Eastern Illinois University

The Keep
Masters Theses Student Theses & Publications

1972

The Effects of the Four Psychological Primary

Colors on GSR, Heart Rate, and Respiration Rate
Keith W. Jacobs
Eastern Illinois University
This research is a product of the graduate program in Psychology at Eastern Illinois University. Find out more
about the program.

Recommended Citation
Jacobs, Keith W., "The Effects of the Four Psychological Primary Colors on GSR, Heart Rate, and Respiration Rate" (1972). Masters
Theses. 3911.
https://thekeep.eiu.edu/theses/3911

This is brought to you for free and open access by the Student Theses & Publications at The Keep. It has been accepted for inclusion in Masters Theses
by an authorized administrator of The Keep. For more information, please contact [email protected].
PAPER GER TIFICA TE #2

TO: Graduate Degree Candidates who have written formal theses.

SUBJECT: Permission to reproduce theses.

The University Library is rece1v1ng a number of requests from other

institutions asking permission to reproduce dissertations for inclusion
in their library holdings. Although no copyright laws are involved,
we feel that professional courtesy demands that permission be obtained
from the author before we allow theses to be copied.

Please sign one of the following statements.

Booth Library of Eastern Illinois University has my permission to

lend my thesis to a reputable college or university for the purpose
of copying it for inclusion in that institution's library or research
holdings.

I respectfully request Booth Library of Eastern Illinois University not

allow my thesis be reproduced because
-----

Date Author
 I

THE EFFECTS OF THE FOUR PSYCHOLOGICAL PRIMARY

COLORS ON GSR, HEART RATE, AND RESPIRATION RATE

(TITLE)

BY

KEITH W. JACOBS

B.A., University of Northern Iowa, 1968

TtiESIS

I·
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR THE DEGREE OF

I
MASTER OF ARTS

IN THE GRADUATE SCHOOL, EASTERN ILLINOIS UNIVERSITY

CHARLESTON, ILLINOIS

1972
YEAR

I HEREBY RECOMMEND THIS THESIS BE ACCEPTED AS FULFILLING

1
THIS PART OF THE GRADUATE ABOVE

.
"

___
________.__3�!1 ?;_
APPROVED BY THE THE�IS COMMITTEE:

( s igned )
 ABSTRACT

Measu�ement of G SR , heart rate, and respi ration rate

were taken on twenty-four male S 's during presentatior. o f
four colo rs ( re d , yellow, green, blue) . Fou� s equences o f
color presentations were used based o n a 4 by 4 latin square
with six subjects receivir.g each presentation. S ignifican t
color effe�ts (p ( .05) were found when G S R was used as the
measure of activatio n . The most arousing color was red ,
followed by green, yellow, and b l ue , with s ignifican t dif­
ferences in arousal value fo und between colors . Significant
trial effects , dis regarding colors , were found in respira­
tion rate . Ne s ignificant effects on heart rate were found
as a function of trials o r color . The arousal values o f the
colors do not support either a l i near or U-shaped function
of wave l e ngth.

iii
 ACKNO WLEDGE YJ.ENTS

This res earch was possible through the efforts o f

D r . Frank Hustrnyer who f o r the past year encouraged me to

pursue this area of research and served as chairman for this

thes is . Grateful appreciation is also given to Dr . Randall

H. Best, Dr. Harold Coe, and M r . Boyd Spencer who s e rved as

committee members and o ffered their advise and assistance.

Finally, I mus t express my gratitude to my parents

whose love and support made this part of my career possible .

iv
 TABLE OF CONTENTS

ABSTRACT •••••••••••• . . . . . . . . . . . . . . . . . . . . .......... . . .. iii

ACKNOWLEDGEMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv

LIST OF TABLES••• . . . . . . . . . . ........ . . . ...... . .. .. .. . .. vi

LIST OF F IGURES••••••••••••••••••••••••••••••••••••••• · vii

CHAPTER

I. INTRODUCTION . . . . ......
. . ........... . ........... 1

A. Historical Background o f Pro blem ••••••••••• 1

3. Statement of Hypothesis . . . . . . . . . . . . . . . . . . . . 14

II. :SXP:SRIM
. ENTAL METHODOLOGY ••. . . . . . . . . . . . . . . . . . . . .

A. Subj ects ••••••••• . . . .... . ....... . .. . .. . .... 15

B. Stimul i Colors ••••. •••••••••••••••••••••••• 15

c. Apparatus •• . . . . . . . . . . . . . . . . . . . . . ........... 17

D. Procedure ••. . . . . ... . ....... . ............... 19

III . RE SULTS••••• . . . . . . . . . . . . . .
. . . . . . .. . .. ... . .... .
. 24

IV . DISCUSSION ••••••••••••••••••••••••••••••••••••• 30

A. Color Effects ••••••••••. . . .............. 30

B. Response Habituation ••••••••••• . . . . . . . . . 33

c. Response Sys tem Specificity ••••. . . . . .. . .... 34

D. Sugges tions for Further Res earch ••••••••••• 37

REFERENCES••• . . . . . . . . . . . . . . .................. . ........ 40

APPENDICES •••••••••••••••••••• . . . . . . . . . . .. . ......... . . 44

I. INSTRUCTIONS••••••••••• . . . . ...... . . ..... . . . . . .. 45

II. STATI STICAL ANA LYSIS OF RESULTS. . . . . . . . . . . . . . . . 47

III . ORIGINAL DATA•••••••••••••••••••••••••••••••••• 51

v
 LIST O F TABLES

Table Page

1 Summary of Physiological Changes . . . . . . . . . . . . . . . . . 26

2 Arousal Values o f the Four Colors. . . . . . . . . . . . . . . . 28

3 Response Habituation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

4 Response System Specificity. . . . . . . . . . . . . . . . . . . . . . 36

vi
 LIST OF FIGURES

Figure P�e

1 Arous al Values o f Visible Hues ( Wi l s o n , 1966) • • • 7

2 Summary cf Previous Res earch • • • • • • • • • • • • • • •••••• 13

3 Stimul i Colors ••••••••••••••••• . • • • • • • . ••••• • • • • 15

4 Order o f S timuli Pres e ntations • •• • • • • • • • • • •••••• 22

5 Latin S quare • • • • • • • . . • • • • • • • • • • • • • •••••••••• • . • • 23

6 Arous al Values o f Four Colors • • • • • • • • • • • •• • • • • • • 27

7 Changes in Respiration Rate by Trial • • • • • • • • • • • • 29

vii
 CHAPTER I

INTRODUCTION

Historical �ackground of Probl em.

The as sumption is frequently made that people are

affected psychologically by colors . This assumption is be-

ing applied in the areas o f advertising, art education, corn-

rnercial art, theater, architecture, and television; and is

bsing appl ied as though the relationships have been proven .

The assumption is further made that these psychol ogical

effects can be measured on a phys iological leve l .

This basic assumption is presented without question

in one of the leading advertis ing references today , which

states that:

Color arous es emotions . Contrary to popular be­

lief, color is not rel ated to vision alone • • • •

As a rul e , red also serves as a s timulant; i t

shakes u p the brain, quickens the pul s e ( Baker,
196 1 , p. 9 6 ) .

A s imilar s t atement has been advanced by one o f the noted

authorities in the area of color and art education, in

which he reports that:

The polygraph ,.or lie detecto r , measures pul s e ,

blood pres sure , perspiration, s kin temperature ,
all o f which may change under exposure to dif­
ferent colors and degrees of brightness ( Birren,
196 3 , p. 176 ) .

This basic tenet is that changes in emo tion, reflected in

changes in the autonomic nervous system, can be measured as

. .

a function o f changes i n color ( hue) and brightnes s .

 2

The amount o f research which directly supports these

s tatements is limited . The l iterature which has given rise

to these as s umptions and has served as a basis for this area

o f investigation is equally limited. This absence of a s i g-

nificant body o f contro l l ed research prompts a review o f the

available s tudies .

I t is obvious that the hypothesis i ts e l f is not new

to psycho logical research. In his early explorations o f the

galvanic skin res ponse ( GS R ) , Fere ( 18 9 2 ) found that one

subj ect produced a marked change in G SR when she was forced

t� l c ok through a blue glas s . Fere ' s concern was primarily

with the variables which produce changes in GSR and he event-

ually reported changes in GSR as a function of "a change in

light" rather than colors thems e lves , but it is interesting

that this research reflects the earliest autonomic nervous

sys tem measurements o f the effects o f e i ther color or in-

tensity of l ight . It has also been reported ( P ressey, 19 2 1 )

that Fere observed changes in respiration and pul s e rates

when hysterical subjects were shown a red l i gh t .

Early attempts to define the concept o f color have

indicated that color produces measurab l e responses and hinted

at differential res ponses for the various hues . One such

definition by the Optical Society o f America i n 19 2 2 has

found its way repeatedly into the psychological l iterature:

Color is the general name for all sens ations

arising from the activity of the retina o f the
eye and its attached nervous mechanisms, this
 3

activity being in nearly every case in ·the normal

individual , a s pecific response to radiant energy
o f certain wave-lengths and intensities • • • •

Color can not be reduced to terms o f any purely

phys ical conception; it is fundamentally a psycho­
logical category (Tro land, 192 2 , p. 5 28 ) .

The earlier research into psycho logical aspects o f

color was concerned with such subj ective as pects as indi-

vidual color preferences . An excellent review o f thes e

s tudies is furnished by Norman & Scott (195 2 ) who point out

the underlying as sumption that affective valu� has been

equated, at least s emantical l y , with intended applications

in industry and schools as he s tudied the e ffects of various

hues and intensities but his findings were based on intro-

s pection and obse rvational s tudieB, limited experimental

s tudies of preferenc e , and s tudies o f task performance while

his subjects were s e ated in front o f l arae colored cards .

He concluded that there was "almo s t absurd difference o f

opinion" in the subj ective areas but that i f "co2.or does

have any fundamental physiological effect • •• the connection

must be o f a very general and e l ementary nature; brightness

may s timulate, or red irritate and dis tract, but more s pe-

cific effects are hardly to be expected . " He found gener-

ally that illumination level would pos i tively effect finger

tapping rate, problems in multiplication, and continuous

reaction tasks but found no effects from hue alone. He did

note, however, variable subj ective reports o f affective

changes with various hues .

 4

Color preference research was refined by Guilford

( 19 3 1) in s e perate evaluations of the three dimensions of

color: hue , chroma, and tint. He found that with hue held

cons tan t , there was an increase in affective value parallel­

ing increase in chroma or tint, or both together . However,

hue was found to be the most important factor in color pref­

erence . This research is responsible for the development of

the P-U continuum ( pleasantness -unpl eas antnes s ) which is

s till po pular in psychological color research today ( Smith,

195 8 ) . Guilford ' s greatest contribution in the area of color

research, however , was the experimental control of chroma,

tint, and hue.

The s tudy of color preferences was further refined

by Eysenck ( 19 4 1 ) with findings that there is general agree­

ment between the sexes in their preferences and that a di­

vision exis ts between subjects in preference for s aturated

or unsaturated colors .

The relationship betweeu color and emotional pattern

has been researched in terms o f : ( a) the effects o f color

on emotions , and ( b ) the effect of emotions on color perce p­

tions . Hevner ( 19 3 4 ) found that col lege students described

various colors in emotional terms , with red usually des­

cribed as happy or exciting, and blue described as s e rene

qr dignified . Similar associations were noted by Wexner

(19 5 4 ) after asking his subjects to s e lect color samples

which depicted eleven "mood-tones" which the experimenter

 5

supplied. Me found red was s e lected as "exciting-s timu­

lating " , blue as " serene-comfortab l e " , orange as "distres s ed­

dis turbed-upset" , blue was also s e lected as "tender-sooth-

ing . "

Colors have also been found to be associated with

musical s e lections ( Odbert, Karwoski, & Eckerson, 19 42 ) .

These researchers reported that when subjects were forced

to rel ate colors to musi c , they g av e responses very s imilar

to the responses g iven by s ubjects who readily react with

vivid visual imagery. They found also that the colors named

with a s e lection were sys tematically rel ated to the named

mood o f the s e lection; and th at when tne st:.bjects were asked

to name the color bes t fitting the same mood-words , the re­

sults were very s imilar to those obtained with the u s e o f

mus ic.

Behavioral correl ates o f color have also been reported�

Smith (19 3 6 ) observed a quieting effect o f blue light on

infants . In his work with organic mental patients , Goldstein

( 19 3 9 ) found red to be disagreeab l e , upsetting , exciting ,

and phys ically naus e ating .

In an experimental s tudy o f the e ffects o f red and

green surroundings on psychomotor tasks and judgement,

Nakshian ( 19 6 4 ) found inconclusive results and noted that

"there has been a remarkabl e dearth o f s c ientific s tudies

dealing with this particular color-behavior problem . " How­

ever, a limited body o f research into the physiological

 6

effects of color has developed in the last 15 years . The

s tudies cited to this point have been concerned with sub­

jective and preferential responses to col o r , and to a lim­

ited extent, the effects of color on gross behavior . They

have a l l carried the underlying assumption that these res­

ponses are potentially measurab l e on a more molecular level

or in terms of a particular body system. S e veral s tudies

have been undertaken to investigate the differential effe cts

of color on GSR , heart rate, respiration rate, and other

autonomic nervous system functions .

In a comparison o f red and green hues on GSR , Wilson

( 1966) found red to be significantly more arousing than

gree�. Wilson presented red and green slides on a s creen

and measured changes ir. skin resistance ( GS R ) . Each o f his

twenty sub j e cts received five presentations of green, in a

random order. He found that both red and green stimuli

produced measurab l e GSR's and showed evidence of habituation

(decrease in magnitude of change over trial s ) and found mean

red s cores to be cons is tently higher than mean green s cores

(p . O S on a one-tailed sign tes t ) . The s ub j e ctive reports

fol lowing the experimental session were cons istent with the

e l e ctrodermal evidence ; the red was various ly described as

more s timulating , awakening, attention-drawing, overpower­

ing , and lively than the green. Based on his s tudies of

these two colors , Wilson h ypothesized that the arousal value

of colors might form a U-shaped function within the visible

 7

wave length . His predicted relationship i s presented in

Figure 1 . I t should b e noted that this prediction is based

on Wilson ' s research with only two colors ( red and green)

and that these colors were identified only as being highly

s aturated but not otherwise des ignated according to any

s t and ard identification system.

FIGUR E 1

Arousal Values of Visib l e Hues

( Wilson, 1966 )

Red Violet

Orange Indigo

Yellow Blue

Green

Visib l e S pectrum

In an attempt to further c l arify the rel ationship

hypothes ized by Wilson ( 1966 ) , Nourse & Welch ( 19 71) s tudied

the GSR as a function of violet and green i l l umination, using

colored bulbs behind a viewing screen. During continuous

measur�ment of electrodermal activity fourteen subjects

received e i ther a series of G V G V G V or V G V G V G

l ights . A significant d ifference between the green and

 8

violet conditions was found for the first trial and that

difference was found to habituate to insignificant differ-

ences on later trials . Nourse & Welch (19 7 1 ) concluded that

viol e t was more arousing than green in support of Wilson ' s

(19 6 6 ) hypothesis . I t must be noted however that a spectral

analysis of the two light sources showed that the violet

'

condition emitted peak wave lengths at both blue (45 5 nm)

and red (677 nm ) while the green source produced a peak at

green only (5 35 nm). It was also found that the two condi-

tions produced different configurations o f illumination

within the presentation bo'x , a cross and an x, due to i:he

pl acement of the bulbs . It is possibl� that the differences

found were actually differences due to visual configuration

or a significant difference between red and green rather

than between blue-violet and green, in which case Wil s on ' s

(19 6 6 ) �ypothesis is not well supported b y Nourse & Welch

(19 7 1 ) .

The conclusions which were cited earlier in this

paper by Baker (19 6 1 ) and Birren (19 6 3 ) were based on re-

s earch undertaken by Rob �rt Gerard for his unpublished doc­

toral diss ertation in cl inical psychology at Un�versity of

California, Los Ang e l es (Gerard, 1958a ) . This research

was s ummarized by Gerard at the 6 9 th American Psychological

Association convention (Gerard , 1958b ) with implications

for Ro·rschach theory, psychodiagnos is , and color therapy.

Gerard ' s methodology included the projec tion of red, blue,

 9

and white lights of equal brightness on a screen in front

of 24 normal adul t males while measuring palmar conductance

(GSR ) , respiration rate, frequency of eye bl ink , systolic

blood pressure , heart rate, and electroencephalographic

measures ( alpha amplitude and alpha percent) . Gerard ( 19 5 8b )

reported that the autonomic nervous sys tem and visual cor­

tex w�re s ignificantly less aroused during blue than during

red or white i l l umination. A}s o , the subjective reports

from his subjects indicated greater relaxation, l e s s anxiety

and hostility, during blue; more tension and excitement

during red exposure . The following specific results were

found:

1. Using analysis of variance, s igni ficant color

effects were found with all dependent variables ,

e �cept heart rate .

2. Systolic blood pressure, palmar skin conductance

( GSR ) , respiration rate , and frequency of eye

blink s , were significantly lower during blue than

during red or white i l luminatio n . Alpha waves

from the visual cortex were present for a s ig ­

nificantly greater percentage o f the time during

blue than during white or red conditions , indi­

cating less cortical arousal during blue i l lu­

mination .

3. In terms of mean change from res ting level , red

was found to be more arousing than white, and

 10

white was found to be more arousing than b l ue ,

except in terms of al ph a ampl itude.

In an investigation of the rel ationship between ef­

fects on GSR .and subj ective reports when subjects were pre­

sented with three by five inch color plates , Smith (19 58 )

found that the hues rated high in excitory value produced

greater GSR's than the hues rated low in excitory value .

This s tudy us ed the entire ten Munse l l hues previously us ed

by Gujlford ( 1931). One criticism of Smith ' s s tudy is that

he el iminated s ix of his original twenty s ubj ects before

analysis of the data en t.�e apos teriori g rounds that their

GSR change s w��e insignificant. This had the obvious effect

of limiting the generality of his results to those s ub j ects

who show GSR,changes . Even though his results are limited,

Smith found that the least GSR change was associated with

blue, with increasing changes to pur ple , blue-green, yellow­

red, red, purpl e-blue, yellow, green-ye l low, red-purple,

and the highest GSR change with green. These findings are

not in agreement with other s tudies cited (Gerard, 1958;

Wilson , 196 6 ) but may be effected by the procedural error,

the colors us ed as s timul i , or the s i ze of the sampl e .

The effects o f color on capill ary blood-oxygen s at­

uration have been studied by Dous t and Schneider ( 19 5 5 ) who

us ed normal and psychiatric patients as subj ects . They found

that by means of autonomically contro l l e d vasoconstriction

all of the seven colors used, except white , lowered the

oxygen content of the blood, with the differences being less

 11

marked for hues at the blue-green end o f the spectrum and

most apparent with hues at the red end o f the s pectrum.

S i gnificant differences were found between blue and yellow

and between blue and green . They also noted that psychi­

atric patients displayed a lessened reaction to color than

the normal subj ects.

The absence o f a significant effect on heart rate

as a function o f either red or blue hues in the s ingle

study which related heart rate to color ( Gerard, 195 8 ) was

not expected because heart rate changes have been found in

other studies of the autouomic nervous system in which

color was not �sed as the independent variable. A general

autonomic response to environmental stimulation was re­

ported as early as 19 2 2 by Guildmeister ( 19 2 8 ) in which h e

measured changes i n pupilary res pons e , vasomotor functions,

skin temperature, GSR, , and heart rate . This l i st o f de­

pendent variables in the general autcnomic re sponse was

expanded by Landis ( 19 30 ) to include respiration, mus cular

activity, and changes in the pH of serum and saliv a . While

the general autonomic res ponse was not studied specifically

as a res ponse to color, these variables are all under con­

trol of the autonomic nervous system and have been shown to

reflect changes in various psychological processes including

visual perception and changes in affective states .

It should also be noted that a wide variety o f colors

have been used in the studies reviewed and that l i ttle

 12

attention has been paid to the s pecification of colors .

Gerard ( 19 5 8 ) .made an a priori s e lection of colors and pro-

ceeded to s pecify these colors as exactly as poss ible by

measurement. Two other researchers who s pecified their

colors ( Smith, 195 8 ; Guilford , 1931) Jhave done so according

to the Muns e l l notation and have equated their colors for

brightness ( Munsell value) and s aturation ( Munsell chroma ) .

T his procedure produced colors which would not normall y be

des cribed by observers as being red, yel low, green, or blue

even though the Muns e l l notations ( 5RS/ 6 , 5 Y5 / 6 , SG5 / 6 ,

SB5/ 6 ) suggest that they approximate these colors . After

examining these colors it is easy to sugg es t that this is

a potential source of insignificant resu l ts in these studies

s ince the Muns e l l colors used were neither vivid or s trong .

It is readily apparent that the s tudies reviewed do

not provide adequ��e evidence to reliably predict the dif­

ferential effects of color when these effects �re measured

as changes in the autonomic nervous s y s tem . Methodological

problems and lack of agreement in results have marked the

major studies in this area ( Gerard, 195 8 ; Smith, 195 8 ;

\

Wilson, 1966 ) . Figure 2 presents a s umm a ry of these studies

and the effects found .

 13

FIGURE 2

S umm ary of Previous Research

Dependent Independent
Variable Source Variable Results

GSR Gerard ( 19 5 8 ) red, blue , white Greater GSR change

with red than blue.
White located be­
tween red and blue
in terms of arou­
sal val '.�e.

Smith ( 19 5 8 ) ten Muns e l l hues Rank order of

increasing GSR
change for psycho­
logical primaries :
blue, red, yellow,
green.

Wilson ( 1966) red, green Greater GSR change

with red than
gree n .

Heart Rate Gerard ( 19 5 8 ) red, blue , white Ins tgnificant

change .

R �spiratio� Serard (19 58) red, blue, white Respiration in­

creased with red
and decreas ed with
blue when compared
to white i l l umin­
ation .

Even though these results are not comp letely in agreement,

there is ev idence presented which suggests that differen-

tial color effects exis t . Using the four pri mary psycho-

logical colors , there is not enough evioence to predict exact

relationships , except red is usually found to be more arou-

sing than the other hues when GSR is used as the dependent
 14

variable, except in the study by Smith ( 195 8 ) . The s e

s tudies have generally contrasted one c o l o r against another

and measured one variable , usually GSR.

S tatement of Hypothesi� .•

This investigation was undertaken to specifically

s tudy the differential effects in GSR, heart rate, and res­

piration rate as a function o f the four primary psycholog­

ical colors . The use of four colors, rather than a s tudy of

two colors , was used to provide a better understanding of

the relationship between the colors in the visible spectrum.

Rather than l imit the measurement to one variabl e , the var­

iables of GS�, heart rate, and respiration rate were selec­

ted to provide added evidence and understanding to the s tudy

of the effects of color on the human body.

Toe fol lowing research hypo these� were formulated to

s tudy the differential effects of color:

1. Tbe four primary psychological colors will be

associated with differential effects in the ampli­

tude o f the g alvanic s kin response (GSR).

2. The four primary psychological colors will be

associated with differenti�l effects in the heart

rate .

3. Tne four primary psychological colors will be

associated with differential effects in the rate

of respiration.
 15

CHAPTER II

E XPERIMENTAL METHODOLOGY

Subjec ts.

Subjects for this s tudy were 2 4 male sol l e ge s tu dents

recruited as unpaid volunteers from undergraduate Psychol ogy

classes at Eas tern I l l i nois University. The median age for

subjects was 20 years with a range of 17 to 27 years.

S timuli Colors.

The f ol l owing method was used to select the four pri­

mary psychol ogical col ors which served as i ndependent var­

iables in this s tudy. Ten male observers who did not par­

ticipate in this s tudy were asked to select from the ISCC-NBS

Color Nam�� Chart (Inter S ociety Color Council-Nati onal

Bureau of S tandards , NBS S tandard S ample No. 2106) those

col ors which they considered to be the bes t examples of

red, yellow, green, and blue. Eight of the ten obs ervers

u nanimous ly agreed on the col ors listed i n F i gure 3.

FIGUR E 3

S timuli Col ors

Color IS CC-NBS Designation. Muns e l l Renotation

Red 11 v. red 5 . 0R 3 . 9/15 . 4

Yel l ow 82 v . yel l ow 3 . 3Y 8. 0/14 . 3

G+een 139 v. green 3 . 2G 4 . 9/11. 1

Blue 178 s . blue 2. 9PB 4 . 1/ 10 . 4

 16

It should be noted that in all cases the observers sel ected

the most s aturated colors available in each hue ( highest

M�?s e l l chroma) which was at leas t ten s teps away from gray

on the M�ns e l l notation. A wide variation in brightness

( Myns e l l value) was selected, ranging from Muns e l l 3 . 9 for

red to 8 . 0 for yel l ow . Similar results would be predicted

by the research of Granger (195 2 ) . The ISCC-NBS designa­

tions (Kelly & Jud d , 195 5 ) indicate that the colors selected

have been described either as vivid (v . ) or s trong (s . )

e xamples o f these colors according to the IS CC-NBS system.

These s elected col o =s ware then obtained from the

Muns e l l Color Corporation as 9 by 12 inch glossy plates .

These color · pl ates were photographed on E ktachrome-X 35

mil limeter film under artificial lighting balanced to day­

light conditions by using an 85c f i l ter on the camera lens

to obtain an accurate exposure .

The s timulus colors which had been reproduced on

s l ides were projected on a 3 by 4 foot projection screen

located ten feet in front of the subjects by using a Kodak

Carousel projector. A second projector was used to project

a white inter-slide interval between coior presentations .

A dissolve unit (Kodak, model 2 ) was used to dissolve be­

tween projectors , thereby avoiding the potentially disrup­

tive effect of a s l ide change on the screen. The actual

screen area used for tne proj ection o f the colors and inter­

trial white was 2 3 by 33 inches .

 17

Using the diagonal visual ang le formul a of Cornsweet (1970 ) ,

these conditions produced a visual s timulus field of 1 8

degrees .

To insure that the s l ides used as s t imulus colors

produced equal levels of il lumination on the screen, a photo­

graphic light meter (Sekonic S tudio Delux, model L-28c) was

used to measure incident light s triking the screen. All

colors were found to produce 16 foot candles of illumina-

tion. In order to equate the illumination from the pro-

j ector which was used to present the inter-stimulus condi­

tio n , a pair of polaroid filters was placed in frollt o f the

pro j ector lense . By independently rotating these filters ,

it was possible to produce a continuous graduatiou o f trans­

mission. Us ing the s ame procedure as noted abov e , the second

projector was equated with the first projector to produce 16

foot candles of illumination at the screen . After these

conditions were reached, no further alterations were made

during the experimentation .

Phys iological Apparatus .

Phys iological measurements were recorded on a four

channel physiograph (E & M Ins truments , model DMP -4A) which

was located in the experimental room to the right of the

subjects . An exosomatic direct current (40 microamperes )

GSR was measured from the dorsal and ventral surfaces o f the

middle digit of the right middle fingers us ing commercially

availab l e lead ( Pb ) electrodes with a surface area of 3/4 by

 18

1 inch and a s tandard GSR preamplifier. Heart'rate was

monitored by use of a l ocally manufactured photoplethysmo­

graph ( PPG) attached to the left middle or left ring finger

depending on which finger bes t accomodated the apparatus .

Previous experience with this equipment ( Boyce & Jacobs ,

19 7 2 ) demons trated that a reliable measurement of heart

rate c ou l d be �btained by these means when a high gain pre­

amplifier ( E & M Instruments) was used in the EMG mode ( time

constant of 0 . 0 3 seconds ) . Respiration was measured by

means of a thermistor located below the right nostril and

held in place by attachment to eye g l as s es f ra�es from which

the lenses had beer. removed. Respiration was therefore re­

corded as changes in temperature between inhaled and exhaled

air ( Boyce & Jacobs , 19 72 ) . The thermistor circuit was

attached to a secon d high gain preamplifier used in the EKG

mode ( time constant of 2 . 0 seconds ) .

In an attempt t o produce constant conditions during

the recording of the phys iol ogical measurements , auditory

distractions were reduced as much as possib l e . A l l experi­

mental sess i ons were held between 1 PM and 9 PM since this

period coincided with the least use of the psychology l abor­

atory. An air conditioner was installed in the experimental

room prior to experimentation in an attempt to produce con­

s tant temperature conditions . This air con ditioner was

operated at a constant high speed throughout all experimen­

tal sess ions to equate auditory level for all subjects and

also had the effect of masking extraneous auditory cues .

 19

The desired temperature effect was not totally achieved

since the range in temperature for the experimental room was

from 74 to 9 2 degrees Farenheit with a median temperature

of 85 degress Farenheit. However, the temperature for any

s ingle session did .not vary more than plus or minus one de­

gree Farenheit during the s e s s io n . This temperature range

would not be expected to effect the results as long as the

variation is limited within each experimental session.

Procedure .

To preclude the pos s ibility of defective color vision,

each subject was pre-tested for color vision using Dverine

Pseudoisochromatic P lates ( Dverine, 1963a) . Each subject

was required to identify correctly at least 13 of the 14

plates in the Dverine test series , even though this criterion

is more rigid than required in other applications of this

tes t ( Dvorine, 1963b) . This test was necess ary since approx­

imately 8% o f the males in the United S tates have some form

of congenital color blindness ( Judd & Wyszecki , 1963) .

Color vision testing was accomplished immediately

upon arrival o f the subject in the e xperimental room. The

experimental room was a 12 by 15 foot research room which

contained a recl ining type chair, the screen, the air con­

ditioner, a table with physiog raph and tape recorder, and a

thermometer for recording room temperature . The preampli­

fiers and Dverine tests were located on a small table adj a­

cent to the subjects ' chai r . The thermometer was located on

 20

the wal l farthest from the air conditioner and ·closest to

the s ub j e c t . The proj ection apparatus was located in a

closet behind the subjects and was s i tuated s o as to pro j ect

onto the screen from above and to the subjects' l e f t .

Fol lowing the color vision tests, subjects were

seated and s tandard instructic�s were presented by tape re­

cording. These ins tructions indicated that the purpose of

the investigation was to see what effects various kinds of

pictures have on the nervous system. They were asked to

relax for ten minutes , after which time their basal measure­

ments would be taken, and Lhen the s t udy would begin. They

were instructed to �elax, avoi d talki�g, and avoid moving

after the apparatus was connected. The subjects were in­

formed that colors would be presented so that they had some­

thing to look at during the basal measurements but were not

instructed to respond to the colors . These instructions

were designed to disguise the true nat�re of the s tudy even

though previous investigators did not feel this was necess ary

( Gerard, 1958; Wilson, 1966) . The ins tructions g iven are

included in Appendix I .

Each subj ect was connected t o the apparatus at the

end of the tape recorded phase of the instructions . General

room illumination was accomplished by a covered sixty watt

bulb located on the ceil ing and remained at this low level

throughout the experimental sessions . The screen was i l l u­

minated with the inter-s timulus white when the subj ect en­

tered the room and did not change to a color until after the
 21

apparatus had been connected, calibrated, and the subject

had remained resting in the chair for 10 minutes. This

delay was used as a further control to allow the subjects'

heart rate, GSR, and respiration to return to a normal

resting level prior to experimentation. After this ten

minute delay, the experimenter quietly announced that he

was now ready to obtain the basal rate.

Each color was presented one time to each subject,

for one minute, followed by non-colored illumination for one

minute during the inter-trial interval. Four sequences of

color presentations were used to insure that each color

appeared an equal number of times in each position. Each

experimental subject was assigned to a color sequence in the

order of his appearance at the laboratory. In this way one

fourth of the subjects randomly received each sequence of

color presentation. The four color sequences are noted in

Figure 4 .
 22

FIGURE 4

Order of S timulus P resentations

Time S equence S equence S equence S equence

( in minutes ) 1 2 3 4

10 White White White White

1 Red Blue Green Yellow

1 White White White White

1 Blue Yellow Red Green

1 White White White White

1 Green Red Yellow Blue

1 White White White White

1 Yellow Green Blue Red

1 White White White White

1 NVS • NVS • NVS• NVS •

1 White White White White

*NVS = novel visual s timulus

These color s equences were randomly determined as was

required by the latin square design ( Edwards, 196 8 ; Myers ,

1966 ) .
 23

FIGUR E 5

Latin Square

Original Latin Square A B c D

( Edwards , 196 8 ) B D A c

c A D B

D c B A

Assignment o f Colors
to Treatment Groups

Sequence 1 Red Blue Green Yellow

Sequence 2 Blue Yellow Red Green

S e quence 3 Green Red Yel l o w Blue

S equence 4 Yellow Green Blue Red

A novel visual s timulus which was unrelated to the

s timulus col ors was inserted following the presentations of

all colors . This was done to insure that recording appara-

tus was operative in the event that the GSR or other mea-

sured demonstrated complete habituation as the result o f

the repeated trials .

Each s lide change was recorded on the event channel

of the phys iography by connecting both the s l ide change con-

trol and the physiograph event marker to an external s witch

which was activated mannually by the experimenter at one

minute interval s .
 24

CHAP TER III

RES ULTS

The G S R was defined as the maximum change in skin

resis tance which occured during the first 15 seconds f o l ­

lowing presentation of each s timulus color. This change was

converted to a log conductance measure ( the log arithm to the

base ten of the reciprocal of the resis tance ) prior to s ta­

tistical analysis of the data. To facilitate the analysis

of the data, the conductance was multiplied by 100 prior to

log transformation to eliminate unnecessary zeros . To elim­

inate the infinitely negative log arithm which would have

resulted from a log transformation of a zero resis tance

change , one ohm resis tance was added to the resis tances of

a l l scores prior to obtaining conductance . This procedure

was recommended by Edwards ( 19 5 0 ) as the best method o f

obtaining meaningful logarithms when the original data con­

tains zeros . The effect obtained was a logarithm of zero

when the original resis tance change was zero . While these

procedures do not produce strict log conductance measure­

ments they have been recommended and used by Nichols &

Daroge ( 19 5 5 ) . Previous research ( Haggard , 1949 ; Keuchen­

meister, 19 70; Nichols & Darogue, 195 5 ; Schlosberg & S tanley,

195 3 ) suggests that a log conductance measure is the bes t

measure for quantification of GSR data since i t provides :

additivity ( equal interval data) , normal ity ( s cores which

are normally dis tributed ) , homogeneity of variance ( common

 25

variances for variables and error es timates ) , and indepen­

dence of means and variances ( variances are unaffected by

changes in the mean ) .

Heart rate measurements were made during the firs t

30 seconds following the pres entation o f the color and were

transformed to obtain a beat per minute rate by multiplying

these measurements by two . Respiration rates were also

transformed from the first 30 seconds after presentation of

the s t imulus colors to obtain a cycles per minute rate .

S imilar meas urements for heart rate and respiration rate

were made during the last 30 seconds of the inter-stimulus

cond itic�. To s tudy the �ctual �ffects of the colors , the

data analysis was conducted on the difference between the

pre-stimulus condition and the color condition. This data

met the requirements for analysis of variance without fur­

ther trans formation . Howeve r , to e l iminate the necess ity

o f analyzing negative as wel l as pos itive changes , a con­

s tant of 20 u�its was added to each difference figure. This

procedure is consistent with the recommendation of Edwards

( 19 5 0 ) .

T o provide a s tatistical test o f the formal hypothe­

ses , a l atin square analysis of variance was employed

( Edwards , 196 8 ; Myers , 1966 ) . This latin square is s hown

in figure 5 . The replicated 4 by 4 latin square was selec­

ted as the appropriate s tatistic to test the effects of

color , the order o f presentation, and the variance attribut­

able to subj ects . With each s ubject receiving one of the

 26

four s equences of s timulus presentations ( treatments ) noted

in figures 4 and 5 , each latin square accomodated four s ub ­

j ects . The same l atin square was replicated five additional

times to accomodate the twenty-four s ub j ects in this s tudy.

This procedure produced an error term with more than 30 de­

grees of freedom and increased the likl ihood that signifi­

cant treatment differences would be declared significant

( E dwards , 1968 ) . A significance level of . 0 5 was es tab­

lished prior to analysis of the data as the level required

to confirm each hypothes is. This procedure was repeated

using the s ame latin square design for each of the dependent

variables ( GS R , he�rt rate, and respiration ) . The result

o f this procedure produced a seperate analysis of variance

for GSR , heart rate, and respiration to test each of the

respective hypotheses .

The analyses of the effects o f the four colors on GSR,

heart rate, and respiration rate are summarized in Tab l e 1,

with analysis of variance tables for each dependent variab l e

included as Appendix II.

T ABLE 1

S ummary of Physiological Changes

( F ratios )

Measure Color Column

GSR 3 . 35 9 * 2 . 491
Heart Rate 0 . 263 0 . 344
Respiration Rate 1 . 606 3 . 286 *

•p <.05
 27

The four pri mary psychological colors were found to be asso-

ciated with differential effects in ampl i tude of GSR. The

effects of trials (columns ) for GSR was not s ignificant in-

dicating that there was no s ignificant change in the response

across trials as might be expected due to habituation o r

fatigue .

The nature of the differential color effects on GSR

are presented in Figure 6 . It is apparent that the mos t

arousing color was red , fol lo wed by green and yellow, with

blue being the l e as t arousing of the four colors .

FIGUR E 6

Arousal Values ( Conductance ) of Four Colors

J
u
::s
'O
c
0 4 . 000
u
O'l ( 3 . 35 0 2 )
0 .....
r-i..-..
QJ
� rog
3 . 000
-- --".
- - ( 2 . 84 8 1 )
aJ -- '

e +' ( 2 .4466)
.._.. '
'
2 . 000
( 1 . 9204)

1 . 000

o . ooo

RED YELLOW GREEN BLUE

To tes t the s ignificance of the differences between arousal

values obtained , a series of t tests ( two tai l ) for correl ated

 28

data was employed using 2 3 degrees o f freedom . · S ignificant

differences at or exceeding the . 0 5 level were obtained be-

tween red-blue, green-blue, and yel l ow-red . O ther differ-

ences were not s ignificant on a two tail t tes t . Using a

one tail tes t the differences between red-green and yellow-

blue were s ignificant at the .05 leve l . This data is pre-

sented for informational purposes in Table 2 .

TABLE 2

Difference in Arousal Value of the Four Colors

Color Combination Difference t ratio Level o f

( Mean l o g C ) S i lJnific�nce

Red - Blue 1 . 42 9 7 5 .31 .005

Green - Blue 0 . 9277 3 . 0627 .01
Red - Yellow 0 . 9 135 ·2 . 4141 .05
Yellow - Blue 0 . 5262 1 . 76 8 2 . 10
Red - Green 0 .5021 1 . 8180 . 10
Green - Yellow 0 . 4015 1. 7074 NS

A l l comparis �ns based 0n 2 3 degrees cf freedom.

The analys is of the effects of color on heart rate

does not support a differential color effect in that system.

Neither the effects o f color nor the effects o f columns was

s ignificant, sugges ting that heart rate changes are indepen-

dent of experimental effects under the conditions investigated.

The analysis of the effects of color associated with

changes in respiration rate s howed insignificant color ef-

fects . The s ignificance of the column effect suggests that

 29

a rel ationship exists across trials without regard to the

colors presented . These trial effects , disregarding col o r ,

are presented in Figure 7.

FIGURE 7

Changes in Respiration Rate by Trial

+2

c-
0 Q)
-.-i ..µ
..µ :::3
cO c +1
s... ·r-l
-.-i El
0..
fll s...
( +0 . 3 3 )
Q) Q)
a:: 0.. /
0 - - - -
- - - - - - - -
- -

c fll
-.-i Q)
.-I ( -0 . 4 2 )
Q) u
Ol >'1
��
.c
-1
u Q)
..µ
c cO
cO a:: (-1.75)
Q) -2
::?::

1 2 3 4

I t is apparent that the initial s timulus was associated with

a mean reduction in rate o f· respiration (-1.75 cycles per

minute ) , with minimal reductions on the second and third

trials ( -0 . 4 2 cpm) , and a mean increase in respiration rate

( +0 . 3 3 cpm) on the fourth stimulus .

 30

CHAP TER IV

DISCUSSION

Color Effects .

The hypothes ized differential color effects were

found to be s ignificant for the GSR , with changes in heart

rate and respiration being insignifican t . The s trict inter­

pretation of these findings must therefore be that the four

primary psychological colors used in this s tudy are associ­

ated with differential effects in the galvanic skin res pon­

s e s ( GS R ' s ) which are elicited with the introduction of the

stimulus colors , and that comparab l e �hanges i D h e � Lt r8t�

and respiration rate were not found in this s tudy.

It should be noted that the popularity o f the GSR in

psychological research is partially due to its sens itivity

as a dependent variab l e . Heart rate and respiration rate

have received considerably less attention s ince they are less

respons ive measures . This research does not es tablish that

comparabl e changes in heart rate and respiration rate do not

exis t , but only that these effects were not found under the
'
condit ions used in th�s study.

The relationship found betwe en red and blue is con­

s is tent with several previous s tudies ( Gerard , 19 5 8 ; Wilson,

1966 ) in which red has been found to be more arousing than

blue. With the high arousal attributed to red and a l o w

arousal attributed to blue , a t the distant end o f the visible

s pectrum, it might have been hypothes ized that a fairly

 31

linear rel ationship exists between arousal value and wave

length in which the most arousing color is red , with de­

creasing arousal for yellow, green , and blue . As was noted

in Figure 6 , such a relationship was not found in this s tudy

since green was found to be more arrousing than yel lo w .

Even though the l inearity o f this relationship can not b e

inves tigated empirically from this s tudy s ince the spectral

values of the colors are not known exactly, it is obvious

that arousal value is not a linear function o f wave length,

especially with s ignificant differences es tabl ished between

each of the colors .

The f in�ing by Gerard ( 19 5 8 ) that red was more arousing

than blue has been supported in this research. The non­

linear U-shaped rel ationship between arousal values and wave

length ( F igure 1 ) hypothes i zed by Wilson ( 19 6 6 ) is not sup­

ported by this s tudy. Wilson ' s hypothesis was developed

from a s tudy of only two colors and was extrapolated to in­

clude the remainder of the visible spectrum . It should be

noted that the significant difference that Wilson found be­

tween red and green, as wel l as the higher arousal value for

red than green, is supported.

In a comparison with the s tudy by Smith ( 19 5 8 ) , the

present results are not in agreement . Smith ' s arous al values

produced a scatter across the visible spec trum but show green

to be most arous ing , followed by yellow and red , with blue

the least arousing of the colors which were used in both

s tudies . The individual relationships between red and blue

 32

and between green and yellow are confirmed i n this s tudy but

the overall rel ationships between the colors are not com­

parab l e . These differences may be attributed to the colors

used by Smith ( unsaturated and equated ) , to his methodology,

o r may be true when only those who respond to color are used

in the analys is .

It is difficult to relate the findings of Nourse &

Welch ( 19 7 1 ) to the present s tudy s ince their colors were

green and purple, only one of which was used in thls study,

and their purple contained components of red and blue . Their

findings that purple is more arousing than green is supported

only for the red ccmponent of purple with conflicting find­

ings for the relationship between blue and green . The re­

sults of the present s tudy support the criticism that the

higher arousal value found by Nourse & Welch for purple

than green is actually a higher arousal value for red than

blue .

The l ack o f a significant color effect on heart rate

is cons istent with the single s tudy which used heart rate as

the dependent variab l e for color ( Gerard, 195 8 ) . The lack

of a significant effect on respiration however , disagrees

with the significant color effect obtained by Gerard ( 19 5 8 ) .

This difference may be partially explained in that Gerard

used longer periods of color presentatioP and several pre­

sentations of each col o r . In contras t , this s tudy sought to

inves tigate the initial effects of the colors which are as­

sociated with c hanges in respiration f o l l o wing a single

 33

presentation o f e ach colo r . I t is possible that respiration

effects do not occur initially in the organism but require

long exposure to the color .

Response Habituation.

It is wel l known that the responses to repe ated pre­

sentations of a s timulus habituate� that is the responses

diminish with repetition of the s timulus . In an investi­

g ation of the orienting response ( OR ) , Lynn ( 19 6 6 ) reported

total habituation ( to extinction) of the GSR after s e ven o r

eight presentations o f an auditory s timulus , with compl e te

habituatio� of the OR in all sys tems �easured by 10 to 30

repetitions o f the same s timulus .

Again disregarding color effects , the habituation to

repeated s timulus presentations is presented in Table 3

along with the reco very of the response when the novel visual

s timulus is presented . In the interpretation of these re­

sults i t must be noted that the inter-trial intervals be­

tween each tabled value represents a white s timulus . There­

for e , the habituation to the second color is the habituation

to the third visual s �imulus . It should also be noted that

this table presents the percent of subjects s howing total

habituation ( to extinction) not merely decrements in magni­

tude of response as wel l as subjects showing no respo�se on

any trial .
 34

TABLE 3

Response Habituation

Habituation to Extinction by Trial

( Percent of subjects showing no respc�s e )

Novel
Co lor 1 Color 2 Color 3 Color 4 Visual
S timulus

GSR 13% 20% 3 3% 30% 8%

Respiration 25% 54% 5 0% 54% 42%

Heart Rate 38% 25% 25% 50% 42%

One of the implications for this data is that if a

response is absent in a s ignificant perce�tage o f the sub-

j ects due to early habituation, it becomes extremely diffi-

cult to reach s ignificance in the resul ts . The habituation

of respiration and heart rate responses very likely contrib-

uted to the lack of significance in the results obtained in

this s tudy, since respiration rate changes become habit-

uated to extinction for 50% of the subjects by the second

color and heart rate changes reach this criterion by the

fourth colo r .

Response Sys tem Specificity .

Recent research has shown that a biological response

does not necessarily occur in all sys te:ns to the same ex-

ternal s timul ation. In a typical inves tigation of the

 35

orienting response ( O R ) by Voronin & Sokolov ( 1960 ) , 100

human subjects were subjected to a 1000 cycle tone, with

OR ' s observed as EEG changes in 95 subjects , respiration

rate changes in 6 3 subj ects , eye movements in 42 subj ects ,

GSR ' s in 95 subj ects , and only 11 subjects showed responses

in a l l of these sys tems . Comparab l e results were obtained

by Voronin, Sokolov , & Bao-Khua ( 19 5 9 ) in which the first

presentation of a 55 db tone elicited the OR in GSR for

95% of their subj ects .

These findings �re in agreement with the general

patterns of phys iological activation which have been in­

vestigated by Lacey and others ( Lacey, 195 0 ; Lacey, Bateman ,

& Van Lehn, 195 3 ; Lacey & lacey, 195 8 ; S chnore, 195 9 ;

McNulty & Noseworthy, 1966 ) . Marked individual differences

have been reported between subjects with respect to which

phys iological measures show the greatest change under s tan­

dard conditions of s timulation. These authors have also

agreed that the patterns of phys iological activation are

rel atively cons is tent across qualitatively different s i tua­

tions for any particular individual .

When the results of this s t udy are viewed in terms o f

response sys tem specificity, comparable resul ts are found .

Table 4 presents the number and percent of subjects respon­

ding to · the firs t trial of color and to the novel visual

s timulus . It is apparent that the most responsive dependent

variable is GSR , with heart rate as the leas t responsive

 36

system. The presence of heart rate changes in only s ixty-

two pe�cent of the total subjects may be assumed to have

contributed to the lack of significant color effects for

heart rate . The inclusion in the data of measures from

non-responsive sys tems may contribute to the lack of sig-

nificance as noted by McNul ty & Noseworthy ( 19 6 6 ) and is

especially relevant to this · s tudy of respiration and heart

rate.

TABLE 4

Response System Specificity

Response to Novel
Response to 1s t color Visual S t imu1·1.1s

Dependent Number of Percent of Number of Percent of

Variable Subjects Subjects Subjects Subjects
Responding Res ponding Res ponding Responding

GSR 21 88% 22 92%

Respiration 18 75% 14 5 8%

Heart Rate 15 62% 14 58%

The appearance of the results at this point would

suggest that differential color effects might be associated

with changes in heart rate and resporation rate for those

subjects who consis tently res pond in these two. sys tems .

However, since this hypothesis was not a part of the original

design, the s tatistical procedures used do not lend them-

s e lves to this inves tigation . By el iminating non-responding

 37

subjects from the data for each sys tem the subjects would

no longer be randomly assigned to treatment groups and the

latin square would no longer be complete .

Sugge s tions for Further �esearch.

The next logical question to b e asked after the

analysis of the results for this s tudy is whether those

subj ects who respond in terms of heart rate and respiration

rate should show diff crenti�l color responses in thos e sys­

tems . This would require increasing the total number of

subjects and e liminating those who do not res pond in the

sys tem being measured . Even though this procedure is not

consistent with the latin square s tatistic used in this

study, a repeated measures analysis of variance may be used

or a large number of subjects who have been shown to respond

in the sys tem may be independently as s igned to different

color groups to allow analysis of independent measures . The

measurements of heart rate and respiration rate may also be

made more sophisticated and accurate by use of commercially

avail ab l e apparatus designed for precise measurements such

as beat to beat heart rate.

Expansion o f the dependent variables which are asso­

ciated with color effects is also needed . In addition to

more extensive s tudies of heart rate and respiration rate

changes , more s tudies are needed in blood pressure, blink

rate, EEG changes , EMG activity, and capi l l ary oxygen con­

tent, since no more than one s tudy has been located which
 38

reports the effects o f color on each o f these variables . The

results of this s tudy also suggested that colors might be

associated with changes in vasomotor functions . A digital

vasomotor measuring device ( photo-plethysmograph) was used

in this s tudy for convenience in measuring heart rate with

no attempt to establish accurate vasomotor measurements since

the a priori conclusion was made that precise measurements

of vasocongestion and vasodilation would not be obtained

with this apparatus . Another question raised by this s tudy

is whether the colors would be associated with differential

effects in the occurance of spontaneous skin potential res­

ponses . Thi3 area has not been investigated previous l y and

such effects might wel l be found under prolonged exposure

to various colors .

The present line of research using more than two

colors to study the relationships between the colors in

phys iological effects definately des erves more experimental

inves tigatio n . The s tudies reported in the l i terature which

have compared two colors and extrapolated from them to other

hues in the visible s pectrum are necessarily l imited by the

use of two colors . The use of four colors in this s tudy is

not intended to provide defini tive results across the vis­

ible spectrum but only to take one step in that direction.

A further criticism of previous research has been

that only two investigators ( Guilford , 1 9 3 1 ; Smith, 195 8 )

have used identical colors or even the s ame procedure for

obtaining the s timulus colors . Orily one investigator has

 39

fully specified the spectral characteris tics of his colors

( Gerard, 19 5 8 ) and few others have reported the peak trans­

mission wave lengths of their colors ( Nourse & Welch, 1971) .

I£ the results of one s tudy are to be accurately compared

with another s tudy i t is L�perative that the colors be

identical in terms of hue, s aturation, tint, and intens ity

of presentation. The specification of the colors also facil­

itates the investigation of the relationship between colors

and wave length when several colors are used in the same

s tudy.
 40

REFERENCES

Baker , S . Visual persuas ion. New York : McGraw-Hil l , 196 1 .

Birren, F . Col o r : a survey in words and pictures . New

York: University Books , 196 3 .

Boyce , L . & Jacobs , K . Voluntary control o f GSR us ing

continuous visual display with s imul taneous changes
in heart rate and respiration . Unpublished research,
Eastern I l linois University, 1972 .

Cornsweet, T . N . Visual perception. New York : Academic

Press , 1970 .

Dous t , J . W . & Schneider , R . A . S tudies on the phys iology

of awareness : the differential influence of color
on capill ary blood-oxygen s aturatio n . Journal of
Clinical Psychology, 1955 � 11, 3 6 6 - 3 70 .

Dvorine , I . Dvorine psuedo-isochrom�tic pl ates , New York;

Harcourt, Brace, and World, 196 3 a .

Dvorine, I . Quantatative class ification o f the color-bl ind .

Journal o f General Psychology, 1963b, 68 , 2 5 5 -265 .

Edwards , A . L . Experimental des ign in psycholoqical research.

r-
New York : Hol t , Rinehart, anc winston, 195 0 .

Edwards , A . L . Experimental design in psychological research.

( 3 rd ed . ) New York: Hol t , Rinehart, and Winston,
196 8 .

Eysenck , H . J . A critical and experimental s tudy of color

preferences . American Journal o f Psychology, 1941,
5 4 , 3 8 5 - 39 4 .

Fere, c� L a pathologie des emotions . Paris : Alcan, 1892 .

Cited by c . Landis . Psychology and the psy�hogalvanic
reflex. Psychological Review, 1930 , 3 7 , 3 8 1-3 9 8 .

Gerard, R . Differential effects o f colored lights on

psychophys iological functions . u.npublis hed doctoral
dissertation, University of California, Los Ang e l es ,
195 8 .

Gerard, R . Color and emotional arous al . American Psycho­

logi s t , 195 8 , 13 ( 7 ) , 340 . ( Summary)

Guildmeister, M . Handbook of normal an� pathological

phys iology, 19 2 8 , 8 , 6 5 7-702 , 766-784 .
 41

Goldstein, K .. The o rganism. New York : American Book , 19 3 9 .

Goldstein, K . Some experimental observations concerning the

influences of colors on the functions of the organism.
Occupational Therapy1 1942 , 2 1 , 147-15 1 .

Granger, G . W . Obj ectivity of color preferences . Nature,

195 2 , 170 , 788-780 .

Guilford, J . P . The prediction of affective value . American

Journal of Psychology, 19 3 1 , 4 3 , 469-478 .

Hagg ard, E . A . On the application o f analysis of variance

to GSR data: I the s election of an appropriate
measure. Journal of Experimental Psychology, 194 9 ,
3 9 , 3 78-3 9 2 .

Hevner , K . Experimental s tudies of the affective value of

colors and lines . Journal of Appl ied Psychology,
1935 , 19 , 3 8 5 - 3 98 .

Inter Society Color Councj l & NR tional Bureau of S tandards .

ISCC-1.JBS Centroid Color Charts . ( NBS S tandard Sam­
ple N o . 2106 ) Washing ton, D . C . : 1 National Bureau
of S tandards , nd .

Jud d , D . B . & Wyszeck i , G . Color ir. business , sc�ence , and

indus try. New York: Wiley and Sons , 196 3 .

Kelly, K . L . & Judd , D . B . The ISCC-NBS method of desig­

nating colors and a dictionary of color names .
( N ational Bureau of S tandards Circular 5 5 3 ) Wash­
ington, D . C . : National Bureau of S tandards , 195 5 .

Keuchenmeister, c. A . Instrument for direct readout o f log

conductance ( log 1/R ) measures of the galvanic skin
respons e . Psychophys iology, 1970 , 7 ( 1 ) , 128-134 .

Lacey, J . I . Individual differences in somatic response

patterns . Journal of Comoarative and Physiological
Psychology, 195 0 , 4 3 , 3 38-35 0 .

Lacey, J . I . , Bateman , D . E . , & Van Lehn, R . E . Autonomic

response specificity: an experimental s tudy.
Psychosomatic Medicine, 195 3 , 15 , 8-2 1 .

L�cey, J . I . & L3cey, B . C . Verification and extension of

the principle of autonomic response s tereotypy.
American Journal of Psychology, 195 8 , 7 1 , 5 0 - 7 3 .

Lynn, R . Attention, arousal and the orientation reaction.

New York: Pergamon, 1966 .
 42

Landis , C. Psychology and the psychogalvanic reflex.

Psychological Review, 193Q, 3 7 � 128-134 .

Myers , J . L . Fundamentals o f experimental design . Boston:

Allyn and Bacon, 196 6 .

McNul ty , J . A . & Noseworthy, W . J . Physiological response

specificity, arousal , and task performance . Percept­
ual and Motor Skil ls , 1966 , 2 3 , 9 8 7 - 99 6 .

Nakshian, J . s . The effects o f red and green surroundings

on behavior . Journal of General Psychology, 196 4 ,
70 , 14 3-16 1 .

Nichols , R . C . & Daroge , T . An electronic circuit for the

measurement o f the galvanic skin respons e . American
Journal of Psychology, 195 5 , 6 8 , 4 5 5 -46 1 .

Norman , R . D . & Scott, W . A . Color and effec t : a review

and s emantic evaluation . Journal of General Psycho­
logy, 195 2 , 4 6 , 185 - 2 2 5 .

Nours e , J . c . � Welch, R . B . EmotioPal attributes of color:

a comparison of violet and green. Perceptual and
Motor Skil l s , 19 7 1 , 3 2 , 403-406 .

Odbert, H . S . , Karwoski , T . F . , & Eckerso n , A . B . S tudies

in synesthetic thinking : I. musical and verbal
associations of color and mood . Journal of General
Psychology, 1942 , 26 , 15 3-173 .

Pressy, S . L . The influence of color upon mental and motor

efficiency. American Journal of Psychology, 1 9 2 1 ,
3 2 , 3 2 6 -3 5 6 .

Schlosberg , H . & S tanley, W . A simple tes t o f the normal­

ity of twenty-four dis tributions of e lectrical skin
conductance . Science, 195 3 , 117 , 3 5 - 3 7 .

Schnore, M . M . Individual patterns o f physiological activity

as a function of task difference� and degree of
arous a l . Journal of Experimental Psychology, 195 9 ,
5 8 , 117-12 8 .

Smith , R . S . An investigation of the relationship between

phys iological and cogn itive measures of the affective
response to color . ( Doctoral diss ertation, University
of Pennsylvania) Ann Arbo r , Michigan: University
Microfilms , 19 5 8 .
 43

Smith , J . M. Th e relative b right ness va lues of three hues

for newborn babies . Univ e rs ity of Iowa Studies of
Infant Behav ior III , 19 36 , 12(1 ) .

Sokolov , E . M . High er nervous funct ioning: th e orient ing

reflex . A nnua l Rev iew of Phys io logy, 196 3 , 2 5 ,
5 4 5 -5 8 0 .

Trola nd , L . T . Rep o rt o f c ommitt ee o n Col imet ry . Journa l

of th e Opt ica l Soc iety of America , 19 2 2 , 6 , 5 2 7- 5 96 .

Voronin, L . G . & Sokolov , E . N . Cort ica l mecha nisms of th e

o rient ing reflex and its relat ion t o th e c ondit ioned
reflex . In H . H . Jasp er & G . D . Smirnov (Eds . )
Moscow c o l loquium on e l ect roencepha lography of h igh er
nervous act iv ity. EEG Clinica l Neurophy s iology,
196 0 , Supp l . 13 .

Voronin, L . G . , Sokolov , E . N . , & Ba o-Khua , U . Ty pe fea ­

t ures of th e orient ing react ion in ma n . Voprosy
Ps ikh ologi i , 195 9 , 5 , 73- 78 . Cit ed by R . Ly nn,
A.tt �nt io!l , Arousa l and th i:: Orientat ion Ri::- �v: +- i.on .
New York : Pergamon�9 6 6 .

Wex ner , L . B . Th e degree t o wh ich c o l o rs (hues ) a re asso­

c iated w ith mood-tones . Journa l of Appl ied Psycho­
logy, 195 4 , 3 8 ( 6 ) , 4 3 2- 4 3 5 .

Wils on , G . D . A rousa l p rop ert ies of red v ers us green.

P e rc ept ua l a nd Mot o r S k i l ls , 1966 , 2 3 , 9 4 7- 949 .
 44

Appendices
 45

Appendix I

Instructions
 46

INSTRUCTIONS

"This experiment is des igned to s tudy the effects on the

nervous sys tem of various kinds of pictures . The scenes
have been sel ected from great paintings for their variety
of subject matte r . Each picture will b e on the screen for
two minutes . All you have to do is s i t quietly and do not
s ay anything until the experiment is concluded . It will
require a few minutes for me to connect the apparatus and
obtain you basal rates before we begin. While I am obtain­
ing the basal measurements please rel ax and do not talk;
I will project some colors on the screen s o you have some­
thing to look a t . Do you have any questions ? "

( Experimenter connects apparatus . )

" P l ease remain calm a few minutes to allow your heart rate
to retur� to its resti�g level a�d for my equipment to
s tabilize . "

( Fo llowing a ten minute pause , the remainder of the instruc­

tions are presented verbally without the use of tape re­
corder . )

"Now I would like to obtain the basal measurements . P l ease

watch the screen and do not talk or move around . "
 47

Appendix II

Analysis of Variance Tables

 48

Analysis of Variance : GSR

Source df SS I MS F Significance

Total 95 2 2 2 , 6 315

Between 23 105 . 7764

R 3 2 .0517 0 . 68 3 9 0 . 13 19

S/R 20 10 3 . 7247 5 . 18b2

Within 72 116 . 8 5 5 1

Color 3 '1.4 . 0 9 3 3
I1 4 .6978 3 . 3590 p ( .OS
I
Colum11 3 10 . 45 30 3 . 4843 2 . 49 13 p <. . 10

Error 66 9 2 . 3089 1. 3986

I
p= the probabil ity that an F ratio as large as the
obtained F ratio could have occurred by chance
alone .
 49

Analysis of Variance: Heart Rate

Source df SS IMS F Significance

Total 95 1714 . 5

Between 23 304 . 5

R 3 70 . 8 3 3 2 3 . 6 11 2 .021

S/R 20 233 .667 11 . 6 8 3

Within 72 1410 . 0

Color 3 13 . 8 3 3 4 . 611 0 .263

Column 3 18 . 16 7 6 . 056 � . 344

s u 6 3 24 . 16 7 54.028 3 . 076 p <. . 0 1

Res idual 60 105 3 . 8 3 3 1 7 . 564

p = the probability that an F ratio as large as the

obtained F ratio could have occurred by chance
alone .
 50

Analysis of Variance: Respiration Rate

Source df SS MS F S ignificance

Total 95 685 . 6 2 5

Between 23 242 . 6 2 5

R 3 5 . 125 1 . 708 0 . 144 '

S/R 20 2 3 7 . 500 11.875

Within 72 443 .000

Color 3 26 .458 8 . 8 19 1 . 60 6

Col umn 3 5 4 . 12 5 18 . 042 '3 . 2 8 6 p < .05

Error 66 3 6 2 . 4 17 5 . 491

p = the probability that an F ratio a s l arge as the

obtained F ratio could have occurred by chance
alone.
 51

Appendix III

Original Data
 52

Original Data: GSR

( Expressed in log conductance)

I
Subject Red Blue Green Yel low

I
1 3 . 6990 I 0 . 0000 3 . 5 44 1 2 . 4785

5 3 . 3012 I 0 . 0000 3 . 6990 0 . 0000

9 4 . 0414 3 . 6 990 3 . 3012 3 . 6 990

13 3 . 0004 0 . 0000 3 . 3 0 12 3 . 1765

17 3 . 4772 I
I
2 . 6998 0 . 0000 0 . 0000

21 4 . 0004 3 . 0004 3 . 0004 3 .9777

Blue Yellow Red Green

2 3 . 4 772
I
I
3 . 3981 3 . 8 45 1 3 . 3 012

6 0 . 0000 I 2 . 69 9 8 3 . 0 004 0 . 0000

10 3 . 7782 3 . 7782 4 � 0 79 2 3 . 3981

14 0 . 0000 0 . 0000 0 . 0000 0 . 0000

18 2 . 6 998 2 . 6 998 0 . 0000 3 . 3012

I

I
22 3 . 39 8 1 2 . 39 9 6 3 . 0972 3 . 00 0 4
 53

Subject Green
I Red Yellow Blue

3 2 . 9036 3 . 1763 0 . 00 0 0 3 . 3981

7 3 . 5 44 1 2 . 69 9 8 0 . 00 0 0 0 .0000

11 3 . 39 8 1 3 .6021 3 . 00 0 4 2 . 6998

15 4 . 09 6 9 4 . 02 12 2 . 845 7 2 . 8457

19 2 . 6998 3 . 5 441 2 . 6998 3 . 3981

23 4 . 1304 3 . 7 782 3 . 30 12 3 .6533

I

Yellow Green Blue Red

4 3 . 00 0 4 4 .0000 0 . 0000 0 . 00 0 0

8 0 . 00 0 0 0 . 00 0 0 0 . 00 0 0 0 . 00 0 0

12 3 . 7782 2 . 39 9 6 0 .0000 0 . 00 0 0

16 3 . 9421 4 . 0 2 12 3 . 3012 3 . 9 778

20 4 . 19 0 4 4 . 13 8 3 4 . 0414 4 . 3979

24
I
I
3 . 6533 3 . 17 6 3 0 . 0000 3 . 30 12
 54

Original Data: Heart Rate

( Change from res ting rate, in terms of beats per minute )

\
Subject Red Blue Green Yellow

1 0 0 -12 -4

5 -2 -6 +4 .
-12

9 0 0 0 0

13 0 +2 0 0

17 -4 -2 -2 0

21 -6 0 -2 +2

Blue Yellow Red Green

2 +24 -8 -2 -2

6 +2 +2 -2 0

10 +2 +2 +2 +2

14 0 -6 0 -2

18 0 0 -2 0

22 -2 +4 -2 +2
 55

I
S ub j ec t Green Red
I Yellow Blue

3 -10 0 0 0

7 +2 -6 -4 -6

11 +2 +2 +2 0

15 0 +4 0 ·0

19 0 -2 +4 0

23 -2 +2 +2 -2

Yellow Green Blue

I Red

4 -2 0 -2 0
I
8 -6 -2 +4 -4

12 -4 +10 -8 -2

16 0 +2 -2 0

20 0 -2 0 +4

24 +2 -4 -2 0
 56

Original Data: Respiration Rate

( Change from resting rate, in cycles per minute)

Subject Red Blue Green Yellow

1 -2 -4 -6 -2

5 -2 0 0 0
.

9 -4 +2 -2 0

13 0 0 +2 0
-

17 -2 -2 +2 +2

21 +2 0 -6 +2

Blue Yellow Red Green

2 +4 0 0 0

6 -10 +4 0 0

10 0 -2 0 -4

14 +2 +2 0 -2

18 +2 0 0 0

22 -4 0 +2 -2
 57

Subject Green II Red Yellow Blue

j
3 -2 -4 -6 0

7 -2 -2 -2 0

11 0 0 0 +6

15 -2 0 0 0

19 0 0 0 0
I
23 -2 +2 +4 +2

Yellow Green Blue Red

4 0 0 +2 0

8 -6 0 0 0

12 -6 0 0 0

16 -6 -8 -2 -2

20 0 +2 +2 +6

24 -2 0 0 +2