masts:

0715‘]

This is to certifg that the

thesis entitled

An Investigation of the Effects of
Sham Feeding on Motor Endurance

presented by

Edward Barman

has been accepted towards fulfillment

of the requirements for

M.A. degree in :thical Education

, fiéwf‘ W

)4 V
f @fl/proiessor d

[)ate_Mazch_é._1953—__

,q

 

 

 

AN INVESTIGATION OF THE EFFECTS OF SHAM

FEEDING ON MOTOR ENDURANCE

by
EDWARD BERMAN

A THESIS

Submitted to the School of Graduate Studies of Michigan
State College of Agriculture and.Applied Science
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Physical Education

1952

 

“\0

(f

i

I!

AC KN OWLEDGE‘IIENT

I wish to express my gratitude to Dr. Henry J. Montoya for

his help and guidance in this study.

CHAPTER
I.

II.
III.

IV.

V.

TABLE OF CONTENTS

TIE ROBLM .OQOOOOOOOOOOOOOOOOOOOOOOOOOOO0.00....

Statement of the problem ......................

Importance of the study ...... ............ ....
Sources of data ...............................
Method of collecting data .....................
Limitations ...................................
REVIEW OF LITERATURE ............................
METHOD OF PROCEDURE .............................
Materials used ...............................o
Testing procedure .............................
Statistical method ............................
RESULTS .........................................
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS .......
Summary .......................................
Conclusions ...................................

Recomndations 0000000000000000000000000.0000.

BIBLIOGRAPHY 0.00.000.00.000000....OOOOOOOOOOOOOOOOOOOOOO

APPEIqDH .00000000000000000000000000000000000000000.0000.

PAGE

QOOF‘WNNI—‘l—‘H

n) FJ F4 P‘ P‘ F‘ F‘
r4 \o E; .4 .4 .4 x» to

LIST OF TABLES

TABLE PAGE

I. Ranking the Times of the First Trial . . . . . . . 10

II. Method of Equating Groups . . . . . . . . . . . . ll
III. Descriptive Statistics and Significance Ratios in

the Comparison of Ergometer Times for the Con-

1310]. and. Ewerinlenta]. Groups 0 o o o o e o o 9 1h

LIST OF FIGURES

FIGURE PA GE

1. The Effects of Sham Feeding on :1 Motor
EnduranceTest. . .. . . .. . . . . . 15

CHAPTER I

THE PROBLEM

Introduction. It is a popular opinion that a person taking

 

a capsule containing vitamins, phosphates, or certain other dietary
supplements will perform above his normal capacity. This improvement
may be due to the actual beneficial effects of the contents or to

its supposed beneficial effects. It is commonly believed.that a
psychological effect caused by merely taking a placebo will improve

motor performance.

Statement of the problem. 'Ihe problem is to determine if sham

 

feeding‘will delay the onset of fatigue in a motor endurance test.

Importance 93 the sturdy. Many athletic coaches feel that

 

capsule feeding<ioes produce a beneficial effect in the respect
of increasing motor output. 'Whether the capsules contain anything
beneficial, or the increased motor performance is caused psychologi-
cally is of little importance to coaches. Since improvement is
their only interest, no records have been kept in regard to a
psychological effect increasing motor performance.

In several other dietary studies reference is made as to the

possibility of a psychological factor distorting the results. In

2
many of these studies a placebo was used to account for the effect.
However, the importance of this factor has never been determined.
This study will aid in determining to what extent measures should

be taken in future dietary experiments to limit this factor.

Sources gfmggtg. Fifty-six boys enrolled in a physical educa-
tion service class at Michigan State College were given a motor
performance test. These subjects were all selected from the same
service class so that the conditioning benefits derived from Physical
Education by them would all be as similar as possible. Furthermore,
the testing period was toward the end of the term, and it'was un-
likely that their physical condition would change within the two
week testing period after participating in strenuous physical
activity for eight'weeks.

The measure of motor performance was an endurance test on a
bicycle ergometer that each subject rode at 20 m.p.h.'with four

pounds resistance.

‘EEEESQ g: collecting data. The subjects were tested on the
bicycle ergometer at the specified speed and friction, and the times
were carefully recorded. These scores were ranked and divided into
two equal groups. The subjects were tested again two weeks later
so that any conditioning to the machine might be kept at a.minimmm.
This second test was administered to each individual on the same
day and approximately at the same time as his first test was taken.

It is believed that the time of day influences performance and that

3

such means would tend to eliminate some of the extraneous variance.
One group received a capsule containing corn starch and was
instructed that it was a vitamin pill that would probably increase
their performance on the bicycle ergometer. The other group served
as a control and received nothing. In order to avoid bias the
investigator did not know which individuals had taken the capsules.
All the subjects were instructed on both tests to extend themselves

to their physiological limits.

Limitations. On an endurance test of this type there are

 

many factors that might influence the performance of each individual.
The room, atmosphere, and experimental procedure were kept the same
in both tests.

1. Weather, diet, sleep, and studies were sources of inter-
ference that could not be regulated. It is impossible to make
allowances for the unpredictable changes in an individual that
cause him to excel one day and fail on another.

2. The age group involved might have been a limiting factor.

3. Non-athletes were used as subjects and the results might
not be applicable to athletics.

h. Any psychological effect, because of individual diff-
erences, might cause an increase in performance in some individuals

and a decrease in others.

 

CHARTER II

REVIEW OF THE LITERATURE

In reviewing the literature it was found that the term "psy-
chological effect" was mentioned in several studies concerning
various drugs, dietary supplements, and other ergogenic aids. In
most of these studies the mention of this factor was merely inci-
dental. Typical of this type of reference is the study by Staten:l
in which he used commercial yellow corn meal as a placebo to offset
the psychOIOgical effect. He recognized the possibility of a psy-
chological effect influencing strength test performance during the
lecithin feeding period.

Karpovich2 reviewed several studies on ergogenic aids and
stated, "Sham tests in which, instead of the 'real stuff,’ inert
substances are given should be employed, and the psychological fac-
tor should be controlled as much as possible." In his review he
suspected that many conclusions were derived from fault;r investiga-
tions in which no measures were taken to account for the psychological
factor. He cited a study on fruit juices that was reported to increase

the speed of swimmers. Since experimental evidence has shown that

fruit juices are not directly connected with muscular performance,

 

1Wesley M. Staten, "The Influence of Soya Lecithin on Muscular
Strength," Research Quarterly, 22:201, May, 1951.

 

2Peter V. Karpovich, "Ergogenic Aids in Work and Sports,"
Research Quarterly Supp., 12:1t32, May, 19141.

3

he suspected a psychological effect.
The classic experiment of Brown-Saguard in which he thought

he was injecting himself with sex hormones also falls into this

class of faulty conclusions. It is the consensus that Brown-Sequard's

supposed physical and mental improvements were entirely due to a

psychological effect.“5

Karpovich6

reemphasized the possibility of a psychological
effect by citing an experiment by Marbe. Marbe's placebo, a drop
of congo red in distilled water, produced the same effect as
three to five grams of recresal.

It should be pointed out that no one has measured the psy-

chological effect in endurance work.

 

3 Ibid., p. tho.
h Ibid., p. tho.

5 C. Donnell Turner, General Endocrinology, (Philadelphia: W.
B. Saunders Company, l9h9), p,.g6-7.

 

6 Peter V. Karpovich, 23. 532., p. Milt.

(‘

 

CHAPTER III
METHOD OF PROCEDURE
MATFILIAIS USED

The measure of motor performance was an endurance test on a
frictional bicycle ergometer. Using Karpovich's1 description as
a.model, a similar ergometer was constructed for The Michigan
State College Physical Education Department. Friction was developed
by a flywheel rubbing on a length of automobile brake band. The
upper end of this band was connected to a commercial spring scale
‘which was graduated into pounds and single ounce divisions. The
lower end of the brake band was connected to suspended weights.
'When a subject pedalled the bicycle, the flywheel against the brake
band lifted and partially supported the suspended weights. The
reading on the scale was reduced equal to the amount of friction.
Thus the difference between the suspended weights and.the scale
reading was the friction or work.load.

In this study five pounds were suspended from the lower end
of the brake band and the scale reading, while riding at twenty

\

miles per hour, was one pound; thus, four pounds was the work load.

 

1 Peter V. Karpovich, "A Frictional Bicycle Ergometer,"
Eesearch.Quarter1y, 21:210-215, October, 1950.

 

 

. .

ad‘ ll ‘1 ‘3 in

2...

. (U... .11.; '1-
l
m

Occasionally, after continued use of the instrument, the scale
reading would increase several ounces. This meant that the friction
had decreased this amount. To bring the load back to the standard
four pounds the same number of one ounce weights were added.

Mounted on the bicycle was a speedometer and a revolution
counter. The speedometer was necessary to enable the subjects to
pedal at the required twenty miles per hour. The revolution counter
'was covered so that the subjects would not know how well they were
progressing.

One addition to the ergometer used in this study was a sweat
guard preventing drops of perspiration from falling on the flywheel
and causing disturbances in its function.

A stop watch was employed as the measure of duration of work
rather than the revolution counter. The watch was started as soon
as the subject reached twenty miles per hour and stOpped when he
could no longer maintain the pace. The revolution counter recorded
the number of‘tuxns it took each subject to attain the required speed
and also the number of turns made in coming to a stop. Since this
number of revolutions was not constant, the revolution counter tended
to record erroneous results of work done at twenty miles per hour.

Ca sules were usedrso that the subjects could not taste the
supposed energy substance. This allowed the use of corn starch.

At first it'was hoped that the capsules would hold their contents
in tact long enough so that the entire experiment would be finished
before the contents could be exposed in the gastro-intestinal tract.

However, according to the manufacturer, the capsules were dissolved

partially within five minutes and a portion of the contents was
exposed to the action of the body fluids. Since the subjects who
received the capsules were finished within ten minutes after
taking the capsules, it is assumed that there was no physical
benefits derived from them.

It is very unlikely that corn starch can be broken down, absorbed,
and transported to muscle tissue within five minutes. Furthermore,
each capsule contained less than one gram of starch. Since corn
starch yields 362 calories per hundred grams, the number of calories
that could possibly be derived from each capsule was less than five.2
This figure seemed inconsequential in the light of the fact that
several thousand calories are used in daily activity.

It should also be pointed out that no other beneficial effects

are likely to be derived from this amount of corn starch.

TEST ING PROCEDURE

Fifty six male freshmen and sophomore students enrolled in
the same physical education service class were tested on the bicycle
ergometer at twenty miles per hour with four' pounds resistance.
After this first test two subjects were dropped from the experiment.
One had been sick in the hospital and was below his normal condition.

His improvement in the second test would be due chiefly to his improved

2 Agriculture Handbook No. 8, "Composition of Foods," U.S.
Department of Agriculture, June, 1950, pp. h6-h7.

physical condition. The other subject was drOpped because of his
lack of interest and attitude. It appeared quite obvious that he
did not extend himself to his physiological limit.

The scores of this test were tabulated and ranked as shown in
Table I. from these rankings two equal groups were arranged in
Table II. The arrangement was made in such a manner that if a sub-
ject in one group failed to take the second test the nearest score
to him in the other group could be dropped to keep the groups equal.

Two weeks later to the day each individual rode the bicycle
ergometer at the same speed and weight as on the first trial. On
this test one group, called the experimental group, received a cap-
sule containing corn starch. The other group, the control, received
nothing. Each subject reported at his appointed time to one ’of the
laboratory rooms next to the room containing the bicycle. An assis-
tant there gave the experimental subject his capsule, and infomed
him that the capsule would in all probability enable him to ride the
bicycle for a longer period of time.

The investigator did not lmow whether the subjects belonged
in the control or the experimental group at the time of the second
test; and the subjects were instructed not to inform him until after
the results had been recorded. After a brief warm-up period of fifty
slow revolutions, which also preceded the first trial, each subject
was again urged to extend himself to his physiological limit. The
whole procedure, from the time a capsule was taken until the ride

was finished, was less than ten minutes.

10

TABLE I

RANKING THE TEES OF THE FIRST TRIAL

 

 

Hm
(Mine, 8630, and tenthS)

Rmk

Hm

(Mine,8800,and tenthS)

Rmk

 

 

AwhwfiUOUVInvLHRvnvL462nuC21¢71Au1inuCJOgfilAuLHQJOI7IAu

9.9.01 9.951JRVOLRVO/ 9.9.Rv Aunv7lCJ71Au
LuLULH CJCJCJCJnunvnVMW1.1.1‘mu 9.1J LHCJnu951Jnv

.0 O. .0 .0 O. .0 O. O O. O. O .6 O. .0 Q. .0 O. O.

111111111222222222222223339

hQHDHDHDAUQUalLuLuLuRunufiJOQCJ1annUHDQUQUnUHDLUnU

H)

O O
MMnud/AvO/thJOJLHHDolleuO/OaQJAuRunv1i1i1LndH3HDleU
CJfiJCJnu1l1¢1m1a1+1¢1a1l1l9b9bOaOafiJQJfiJQJQJQJQJQJQJ

0. O. .0 O. 00 .0 O. .0 O. O. O. O. O. O. O. O. .0 O. O. 0. O. .0 .6

1*1.1.1.1.1.1.1a1.1.1.1.1l1.1.1.1.1.1.1.1.1.1.

1.9.1JL4C/Av7tRuO/nv1m9512WmCJAv7 :O/nv1l :QJLH Au?!
1l1l1¢1l 1.1.11:9.9.:9.9. 9—9.

 

 

 

 

I Eli-slip I

\.

1.... I...

I'll

.5 u/ in .hr.

1.va admin

TABLE II

WTHOD OF EQUAT I‘JG GROUPS

 

 

Group I Group II
W

l 2

h 3

S 6

8 7

9 10
12 ll
13 114
16 15
l? 18
20 19
21 22
2h 23
25 26
28 27
29 3O
32 31
33 3h
36 35
37 38
to 39
Ll 1:2
m 1:3
1:5 h6
ha 117
1:9 50
52 51
53 Sh

ll

12

Two subjects failed to take the second run on the bicycle.
They were ranked eleventh and fourteenth, and were in the same
group. Rankings twelve and thirteen in the other group were also
dropped so that the groups would remain equal. This brought the

number of subjects down to fifty who completed the whole experiment.
STATISTICAL METHOD

The data derived from this investigation were subjected to
the following statistical treatment:

1. The coefficient of correlation between the scores of the
control group in both tests was calculated.

2. The coefficient of correlation was also calculated for
the scores of the experimental group in the first and the second
tests.

3. In order to check the statistical significance of the diff-
erences in achievement shown by each group in both.tests the large
sample "t" values were calculated between the scores of the first
and second tests of the experimental group, and the scores of the
first and second tests of the control group.

h. ‘A small sample “t” was also calculated using the differences
in the scores of the first and second tests of the control group,
and comparing them to the differences in the scores of the experimental
group.

These data and calculations are included in Chapter IV.

CHAPTER IV

RESULTS

The results are presented in tabular form on page lb, and
are also represented graphically in Figure 1 on page 15.

The coefficient of correlation between the first and second
tests of the control group was .959. This correlation indicates
high reliability of the bicycle ergometer and the consistency of
the control group. It was thought advisable to correlate these
tests excluding the twenty fifth subject who scored unusually
high on both trials. It was felt that this subject's scores could
possibly influence the correlation unduly. 'Without his scores
the coefficient of reliability dropped from .959 to .901 which is
still high for this kind of measurement.

The coefficient of correlation between the first and second
tests of the experimental group was only .7h3. The reason for the
experimental group correlation being lower than that of the control
group may be explained.by the variable, the placebo. The placebo
probably influenced some of the experimental group to either improve
or lower their scores in the second test. These changes, however,
were not great enough to show up in the "t" scores. (See Table III)
These coefficients were converted to Fisher "2" scores and the sig-
nificance of their difference in magnitude was calculated. The "t"

value was h.3 which was significant at the .001 level.

‘1h

TABLE III

DESCRIPTIVE STATISTICS AND SIGNIFICANCE RATIOS III THE COMPARISON
OF ERGOIIETE’R TIRES FORTHE CONTROL MID EXPEREIENTL GROUPS

 

 

 

 

mean SD N’ Range Signif. P
(sec.) Ratio
Control 115.83 51.3 25 50.5-306.6

(first test)

Control

)

)

3 .7h3 .227
(second test) 118.50 53.9 25 60.0-350.h )

Experimental 113.18 u1.8 25 hh.9-217.7 )

(first test) )
) 1.090 .138

Experimental )

(second test) 119.83 h1.6 25 hh.6—l95.l )

 

III-uni: :='

. . . -- E1111": “a1

$1511-11'1111" it"‘imr- "fisflfihm’fifi- 1111;“

111111111"; first .11" '13.... .1
.111 was mania-swap

:SEE ui-pifl
1111-11. 11111? -.111

131.11.“ _
' 5‘1” «£59111
"-Twigrm 1151' $1.11]

mesa=
u- ”1:3,:

==idmwmwsg= “P @-
-.-.._ 5;?» :2qu mass...
111.111 11 - .111saars _ ass

 

16

A large sample "t" value was calculated for the control and
experimental groups on the first test. The coefficient of correla-
tion was .9h3; the "t" value was .710; and the probability was .239.
This showed that the groups were fairly equal and any differences
between the two groups at the start were insignificant.

Both.the large sample "t” and the student's "t" values were
calculated because the number of cases seemed to be on the border-
line as far as rules governing the use of the.matistics. The
student's "t" was calculated by using the differences in the scores
of the first and second trials of each group. Its value was .552‘whiCh
was also insignificant at both the 1% and 5% levels which were

2.797 and 2.06h respectively.

CHAPTER V

summRY, CONCLUSIONS, AND REconmmATIONS

SU WY

The purpose of this investigation was to determine if sham
feeding would produce a psychological effect that would delay the
onset of fatigue in a motor endurance test. Fifty college freshmen
and sophomores were pretested on a bicycle ergometer and arranged
in two equal groups according to their scores. One group received
a capsule containing corn starch and was instructed it was a
"quick energy pill." The other group served as a control and re-
céived nothing. Both groups were again tested on the bicycle er-
gometer at twenty miles per hour and against four pounds resistance
load.

All the subjects were selected from the same service class so
that conditioning would not be a disturbing factor. Each subject
was tested the second time two weeks after his first test and at
the same time of day.

The results were statistically analyzed and large "t" and

students "t" were calculated.
CONCLUSIONS

On tie basis of the data derived from this investigation the

following conclusions are set forth:

18

1. No significant differences were demonstrated in the
improvements in motor endurance of flue experimental group over
the control group.

2. Since the mean of the experimental group Started lower
and ended higher than the mea-s of the control group, a trend
was indicated that sham feeding does increase motor endurance.

3. The differences in correlations between the first and
second trials of each group further indicated that some addition—
al factor, presumably a psychological factor, was present in the
experimental group but not in the control.

h. The reliability of the frictional bicycle ergometer used

in this study was proven to be quite high.

RE .212 I‘ENDATICI-J S

The following recommendations are made for additional inves—
tigation into the effects of sham.feeding:

1. This experiment should be repeated using athletes, former
athletes, or physical education majors for subjects.

2. It is recommended that a similar experiment be set up

using a placebo that would appeal to the sense of taste.

BIBL IOGFJKPHY

20

Books

Turner, C. Donnell, General Endocrinologg. Philadelphia: W. B. Saunders
Compam', 191:9. EOE pp.

 

Periodical Articles

Karpovich, Peter V. , “A Frictional Bicycle Ergometer," Research
Quarterly, 21: 210-215, October, 1950.

Karpovich, Peter V., "Ergogenic Aids in Work and Sports," Research
Quarterly Supp., 1?: 1:32—1:50, May, 19141.

Staton, Wesley M., "The Influence of Soya Lecithin on Muscular
Strength," Research Quarterly, 22: 201-207, May, 1951.

 

Government Publications

Agiculture Handbook fig. _8_, "Composition of Foods," U. S. Department
of Agriculture, June, 1950.

.x

 

- -n‘ r- ‘4“X'. 411'": ”I

7!
Al

{D

 

1»...!Irqla\.1i.ufiLFt . I. ‘4 .

APPEND II

22

BASIC DATA FOR EXPERIMENTAL GROUP

 

 

 

Subject Time of Time of Difference in
First Run Second Run Scores
Jeffery M3 hh.6 -.3
Wright 56.5 83.1 {26.6
Crowe 6905 Shell .1591
C00!) 7308 8001... £06
Irish 7’40]- 8506 05
Chambers 83.2 91.3 .1
Barnes 86.5 108.5 22.0
Terpeny 91.0 89.1; .1.6
Wintemut. 9105 7,403 .1702
Jolmson 95.0 133.2 {38.2
Houdek 95.5 127.7 /32.2
Hinesley 101.6 109.8 {8.2
Harris 102.0 99.14 -2.6
Zick 111.7 182.9 {71.2
Gavan 112.0 97.6 -1h.h
Hoffman 118 .0 113.9 -h.1
Bertram 122.h 127.2 fins
Dutcher 129.5 162.2 32.7
Pope 132.1 1711.5 $2.1;
White 1h1.1 175.0 $33.9
Height 11.2.8 1119.9 7.1
Summers 16h.7 192.0 7.3
Oleson 166.6 105.8 -60.8
Gowan 205.9 138.0 -67.9
Chapman 21707 19501 -2206

 

 

 

 

 

BASIC DATA FOR CONTROL GROUP

23

 

 

 

 

 

Subject Time of Time of Difference
First Run Second.Run in Scores
Karamazrak 50.5 60.0 £95
Content 55.14 71.5 g 6.1
Sheffey 70.5 69.0 -1.5
Touhey 75.1. 107 .1. {32.0
- Roberts 82.5 67.7 -1h.8
Pyman 88 .1 8901‘ £03
McClurg 90 .0 10h. 2 . 2
Hawley 91.8 82 .1 -907
Stratton 92 .8 81.14 -1l.h
Segura 97.6 10303 #507
Warwich 98.0 99.1: .
Hemingway 102.3 118.3 6.0
Hamilton 109.9 97.8 -12.1
Anderson 112 .h 115 .3 {2.9
Dflsey 113.8 10105 .1203
Smith 128.5 113 .3 -15 .2
Sprott 132.7 122 .3 -10.h
Beardsley 138.6 125.3 ~13.3
Pearce 1,4505 16703 $2108
Swick 153.2 17h.6 {21.1.
Whitehead 170.). 198 .9 {28.5
Walker 187.3 166.3 -2l.O
McDonald 306.6 350.1. {10.8

 

 

 

"J. '. .1 o' . ‘

 

1,1

   

Pfi-

ROOM “SE 0801,

 

It." 's‘
‘-