bands were used in this portion of the investigation.

They were 437-490 mp, 472-576mu, and 528-688mu.

RESULTS

The author's photOpic luminosity curve has been previously de-
termined and is available in the literature (1). It closely approximates
the standard I.C.I. photopic luminosity curve. The transmission
curve of a similar instrument is also available in the literature (1).

The instrument transmits better in the longer wavelengths than in the
shorter wavelengths. Thus, as would be expected, the peak of the
luminance distribution curve derived from CFF data will be shifted to
a higher wavelength (575-580mp).

Figure 2 illustrates the effect on brightness at five points of the
spectrum when PCF is varied. It is readily seen that only a PCF of
1/8 or 1/4 gives brightness substantially above the Talbot level (62)
and even here only for the wavelength of 500mg.

A statement should be made here concerning the ordinate values.
The values 0 through 70 are percentage transmission through the rotary
polaroid arrangement over the steady half of the bipartite field. Each
point recorded on the graph is a mean value of six experimental bright-
ness matches (three ascending and three descending). When the two
halves of the bipartite field are both steady (episcotister not running)
they are matched in brightness for all wavelengths at a rotary polaroid
setting of 62. 5 per cent transmission. Thus, any brightness value of
over 62. 5 is an enhanced value and the Talbot level value is found by
multiplying 62. 5 by the PCF. Luminance of the test stimulus throughout
the experiment is maintained at 50 foot lamberts except where specifically
noted otherwise.

The question has arisen as to the best method of presenting the data

of this investigation. In most previous work luminance of the matching

12

13

stimulus has been-the ordinate and rate of intermittency the abscissa.
This investigation, however, is greatly concerned with the brightness
effect pattern as it relates to wavelength so that the author has decided
to present much of the data with wavelength rather than rate of inter-
mittency as the abscissa. However, in Figure 3 data are presented in
the usual way to facilitate comparison with results of previous studies.
Here rate of intermittency is the abscissa and a separate curve is
plotted for each wavelength. vThe substantially increased effectivity of
wavelength 500mg over all longer wavelengths is quickly seen. These
data are for a PCF of 1/4 and target size of l. 250.

Each figure from Figures 4 to 8 represents data for a given com-
bination of stimulus size and PCF. The various curves on each graph
represent data for specific rates of intermittency. The lower portion
"A" of each figure shows for each waveband the luminance of the match-
ing stimulus plotted as a function of wavelength. The upper portion. "B"
of each figure shows the effect of wavelength on saturation or "washout. "
A multiple point symbol is used when points from two or more lines
fall at the same place on the "washout" graph. In each case all of the
coincident points are drawn directly above the multiple point symbol.

Figure 4A shows there is not much increase in brightness over the
Talbot level with a PCF of 1/16 although there is a hint of greater bright-
ness for wavebands around 500m}; andlower rates of intermittency.
Figure 4B shows very little "washout" at PCF 1/16 but it is greatest at
500m}; and 6. 5 cps rate.

Figure 5 presents data for PCF 1/4, target 1. 25°, and seven dif-
ferent rates of intermittency between 6. 5 and 20. 0 cycles per second.
Figure 5A shows the increased brightness effect of PCF 1/4, the great
increase in apparent brightness of wavebands 500mg and 510 mp. over
longer or shorter wavebands, and the increased brightness effect for

rates of 10 cps or less over the faster rates.

14

Figure 5B follows by showing a strong "washout" effect at 500mg
which is greater for rates of 10 cps or less. Because a marked in-
crease of the enhancement effect also occurs at 500mg, it might be
supposed that there is some interdependence between "washout" and
brightness enhancement. However, the occurrence of the second "wash-
out" hump at 600-620mp as shown in Figure 5B illustrates that the
brightness enhancement tendency and the "washout" phenomenon do not
possess a simple invariant relationship. This increase in "washout"
in the red-orange and red regions has also been described by Bartley
and Nelson (51).

Figure 6 presents the same kind of data as Figure 5 but for a
target size of 40. This is no longer as strictly a foveal target as is
the l. 250. . Essentially all of the same tendencies are present here as
with the smaller target but of a decreased magnitude. Of significant dif-
ference is the increased brightness effectiveness of the 12.0 cps rate
and the decreased effectiveness of the 7. 8 cps rate. In "washout" the
rates of 10 cps to 13 cps are most effective rather than the rates of
10 cps and less for the smaller target.

.. Figure 7 shows the decreased brightness and "washout" effective-
ness when a PCF of 1/2 is used. Also, the greatest brightness occurs
at 480mg rather than 500mg.

Figure 8 shows the extreme ineffectiveness of PCF 3/4 in elliciting
either the brightness enhancement or the "washout" phenomena.

Figures 9 through 16 display the effects for five wavebands be-
tween 500mp..and 680mg when luminance level, PCF, and target size
are systematically varied. In all cases, the rate of intermittency was
maintained at 9. 8 cps. Results for four levels of intensity are shown in
eachgraph. The highest level (100%) is the level used in all previous
figures (50 foot lamberts). . For the other levels neutral density filters

of 0. 5, 1.0, and 2. 0 were added. . In terms of the filter free level, the

15

filters provide levels of 32%, 10%, and 1% or 16, 5, and . 5 foot lamberts.
By observation it appears that the 2. 0 N.D. filter reduces luminosity to
not far above the photOpic threshold.

Figure 9A shows that for the 100% and 32% levels of luminance of
the intermittent stimulus the luminance of the matching stimulus is
above the Talbot level with PCF 1/8 but that the 100% level at 500mg
gives the greatest apparent brightness. The 32% level gives somewhat
less apparent brightness at 500m}... and the 10% and 1% levels give essentially
no difference in apparent brightness for the five wavelength bands and
essentially no difference from the Talbot level. Figure 9B shows that
r "washout" is strong only at 500m}; at the 100% level.

.Figures 10 and 11 show the results for PCF 1/4 and target sizes of
1.250 and 40. Here it can be noted that the 1. 250 target gives more
apparent brightness than the 40 target for the 100% level 500mg waveband.
This is the same situation that was noted in comparing Figures 5A and
6A. As in the previous figures, the 500ml; waveband is much more
effective in producing apparent brightness and this tendency decreases as
the intensity level of the stimulus is decreased. Here the 32%, 10%,
and 1% levels give essentially no brightness difference for the five wave-
bands.

Figures 10B and 11B show the strong "washout" at 500m“ full
intensity and also the existence of the previously mentioned second
"washout" hump at 620mg. "Washout" decreases as intensity decreases.

Figure 12 shows the loss of the apparent brightness hump at 500mg
whenPCF is 1/2 but that the full intensity "washout" remains strong at
500mp. This again illustrates that brightness enhancement tendency
and "washout" do not always go hand in hand.

Figures 13 through 16 utilize the same experimental conditions as
Figures 10 and 11 but for two different observers. This is probably an

appropriate place to interject a statement on what will certainly be

16

considered a weak point in this investigation; the small number of
observers involved in the data collections Only three observers were
used and two of these in only limited phases of the study. Thus, the
great majority of all the data discussed were obtained from one ob-
server;"the£ authorr It is difficult to find or produce trained observers
who can participate in this kind of experiment.

For this study, the word "trained observer" should be italicised
and in double quotes. In the majority of the experimental conditions in
this investigation apparent brightness, hue, and saturation are all vary-
ing simultaneously. The problem for the observer is to view a bipartite
field where the flickering half is different in brightness, hue, and satur-
ation from the steady half. Then the observer must separately make a
qualitative hue difference judgment, a semi-quantitative "washout" judg-
ment on a five-point scale, and a quantitative brightness match judgment
while never allowing the other two attributes of color to influence his
judgment on the one being measured. It requires a fantastic amount of
practice to make valid repeatable judgments.

The author collected‘data by countless observations during a period
of over one year. Previously, this author had spent a year in color
vision research under the direct supervision of Dr. Glenn A. Fry where
the research on lines of constant hue (2) required a similar type of
observational task of making judgments on one variable in a complex
brightness, saturation, hue stimulus situation. Even with this total of
two years of intensive training in making these judgments, all of the data
presented in this study were collected during the last four months of this
period. The earlier data showed exactly the same trends but with much
greater variance in individual readings. The author is now able to make
a set of three ascending and three descending brightness matches and

rarely have a spread of over 5 per cent in the six readings. . Thus,

l7

differences between wavelength bands illustrated by the graphs certainly
cannot be attributed to chance variance in individual judgments.

The two other subjects used, Dr. Thomas Nelson and Dr. Charles
Bourassa, have had extensive experience as observers in visual per-
ceptual experiments but not in this particular type of experiment. I As
expected, the variance in their individual judgments is much greater but
still only in the area of 10 to 15 per cent spread for groups of six read-
ings.

In general, the data collected from these two subjects confirm very
well the data collected with the author as observer. Some of the data
are depicted in Figures 13 through 16. The same hump on the graph at
500mg. for both brightness and "washout" is apparent. Also, the second
"washout" hump at 620m}; can be noted and the same effects of decreasing
intensity. The greatest difference is in the effect of target size. Bourassa
concurred with the author in finding the 1. 250 target more effective in pro-
ducing apparent brightness but Nelson found the 40 target more effective.

In Figures 17 and 18 a new variable has been added with very
interesting results. Here the surround field is not dark as it has been in
all previous figures. The experimental condition where the greatest
differential wavelength brightness effect has been obtained is utilized to
study the effect of adding surround luminance. This is the condition of
PCF 1/4, target 1.250, and rate 9.8 cps.

Figure 17 depicts the results for a surround luminance witha
dominant wavelength of 540mg. The surround is quite desaturated since
it is obtained with Corning filters rather than with a monochromator.
Four curves are shown; the first a repetition of the dark surround con-
dition; the second with the surround illuminated but with the luminance
definitely less than that of the target; the third with the luminance of
the surround equal to that of the target when the target is not intermittent;
and the fourth with surround luminance definitely greater than that of the

target.

18

Several interesting things should be noted on Figure 17A. Increas-
ing the surround luminance decreases the brightness hump at 500mg
and increases the apparent brightness of the longer wavelength bands. Of
extreme interest is the appearance of a large apparent brightness hump
at the dominant wavelength of the surround when the target and surround
are of equal luminosity. This is of almost an enhancement level and
occurs at a wavelength that with dark surround consistently gives low
apparent brightness.

Figure 17B shows that the "washout" hump at 500mg is decreased
as surround illumination increases and that the second "washout" hump
is shifted from 620m“ downward to 580-600mu.

Figure 18A illustrates the same conditions as Figure 17A but with
a surround of dominant wavelenth 600mg. Very similar trends are seen
in the results except that the interesting second hump in apparent bright-
ness shifts to the surround dominant wavelength of 600mg. In Figure 18B
the second "washout" hump occurs at the surround dominant wavelength
of 600mp. rather than at 620ml). as in the dark surround condition or 580-
600mp as in the 540m}; surround condition.

Figure 19 is presented to illustrate that the results can be repli-
cated. The condition which yielded the greatest differential brightness
effect is graphed as it gives the greatest variance in replication. This
is the PFC 1/4, rate 9.8 cps, and target 1. 250. All other conditions
can be replicated much more closely. These four curves are from data
collected at different times for different sequence conditions. The four
curves in Figure 19A can be noted as the apprOpriate curves appearing
in Figures 5A, 18A, 17A, and 10A. The curves in Figure 19B can be
noted as appearing in Figures 5B, 18B, 17B, and 10B.

Figure 20 shows the results when a wider spectral band is utilized
as a target. Three wavelength bands of 437-490mp, 472-576mp., and

528-688mp were used. The graphs are plotted as luminance against

19

rate of rotation. It can be readily noted from Figure 20A that only the
472-576mp waveland gives brightness even approaching an enhancement
level and then only for rates of 10 cps or less. These findings on
effectivity of rate of rotation agree very well with the preyious figures
which utilized narrow wavebands throughout the visible spectrum (com-
pare Figures 3 and 20A). By taking into account the overlap of the three
bands used the enhancement tendency appears to be caused by wave-
lengths somewhere between 490mp and 528mg. This agrees extremely
well with the previous figures which show the 500mg and 510mg. wave-
bands as being the primary contributors to brightness enhancement
tendencies (compare Figures 3, 5A, and 20A).

Figure 20B shows the correspondence with previous figures in
comparing "washout" with rate and wavelength. A comparison of Figures
5B and 20B shows that "washout" is moderate to total for rates of 15 cps
or less if the waveband contains the 500mg wavelength. Rates over 15
cps or wavebands not including the 500ml; wavelength give very little
"washout. "

Data were collected on all the three variables of brightness,
"washout, " and hue. Up to this point only the results of the quantitative
brightness judgments and the semi-quantitative five-point scale "washout"
judgments have been presented. Figures 21 and 22 present the quali-
tative hue shift judgments for all of the experimental conditions. Figure
21 shows experimental conditions where thirteen wavebands are utilized
and Figure 22 for those conditions where five wavebands were used.

A number of interesting and extremely consistent things can be
noted from these figures. For wavebands 500mg, 510mg, 520mg, and
540m}; a hue shift occurs for a great many of the experimental conditions
and it always shifts the apparent hue of the intermittent target to a longer
wavelength. For wavebands 600mg, 620mg, 640mg, and 680m“ there is
usually a hue shift and it always causes the apparent hue of the intermittent

target to shift towards a shorter wavelength.

20

Only rarely is there a hue shift for wavebands 560ml; and 580mg
but when it occurs it is always upward at 560m}; and downward at 580mg.
This places the invariant hue point at about 750mg. A very different
situation appears for wavebands 480mg and 460mg. When they do shift
in hue they become purplish and shift downward away from the 570mg
point. Also, it can be noted that total "washout" never occurs except
at the 500mg waveband.

In all of the results it is noted that while there are great differences
in brightness obtained there is very little true brightness enhancement
where brightness of the intermittent stimulus is actually higher than the
steady state stimulus. Most of the values obtained are of an inter-
mediate brightness where they fall between the enhancement and Talbot
levels. Bartley and others have, on occasion with achromatic stimu-
lation, reported much greater magnitudes of brightness enhancement.
Some explanation needs to be made as to why large degrees of enhance-
ment are not found in this present investigation. While narrow waveband
stimulation is certainly a very different situation from achromatic stimu-
lation one would expect more enhancement. Part of this lack of enhance-
ment may be due to a problem with the apparatus. Bartley's findings
indicate this to be a definite possibility. Bartley (14) notes that "In using
intermittent stimulation in the comparison target and steady illumi- I
nation in the standard target, the retinal area supposedly at rest during
the 'dark' periods of the intermittency cycle will be stimulated by the
stray illumination of the steady target. " Bartley (21) further found that
"By putting more light into the intervals between intermittent pulses, we
pre-empted the usilization of a portion of the total number of otherwise
available parallel circuits by the light pulses themselves. "

The biprism arrangement, unfortunately, does not yield a com-
pletely on-off stimulus situation. There is some light leakage between

the halves of the field in addition to the expected stray light effects.

21

This means that the intermittent half of the target does not go completely
dark. If a total cessation of stimulation rather than an almost total
cessation had been possible greater enhancement levels might have been

obtained.

Summary of Results

 

This section will attempt to assess the "expectations" originally
made in the light of the results of this investigation. We will take these
expectations in the same order in which they were made.

1. A PCF of approximately 1/4 will give maximal enhancement
effect. This has been definitely shown. Ability to cause brightness en-
hancement falls off quite rapidly as PCF is either increased or decreased
from 1/4. Figure 2 illustrates this.

2. A rate of intermittency of approximately 10 cps will give maxi-
mal enhancement effect. The data show this to be substantially true
although it does not seem to be as critical a variable as some of the
others investigated. Rates much over 10 cps and especially over 12 cps
decrease apparent brightness tendency but rates of 6. 5 cps and 7. 8 cps
appear almost as effective as 9.8 CpS. Thus a range of rates from 12.0
cps down to 6. 5 cps or possibly lower seem to give an increased apparent
brightness tendency.

3. Greater intensity of stimulus will give maximal enhancement
effects. This has essentially been substantiated by the data.

4. Varying wavelength of source as long as luminosity is kept
constant should have no significant effect on brightness enhancement.
The results show this "expectation" to be completely wrong thus forming
one of the interesting and, I hope, significant contributions made by this
study. Apparent brightness tendency has been shown to be extremely

dependent upon wavelength of stimulation. A narrow and very specific

22

range of.' wavelengths around 500mg have been shown to be extremely
subject to increased apparent brightness when intensity and temporal
variables are properly manipulated.

5. Dark target surround should give greater brightness enhance-
ment effect than illuminated surround. This is found to be true.

6. As intensity of surround is increased brightness enhancement
effect should be decreased. The results show this to be essentially
the case.

A ,-_7. Chromaticity of surround, as long as luminosity is kept con-
stant, should not give any significant effects on brightness enhancement.
Here again the "expectation" is incorrect and, thus, provides another
very interesting result. The data show the existence of a great increase
in apparent brightness of the intermittent target when the surround and
target are of the same wavelength and intensity.

8. Larger target size should give greater brightness enhancement
effect. This is supported by the results of one subject but opposed by

the results of two subjects.

CONCLUSIONS

I know of no better introduction to this section on conclusions
than a quotation from Troland (64) written in 1921 in the first concrete
formulation of a modulation type color vision theory. Though written
over forty years ago, this quotation is just as poignant for color
vision theory today as it was then. "The actual task which we have
before us in constructing a rationale of the data of physiological optics
is a tremendous one, making more rigorous demands upon the intellect,
I fear, than the formulation of certain far more cosmic theories, and
one upon which our present, neat little academic explanations form
only a burlesque. "

It appears that this investigation may have revealed a new ramifi-
cation of a vi sual function or possibly even a complete new function
concerning brightness transmission. A brightness phenomenon is
indicated which for intermittent stimulation is extremely dependent on
wavelength of stimulus, quite highly dependent on PCF, and moderately
dependent on rate and intensity of stimulation. Bartley (6) prophesied
this in part when he stated "we should expect monochromatic light to ,
be somewhat more effective than white, and expect wavelengths toward
the blue end of the spectrum to be most effective. "

One of the most striking features about this function is the highly
restricted wavelength region in which it Operates with an enhancement
tendency. When one looks at the curve in Figure 5A one must note that
its peak wavelength falls near to that of the standard scotopic luminosity
curve. But this is obtained with a 1. 25° target which is strictly foveal.
Thus, we have cones with an enhanced response at 510m}; when the proper
form of the intermittent stimulation is used. Can it be that these specific
temporal-spatial stimulus patterns have isolated a cone brightness

23

24

response? Is this hump due to lack of inhibition or favorable sum-
mation for these specific wavelengths?

This would agree well with Bartley's Alternation of Response
theory. That is, brightness is dependent on the total number of separate
channels in the visual system which are being activated at a given
instant. The number of channels can be increased and thus brightness
increased either by increasing the stimulus intensity or by proper
timing of the stimulus input. The interesting new feature is that photopic
brightness function is dependent not just on the intensity level and the
timing of the stimulus but also is highly dependent on the spectral
composition of the stimulation. This is a new piece in the puzzle which
ultimately must find a place to fit. There are those intensity-timing-
wavelength combinations which produce an optimum stimulus pattern
to the visual system and thus give brightness enhancement.

When the area of stimulation is increased so that a strictly foveal
target is not used (40 rather than 1. 250), two of the three observers
found that the 500mg brightness hump is markedly decreased. See
Figures 10A, 13A, and 15A. A dangerous simplification of these results
might be, that under proper patterns of stimulation, cones will give a
brightness response curve much like that of rods but that when target
size is increased to include rods the rod response inhibits the ability of
cones to act like rods.

An important aspect of these data concerns the effects of surround. ‘
The finding that the surround illumination decreases the brightness hump
at 500mg is not too surprising. See Figures 17A and 18A. Steady
stimulation of adjacent channels surrounding a small intermittent area
(1. 25°) might well be expected to disrupt the intensity-timing-wavelength
interrelationship necessary for enhancing brightness. Also, the de-
creased effectiveness of increasing area of retinal stimulation as shown

in Figures 10A, 11A, 13A, 14A, 15A, 16A by two of the three observers

25

might come into play here. With the same reasoning one should expect
the depressive situations to also be disrupted. See 600m}; on Figure 18A.
The striking feature is the emergence of the brightness hump at the
dominant wavelength of the surround when surround and target are equal
in luminosity. This means that a surround illumination where intensity
and wavelength are matched to the pulse of the intermittent target acts

as an enhancer of brightness rather than a depressor. Furthermore,
intensity alone in the surround illumination acts to decrease both the
functions of brightness enhancement and brightness depression but if the
proper surround intensity-wavelength interrelationship is used then a
condition of strong brightness enhancement tendency is created at a previ-
ously depressed waveband.

One would hope to be able to relate the results of this investigation
to color vision theory. In this respect, the most striking feature of
the result is the strong effect of temporal stimulus variables. This would
tend to support some sort of modulation type color vision theory. I do
not believe that the known facts concerning photochemistry can alone be
used to explain these results.

The results of this investigation would tend to support a modulation
color vision theory where brightness is dependent on total number of
visual channels activated and color is dependent on some subtle modu-
lation of impulse transmission among a group of adjacent channels or a
superimposed modulation on the impulse in a single channel. Here one
could speculate that maximum brightness would be achieved when all
channels within the area of stimulation are activated maximally at the
same instant. This, however, would leave no chance for a temporal or
amplitude variation among the channels or a modulation effect within a
channel impulse that could transmit a modulated color response. The
results of this investigation support this possibility. As brightness in-

creases substantially (at wavelengths around 500mg) there is also a strong

26

"washout" or desaturation effect. See Figures 5A and 5B. This could
mean that as brightness increases more channels are being maximally
activated at a given instant and at the same time the possibilities of
interchannel modulation are decreased so that hue progressively washes
out. At peak brightness the results show total "washout. " This might
be approaching the condition where all channels within the area of stimu-
lation are being maximally activated simultaneously with no interchannel
modulation possibilities remaining. Thus, a very bright achromatic
target results.

This, of course, creates a situation where brightness and color
sensation are highly related but not totally interdependent. . Brightness
sensation would of necessity be present when color sensation occurs
because any modulated interaction of channel activation would necessitate
some channels being activated at a given instant. However, brightness
sensation would not necessitate color sensation for two possible reasons.
The first which could hold true for cones and the fovea is that maximum
channel activation leaves no room for modulation and an achromatic
sensation results. The second possibility is if there were not a fine
enough mosaic of visual channels in a given retinal area to allow for
proper interchannel timing and/or amplitude of activation that would give
the necessary modulation. This might be the reason that rods, with
.many being connected to a single gangliontcell, give essentially no color
response.

Something should be said concerning the possible site of a modu-
lating mechanism. Here the author is apt to wander into that fascinating
world of "pure speculation. " If a modulating mode of stimulus trans-
mission is the actual means of color vision it must exist in some form at
all levels of the visual system. Too many investigators in their fervor
for a particular concept have seemingly forgotten this. You cannot have

unmodulated impulse trains from the retina to the cortex and then

27

suddenly modulate them in the cortex. Equally impossible would be a
situation where the retinal elements produce a modulation stimulus
pattern and the cortex has no mechanism capable of utilizing a modu-
lated input. The third impossibility is that one could have a modulation
mechanism at the retinal and cortical levels but no means of trans-
mitting a modulated message along the optic nerve. This means that
if stimulus modulation on a temporal and/or amplitude basis is the
means of color vision then the visual system must have the mechanism
capable of modulating, transmitting, and decoding this message at
every step of the visual pathway.

In this paper, speculation has primarily been concerned ,with how
a modulated input might be transmitted from the retina to the lateral
geniculate and/or occipital cortex. This has utilized a combination of
the Bartley Alternation of Response theory and a modulation type color
vision theory. Recent work by DeValois and others (24, 27, 28, 29, 39) on
the lateral geniculate in monkeys has shown layers of cells which can
be differentiated on a spectral input basis. If the gap can be lumped
and the inference made that a human lateral geniculate operates in the
same manner then the lateral geniculate could be thought of as a way
station in the visual system. It could be that the modulation of the optic
nerve transmission is decoded here and then impulses sent to other
brain centers for various visual reflex and associative functions while
a recoded message is sent to the visual centers in the occipital region.
Or it might be that the lateral geniculate in man only skims off a portion
of the input for transmission to other centers while the major portion of
the modulated impulse train continues on to the occipital cortex where
it is decoded by a cortical mechanism which yields the sensation of
color. . In either case both the input and the output of the lateral geni-

culate must be coded.

28

A large remaining speculation concerns how a modulated message
might be initiated at the retinal level. That this coding of some sort
takes place is essential for any color vision theory. This author does
not feel that a photochemical basis or a three-component basis best
describe the results of this investigation. But if modulation occurs how
is it initiated? One could speculate variance in cone diameter or nerve
fiber diameter as being selectively effective for different wavelengths.
Cones of relatively small differences in diameter could be randomly
distributed and thus different wavelengths of stimulation could set up dif-
ferent interchannel or intrachannel temporal and/or amplitude modulations.

The author would consider a better possibility to be the effects
created by the cells in the bipolar retinal layer. In this layer are cells
such as the amacrine and horizontal cells whose function has never
been adequately explained. They must have a definite function as nature
does not operate otherwise. Is it possible that these cells form the basis
for a modulated visual message? Can they provide the facilitation,
inhibition, and modulation between channels in the visual system at the
bipolar cell level by being selectively responsive to intensity and wave-
length photic stimulation?

The foregoing is not intended to intimate a formalized color
vision theory in any way. The author possesses neither the information
nor the insight necessary for such a cosmic undertaking at this time.
The author merely feels that this investigative procedure opens fruitful
new avenues for color vision research. The speculations are presented
only as an attempt to stimulate thought and comment both pro and con.
Only by diligent, continuing, and exhaustive efforts can the complexity

of color vision finally be unraveled.

FUTURE PLANS

As was stated at the outset I hope that this is not a finished tOpic
but merely a beginning. The results of this investigation seem to
indicate that future work in this area is warranted and could prove
extremely fruitful. Investigation can well be expanded. and more
exhaustive on all of the variables already dealt with. Surround illumi-
nation in particular has only been touched. Also, target size has not
been systematically varied.

In addition, several new variables might well be added. For
example single pulses or‘limited trains of pulses can be utilized as
stimuli. Stimulus input can be modulated in various ways other than
the on-off situation of the sectored disk. For instance a rotating
polaroid disk would give a sine wave input.

Many of these present studies might well be replicated on indi-
viduals with various types of color vision deficiencies. These are just
a few of the possible avenues of future investigation in this area.

The author can only hope that he has been fortunate enough to stumble
onto a means by which significant additions can be made to man's

knowledge concerning his sense of vision.

29

10.

11.

12.

REFERENCES

Bailey R. W. The effect of chromatic adaptation on color-blind
observers. -Am. J. Optom. and Arch. Am. Acad. Optom.,
1960, 31, 499-516.

 

Ball, R. J. , and Fry, G. A. Lines of constant hue in the blue corner
of the color mixture diagram. Am. J. Optom. and Arch. Am.
Acad. Optom., 1959, a, 200-212.

 

 

Bartley, S. H. The features of the optic-nerve discharge underlying
recurrent vision. J. Exper. PchhoL, 1942, 32: 125-135.

 

Bartley, S. H. Intermittent photic stimulation at marginal intensity
levels. J. Psychol., 1951, 32, 217-223.

 

Bartley, S. H. The psychophysicology of vision. In S. 5. Stevens
(Ed.), Handbook of experimental psychology. New York: Wiley,
1951.

 

Bartley,'S. H. Brightness enhancement in relation to target
intensity. J. Psychol., 1951, _3_2_, 57-62.

 

Bartley, S. H. Brightness comparisons when one eye is stimulated
intermittently and the other eye steadily. J. Psychol. , 1952,
_31, 165-1670

 

Bartley, S. H. Light adaptation and brightness enhancement.
Percept. 81 Mot. Skills, 1957, _7_, 85-92.

 

Bartley, - S. H. .Some factors influencing critical flicker frequency.
J. Psychol., 1958, 46, 107-115.

 

Bartley, S. H. Principles ogperCfltion. New York: Harper, 1958.

 

Bartley, S. H. Some facts and concepts regarding the neurophysiology
of the optic pathway. Arch. Ophthal., 1958, 62, 775-791.

 

Bartley,-S. H. Central mechanisms of vision. In Field, J. Magoun,
H. W., and Hall, V. G., (Eds), Handbook of Physiology, Sec. 1,
Neurophysiology, Vol. 1. Washington, D.C.: Amer.-Physiol.
Soc., 1959.

 

30

13O

14.

15.

l6.

17.

18.

19.

20.

21.

22.

23.

31

Bartley, S. H. Some relations between pulse-to-cycle fraction and
critical flicker frequency. Percept. and Mot. Skills, 1960, g,
3-8.

 

Bartley, S. H. A clarification of some of the procedures and con-
cepts involved in dealing with the optic pathway. In Jung, R. ,
and Kornhiiber, H., (Eds.), The visual system: Neurophysiology
and psych0physics symposium, Freiburg. Heidelberg: Springer-
Verlag, 1961.

Bartley, S. H. , and Nelson, T. M. Certain chromatic and brightness
changes associated with rate of intermittency of photic stimulation.
J. Psychol. , 1960, 52, 323-332.

 

Bartley, S. H., and Nelson, T. M. Equivalence of various pulse-to-
cycle fractions in producing critical flicker frequency. J. Opt.
Soc. Am., 1960, 50, 241-244.

Bartley, S. H., and Nelson, T. M. A comparison of three rates of
pulse onset and decline in producing critical flicker frequency.
J. Phychol., 1960, 49, 185-194.

 

Bartley,-S. H., and Nelson, T. M. A further study of pulse-to-cycle
fraction and critical flicker frequency. A decisive theoretical
test. J. Opt. Soc. Am., 1961, :1, 41-45.

 

Bartley, S. H., Nelson, T. M., and Ranney, J. E. The sensory
parallel of the reorganization period in the cortical response in
intermittent retinal stimulation. J. Psxcholn 1961, 52, 127-147.

 

Bartley,.S. H., Paczewitz, G., and Valsi, E. Brightness enhance-
ment and the stimulus cycle. J. Psychol., 1957, _4_3_, 187-192.

 

Bartley,.S. H. , Wilkinson, F. Brightness enhancement when
entoptic stray light is held constant. J. Psychol., 1952, i3,
301-305.

 

Baumenn, C. Bertr'age zur physiologie des sehens: VII. Subjektive
farbenersheinungen. Subjektives und objektives empfinden.
Pfliig. Arch. ges. Physiol., 1918, 171, 496-499.

 

Briicke, E. Uber die nutzeffect intermitterender netzhautreizungen.
Sitzungsberichte der Mathematisch-Naturwissenschaftlichen Classe
der Kaiserlichen Akademie der Wissenschaften, 1848, _42, 128-153.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

32

Clark, W. E. LeG. Brit. J. OJDhthal., 1942, _5_6_, 264.

 

Cohen,. J., and Gordon, D. A. The Prevost-Fechner-Benham
subjective colors. Psychol. Bull., 1949, :16, 97-136.

 

DuBois, R. Recherches expérimentales sur le r61e de la contractilité
dans les méchanismes sensoriels chez les mollusques. J. de
Psychol., 1922, _5_1_, 321.

De Valois, R. L. Color vision mechanisms in the monkey. J. gen.
Physiol., 19 , :13, 115-128.

DeValois, R. L., Smith, C. J., Karolz, A. J., and Kitai, S. T.
Electional responses of primate visual system I. Different layers
of macaque lateral geniculate nucleus. J. Comp. & Physiol.

Psychol., 1958, _5_1_, 662-668.

 

DeValois, R. L., Smith, C. J., and Kitai, S. T. Electrical
Responses of primate visual system 11. Recording from single
on-cells of macaque lateral geniculate nucleus. J. Comp. 8t
Physiol. Psychol., 1959, 22, 635-641.

 

 

Fry, G. A. Modulation of the optic nerve-current as a basis for
color-vision. Am. J. Psychol., 1933, 42, 488-492.

 

Fry, G. A. Color phenomena from adjacent retinal areas for dif-
ferent temporal patterns of intermittent white light. Am. J.
Psychol., 1933, i5, 714-721.

Fry, G. A. New observations related to the problem of color-
contrast. J. Exper. Psychol., 1934, 17, 798-804.

 

Fry, G. A. Color sensations produced by intermittent white light
and the three-component theory of color-vision. Am. J. Psychol. ,
1935, :11, 464-469.

 

Fry, G. A. Color sensations produced by intermittent spectral
stimuli. Am. J. Psychol., 1936, 48, 326-330.

 

Gebhard, J. W. Chromatic phenomena produced by intermittent
stimulation of the retina. J. Exper.-Psychol., 1943, _3_3_, 387-406.

 

Granit, R. , and Wirth, A. A scotopic "blue shift" obtained by
electrical measurements of flicker resonance. J. PhjSlOl. , 1953,
122, 386-398.

 

37.

38.

39O

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

33

Halstead, W. C., Knox, G. W., and Walker, E. Cortical activity
and photic stimulation. J. Neurophysiol., 1942, _5_, 349-356.

 

Halstead, W. C.,. Knox, G. W., Woolf, J. I., and Walker, A. E.
Effects Of intensity and wave length on driving cortical activity
in monkeys. J. Neurophysiol., 1942, _5_, 482-486.

 

Hartridge, H. Recent advances in the physiology of vision. London:
Churchill, 1950.

Ingvar, D. H. Spectral sensitivity as measured in cerebral visual
centers. A study Of responses to monochromatic flicker in the
unanesthetized cat. Acta Physiol. Scand., 1959, 46 (suppl. 159),
105 pp.

 

Landis, C. An annotated bibliography of flicker fusion phenomena
covering the period 1740-1952. Armed Forces--National Research
Council, 1953.

Landis, C. Determinants Of critical flicker-fusion threshold,
(Physiol. Revs., 1954, 34, 259-286.

 

LeGrand, Y. , and Geblewicz, E. Sur 1e papillotement en vision
latérale. C. R. Acad. Sci..Paris, 1937, 205, 297-298.

 

Lennox, M. A. Geniculate and cortical responses to colored light
flash in cat. J. Neurophysiol., 1956, _1_2, 271-279.

 

Lennox, M. A. Single fiber responses to electrical stimulation in
cat's Optic tract. J. Neurophysiol., 1958, _2_1, 62-69.

 

Lennox, M. A. The on responses to colored flash in single Optic
tract fibers of cat. Correlation with conduction velocity.
J. Neurophysiol., 1958, _2_1, 70-84.

 

Lennox-Buchthal, M. A. Single units in monkey, cercoecebus
tomuatus atys, cortex with narrow spectral responsiveness.
Vision Research, 1962, _2, 1-15.

 

 

 

Lennox,.M. A. , and Madsen, A. Cortical and retinal responses
to colored light flash in anesthetized cat. J. Neurophysiol. ,
1955, 1_8_, 412-424.

 

Madsen, - A. , and Lennox, M. A. Response to colored light flash
from different areas of optic cortex and from retina of anesthe-
tized cat. J. Neurophysiol., 1955, _1_8_, 574-582.

 

50.

51.

52.

53.

54.

55.

56.

57.

58..

59.

60.

61.

62.

34

Nelson, T. M. Unpublished N. I.H. research proposal, 1962.

Nelson, T. M., and Bartley, S. H. The role of PCF in temporal
manipulations Of color. J. Psygholu 1961, 22, 457-477.

 

Nelson, T. M., Bartley, S. H., and DeHardt, D. Considerations
involved in Obtaining averages of CFF data. J. . Psychol. , 1960,
49, 263-277.

 

Nelson, T. M., Bartley, S. H., and Mackavey, W. R. Responses
to certain pseudoisochromatic charts viewed in intermittent
illurninance. Percept. 81 Mot. Skills, 1961, _1_3, 227-231.

 

Pauli, R., and Wenzl, A. Uber farbenempfindungen bei intermit-
tierendem farblosen lichte. Arch. ges. Psychol., 1924, 48,
470-484.

 

Piéron, H. Le méchanisme d'appartion des couleurs subjectives
des Fechner-Benham. Année Psychol., 1922, _2_3_, 1-49.

 

Piéron, H. Des lois du d'éséquilibre chromatique initial et de la
prépondérance de la diffusion chromatique dans l'excitation
lumineuse de la rétine. (Méchanisme de production des couleurs
subjectives de Fechner-Benham.) C. R. Soc. Biol. Paris, 1922,
86, 922-925.

 

Piéron, H. Le méchanisme des couleurs subjectives de Fechner-
Benham. J. de Psychol., 1923, 29, 85-90.

 

Piéron, H. The sensations, their functions, processes, and
mchanisms. New Haven: Yale Univ. Press, 1952.

Roelofs,. C. O. and Zeeman, W.~P. C. Benham's tOp and the color
phenomena resulting from interaction with intermittent light
stimuli. .Acta Psychol., 1958, _1_3, 334-356.

 

Segal, J. Une nouvelle manifestation de couleurs subjectives.
C. R. Soc. Biol.~Paris, 1940, 133, 259-261.

 

Smith, F. J. Apparent decomposition of sunlight by intermittent
reflecting surfaces. Nature, London, 1881, _2_4, 140.

Talbot, H. F. Experimenta on light. Phil. Maj. (3rd series),
1834, _s, 321-334.

 

35

63. Tomiyasu, K. Laser magic. Time, April 20, 1962, 103-104.

64. Troland, L. T. The enigma of color vision. Amer. J. of Physiol.
Opt. 1921, 3., 23-48.

 

65. Valsi, E., Bartley, S. H., and Bourassa, C. -Further manipulations
Of brightness enhancement. J. Psychol., 1959, 48, 47-55.

 

APPENDIX

36

L mmDoE

 

LUMINANCE
OF THE MATCHING STIMULUS

38

TARGET lg

 

 

 

RATE 9.80PS DARK SURROUND OBS: RJB
70-1 DOTTED Lms ARE PCF
I ENHANCEMENT TALBOT LEVEL roe _3
I EACH PCF -——-o-l/I6 4
60" ~ LEVEL 04/8

 

 

 

 

 

FIGURE 2

LUMINANCE
OF THE MATCHING STIMULUS

O
TARGET H; pg; L

39

4 DARK SURROUND OBS: RJB

 

 

 

 

 

 

 

 

 

RATE IN CYCLES PER SECOND

FIGURE Q

7

OI ENHANCEMENT

60.1 LEVEL WNELENGTH
+500 ;
+540

50. +580
+620
+680

40--

m...

20- TALBOT

LEVEL
IO-
0'6 635 738 9'3 lip 151 :50 2030

40

 

  

 

 

 

 

 

 

 

 

A

 

.- Tm‘ oss: RJB
:3
g C MULTIPLE POINT
a, exTRENE-
<
3
”MODERATE-
? /
t- a /
g SLIGHT— ‘T'fl-{n—i
5
O/’
uous- c- —-- -—--~ .
(f) 4:20 430 «'50 480 so'o 5Io 5'20 530 560 560 660 650 6:00 660 site 160
3 RATESINCYCLES PER same
:3 90_ DARK SURROUND 4*“
o +9.8
E TARGET 1% *‘5-0
5
PCF J-
2? 70. I6
EE ENHANCEMENT u a
,9 ‘°- LEVEL
<1
2 so.
LlJ
I 40.
F—
5 30.
L5, .._
Z
<[
2 IO.
‘2‘ TALBOT
D 0 LEng I l I 1 T I I l I I T I
-J 420 «o «0 Jo 500510520 540 560 soo 600 620 «o «0 600 Too

WAVELENGTH

IN MILLIMICRONS

 

FIGURE 4

 

41

 

Tom-

EXTREME-

MODERATE-

SLIGHT—

SUBJECTIVE WASHOUT

NONE-

 

O MULTIPLE POINT

 

 

OBS: RJB

 

 

9

I

80..
TO-
60-
50__
40.
3 O-

20_

ENHANCEMENT
LEVEL

LEVEL

‘!0 do 470 450 560 SIO srzo 550 530

DARK SURROUND

TARGET 1,1,3
per i
' 4

560 060 do 030

RATES IN

CYCLES PERM

60.5
+7..
+9..
'O-IZD

 

T U T
GOO GOO 700

+5.!
+5.0
+809

A

 

 

 

LUMINANCE OF THE I.CATGHIPJG STIMULUS

420

I l I
440 460 480

I l T T T
500 SIG 520 540 560

l l I I
580 600 620 640

WAVELEIIGTH II.‘ I‘XJLLHUICROIJS

I l
660 680 TOO

FIGURE 5

SUBJEGTIVE WASHGUT

TOTAL— 1 0885 RJB
‘. MULTIPLE POINT
EXTREME-
MODERATE— ._ / /9t_—\
: ‘/ . s \;
SLIGHT— ./ ' z ; ;
NONE- B
,3 4:20 430 «'50 460 sob 5I0 53.0 5710 ago 530 BOO 650 6'40 6E0 300 700
D
_J. DARK SURROUND RATES IN
:3 90- CYCLES PER SECOND
é TARGET 4° see +131
f... 80 +7.8 +5.0
U9 - | "9'93 +209
pop -— +120
(FT 4
_v_ 70-
E ENHANCEMENT
0 so-
I“ LEVEL
<1.
i; so.
'.;_I
I 40_
'r__
L1.
0 30-
LL]
. 2c_
2 TALBOT
(fl IO- LEVEL
2:
'3'
:3 o -A
v r I I I l I T T I T— I I f T I
__, 420 440 460 480 500 510 520 540 560 500 600 620 640 660 600 700
WAVELENGTH IN MILLIMICRONS FIGURE 6

42

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I.-<.-I\G~1a “\a-Fcu_.a—~m

43

 

 

OBS’ RJB

 

 

 

 

 

 

._ Tom.—
3
g c, MULTIPLE POINT
a, ExTREuE-
<
3
meMTE- 8 3 r» 3 p
Z I 3. 3 L t
" t 1 i i
g1 smuT- 9 2 V O
A C.
m i
a .o... s. v V\ B
420 470 460 450 560 sio $20 540 050 530 000 350 :40 050 0'00 700
DARK SURROUND RATES IN
90. CYCLES PER SEOCND
O '
TARGET l-L ea: +5.1
4 +7.. «no-15.0
”- , +93 +200
PCF E- +I2.0

LUMINANCE OF THE MATCHING STIMULUS

7°- ENHANCEMENT
LEVEL

 

  

 

 

 

 

50. \‘i
40.
\‘
TALBOT . -- Ag , n
30. ' 1--.} ., \
. ' " \V""§‘2—- -z!
.._ ' \.><
'9-
. ,_ A
420 440 450 4B 5001505120 540 500 530 000 65 :3 «F0 «'10 700

WAVELENGTH

IN MILLIMICRONS

F'GURE 7_

44

 

SUBJECTIVE WASHOUT

 

 

 

 

 

 

 

 

 

A

 

 

rom— OBS: RJB
EXTREUE- CMULTIPLE POINT
MODERATE—

SLIGHT- ~.__r-' A ~ A J‘ ,1. 2?,

L/ : .2/ E E; ‘3 fl/\__‘

NONE- .‘ , A it \ , 7, B
U) 4110 440 460 4E0 560 5I0 5720 540 560 530 560 sic J10 0110 ciao 700 '
3 DARK SURROUND RATESNCYCLES PER SEOND
I) 90. +6.5

0 '9-98

E TARGET 1J4— *‘5-0
1‘- BO-
“3 3
o PCF 4
_ 70.
E
3:) ENHANCEMENT

so.
'__ LEVEL
E! " /\

TALBOT W .//\

LLI LEVEL
I 40_
F.
‘5 30.
g .._
Z
<1
2 IO.
2
3 0
-J 420 440 460 450 500 $10 $20 540 5E0 Jo 650 1550 6'40 ssro «'0 700

WAVELENGTH

IN MILLIMICRONS

FIGURE £3 ._

T |
x

F. s MOP—(2.2314

M
FDOIWA‘; w>-Fom7m\.-W

ODE

45

 

  

 

 

 

 

 

 

 

1— TOTAL— 088: RJB

8

<

3

11.1 MODERATE-

Z

.—

8 SLIGHT-

a

I?"

NONE— B
7330 540 580 020 6&—

-0— FULL INTENSITY
DARK SURROUND _._ 0.5 m, .0...

6°- -°- 1.0 1110. ADDED
RATE 9.8m —.— 2.0 11.0. ADDED
O
50- TARGET 121,-
I
40- —
PCF s

LUMINANCE
W THE MATCHING STIMULUS

IO- G-\9’ _A
\\

 

 

 

 

 

0- L n n n n A
500 540 530 oio 0J0

WAVELENGTH IN MILLIMIORONS
FIGURE 9

m
t—u—JOIWA‘)’ w>_1—10U\Jmfidm

moz<2=\(31—

46

 

 

 

 

 

 

 

 

   
  
 

 

 

 

 

 

 

Tom— OBSI RUB
'—
8 — T
f— : 1'
5:1 ennub -. MULTIPLE rCIN
<
3
muooenATE-
Z
1— Q
8 SLIGHT- 7; c
_,
m
NONE- A * 3:2-
7o_soo 540 580 , 620 680 _1
-o—
DARK SURROUND .9. gamgmggg
60. 4- 1.0 N.D. ADDED
:3. RATE 93993 -- 2.0 N.D. ADDED
0
g 50_ TARGET 12',-
r.
m
g 0 .0. PCF _L
a z 4
z 5
2 I; 30-
3 2
g 20- \
'—
8 10-
°\r/_a¥ '"' '1
O— [L ' A A ' \ A
560 540 500 :50 Geo

WNE ENGTH N M MICRONS
L ' 'L” FIGURE IO

M “CF-(2.5:. .
FDOImd‘)‘, U>_.—10U7m3w

47

 

 

 

 

 

 

 

 

 

.— TOTAL- 088: RJB
8 OMULTIPLE POINT
a EXTREME-
<
3
”MODERATE-
Z
.—
8 SLIGHT-
_J
m
a NONE-
70—50'0 540 550 gig 6L
-0- FULL INTENSITY
Go- DARK SURROUND _-: 0.5 ND. ADDED
1.0 N.D. ADDED
RATE 939” -— 2.0 110. ADDED
50_ TARGET 4°
I
40- PCF 2F

LUMINANCE
07 THE MATCHING STIMULUS

 

 

 

 

 

 

 

 

Bio B40 540 Bio 060

WAVELENGTH IN MILLIMICRONS
FIGURE II

IF— _CIU..<)> hu>-ch1m3w

SUBJECTIVE WASHOUT

LUMINANCE
OF THE MATCHING STIMULUS

TOVL-

EXTREME-

MODERATE-

SLIGHT-

NONE-

48

 

 

 

OBSI RJB

CI MULTIPLE POINT

 

 

I?!

 

50-

20-

IO-

 

540 séo 620 - 6 'Bo

 

D... sumo... $29.“..3151‘SE‘JJ

+ 1.0 N.D. ADDED
RATE 9.80PS + 2.0 N.D. ADDED

TARGET I-E

_I-
PCF 2

 

 

A‘ M
Y‘T— “
1L \ A 43

 

560

540 560 650 680
WAVELENGTH IN MILLIMICRONS
F IOU R E 12

a m I U02<Z.V<3..
FDOIWA‘)? N>.Ir0mu..am.I-m

49

 

?

EXTREK-

MODERATE-

SLIGHT-

SUBJECTIVE WASHOUT

 

.900

CBS! TN
OMULTIPLE POINT

 

61»
119

 

   
 

 

 

560 $550 550

+ EULL INTENSITY
DARK SURROUND -°- 0.5 N.D. ADDED

  

 

 

 

 

 

 

(I) 50.. + I.O N.D. ADDED
3 RATE 9.8cPs -- 2.0 110. ADDED
D o
g 50- TARGET 1-},- PCF L
1- 4
a:
111
o o 40—
z E
5 s
3 I- 30-
3 §
11! 20-
.- (\9//\;\
8 10-
F.— _‘/£"_ B m
0- A A A A \ A
530 540 550 550 560

WAIELENGTH IN MILLIMICRONS
fflGUREI3

N
FDOIWA‘; m>.I—10mqamfidm

50

 

Toum— OBS=‘TN
O MULTIPLE POINT

EXTREME- .
“ODEMTE- A

 

 

 

SUBJEOTIVE WASHOUT
01

l'

8

1

T

.

>

 

 

 

500 540 550 Géo 650

g

D... sum... : Si”..':.’i%§£."

‘5‘ 1.0 N.D. ADDED

    
     
 
 
  

50- I
PCF I "‘P 2.0 N.D. ADDED
,o_ TARGET 4°
RATE 9.8cPs

20-

L}_ ,-
r‘r—fi "
I0-

I; 7‘. r‘ A - - t
A A A x _ - .. E
O—

560 540 560 650 680
WNELENGTH IN MILLIMICRONS

 

 

LUMINANCE
(I: THE MATCHING STIMULUS

 

 

 

 

 

 

 

 

FIGI.RE .4

w moz<2.23l~
#3371.qu w>_LI0Mquu-JMW

51

 

 

 

 

 

 

 

 

TOTAL— 0353 CB
5
(:2 C MULTIPLE POINT
m EXTREME-
<1
3
LUMODERATE-
Z
P—
o
111 SLIGHT-
j
m
D
m NONE- B
500 540 550 650 650
70- r
+ FULL INTENSITY
DARK SURROUND : 0.5 N.D. ADDED
_ RAT 1.0 N.D. ADDED
8 6° E 9'85” -- 2.0 N.D. ADDED
51 TARGET 11',-
g 50-
'—
m i.
8 <9 40- 4
‘3 E
z 5
2 ’4‘ 30—
3 2
11.1
I 20' (a
'— \ ,4;
;—— \\ ‘J
8 IO- VT
~ ‘9‘ A
ref ” ‘9 4
0- L A A A 4‘
560 540 560 560 560
WNELENGTH IN MILLIMICRONS

  
  
 

 

 
   

 

 

 

 

 

 

FIGURE I5

s

>-.h|0wfiamfi.dm

m0242—23I—

52

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TD‘IAL— 0883 GB
.—
8
£5 EXTREIAE- C MULTIPLE POINT
<1
3 2
mquRATE- %
Z
l—
8 SLIGHT-
5
m
a
NONE-
560 540 séo 5210 550
70-
+ FULL INTENSITY
DARK SURROUND '9- 0.5 ND. ADDED
‘5'- LG N.D. ADDED
‘3 GD- RATE S'BCPS -- 2.0 N.D. ADDED
5' TARGET 4°
g 50—
*- O
m "’ TARGET Ii-
0 o 40-
3 E
z 25
2 I; 30-
3 z
E 20-
F 0
3 9f , -
Io-
G—i _¢‘r {_fi
0 A A ' A ' A I
560 510 680 620 630

WNELENGTH IN MILLIMICRONS
FEURE

53

 

 

 

 

 

 

 

 

 

 

 

 

 

*- Tm_ Des: RJB
3 cMULTIPLE POINT
S EXTREME-
3
“DIODEMTE-
Z
.— F.
I131) suGRT— ’3
”8°
NONE- " B
U') 420 430 480 430 2530 Ho 03.0 530 550 030 050 GR 030 0:0 0730 700
3 SURROUND 540““ SURROUND LUMINANCE
g 90. RATE 9.8093 137——
.L '9-Less THAN TARGET
; TARGET l4 +EGUAI. T0 TARGET
(D 30‘ +GREATER THAN TARGET
I
P F —
o ,o_ C 4
.2.
E ENHANCEMENT
Go.
'__ LEVEL
<
2 50.
DJ
I 40.
l.—
LL
0 30.
LIJ
2°.
‘2’ TALBOT
LEVEL
<21 IO.
2
3 O , '7 I I I I T T l l T I T I I
420 «0 400 450 00050520 540 500 500 000 620 640 660 680 700

WAVELENGTH

IN MILLIMIGRONS

FIGURE I7

54

 

  

 

mm Des: RJB
.-
3 c MULTIPLE POINT
g EXTnfi-
3
“WHAT!-
2
g SLIGHT-
5
m- B

 

 

v v v I I I I r I I I
420 «0 «0 400 000 no 020 040 «'0 500 000 050 030 660 600 700

 

 

 

 

 

 

 

8 SURROUND 600w
_I RATE 98m SURROIND LUMINANCE
é) 90L .J: +2ER0
~9-LEss THAN TARGET
; TARGET '4 +EOUAL T0 TARGET
U) 00— ' +GREATEN THAN TARGET
PCF —
‘2” "L 4
if ENHANCEMENT
E ”- LEVEL
<[
2 so.
LIJ
I 40. ' ’
U. I
3
O' 0. . \, [A
8 2°. \gfi’l \- 4.0
Z TALBOT -
g "L LEVEL ‘
S
D o I A
—I 420 «0 430 4‘0 SBOEEO 010 560 do 050 0130 050 6&0 «'0 700

WAVELENGTH IN MILLIMICRONS FIGURE I8

 

55

 

TOTAL—I -
I 0 883 R JB

 

C»- MULTIPLE POINT

r. ‘.

SUBJECTIVE WASHOUT
3
8
"I
S
'I"

 

In

 

 

 

500 530 5&0 650 6.80

 

.1 ENHANCEMENT

 

 

 

 

 

 

 

 

8 6°- “ LEVEL DARK SURROUND '9' 4 MARCH 62;PM
5’ RATE + I MAY 623AM
9.80Ps "" 22 MAY 62;PM
g 50— . _._ 24 MARCH 62;AM
'5 TARGET Ii-
In
E E 4°~
< i
3 o
E 2 =0-
..I 2
2 2°' TALBOT
8 LEVEL
I0-
0- __ A
5&0 030 560 650 660

WAVELENGTH IN MILLIMICRONS
FIGURE I9

 

56

 

    

 

 

 

 

 

 

'— mm_ 088: RJB
8 C MULTIPLE POINT
fi EXTREME-
<
3
“JNODERATE—
2
I- , _‘ 2
8 SLIGHT— ' - , ., Q
j
g \
NONE- 5::
II F I I I I I I I
0 6.5 7.0 9.0 I2.0|3.I I5.0 20.0
70-
ENHANCEMENT RATE 9.86%
00- LEVEL

DARK SURROUND WAVELENGTH BANDS
-€"‘- 472 T0 576

50- -B- 528 TO GBB

-ér- 437 TO 490

C,

40- pop 7:; TARGET If

 

20-

 

TALBOT
LEVEL

 

 

LUMINANCE
(I: THE MATCHING STIMULUS

IO-

 

 

 

6 035 730 9'0 I50 I31 ISO 20.0
RATE IN CYCLES PER SECOND FIGURE 20

O- I

 

 

 

480. 500 510 520 540 560 580 600 620 640 660 680

460

Size Sur. Sur.

Rate

PCF

I

 

QQQQQQQQI

QCICI

GOO

GOO

QQQ

OO

QQQQQI

QQQQQ

QQQQQQ

QQQQ

DDDIDI

DDDDD

DD

333.

00

U)

13.1

I Q

QQQ

O

57

QQQQQQQQ

Q

Q

Q

QQQQQQ
QQQQQ
QQQQQ

QQQQQ

DDDDD.

DDDDD

DDDDD

IBBBII

QQQQI

L3.0

g-‘In—‘I

Q

Q

Q

Q

:3

D

D

QQQQQQQQQ

QQQQQQQ IQ

QQQQ IQQQ
CIQQ

QQQ

OOOOO

mmmmm

15.0

Less 540

UIQQQ'I
IIQQQT
QIQQQH
QICIQQQ
QIQQII
IIIQII
""'I
DIDDI'
DIDDI'
IIDDI'
IIDII'
""|l
$3- .888
mm 000
HEDGE)";
(DCDUJIDCDW
RREREE
O~O~O~O>O~O~
§§§§§§

 

58

 

 

 

 

 

 

 

 

 

‘Figure 22. Hue shift for experimental conditions using five wavebands.

PCF Rate Size Obs. Sur. ~N.D. 500 540 580 620 680
Added

1/8 A S RJB A O U U - - D
1/8 -S RJB . 5 - - - D -
1/8 s RJB 1.0 - - - - .-
1/8 s RJB 2.0 - - - - -
1/2 S RJB O U U - D D
1/2 S RJB 0.5 U U - D D
1/2 S RJB 1.0 - - - D -
1/2 S RJB 2.0 - - - - -
4‘ S RJB 0 W U - D D
S RJB 0.5 - - - D D

S RJB 1. O - - - - -

S RJB 2.0 - - - - -
L RJB O W U D D D
L RJB O. 5 - U - D D

L RJB 1.0 - - - - -

1/4 9, 8 L RJB 0 2.0 - - - - -
S TN 0 U U - D D
S TN 0. 5 U U - D D

S TN 1.0 U - - D -

S TN 2.0 - - - - -
L TN 0 W U - D D
L TN 0. 5 U U - D D
L TN 1.0 U U - D D

L TN 2.0 - - - - ‘-
L CB 0 - U - D D

L CB 0.5 - U - D -

L (SE 1.0 - - - - -

V V L .CB V 2.0 - - - - -