The relation between critical flicker frequency and several psychological variables

Material Information

The relation between critical flicker frequency and several psychological variables
Colgan, Carroll Monroe, 1925-
Publication Date:
Copyright Date:
Physical Description:
40 leaves. : ; 28 cm.


Subjects / Keywords:
Academic aptitude ( jstor )
Experimental psychology ( jstor )
Gestalt psychology ( jstor )
Memory ( jstor )
Personality psychology ( jstor )
Psychology ( jstor )
Psychometrics ( jstor )
Psychophysiology ( jstor )
Scotophase ( jstor )
Standard deviation ( jstor )
Dissertations, Academic -- Psychology -- UF ( lcsh )
Psychological tests ( lcsh )
Psychology thesis Ph. D ( lcsh )
Psychophysiology ( lcsh )
City of Jacksonville ( local )
bibliography ( marcgt )
non-fiction ( marcgt )


Thesis (Ph. D.) -- University of Florida, 1954.
Bibliography: leaves 37-39.
Additional Physical Form:
Also available on World Wide Web
General Note:
Manuscript copy.
General Note:

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
000570704 ( AlephBibNum )
13721502 ( OCLC )
ACZ7685 ( NOTIS )


This item has the following downloads:

Full Text






A DI'FPFTTIO*N rPCFi F Nr I T. Tv 7 iii. 7:f.t.' A COLt CiiL :-r
T iH E U ,l' i.'P :17, .*- t ,., [..*.
[,[* PE L ,.J [,,-.**. T.: ,., .i iL.,.,. l ,

Augu'r. I',4


3 1262 08552 2323


ihe author is Indebted to Lr. E. Porter Home, the

Chairman of the .upcrvicory Cor.ittee, for the generous porti-n

of his tLLo and anorgy that uent into the study. heartfelt thanks

ar- also duo ErT. E. L. Kincicley and Z. J. Anderson, representing

the najor, and Drs. G. R. LarLlett and J. V.. Oliver, representing

the minor.

For help of a special nature, the cutnor wishes to thank

C[ean e.. k,. Little and Dr. ri. E. Iooro of the Univerrity College

iad Dr. J. V. rcquitty, the Univcrsity Exan-inr. Credit is also

dcL Lrs. H. A. t hycr and A. K. K(rtz for aid xid critical advice

on tre statistics.

Thnl:s to, a-lso, to trs. 2. U. Ilemneier, 5. EI.imboriy,

and Ilenry ivundorlich.



ACiKNOLEDGM NT . . . . .




III. RSULTS . . . . .



1. 5A;Jo aDh STi-JDA.) DEVIATI . . .

2. LiT.aCORR!!L TIOiS. . . ....

PLATE I. Subject and Experimenter at the Apparatus

WIlLIOGRAPHY . . . . .

aIOGRAPHY . . . .

. . 21

. 23

. 15

* 36

* 39

. . 12

. . . 20

. . . 25



The present study as desiaiod to investigate the relation

between critical flicker frequency and a nunbor of psychological

variables. These were attention, ameory, perceptual ability, scho-

lastic aipitde and scnolastic achievement.

A Licht flashing on and off may ap;.car to fLic,:er or %ay

appear to luse, depending on a nurejr of en7iron:'ental conditions

and the condition ol tne subject. The frequency of intenrittence at

which some change in the physical stimulation is paralleled by a

change in tne sensation from flicker to "stencay" or from steady to

flic.ier) ic Icoon as the critical flic:er frequency (CFF). Tno

CFF of a civen individual represents his fliccer-fusion threshold

for the conditions under which it uas measured.

According to lateau, as reported in an article by Landis .17),

the first reported investigation relevant to flic.:er -as done by

Ptolon y-Roglonontanus about ,.. D. 150. ?tolmcy noted that if a disc

Ulth sectors of different colors is spun rapidly the colors seam to

fuse anrd become a new and dirffro:Lt color. oince tnat tirLe, the

phenomenon has inturosted many i jvostigators so that much is lkown

about the conditions under which it occurs. ai colorful and detailed


history of CFF, including a survey of the osctblished factual know-

ledge, is given by Landis (17).

The information available on CFF may be divided into three

categories. These are: (a) psychophysical--concerning the relations

betwwn CFF and various stimulus conditions, (b) physiological--

concerning the relations of CFF with various bodily conditions, and

(c) psychological--concerning the relations with the various measur-

able components of mental abilities, characteristics of personality,


Most of the early work on CFF and much of present day research

involves psychophysical investigations. r.ithin areas of agreement

it can be said that CFF is a function of the size, shape, and posi-

tion of the patch of light, that it varies with brightness and dis-

tance of the light. It is influenced by the surrounds of the light

patch as well as the ratio between the length of the light period

and the length of the dark period in each cycle of intermittence.

Complete discussions of these relations and others are included in

the writing of martley ( 2), Landis (17), and Sinonson and Bro ek (31).

The physiological data, while not as numerous as the psycho-

physical, permit wider generalization. Most, if not all, central

nervous system stimulants cause a rise in CFF, i.e., the light is

seen as flickering above the previous fusion-point and a new, higher

fusion-point is established. Most, if not all, central nervous

system depressants result in a lower CFF, i.e., the light previously

seen as flickering now appears steady and there is a new, lower


fusin-point (31). Further, in tho words of Landis:

. any condition or agent which acts to decrease the avail-
able blood msuar and/or oxygen availaolo to the retina or to the
brain decrenses CFF, whereas conditions increasing the efficiency
of the vascular supply increase CFF. In other words, CFF seems
to be a measure of the functional efficiency of tho retina and/or
the visual cartex. ... 17)

ihe psychological data on CFF are not as numerous as the

pszchophyaical and in the light of present knowledge may appear

mnro limited in value compared rith the physiological. The affects

of practice, and t!Leroforc loarnin,, zermi ts be negligiblc provided

that care has been ta-en to insure that the scbjuct is thoroughly

acquainted costcn-ively) with tne phenceaioon .17). CFF va-rics wiut

the dircction:s Civen the s-uject, such that thel- set to see 'fLic:;er"

gives higher value than the ect to see "fusion". studies oy

ic:lcmar 30) and by hroiek, et al. ( 3) show the reliability of CFT

tl be hiCh--on the order of 0.97. Several personality studies have

shown that "introverts" have higher CFF2 than do "extroverts" (31).

Halstead includes a flicker test in a battery of tests designed to

neacure "bitolical intelligence" (il), but he foJnd little relation-

:!up between th'e test with a lignt-dar, ratio (LD.1) of 1:5,000

and the Ilennon-lelson est: of i' mental Auiiity. manner, on the other

hand, fou..d C. wiUt an a.craie LD2 of 28:1 to vary significantly

witn the hACL (3-). COla3n, iusi 3 LU2 of 1:1, fc nd a a iificant

relatinshi- iuith the Wechslir-bellerae oale, in a :tuuy of acujccts

ranging Li aIg f-rm 67 to 92 years ( 5).

/. oiological theory of lo.i, standJin i3 that mental functions

are intLaateia conrnecced with the functions of the orain. In his

book, Man on his Nat.u:'e (30), ShcrrLnct)n, the physiologist, develops

this idea at great length, citing experimental evidence to demonstrate

the correlation. A similar basic view is held by the psychologist,

Spearman, although in 'ie Abilities of Man (32) he expresses concern

over the inadequacy of present day kPiowledge of physiology for a

satisfactory close fit with established psychological facts. Others

holding to this theory include Kdhler (16) and Lashley (18). As a

matter of fact, the Gestaltists (including Kdhler) assume a relation

of isomorphism (16) between physical events of some degree of complex-

ity in the brain and mental events occurring in the realm of experi-

ence. There is then, a desire on uoth sides to further the integra-

tion of psychological and physiological facts.

It should perhaps be noted that both the physiologists and

the psychologists consider the study of vision important not only

in itself but, by the former, as a means of understanding the central

nervous system and, by the latter, as a means of understanding a vast

array of mental functions and personality characteristics. In

biological symposia held at the University of Chicago, Gelihorn


It is. . not inappropriate to study the functions of sense
organs not as a purpose in itself, D,.t rather as a means of
eliciting fundarmental processes of the central nervous system.
The senso organs including the visual system) represent natural
avenues of approach to the central nervous system. They allow
one to perform experiments on the central nervous system in the
human, under the most physiological conditions. . it ought
to be possible to illustrate fundamental laws determining the
activity of the central nervous system by the study of. the
visual system. (10)


The importance of vision in theoretical psychology c-n be illustrated

by the fact that two present day theories, as presented by Koffka ( i)

and by Hobb ( L), endeavor to explain all of behavior in terms of, or

by analogy to, visual perception. The use of the Rorschaclh sycno-

diagaostic Lest in personality research illustrates this still further.

lo it is that CFF, already considered of wide si -tilicanco by the

physioloist could just as easily oe accepted as of far-reaching

import.Lce, psychologically, if tnh facts warranted it.

It was mentioned anove tnat CFF reflects many major changes in

the brain. This point is made cloar in the following comment of

Si oason and broeck:

In such clearly defined stresses as anoxia or carLon dioxide
inhalation Lboth probably producing Local acidosis) the flicker-
fusion frequency is a sensitivc method for characterizing, in
a quantitative way, the C.S involvement. (:!)

In accordance with oioloical theory, then, CFF should show

definite relationship %rith some psychological function. If it is

as importantly related to the metaialic activity of the brain as

Landis i7) thinks, thon it should be significantly related to an

important relatively oasic psychological ability such ab attention.

Tne present study was designed to investigate sucn a possi-

oillty. In addition to attention, several oUler psychological aoil-

ities were included so that the relationship of CFF to each might

be determined. These were memory, perceptual ability, scholastic

aptitude, and scholastic achievemer:t. because of the imRrensity of

the task of relating CF to all pos-:iule psychological variables,

limitation was necessary and those chosen are of practical importance.


Attention, as a nae for a class of mental phe.ino.mna, has

had a varied career in the history of theoretical psychology.

Spearman, in discussing attention in his c'chio, Do.In the Ages

(33), states: "In truth, there is hardly an element, or aspect, or

relation, of consciousness wherein this 'attention' has not been

located by some one or other." -.cert*ileos, after a thorough review,

he finds that it has a saving grace--"This is that, however diversely

defined, it has nevertheless been uniformly applied." It would

seen that Pillsoury is in agreement, for in his book, Attention.

he says:

e may say then, in conclusion of our examination of the various
theories of attention from the side of consciousness, that each
has picked out some more or less important coneonitant process
or some aspect of attention and regarded it as the explanation
of all the remaining parts of aspects. . Attention is not
any one of these things alone, but it is all of them taken to-
gether and more. (25)

Hebb, in proposing a new theory of behavior (13), thought it neces-

sary to face the problem of explaining attention squarely. In his

opinion, the confusion and hesitancy about attention derive from in-

safficient realization of the effect on current theory of admitting

the facts of the phenomenon. He states (13, p. 4):

'he an experimental result maies it necessary to refer to "set"
or "attention", the reference means, precisely, that the activity
that controls the form, speed, strength, or duration of response
is not the immediately preceding excitation of receptor cells
alone. . Almost without exception psychologists have recognized
the existence of the selective central factor that reinforces
now one response, now another. The problem is to carry out to
its logical conclusion an incomplete line of thought that starts
out preoccupied with stimulus or stimulus configuration as the
source and control of action, eventually runs into the facts of
attention and so on, and then simply agrees that attention is an


important fact, without recorgnzing; that this is inconsistent
with one's earlier ascunntions.

Attention then, is the keystone for psychological theories that would

go Dcyond single stinulus-respons.a association.

In line with Hcbb's contention that attention denotes a

"selectivity of response" originating within the individual is the

working definition given by "hilip in 19A8 1',). After aiL extensive

review of the literature on methods of maazurin, attention, Philip

decided that the w.orklin. definition waic.n best fitted the varied

approaches :ras that attention refers to .a -ubject's power

of holding the :niind on a task in spitc of fati.guo and accidenta.l

disturbu-nce.." Using a critorion proposed by hloodrow 1u), he

studied the rcl:tiie ncrit of fourtecn moasroez of attention derived

partly from the literataro and partly from c.ntribations original. ,ith

Ilia. Ho selected five of the sest of' those measursr to onr a battery

suitable for act-itiistratilon to groups.

In severE-l investigations this battery, Knmun as thiiLp's

Test of Attenit~n, has oe@& used with consistent findings. ;aeinku rt

(26) nade use of this battery in her study, "A scale for measuriig Lihe

g-factors in intelligence." aho co;icludes: ". .that underlying

the atLentioni tosts of LroLbuer LhilLip tnuer is a general factor. .."

(2o, p. 7). In 1935 Monachen included ilhilip's Lest in his study (.),

major r factors in cociiti n." lie also found a goneral factor under-

lying the attention tests, Lrt this general factor for attenti:. uwa

intercorrelated with t;ho g;CerJal factors for theo memory and the coni-

tive tests so that he concludes that thero is a "supcr-uneral factor'

underlying all of them. In a work of tre- adous scup the Speam-

Holzinger Unitarl Trait Study (6 ,3h) has reported similar results.

Attention (including the test by Philip, see34, p. 73) is listed

as a "unitary trait" along with eight others (14, p. 1).

Hames (12) has compared GFF with four of the five sub-tests

from Philip's battery, finding two of them, number span and cancel-

lation, related to CFF. Unfortunately, his results are of doubtful

significance because of the distinct possibility for systematic

errors in the UFF determinations. For his source of illumination

he used a General Radio "Strobotac" which gives an extremely short

flash of constant duration. This instrument consequently produces a

flashing light whose light-dark ratio varies with the frequency of

intermittence. In a vast majority of the studies of CFF the time of

the flash has a constant ratio in the flash cycle. Since CFF is in

part determined by the light-dark ratio, Hamnes's results are not

directly comparable to those of other studies and of greater iEportance

is his omthod of measuring CFF. HIames says of his procedure that

the subject was not dark-adapted, but instead his CFF was immediately

obtained. CFF was recorded in one direction only, from fusion-to-

flic;ker. Five readings were taken with the first discarded as practice,

and a mean CFF was computed from the remaining four. The really serious

source of error here is the lack of care taken to insure that each

subject's CFF was valid for him under the testing conditions. This

need for care can hardly be over emphasized. Landis states:

Finally, one must be sure to instruct and demonstrate to


the observer just what a steady liCht looks like at each level
of intensity, and what a fliccering field loo;:s like at each level.
Crosier I 6) hcs described the appearance of a white field flicker-
ii:; slightly below the point of fusion at low-brightness intensity
as smooth, bluish-grey rp'lsation, at lIn-randiun intensity as
spec:od with yellow spots, at medium intuisity as a granular
ruricco, at nedium-hiLih intensity as frostoi, and at that much
that has Lecn cal.le 'practice', as well as much of the variaDility
in results, is dUo to the fA'ct that untrained ooservcrs do not
know urat to look for. (17)

r4emory, more than any other mental. .unction, is the most widely

acccptnd and the most difficult to explain. As bpea-rman L(.3, ,. 161)

notes: 4Alen talking of the act itself, everyone, oven the plain

nan, seems to mern and always to have meant much the same tning."

Explanations however, have centered around t.o radicalLj different

.lnuw: (a) that reonenorinZ so-not'ing from the past is accomplished

by access to tracks stored in the nervous system of the original

perception or tnoaJzht, etc., and (0) tiat tnere is an "'unc.nsciousness"

.bcra p-st exporicnces aro neld, access beilnL dotcrnined ry a caopli-

catol syster. of mental dynamics. both of teset oxplanations date

at least lroea the middle Aces an d both have their modern-day exponents.

"owuovcr oxplain id, tne experinental investigtii3: of0 t:io conditions

inliuoncing nomory stands as one of the Drigntost achiev. nnts of

mnden-r pjacholo y. ho relation cet.eenc CFF and nemory has nat ueen

studied apparently.

Although pcrcaption itself has long; ecz- on the psyc.olo.ical

sco:o perceptual abilltj is a raticr recent innovation. It cec.e3

to have ocplicit as a recul of the earlier ..or. of tne factor

analysts, notably J.pearnan, idoinanicr, and Iharsto.'ic (uc). Their

General approach is, periha.s, guest ciar-ctoriLed Dy tie uar.i of


rnurstrin (38). In 1944 Thurstone published A Fact:.rial Stur- of

Perception (37). Included were five of his "primary mental abilities"

which did not correlate with the test of CFF used in this battery.

His test of perceptual ability was not used. There were, however,

numerous approaches to perceptual responses. In general the meas-

ures of perception did not correlate with CFF. Thurstone did not

relate CFF to tests of attention as defined by Philip. It seems

from Taurstone's description of the measurement of CFF in his study

that the CFF determinations were suoject to the sources of error

referred to above in connection with Hammes's study.

Scholastic aptitude is the term used to denote the capacity

to learn academic subject matter. Even though it specifically refers

to academic subject matter, it is clear that that which is measured

oy scholastic attitude tests is, to some extent, measured by such tests

as the techsler-Bellevue and the Stanford-Binet. In practice, however,

the two seem to diverge. scholastic aptitude tests are primarily

verbal in nature (in keeping with their validation criteria) whereas

the major intelligence tests (such as the wechsler-zellevue) give

approximately equal weighting to both the veroal and the performance

test items. This difference in the composition of intelligence tests

and scholastic aptitude tests plus differences in use and therefore

validating criteria have led authors of scholastic aptitude tests to

point out that, at best, their tests give only rough indications as

to the intelligence of testees.

Biological. theory posits a close correspondence between


neurological events ani psychological functions. At the present time

the experimental findings of neurophysioloLy have some correspondence

with psychological findings, uut much remains to be done Lsee ieob,i3).

As new information iaoot toe functioning of the nervous system is pro-

vided by the physiologists, it is i-rortant that this Liortmartin be

studied in relation to psychological Iunictioan ius 1 id vice versa.

Recently CFF has ocen found to Do related to widespread general

activity of the central nervous system. CFF represents the best

sinnie indicator the physiologists now have for estimating the func-

tioninr, efficic-ncy of the central nervous system. The relating of

CFF to psycholo-ical functioning hao now becomnc an important tas:

for psycholojir'. The present study was undertaken as part of that

_reatcr L-s;:. As a result of this study it should Le posaiole, Dy

means of co,.parin. the relations between CFF and the various psycho-

loical variables, to decide whether CFF is a measure of attention

or memory or perceptual ability or scholastic aptitude or, perhaps,

none of those.

Ci:'rtJ II


The general procedure was broken into two parts, a group

session and an individual session. Sixty-seven subjects took part

in one of the six group sessions but only fifty took part in the

individual sessions. Seventeen not taking part in individual sessions

failed repeatedly to keep their appointment. From the fifty there

was a final reduction to forty due to the following: two had scores

that were considered invalid because of difficulty with the English

language on tests, two -omen were dropped in view of difficulty of

scheduling, one subject was hospitalized, one subject never saw fusion,

several subjects failed to keep further necessary appointments. The

forty subjects who satisfactorily met all the requirements of the

study were male college sophomores with three semesters of college

work completed. They were all in good standing and in good health.

The tests suitable for administration to groups (attention, percep-

tual aoility) were given at the group session, while the tests limited

to individual administration (sensory attention, memory, CFF) uere

given at the individual session. Scores on scholastic aptitude and

scholastic achievement were obtained from university officials. At

the individual session each subject's height was recorded so that a

gross physical measurement might be included in the intercorrelations.



The light source used for the CPF measurements was the glow

modulator tubo, Sylvania R-1130-;, modulated by an electronic snitch.

The wave-torn was essentially rectangular, thus linitino the latency

and decay of the liflt source to the time required for ionization

and de-ionization of the gas contained therein .nhich was approximately

5 nicro-secoads.

This tube was part of a circuit unicn allows variation of the

Brightness of the light source. The circuit was stable and noise-

less in operation. it had been used in several previous studies,

see Colgan (Li ) and Saucer (2U,29). Suitable regulating devices

were included to filter out fluctuations in the 110 volt, 60 cycle


The circuit consisted essentially of a positive bias multi-

vibrator with symmetrical time constants. Large frequency increments

were controlled by a switch which places the correct value of C in

the grid circuit of eacn tube of the multiviurator. Small frequency

chances of the order of 3:1 were made oy the adjustment of the posi-

tive Dias potentiometor.

The signal output of the multivilrator was developed across

a 5,000 ohm resistor. A 6587, functioning as a wide-band grounded

amplifier, is connected across the same resistor, so that the 65H7

signal input was transmitted via the cathode in accordance with con-

ventional practice in grounded grid circuits ,hi). ahis eliminates

phase shift and other forms of wave distortion associated with RC

coupling networks. In addition, these tabes served as clippers and


wave shaping circuits, so that the gatin, signal applied to the Eon-

trol tube had an extremely short rise time.

A 6V6 was used to control tle current throoii the R-l130-i.

In order to control this tube it was connected as a direct-coupled

amplifier across the plate load resistor of the 6j7il, so that when

the 6Si7 was conducting the 6V6 was cut off by apiro:icnately 125

volts negative bias. ..hen the 6&H7 was not conducting, the bias

on the 6V6 was the value imposed by a 100 ohm cathode ra~istor.

Further intensity control was secured by the use of a voit-ge divider

in the plate circuit of thc 6V6.

The sat.uiu3-patch was a 10-sided polygon, white in color,

4 mm. across, and 1000 mm. from the suoj-cts', pyes. hcn flas&irg,

the stimulus-patch had a orignl.ross of 23.71 ioot-lanDerts and a Li,;ht-

dark ratio of 1:1. Light and test-patcn were enclosed in a large

rectangular box, the insides of which ioer flat-blac... he szoject

observed the test-patch by means of a vicerinv hood attached to one

end of the box (see Plate I).

In general the procedure for the measurement of CFF was the

sauce for all subjects. Each spent approximately 50 minutes prior

to the testing in a room illulinatcd by relatively low-level arti-

ficial illumination. For the measurement itself, the .cauject fixated

the patch binocularly. Flicker and the fusion-point were given

ostensive definition for the subject. Following several preliminary

trials, five ascending and five descending trials were given the

subject who was instructed to report when the flicker disappeared






and reappeared. The average of these ten determinations was taken

as his mean CFF. The rate of change of the flash rate was rapid

and irregular. The initial rates of repetiioL were rand-irizod

and the exposure was continuous for each trial, varyin-i irom 2 to

6 seconds. The light was off between trials. The exierinreintal

room was darkened throughout all determinations.

The preliminary trials served to increase the reliability of

the CFF determinations. Investigation had shown that experienced

subjects gave initial judgments with little or no scatter, whereas

among naive subjects, some gave highly consistent initial judgments

and others gave widely scattered judgments initially, but after four

or five detenninations they became consistent also. This variability

or practice effect was controlled by taicing as many ascending and

descending determinations (fron 2 to 10) as was necessary to insure

that each subject was civbi; consistent judgments. For 19 subjects

excessive variability necessitated a retest. The preliminary results

were not used in the determination of CFF for any subject.

The test selected as a measure of attention was tPilip's

Test of Attention. As mentioned in Chapter 1, this test has been

found to be a suitable test of attention. It is coaoosed of five

subtests and requires slightly over an hour to administer. Special

ansujer sheets are available and the test is suitable for administration

to groups. Complete instructions as well as scoring keys and procedure

are given by rhilip (2t). Several of the subtests are timed. The

use of a blackboard or similar device seemed nocossary to make sure


thnt -al of the subjects understood the instructions for sone of the


The five subtests were named by Philip as: (a) Number span,

(b) Iultiplication, (c) Alphaoct, (d) Addition, Laid (e) Cancellation.

The Nu.ioer ispan test. consisted of twenty-four numbers, from one to

nine digits, with two to three letters randomly placed ameao them.

These were continuations of letters and digits dictated uy the ex-

aminer writh the subjects required, at a signal froa the eyamniner,

to rite only the digits in their proper order and to disregard the

letter. rho Ahiltiplication test requires mental multiplication of

relatively lonj numbers, itkewioe dictated by the examiner, and the

retention of tne answer in the mcanry until the si3ial is given to

Twrlte down only the answers Ln their proper place. Dhe Alphabet

test was original with ?hilip and consists of the olloUiing: A

letter of the Japhaoet is dictate, follow cu from one to five

digits. The s-oject, at a given &ignal, is re.uirea ta writo cJwn

tre letters of tne alphabet, in the sequence in which the aiitG were

givo:i, counting from the initial letter. Thus if' F-35 is dictated,

only I N R is to be written down. The uore: must De done mentally.

The Addition test consists of numbers of from one to six digits pre-

sented on the answer sheets writh the subject required to adJ them

horizontally in his hnad. The numuors on any one line varied in

length so that care is needed to plane heo uiits, tons, hundreds,

otc., in their proper column. Tie Cancellation test represents a

modification by Phiip of a test by 6tcrziLuer. It consists of


seventeen lines of randomly assorted letters of the il-,habet in groups

of from one to four letters. The groups are also randomly arranged.

In a ten minute time period the subject is required to cancel letters

according to their position and the &ailic.ioiLtit of one or more of

three rules given for this purpose.

In his study Philip called his test a test of "abstract"

attention and used a modification of a test by Nfnsterberg for the

measurement of "sensory" attention. It was thought desirable to

include this test in the present study to investigate the possibility

that CFF represents more a measure of "sensory" rather than abstract"

attention. Describing his test of sensory attention, Philip stated:

A pack of thirty-two cards had marked on each the four colors
black, red, green and purple in thirty-six randomly distributed
dots, according to the following scheme: Two cards had sixteen
olaci dots and of the remaing colors the following numbers,
seven, seven, six. Two cards had fourteen blacks, and seven,
seven, and eight respectively of the others. Two cards had twelve
slacks, and eight for each of the others. Two cards had eleven
blacks, and eight, eight and nine for the others. This scheme
was followed for each of the four colors. (24, p. 37)

The subject is required to sort these cards into four piles, accord-

ing to the predominant color. Three trials are given, with means

for time and for errors recorded.

As a measure of perceptual ability, the (Serics it) Likert

and Quasha revision of the Minnosota Paper Form board Test was used.

This test has been widely used and publicized (19, 7 ). It is suitable

for group testing.

As a test for memory, the standardized Memory Scale of *,echsler

(39) was adopted. This test was designed primarily for clinical use


and consists of seven subtcsts: (a) Porsonal and Current Infonrmtion,

(b) Orientation, (c) Mental Control, (d) L)gical M iory, (e) memory

Span, (f) Visual Reproduction, ([) Associate Learning. The first

three were included, Wechsler states (39, p. 87), because of their

value for clinical diaGnosis--even though thny discriminate "...

very little or not at all Detween normal or even near-normal subjects."

.inco the subjects in the present study wore at lost in the Latter

cateeor7 tho first three subtests were not used. Appropriate points

wore allowed all subjects on these tests, however, so that tt.-al

: eichted scores could be attained for comparison with Wechsler's norms.

The other subtests .e*re scored according to the toat manual.

Scholastic aptitude was measured oy meruis of the American

Caoncil on -ducation Psychological Exainination (ACE), 19L9 edition.

This test had been taken Dy all subjects when they first entered the

university as freshman. The scores were outalnod from the University

Sxa2iiner's oilice. Three scores are derived front this test: ka) a

,uantitative or 4 score, (b) a Linguistic or L Scoro, and (c) a Total

or r score. All three have extensive norms for comprrison purposes.

Scholastic achievement was determined uy computing the honor

point average (HPA) for each subject on the basis of course grades

since entering college. A grade of "A" receives four honor point,

a crade of ".;" three points, a grade of "C" two points, a grade of

."" ono point, and a grade of "S" no point. Access to these grades

was furnished by the Lean of the Ualversity Colloge.



As a means of reporting the relationships between variables,

Pearson product-moment correlations were used. The variables fell

into three classes: (a) experimental variables, (b) test variables,

and (c) subject variables.

The average of ten determinations of a subject's fusion-

point was taken as his mean FF--rocorded in cycles per second (cps).

A subject's score on the sensory attention test was obtained by add-

ing the average errors to the average time in accordance with a

formula used oy thilip I2h).

The Minnesota Paper Form board (P. Fm. Ld. ) and the ACE were

scored according to their respective test manuals, scores weighted

for age were obtained from the memory test, while the perceptual test

scores were raw scores. The attention scores for the Philip's battery

of subtests were weighted according to a formula given by Philip (24).

Each subject's height was recorded to the nearest centimeter.

His HPA was taken as the average number of honor points per semester

hour credit. The course work covered general introductory freshman

and sophomore subjects.

In Taole 1, measures of central tendency and variation and

also a measure of variability of the central tendency are presented.



As representations of these statistics, the following were obtained

in order: mea; (M), standard deviation (3), and the standard error

of the noan (rw).


LuLt 1

DS- .TIOiiOa, ,.l;u Si'~;..i, Oi.0u OF D-6 KrA.-
(1- = 40)

Measure I o oM j eauure 1 o aa

CFF (cps) 36.9 1.8 0.3 HPA 2.5 0.8 0.1

Attention 89.6 12.4 1.9 Sensory 155.0 35.0 5.5
Total) Attention

.Jo. span I1S.h 5.8 0.9 AC1. (total) 115.7 17.0 2.7

.tult. 6.5 1.6 0.3 .uant. 4h.0 8.2 1.3

Alphaoct 11.8 L.O 0.6 Linguistic 71.7 12.1 1.9

AdnitLion 13.9 h.7 0.7 ilcmory 103.2 4.9 0.6

Cancel. B.2 2.6 O.h A. Ld. 17.h 7.3 1.2

Hcidit (km) 179.9 6.1 1.0

To the relation DetweLn CFF and each of the other

variaoies, correlation coefricients were com.juted. Ihis tecclique

of statistical analysis was iaso usoe to determine the ma6nituce and

direction of tihe interrelations among the scores on the tests ol

attention, perceptual aoility, memory, scholastic aptitude, ual

scholastic achievement. The measurcments of height cwre correlated

with scores on all of the auovn tests.

Correlation coefficients are presented in Taole 2. These


coefficients permit evaluative inferences about GFF--both aoout

its psychological and its physiological explanations. In addition,

they permit interpretations aoout the relations of attention, as

it is measured by Philip's test, with the other psychological aoil-

ities. Relations between attention and other psychological abilities

have frequently been hypothesized or discussed out seldom studied


In interpreting these coefficients of correlation it should

be noted that for an N of 10 a correlation coefficient (r) of approxi-

mately 0.30 is considered significantly different from zero at the

5 per cent level of confidence*. Similarly an r of approximately

0.40 is considered significant at the 1 per cent level of confidence.

Because a high score on the sensory attention test means a

poor performance and a low score signifies a good performance, a

negative correlation with any other test should have a positive sign

to be readily interpreted. Therefore the signs of the correlation

coefficients between this test and each of the others have ueen

changed accordingly.

*'he standard error of r for various values of r is as follows:
for an r of 0 to .20, .16; r of .21 to .30, .15; r of .31 to .39,
+ .lh; r of .o0 to .45, .13; r of .46 to .52, 1 .12; r of .53 to .58,
S.11; r of .59 to .63, 1 .10; r of .6h to .68, 1 .09; of .69 to .72,
.08; r of .73 to .76, + .07; r of .77 to .80, .06; r of .81 to .84,
S.05; r of .85 to .87, 1 .Oh; r of .89 to .91, 1 .03.



-easure 1 2 3 4 5 6



lIuiber Span





sensor r Atte-tiona

AC0 ToLal



I-inn. P. Fa. Ad.



. . +0.23 -0.27

+0.23 . +0.67

-0.27 +0.67c . .

+0.12 -3.51c +0.10

S0.32 +0.703 +0.31'l

40.07 +0.76c <0.29

+0.16 V3.23 -0.26

-0.15 +0.37) -0.28

40.02 40.550e 0.16

+..05 0.50c +O.J7

-0.02 +0.60c +0.17

+O.ObL 0.3)0 +0.07

-0.01 +0.23 +0.01

+0.12 +3.LOC +0.18

-0.0h -0.03 -0.11


+4. 51c


. 2

















*0. 10





















aThe correlation coefficients are rcrersed in sion--see
page 22.

jaicnificant t t o 5 level of caLlidea.Le.

c.lgnificanL t t he 1% Level of confidence.

TABLE 2--GC.-.binuid

7 8a 9 10 11 12 13 1l 15

40.16 -0.15 +0.02 +0.05 -0.02 +0.04 -0.01 +0.12 -0.Oh

+0.23 +0.37b +0.550 +0.54c 0.60c +0.39b +0.23 +0.)00 -0.03
-0.28 -0.28 +0.16 +0.07 +0.17 +0.07 +0.01 +0.18 -0.11

+0.12 +0.36b 0.33b +O.lIOc +0.20 +0.19 +0.00 +0.10 +0.13

+0.27 +0.10 +0.27 +0.29 +0.18 +0.11 +0.14 +0.25 -0.05

+0.25 +0.310 +0.62 +0.60 0. +0.6 +0.l3c +0.25 +0.h2C -0.O0l

. +0.06 0.32 +0.38b +0.19 +0.32b +0.04 +0.03 +0.15
+0.06 ... +0.h6 +0.36 +3.0c +o0.38b +0.15 +0.14 -0.05

0.32 0.6 0.76 0.+0.320 6C + 76 9C 0.32b +0.45c -0.17
+0.38b +0.360 +0.76c 4038b +0.4l1 +0.29 +0.38b +0.10
+0.19 +0.0Oc +0.89 +0.38b .. +0.27 +0.24 +0.38b -0.30b

+0.32b 10.38b 0.39b +0.41c +0.27 . +0.8c -0.05 -0.20
0+.04 +0.15 +032b +0.29 +0.24 +0.8c0 . -0.27 -0.26

+0.03 +0.1L +C.45c ,0.38b +0.380 -0.05 -0.27 . +0.16

+0.15 -0.05 -0.17 +0.10 0.30b -0.20 -0.26 +0.16 . .



Discussion of the results of this study is presented in three

parts. First, the particular method of analysis appropriate to the

study is discussed. Then follows a general discussion of the group

results on each test or measure considered separately. The last

section deals irith the presentation and interpretation of the inter-

relations among the variables, including the conclusions to be drawn

from the study.

A method of analysis suitaulo for the present study is derived

front one referred to by J. S. Mill a: the joint method of agreement

and difieronce (3, p. 220). Lnployinc this metnod in a now applica-

tion, Spearman (J., p. UI) demonstrated its usefulness as a criterion

for ovaluatin interrelations expresuso in corrolational form. To

meet tno agreement criteria, CFF should correlate Aignificantly with

not only one variaDle--such as sensory attention--but also with each

of the other variables included in the study that can Oe adequately

described as measures of a closely related function--in this erxaplie,

the attention battery of lhilip. Following the ex-aple, in order for

the difference criterion to Do satisfied, CFF should correlate little

or not at ill i.ith each of the variables that can be nde.;uately

described as not measuirin attention or any closely related function.



CFF should meet both criteria to satisfy the method of agreement

and difference.

Before considering the interrelations among the variables,

the group results on each of three separately should be noted. The

average OFF was 36.9 cps, with a standard deviation of 1.8 cps. 'his

average value is about the same as that found by several investigators

S31). The standard deviation found here, althouEh it is less than

that reported by Simonson and brolek (31) for a sample of 32 men,

is sufficient in view of the high reliability associated with CFF,

especially when the latter is carefully determined, to siow whether

a relation exists between CFF and any of the variables included in

the present study.

Means and standard deviations for Philip's Test of Attention

are comparable to those Philip obtained for the 18 and 19 year old

boys and girls in his study (24). This is trae whether the comparison

is between total scores or between scores on any one of the subtests.

For the sensory attention test, Philip reported a mean and standard

deviation for 49 college males which are similar to those found in

Table 1.

Comparisons are also possible for each of the other variaoles.

The mean value found for the 'irinne:za taper Form board, 47.4, is

approximately the sane as that given in the test manual for the 50th

percentile, third year engineering students or for the 85th percentile

for liberal arts freshmen. Mean and standard deviation for the memory

test are similar to the values reported by Wechsler for subjects of


comparable ase. iccordi ng to the norms for the AC., the ncans found

in Table I for Total, j. and L, have percentile ranks of 62, 65, and

6L, respectively. The average IRA is higher than the average for

University of Florida males and the average height of the group is

greater than that of the average American male.

Against tiis background, the interrelations aaong the v:ri-

ables may be discussed. CIF correlated with the attention test of

Philip *0.23 and with the teat of sensory attcution -0.15. r1either

of these coefficients is siuiilicantly dilfforent from zero. One of

the subtests of rhiLip's test, Alphabet, correlated with CFF tb the

extent of +0.32. On the basis of chance fluctuations, an r of this

magnitude could be expected to occur approximately five times out of

100 if the true r were zero.

Applying the criteria of agreement and difference to these

interrelations, it is concluded that for subjects and testing, con-

ditions similar to those that here used in the present study CFF is

not relatoi to attention as measured by either rPilip's test or the

test of sensory attention. These test-, as measures of attention,

dopond on the subjects concentrating on the t,-t natter to a maximum

extent and cherefore reflect individual differences in this maxianm.

There are other ways of describing the variation of attention. It is

possible for attention to vary from time to time within the sane

individual. The lack of a suitable measure of this aspect of atten-

tion prevented its inclusion in the present study.

OFF correlated nearly zero with the *j, L, &id T scores of


the A.-. Interpreting the CE as a measure of intelligence, Tanner

(36) reported significant correlations between CFF and the ACE.

Colgan ( 5) found similar results with the beehsler-Bellevue Intelli-

gence Scale. Since both Tainer and Colgan used testing conditions

for the CFF determinations siLilar to those used in the present study,

the apparent conflict in results deserves discussion.

To begin with, Tanner measured the shortest noticeable dark

period for several fixed light periods. This means that for a given

fixed light period, the dark period of each cycle was reduced in length

until the subject reported the absence of flicker and then the dark

period was increased in length until the subject reported fliceter.

The length of the dark period at the point of transition tas determined

by one ascending and one descending determination) was taxen as the

shortest noticeable dark period. Thus the light-dark ratio (LDR)

was variable. In the present study the LDR was constant at 1:1,

i.e., the light and dark periods in each cycle were equal and were

shortened (or lengthened) by the same amounts whenever the frequency

was changed.

At the fixed light period of 16 milliseconds, Tanner's 25

subjects gave an average LDR of approximately 1:1. Since this is the

constant LDR used in the present study it is interesting to note that

Tanner did not obtain significant correlations with the ACE at this

setting (35). There is, then, no conflict with the present findings

but rather corroboration. His positive findings with the 38 and the

8h nillisecond light periods stand in need of corroboration also;


tlero are reasons for believiLi that the CFF of a given sauject is

different when measured with a constant LULR from that obtained when

usLng a variable LDR. Coasentiji: on this, Landis i;) states that

u.iore the LD1 varies with freuency ". .tne contour found by Lraph-

Lng frequency against intensity has a different position from that

were the time of the flash has a constant ratio in flash cycle."

The app-artus used in the Moosehaven study 5) is the sane

as that used in the present study, with a LDR constant at 1:1. The

subjects studied ranged in age from 67 to 92 years and they were

trstedJ with the ihecaler-Bellpvue Intplli.inco Scale (Ui-). by

neoois of partial correlation, Colgan demonstrated tLat for his L,0

subjects CFF was related to the v-b and not to aoe except in so far

as the latter was possibly .un underlying datorminant of both tne W-b

ccores and the CFFs. lhe high correlation oetw:een CFF and esca of

the performance sutoests accouita, for the most part, for the bish

correlation betwoun CFF and the total score on the W-L.

In the present study, CFF was found to De unrelated to the

/.C. Studies ny Sartain (-,) and oy n/derson, et al. k i) have sh:un

the h-B performance IQ to have little or no relation with the ACE for

college students. It is not surprising that CFF, which correlates

highly i.ith the 'W-3 perfor= noc scale, does not correlate sicnifi-

cantly with the I.C.

In view of the above, it is not unexpected that CFF should

not correlato siiiAif1catly l.ith ViA. Tho purpose or function of

the AC; is to predict ILr,--at least to sorle extent. In. the present


study, ACE is definitely related to HPA and just as definitely not

related to CFF. In addition, the W-1 performance scale was found

by Anderson and by Sartain to be only slightly related to HPA.

The bochsler Memory Scale correlated to an insignificant

degree with CFF. Wechsler found a relation of sufficient magnitude

between raw scores on his memory scale and age to warrant weighting

these scores in terms of the W-B full scale, but points out (39, p. 90)

there is no relation between the memory scores and the W-B without

variation in age of appreciable scope.

The relation of CFF to the W-B Performance Scale deserves

special attention. Colgan found a correlation of 40.48 between scores

on these two variables, using elderly men as subjects. A coefficient

of this size, for an N of h0, is significant from both a statistical

and a practical standpoint. It is possible, however, for the W-B

Performance Scale to be related approximately to the same extent

with some third variable totally unrelated to CFF and vice versa.

Nevertheless, it is significant the way the relation between CFF and

the performance scale is paralleled by their relations with several

other variables. In Colgan's study both CFF and the performance

scale iere significantly related with age. In Anderson's and Bartain's

studies, using college students as subjects, the performance scale was

found to have very little relation to either the AGE or the HPA. In

the present study, again with college students as subjects, CFF was

found to have little or no relation with the ACE or the IPA. echisler

found the performance scale to have little relation with scores on his


emoary test for subjects of about the same age. In the present study

the subjects were all approximately the same ago and CFF was found to

have little or no relation to their scores on Wachsler's memory test.

nLds is either a remarkable set of coincidences or further evidence

of definite relationship between CFF and tre W-b perfonqcnce scale.

RFrther research along tnis line is indicated. This nuch can be stated

at present: none of the findi-as in the present study ii any way

contradict the observation made Dy Colgai that CFF and thLc; -5 per-

formance scale are related.

CFF did not correlate significantly with the Minnesota Paper

Fonr board or with the height of the suojecta. It is concluded

that CFF is unrelated to perceptual ability or to height. The

specific aspect of perceptual ability measured by the Minnesota Paper

For, board was found by tMurphy (22), in a fcctorial study, to oe tne

ability to manipulate spatial relations. The lack of relation between

CFF and height. does not preclude possible relations with other body

sises or proportions not studied.

To sumarize the findings of the present study, CFF was found

to be unrel:-toa to attention--oither abstract or sensory. Neither

was CFF related to scholastic aptitude or scholastic achievement. In

addition, there was little or no relation ooservaolo uetween CFF and

perceptual ability, memory, or height. In regard to some of these

variaules CFF exhiDitod a lac, of relation sirdilar in pattern to that

reported by other investigators for the r-b performance scale.

Iriarnuch as attention has seldom been studied in relation to


other psychological abilities as measured by modern standardized

tests, the results of the present study that bear on this point are

of great interest and practical importance. both Philip's Test of

Attention and the measure of sensory attention correlated very signi-

ficantly with the ACE (+0.55 and +0.46, respectively). Philip's test

correlated with HPA (+0.40) almost to the same extent as did the

ACE (40.45). By means of multiple correlation, HPA was correlated with

ACE and the attention battery of Philip. The multiple correlation

coefficient was +0.54. To provide a means of evaluating the influence

of attention on the relation beoteen the ACE and HPA, a partial corre-

lation was conmuted between ACE and IIPA with attention statistically

held constant. The resulting coefficient ias +0.30. The relation of

attention to HMA with the ACE held constant is represented by a coef-

ficient of +0.20. The test of sensory attention, Philip's Test of

Attention, and the ACE all correlated with perceptual ability to about

the same degree (+0.38, +0.39, and +0.39, respectively). Neither the

attention test nor the sensory attention test correlated significantly

with either memory or height.

On the basis of the intercorrelations between then, the test

of scholastic aptitude seems to be measuring to a considerable extent

the same function as does the test of attention. If Philip's test

is measuring attention in the sense defined by him, this finding is

entirely reasonable. Any careful observation of the ability require-

ments for acadunic success will show attention to be one of the traits

differentiating good from poor students. FPrther, the correlation


found between the test of attention and IIFA substantiates this obser-

vation. That attention is not the only ability needed for scnolastic

aciievo,.ient is demonstrated by the significant r Oot.oan ACL and ILPA

with attention statistically held constant. The r between attention

and H.iA, o0. thu other htr.1, loses significnce iuhna the ;.C is held

coA.stuit. Anlothor ability, besides attention, noasurod by the ,.C

is nemory. This is demonstrated by the ftnding that menoxry correlated

sELtiiLfcantly ulth the AGE uut not ith either of the attention tests.

The ACE exhiDited an interesting pattern of intercorrelations.

Lesidos correlating highly with its own parts, the ACi correlated

to a si[i[ticant degTee uith almost all of the other variables.

Lcholastic aptitude was trhus related to memory, prceptual ability,

abstract LnJ sensory attention, and scholastic achLevement. Cnly

CrT, height, r-nd two of tre attr iti .n subt.sts (!iumer Span and

Alphabet) did not corrolato a-igiific:uLt1y id u the L.C. Lyon so

hei;ht wan related to the LinLuistic part, with tho r barely reach-

Ln. the 5Z lovol of confidenco--a finiiin co.,siderea of no L.portance.

To return to a discussion of CFF, it is apparent that this

". msinsitive indicator of neurological efficiency" (!Landis, 7)

is no lssitive indicator of any of the joveral psycholo.icaCl functions

included in the pruueat study. At is of course posaiule that CFF

is sucij an indicator for soao psychological uniction nat ctudled

here, bit tictler this is so or not furtihr specificatio. of the

physiological nature of CFF seems desirable. There i. the questionn

of whether a high CFF for one individual neans that his nurrous system


is more "cfL 'jricnt" than that of an individual with a low CFF. Or

perhaps a high CFF is "normal" for one person, while a low CFF is

proper for another--with CFF being a "sensitive indicator" for varia-

tions (intraindividual) from this norm. either specification of the

physiological nature of CFF is more desirable than the present



For suojects and testing conditions similar to those used

in this study, the following conclusions are presented:

1. CFF is correlated with attention, as defined oy Philip,

to the extent of a Pearson r of +0.23. This r falls within

fiducial limits (1%) +0.58 and -0.19. CFF is correlated

with sensory attention to the extent of an r of -0.15.

This r falls within fiducial limits (1%) -0.52 and +0.27.

Ho definite relationship was found in this study between

CFF and attention.

2. The results of this study indicate that it is probable

that the correlation between CFF and scholastic aptitude

as measured by the psychological Examination of the ACE

is low. The r of +0.02 with its corresponding fiducial

limits of +0.42 and -0.38 indicates the extent. The

absence of relation found between CFF and the ACG. in no

way contradicts the relationship previously found by the

writer between CFF and the Wechsler-bellevue Intelligence



3. Teio Penrson r toetoeen ho.or poi-t a--orace -and CFT wac

40.12 ..ith fiducial limits of .0.50 and -0.29. .colastlc

ac;ievo-aent as representca btr h3.ior point average and CFF

are probably not relatea to any appreciaole ext.nt.

I. here is prooaoly little or no relation Jot:.,an CFF and

perceptual aluiity as measured by the ilinnesota Paper Fona

board nor between C."F and memory as moas-red by the :.echsler

1 Coory Scale. The fiducial limits for the r Detween CFF and

the Miinnesota L'rer Forni board (JO.0) are +0.143 nd -0.36.

The corresponding values for the r betccwnO CFF and th 4 ech-

alor 'Memory Scale L-O.0l) are *4O.1 and -0.39.

5. 'l:iiCt anjd CFF are probably unrelated for a..ult szDjccts.

T-o exb-nt is iidicate.l uy an r of -0.04 i th its fiducial

linits of -3.13 aid +0.36.

6. L'hilip's Te't of Atterntioli is sLTii'iccALtiy related to the

hAC'. Die r of ).55 with its fiducial Lirts or ).70T ao:a

r0.19 indicates the extent. acniolastic aptitcde iz a lunc-

tioL of attention, memory, perceptual ability, and possibly

ot:er metal aDitities not represented in the present ztudy.

7. The measure of scholastic aptitude, the AC:, was foand to

relt., to the measure of scholastic achiev.eent, H.r, to

about the sano extent as t.nat reported in tU

8. '3oro is need for further clarification by phycio ogists

of the sopcific status of CFF in neurop.nysioloiceal theory.


Further study of the relation of CFF to other psychological

variables is indicated since many experiments have shown

C.F to be importantly related to events occurring in the

central nervous system.


1. Anderson, E. E., et al. iilson College studies in psycholo/"i
I. A corparisorL of the Lochclor-bellovue, Revised Stanford-
binet, ard Americ.n Council on Education T3ets at the College
Lovcl. J. syrczol., 19.2, U1, 317-326.

2. bartley, 3. H. Vision. New iork: Van Nostrand, i91..

3. broke!:, J., SL-ionson, E., and Fr i:Lin, J. C. A note on aothod-
olo ical evaluation of oclected visual tests. An. J. Ophth.,
1940, 31, 979-984.

h. Colo;an, C. H1. Vi-sual oDrzG tncss in relation to the frequency of
intornittent stimulation. UnrubliUtned Mastcr's Thesis,
Univer. of Florida, 195i.

5. Colgan, C. M. Critical flic,.or frequency, ane, and intelllgence.
Amor. J. 1'scaol., in press.

6. Crauford, A. L. and ournham, P. 3. Forecasting, college achieve-
ment. tart i. General considerations in tne measure-ent of
t.cajdaic oronise. flew iaven: isle Univer. tress, 19b6.

7. Gronuocn, L. J. iEsentials of psychological testing i;ew York:
Harper, 1949.

8. Crozier, W. J., and ~.olf, L. J. Theory anJ measuroment of visual
mechanisms. V. Flash duration and critical intensity for
resporso to flicker. J. cn. Physiol., 19 2h, 635-65L.

9. Crults Lnaboratory. ilectroniic tuoes and circuits. !ew Yor.-::
McOra,-Hill, 1907.

10. Gellhorn, L. hnoxia in rulation to the visual syoatc. In
H. lidvcr (Ld.) Visual ilecianisnms. Vol. 7. Aiol. ajinos.
Lancaster: Jaques CatLelT, J19k. ip. 73-05.

11. Halstead, n. rain and intnlliLcnco. CnicaCo: Univer. Cnicago
Press, 19h7.

12. Haames, John A. Critical flicker frequency an'd attention.
Unpublicheoa master' Thesis, Cathol. Unrver. Amer., 1950.

13. .Ieoc, D. 0. The org.aniiation of behavior; a neuro,.-ychjio.Lica
theory, ac.J ortc: .iiy, 19Ly.

14. Holsinger, Karl J. Preliminary report on ASear-an-Holainger
Unitary Traits Study, ho. h. Factor patterns a-,d rezidu-a
correlations for Thorp and Mooseheartdata. Dept. of Educ.
kull., 1936, Univer. of Chicago.

15. Koffka, Kurt. Principles of gestalt psychology. New York:
Harcourt, urcce, ,135.

16. Kdhler, .0olf~a.;. ., place of value in a 6,orld of facts.
New York: Liver-ight, 193.

17. Landis, C. Something about flicker-fusion. Sci. ilon., 1951,
73, 308-31h.

18. Lashley, i. S. brain isl and intrilli-cncc: a-Lrit.tive
stud'r of inurie3 to the brain. L;iica.go: Uilivr-. Chicago
rress, 1929.

19. Likert, it. and 4uasha, ,. H. devised Mdinnesota Paper Form board
Test (i'lanual, 19h8). The esycaological Cor:orration, le. York.

20. McNemar, 0. W. Ordering of individuals in critical flicker
frequency under different measurement conditions. J. Psychol.,
1951, 32, 3-2h.

21. Monaghan, E. A. lajor factors in cognition. Stud. .'sc.,ol.
:.chiit. Cathol. Univwr. Amer., 1935, 3 No. -, b p.

22. Murphy, L. W. The relation between mechanical ability tests
and verbal and non-verbal intelligence tests. J. Psychol.,
1936, 2, 353-366.

23. Nagel, E. (Ed.) John Stuart Mill's Ahilosoohy of scientific
method. Aew- Iorkt-? ner, 1950.

2L. Philip, B. R. The measurement of attention. Stud. Psycnol.
Psychiat. Cathol. Univer. Amer., 1928, 2 No. 1, vii + dl p.

25. Pillsoury, w. 1j. Attention. hew York: Mlacillan, 1908.

26. Reinhart, M. A scale for measuring the g-factors in intelligence.
Stud. i'sychol. Psychiat. Cathol. Univer. Amer., 1931, 2 Do. 5,
iv + h2 p.

27. Sartain, A. Q. A comparison of the hew Rtevised Stanford-binet, the
bellevue bcale, and certain group tests of intelligence.
J. soc. Psychol., 1946, 23, 237-239.


28. Saucer, t. i. 'Ihe cfect of dark adaItatiorR upon t;:c perception
of apparent motion. Un.julishel hastor's Thesis, Univer. of
Florida, 1950.

29. Laucor, R. r. The nature of perceptual processes. aciencc,
1r53, 117 lJo. 30O7), ,56-55U.

3J. ~herringtorL, Charles. tan on 11iri5 ture-.2nd Ed.) Camurid;e,
t.igli and: Cambridgc Univer. rra!s, 1951.

31. aLmonson, Ernt ania rolek, Josof. Flic.oer lusicn rrequcecy.
ac;g;onjid and applications. Physiol. uav., 1952, 32, 349-378.

32. 6poarnan, Charles. The abilities oi man. e-%, lork: Macmillan,

33. Spoarman, Charles Psycholory down the arcs, Jol. I. Lo.don:
Macnilla1, 1937.

31,. S.ear.nn, C. and Jns, a. liiuman aAlit. ioadon: liacailian,

35. Ta-cner, ;. '., Jr. h relationship bLotev-n via;al fusion and
inteLlicaucc. Unpuolishedl master's l'heis, Univcr. of Florida,

36. Tanner, Fr., Jr. A preliin nary Livestigation of tre relationship visual fusion of interAittent liht and irLtoiligunce.
Lcienco, 1950, 112 IUo. 2903), 23)i-J33.

37. rharstone, L. L. A factorial study of perception. Psycnometr.
Atnogr., 1911, uo. b.

30. fTurstone, L. L. The vectors of tie .mind. Chicago: Uriver. of
Cnicago tross, 1y35.

39. niechsler, Lavid. A standardized memory scale for clinical Luse.
J. '3s cnol., 1945, 19, 87-95.

hL. >.oodrow, Herbert. The measurement of attention. syciiol. Iono .,
191L, 17, I;o. 76, h.

hl. neforenco data for ra .o c.-!inacrs. Federal telephone and eUio
Corporation, feuw orki 19 h5.


Carroll Monroe Colgan was born May 12, 1925 in Jacksonville,

Florida. He attended elementary school in Jacksonville and Marianna,

Florida, and graduated fror Andrew Jackson High School--receiving

his diploma in absentia. Upon his discharge from the Army Air Forces

he entered Jacksonville Junior College and received a diploma in

May, 1947. He then enrolled at the University of Florida and graduated

with honors and a Bachelor of Science degree in June, 19h9. Entering

graduate work in Psyc'oloi also at the University of Florida he

received the degree Master of Arts in June, 1951. He was employed

during the same period as a graduate assistant in the Psychology

Departnont. Upon receipt of a Fellowship from the United States

Public Health Service he continued his studies in 'syc-ology with

Philosophy as a minor while engaged in several research investigations

at the Moosehaven Research Laooratory in Orange Park, Florida.

The author is married and has one son. He is a member of

Phi ueta Kappa, ihi Kappa Phi and an associate member in the Society

of the Sigma Xi and the American Vsychological Association.


This dissertation was prepared under the direction of the

chairman of the candidate's supervisory committee and h?.s been

approved Dy all members of the coaiittee. It v;as submitted to

the LDan of the College of Art r.nd Sciences and to the Graduato

Council and was approved as partial fulfil-ment of the rduiroments

for the dgreeo of Doctor of Philoaophy.

August 9, 195h

Dean, CotiLet of Arts ana sciencess

Dean, Graduate school



/ I : .-,

/.' ~ ..7 ?,