«a w .... ..... .......... ....,....... <
w. 9‘ ._. .q- . _ ~
1. {2" _ _ _ ,..._ ‘ .37....“ 7&3”: Z‘
‘ ”5% ‘ - ' "£213.; m...“
u .-:,.....':.. L . . m. m..." ..
‘ d‘ I... ~ g...- ‘0‘ I.- M
| GI”
{f ‘. v A; ~ ”L 1-3.; ._.‘_ ‘u-
z: - ‘ .- “ A~--t'«....m?.~.....
~..,..., . 4 ._ - . -u‘ .
'”"‘*'*’v"r—-"‘r~:«:‘ , 2:... ..., ‘ . M.“ Mao- ‘
O”. I".~‘g" am ‘ I. 3‘ Him “0 _ ~ ‘ .
.. .., . - . ”,1 a, . .., . - ,1
.- an {Una ‘ v . -: ‘ 0'
”T 3-— _.g.;.-n “1

”3:11“:

‘ ¥._

.1 ur—

-v'-v~‘:!-~
an; ,

a

A W ,
””3132'72.‘ » . v
“N“«v-iv-iizvxwf-r
‘- " -.'—'.. 47-12%:
I‘ . sf ‘ , ' ‘
4: ga,
Isl.

19‘ “an :2”

 

 

 

‘h
M“.

 

7L1

 

‘ 3

51.13
-L

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIILIIIIIIIIII

3 1293 1038

This is to certify that the
thesis entitled

"The Quantification of Drive III:
Privation of Food and Water"

presented by
Richard A. Behan

has been accepted towards fulfillment
of the requirements for

M}..— degree in PSQ’Chol oay

22%;»?

Majogprofesso

Date /°2'6 'fé/

 

0-169

 

 

MSU

LIBRARIES

 

 

RETURNING MATERIALS:
RIace in book drop to
remove this checkout from
your record. FINES W111
be charged if book is
returned after the date
stamped be1ow.

 

 

 

 

 

THE QUANTIFICATION OF DRIVE III.

PRIVATION OF FOOD AND WATER

BY

Richard A. Behan

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

DOCTOR OF PHILOSOPHY

Department of Psychology
195M

\"\

g,"
\

ACKNOWLEDGEMENT

The author wishes to eXpress his grateful thanks
to Professors M. Ray Denny and Henry S. Leonard
for their help and encouragement throughout the

course of this investigation.

{J
Li:
gh
a»
" }
3

THE QUANTIFICATION OF DRIVE III.

PRIVATION OF FOOD AND WATER

BY

Richard.A. Behan

AN ABSTRACT

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

DOCTOR OF PHILOSOPHY

Department of Psychology
1954

/
Approved .

 

 

‘V

The present thesis was concerned with a test of the

Hullian theory of drive, using privation with reapect to e'q'

food and with reapect to water. The position was formalized,
and consequent theorems were tested.

A combined activity box and panel-pushing device was
used in the present study. The panel-pushing device was so
arranged to provide a measure of force of reaponse, and of
the number of trials to extinction.

Forty-eight male albino rats were used as subjects in
the study, and were divided into two major groups:

1. Twenty-four animals deprived of water;

a. Twelve animals given a large reward (0.20 cc)
b. Twelve animals given a small reward (0.12 cc)

2. Twenty-four animals deprived of food;

a. Twelve animals given a large reward (0.32 gm)
b. Twelve animals given a small reward (0.08 gm)

All animals were given habituation training in the
apparatus during the first five days of the eXperiment,
while on an ad lioitum feeding schedule. Activity level
was recorded.

On days 6 through 13 all animals were trained on the
panel-pushing task, under 22% hours of apprOpriate privation
and reward. Each animal was given ten trials per day.
Activity level was recorded on all 8 days, and force of
response was recorded on the last 3 days.

At the completion of the training series the animals

went immediately to the testing series, where they were
tested after different numbers of hours of privation, 1.8.,
on 1, 2, 6, 12, 22% and hB hours of apprOpriate privation.
The order of presentation of privation levels was randomized
for each animal. Each animal was given h trials per day.
Activity level and force of reaponse were recorded each day.

At the conclusion of the testing series each animal
was extinguished on one of the above mentioned hours of
privation.

The results are as follows:

.la Activity is not a reflection of privation, ESE s3,
but is rather dependent upon the amount and kind of reward
in interaction.with privation, and on a learned anticipation
of reward.

_2. A general concept of drive is not tenable, because
the correlation between the food and.water groups, when
activity is constant, was not significant.

'3. A drive concept is not tenable, because the water
groups did not show the increase in behavior measures which
the large food group showed.

h. There is a significant interaction between the kind
of reward substance and the privation level.

.5. There is a significant interaction between the amount
of reward substance and the privation level.

The implications of the study for activity as a measure

of drive, and for drive as an eXplanatory construct, were

discussed.

 

 

The category of need was discussed, and an alternative

interpretation of need, as interaction, was suggested.

INTRODUCTION .

THE THEORY OF DRIVE.

STATEMENT OF PROBLEM

PREDICTIONS.
SUBJECTS . .
APPARATUS. .
PROCEDURE. .
RESULTS. . .
DISCUSSION .

TABLE

SUMMARY AND CONCLUSIONS.

REFERENCES .

APPENDIX . .

OF CONTENTS

Page

13
33
3h
36
37

I-l-7
7o
83
86
9O

Table

10

ll

12

13

11}

LIST OF TABLES

Analysis of variance of the habituation
activity for the four groups-to—be. . . . . . .

Analysis of variance of the activity data
on the mean of the last two days of habitu-
ation, and on the first day of training . . . .

Analysis of variance of the activity data
for the training series 0 e e e e e e e e e e 0

Analysis of co-variance of activity level
on the testing series . . . . . . . . . . . . .

Pattern of differences, significant at the
one percent level of confidence, for the
testing activity for four groups. . . . . . . .

Summary of the analyses of variance of
ECtIVity at GXDinCtion. a e e e e e e e e e e 0

Analysis of variance of force of response on
the last three days of training for the
fOOd gFOUpS e e o 0 e e e e e e e e e e e e c 0

Analysis of variance of force of reaponse on
the last three days of training for the
water groups. 0 e e e e e e e e e e e o e e o 0

Analysis of variance of force of reaponse on
the testing data. 0 e e e e e e e e e e e e e 0

Pattern of differences, significant at the one
percent level of confidence, for the testing
force of reaponse, for the four'eXperimental
gPOUpS. e e e e e e e e e e e e e e e e e e e 0

Summary of analyses of variance of the
number of trials to extinction. . . . . . . . .

Analysis of variance of deviations from
linearity for activity on the testing series. .

Analysis of variance of deviations from
linearity for the force of reaponse on the
testing series. 0 e e e e e e e e e e e e o e 0

Parallels between the notation of the general
theory of measurement, and the present theory
of drive, in the case of food privation . . . .

Page

I48

50
52

5h

57

S9

59

60

63

66

67

69

69

18

LIST OF FIGURES
Figure Page
1 Apparatus e e e e e e e e e e e e e e e e e e e 38

2 Activity level on each of the five days
during the habituation series, and on the
first day or training . e e o e e e e e e e e e #9

3 Activity level on each of the eight
training days for four eXperimental groups. . . 53

h Activity level on each of six levels of
privation for four eXperimental groups. . . . . S6

5 Force of reaponse on the last three days
of training for four eXperimental groups. . . . 61

6 Force of response on each of six levels of
privation for four eXperimental groups. . . . . 6S

INTRODUCTION

The present thesis has two objectives: first, a
formalization of the Hullian theory of drive; and, second,
a test of the empirical adequacy of this theory of drive.
This introductory chapter attempts to do three things:
first, to isolate the Hullian theory of drive; second, to
consider the theory of drive in the context of the larger
theory of which it is a part; and, third, to consider the
empirical research which bears on the task of testing the
part-theory of drive.

The part-theory of drive finds its most complete
elaboration in the Principles 2: Behavior (16), and this book

 

will be used as the source for the drive theory. The latest
revision of the general behavior theory, of which the drive
theory is a part, is presented in A Behavior System (17).

It is unfortunate that we have to go to two sources, but
Hull did not deal with drive as completely in his System

as he did in his Principles.

Hull's theory of drive begins with the assumption that
animals behave in such a way as to provide Optimum conditions
for survival. From this general assumption Hull arrives at
the notion of 233g. Hull says:

" ... when a condition arises for which action on

the part of the organism is a prerequisite to
optimum probability of survival of either the

,h_,.

 

individual or the species, a state of need is said

to exist. Since a need, either actual or potential,
usually precedes and accompanies the action Of the
organism, the need is often said to motivate or

drive the associated activity. Because of this
motivational characteristic of needs they are

regarded as producing primary animal drives." (16, p. 57).

Drive is an intervening variable, and as such is not
directly observable. However, if it is to be a satisfactory
intervening variable it must be defined by reference to
events which are themselves observable. Hull defines drive
in terms of privation.with respect to a need and the amount
of energy expended by an animal in an effort to get the
needed substance. Hull says:

"Specifically, the amount Of food need clearly
increases with the number of hours elapsed since
the last intake of food; here the amount of
hunger drive (D) is a function of observable
antecedent conditions, 1.6., of the need which
i§fimeasured by the number of hours of food
privation. 0n the other hand, the amount Of
energy which will be eXpended by the organism
in the securing Of food varies largely with
the intensity of the hunger drive existent

at the time; here the amount of 'hunger' is a
function of observable events which are its
consequence." (16, p. 57-58).

 

Drive abstracts from the specificity of need. Needs
are specific for certain classes of environmental supports.
Drive abstracts away from this specificity and is the
resultant Of all Of the needs Operating at a given time.
Hull says:

"The drive concept, ... , is prOposed as a common

denominator of all primary motivations, whether

due to food privation, water privation, thermal

deviations from the Optimum, tissue injury, the
action of sex hormones, or other causes." (16, p. 239).

If

x
\
‘VI
‘Q
._, -m—o
‘
g _

I.

 

 

individual or the species, a state Of need is said

to exist. Since a need, either actual or potential,
usually precedes and accompanies the action Of the
organism, the need is Often said to motivate or

drive the associated activity. Because of this
motivational characteristic of needs they are

regarded as producing primary animal drives." (16, p. 57).

Drive is an intervening variable, and as such is not
directly observable. However, if it is to be a satisfactory
intervening variable it must be defined by reference to
events which are themselves observable. Hull defines drive
in terms of privation.with respect to a need and the amount
of energy expended by an animal in an effort to get the
needed substance. Hull says:

"Specifically, the amount of food need clearly
increases with the number of hours elapsed since
the last intake of food; here the amount Of
hunger drive (D) is a function of observable
antecedent conditions, 1.6., of the need which
IESmeasured by the number Of hours of food
privation. 0n the other hand, the amount of
energy which will be eXpended by the organism
in the securing of food varies largely with
the intensity of the hunger drive existent

at the time; here the amount Of 'hunger' is a
function of observable events which are its
consequence.” (16, p. 57-58).

 

Drive abstracts from the specificity of need. Needs
.are specific for certain classes Of environmental supports.
Drive abstracts away from this specificity and is the
resultant Of all Of the needs Operating at a given time.
Hull says:

"The drive concept, ... , is prOposed as a common
denominator of all primary motivations, whether

due to food privation, water privation, thermal
deviations from the Optimum, tissue injury, the

action of sex hormones, or other causes." (16, p. 239).

,s

 

The reader will note the use of the word energy in
last quotation but one. This word energy will be interpreted

to mean general activity in the present discussion. It is

 

difficult to know just what Hull had in mind when he wrote
this word, and an adequate test of the theory depends upon
a correct interpretation. Whatever was meant by the word,
it must be something observable, because it is to be one

of the observables in terms of which the presence and amount
Of drive are to be determined. Furthermore, it cannot mean
one Of the four observables in terms of which reaction
potential is measured, because this would introduce into
the theory a nice tight circle. he word energy must refer
to some observable which is independent of the remainder

of the system.

By the term general activity we refer to what an animal
does when placed in a specific apparatus for a determined
length of time. The apparatus used will be the MSC activity
box, and the animal will be the white male albino rat. It
is necessary to define activity in terms of the situation,
for as Reed has pointed out at the start Of his review:

Much Of the research to be reviewed depends upon

a general concept Of Spontaneous activity without

regard to how the activity is measured. It will

become evident in the course Of the review that

our concept Of activity must be tied to the measure

Of it which we have used, ... . (27, p. 393).

The elements of the Hullian theory of drive may be

summarized as dummy equations% as follows:
Let: "f", "g" and "h" denote functions,
"a" denote magnitude of activity,
"d" denote magnitude of drive,
"n" denote magnitude Of privation with respect

to a need, then,

d = NM (1)

a = g(d) (2)
An interesting consequence Of this position is that

a = MM (3)

We may now consider the relation of drive to other
concepts of the Hullian system. Hull considers that the
performance of a.particu1ar pattern Of behavior is dependent
upon the strength of the effective reaction potential for
that pattern of behavior. There are five variables which
determine reaction potential in the Hullian system. One
Of these, drive, is the subject Of the present thesis.

The others are: 2, habit (SHR)’ which is determined by the
number of reinforcements to the reaponse being learned;

3, incentive motivation (K), which is dependent upon the
quantity of incentive given as reward; h, stimulus intensity
dynamism (V), which is dependent upon the intensity of the

stimulus; and S, delay of reinforcement (J), which is

 

”We speak of dummy equations because the expressions used
are not equations. The letters "f", "g", "h", etc. represent
functions which are unknown; hence the expressions make no
assertions. They are, rather, assertions Of dependence, and
the hOpe that some mathematical function will describe this
dependence.

dependent upon the time interval between the making of a
response and the receipt Of reward. (18, p. 6-8).

Hull's basic equation for reaction potential,
“15%.

may be written as

SER=CxDxK (14-)
where C is a constant determined by lumping all Of the
variables except D (drive) and K (inceptive motivation) and
holding them constant. We wish to point out here that C
is a constant determined by holding each of the components
of C constant. C could remain constant even if all of its
components were not constant, if their variations compen-
sated for each other. This latter method of holding C
constant is not what is meant in the present formulation.
This second method of holding C constant is fundamentally
weaker than the method adOpted for the present interpre-
tation. It would have to be shown empirically that there were
no interactions between the variables SHR’ V and J before
it could be legitimately used.

The equation for reaction potential as it will be used
in the present study is represented by (h). D and K will be
variables. D will be varied in the test of the part-theory
of drive. K was chosen as a variable because of the possi-
bility that there may occur interactions among the different
values of the variables. As the Hullian system is set up

today there is implicit the assumption that the components

of reaction potential are independent. However, this
assumption may not hold when we deprive animals of food to
set up drive, and then reward the animal with food for the
performance of a given response.

If we now find some magnitude of water reward which is
behaviorally equivalent to a given magnitude of food reward*,
we may eliminate K from the above equation (A). If we,
then, let "e" denote the magnitude of reaction potential we
may rewrite (h), in terms of our previous notation as

e = cf'(d) (5)
where c is a constant. An interesting consequence Of
(1) and (S) is

o = cs'(n) (6)

Drive as a function of privation

The studies of Herenstein (15), Kimble (l9), Yamaguchi
(hl) and Cotton (7) were direct tests of equation (6).
These investigators all used food privation and food reward,
and their results in general support the assertion of (6).
Cotton's results, however, suggest a limitation on the
generality of the Hullian theory. Cotton measured running
time in a straight alley, and when he eliminated trials on
which competing responses occurred, he showed that the

decline in running time with increased privation approximates

 

*Davis (8) has shown that 0.08 gms. of food reward was
behaviorally equivalent to 0.20 cc. Of water reward, using
the panel-pushing device.

a straight line, rather than a negatively accelerated
decreasing curve. Hull has used the results Of'Yamaguchi's
study as a first approximation to the drive function, at
least that based on food privation. The writer was unable
to find any studies on water privation, comparable to these
based on food privation.

E. E. Anderson (I) conducted an extensive correlational
study on the interrelationship of drives in the male albino
rat. He obtained intercorrelations among A7 different tests,
using 51 male albino rats. Anderson concluded that in
general neither the thirst nor the hunger tests correlated
significantly. He states

"intercorrelations between measures Of different

drives, on the whole, are somewhat Sporadic in

occurrence, and there is little evidence of any
important 'general drive' factor influencing
performance upon a large number of tests.

There is however, some slight indication of

relationship between such direct tests of

different drives as eating, drinking, and

Opulation tests."

As the term 'food privation' and the term 'water pri-
vation are used in the eXperimental literature, one receives
the impression that the two are independent. This is not
the case. Finger and Reid (10) and Verplanck and Hayes (37)
have shown that when animals are deprived of food they
automatically deprive themselves Of water, and conversely.
The results of the rather extensive study by Verplanck and
Hayes hae been confirmedby (31) and (5). It is thus seen

that to deprive animals of food is to simultaneously deprive

them of water, and conversely.

Activity as a function ofgprivation

Equation (3), 1.6.. a'= h(n),
has been the object of a large number of experimental
researches. The literature on activity as a function of the
number of hours of privation is very extensive. Since there
have been a number of excellent reviews of this material we
will be primarily concerned here with studies which have
been directly concerned with activity as a function of
differences in privation. Of the general review articles
we may mention the Bulletin article by Shirley (29) which
includes work up to 1929, and the later Bulletin article by
Reed (27) which covers the work between 1929 and 19h7.
Richter (28) has summarized the research on activity which
has been done under his direction. Munn (26) has a summary
of the work on activity which was completed up to the I
publication of his book.

The two methods that have mostly been used for the
study of activity have been the running drum and the tambour-
or spring-mounted cage. The running drum consists of two
circular boards mounted on an axel shaft and separated by
a sheet of wire mesh wound around their periphery. This
basic design has been modified in a number of ways. One
may vary the diameter of the boards, arrange the living cage
so that the animal may enter the drum at its leisure, or
confine the animal to the drum for given periods of time.

One may record the total activity by attaching a counter to

the drum supports in such a manner that each revolution of
the drum advances the counter one unit, or one may record
activity as a function of time. Reliability measures using
this device are on the order of .95, but it is unknown
whether this value reflects the reliability of the measure-
ment of activity or whether it reflects the consistency of
different drums.

The tambour- or spring-mounted cage is a small cube
of wire mesh mounted at three points on either a spring or
an air tambour. The latter is the more effective because
the air pressure provides some dampening effect. This
piece of apparatus has also been modified in a number of
ways to study various aspects of activity. Other devices
that have been used to study activity are tilting cages,
utilizing the movements of the animal to interrupt the beam
of light which activates a photoelectric cell, a horizontal
turntable, the pedometer, and various observational methods.

The most interesting thing about all of these methods
is that they give different results, and modifications of
one apparatus give still different results. It is thus
necessary to know what kind of an apparatus was used to
record activity before one may interpret the results.

With respect to food and water it is commonly accepted
that privation will increase activity up to a point, after
which further privation is accompanied by a decrease in

activity, probably due to physical weakening of the animal.

 

 

 

10

Most of the attempts to quantify the relationship between
the number of hours of privation and activity have been done
with food privation. Using the Columbia Obstruction
Apparatus,‘Warden (39) has shown the number of crossings of
the charged grid drOps off faster with water privation than
with food privation.

Siegel and Steinberg (30) have utilized the movements
of the animal to interrupt a beam of light which activates a
photoelectric cell to study activity as a function of food
privation. These investigators used privation intervals of
0, 12, 2h, 36 and MB hours, and their results indicate that
activity increases as a negatively accelerated function of
the number of hours of privation. 'With reapect to this study
it should be pointed out that the animals remained in their
home cages all during the study and it is possible that the
activity was influenced by expectancy of food.

Hall, Smith, Hanford, and Schnitzer (13) report the
results of a study designed to determine the effects of a
restricted feeding schedule on activity level. These in-
vestigators used 10 standard Wahmann activity drums, each
mounted separately, and each provided with a small stationary
living cage. All wheels were equated for frictional torque.
Control animals had unrestricted access to food and water,
while the eXperimental animals had unrestricted access only
to water, being allowed access to food for one hour in the

morning. The eXperiment was continued over a period of 20

11

days. These investigators report:

"Although activity during the control condition

remained relatively stable, mean daily activity

during the experimental period rose to more than

l,hOO percent of normal, reaching this level on

the twelfth day of restricted feeding. The rise

seemed to take the form of an exaggeration of the

normal daily activity cycle, imposed upon a

rising base line."

Campbell and Sheffield (h), utilizing an activity
recording device constructed by Campbell (2), report the
results of a study from which they conclude that "Starvation
does not instigate activity; it only lowers the threshold
for normal stimuli to activity." The apparatus was a small
round wire mesh cage, pivoted at the center of the base so
that it would tip a maximum of 1/8 inch. Four sensitive
microswitches were placed at each of the four quadrants.
Activation of any of the microswitches advanced an electro-
magnetic counter. The eXperimental animals were placed on
an ad libitum diet for four days while in the apparatus.
Then on the succeeding three days the animals were deprived
of food, but not of water. On each of these three days the
activity of the animals was recorded for ten minutes at
noon, after which an environmental change was introduced
and the animals' activity recorded again.for ten minutes.
An environmental change consistently produced an increase
in activity, and the magnitude of the change in activity was

correlated with the change in the environmental condition,

but not with increased privation. The results of this study

12

stand in striking contrast with the results of other studies
in the area, and the reason for this may be found in the
fact that "The apparatus usually fails to record certain
small movements like scratching, but it records larger
movements such as moving from one quadrant to another or
shaking the cage." (h, p. 320). In other words, the
method may not be very sensitive.

Thompson (36), using a rectangular elevated.maze, has
studied the exploratory activity of maze bright and maze
dull rate, under three conditions of privation: 0, 2h and
hB hours of food privation. He reports that while exploratory
activity in an unfamiliar situation declines as a function
of time, food privation increases the amount of eXploratory
activity in which animals engage. Male rats show a steady
increase in eXploratory activity as privation increases
from O to AB hours, while for female rats the maximum of
exploratory activity is reached after 2h hours, as measured.
Maze bright and maze dull rats did not behave differently.

Montgomery (25), using an enclosed Yhmaze has reported
the results of a study which show that food or water pri-
vation significantly reduces the amount of exploratory
activity, the maximum reduction occurring at 2h hours of

food privation.

THE THEORY OF DRIVE

Preliminary considerations

It will be the purpose of the present chapter to set
up the theory of drive in the notation of symbolic logic,
so that testable consequences of this theory may be deduced
as theorems. In this section on preliminary considerations,
we shall attempt to relate the theory to conventional experi-
mental methodology, to relate the theory to the more general
problem of’measurement, and in general to explain what the
task of formalizing the theory of drive amounted to.

The first task is to relate the theory of drive to the
conventional methodology of experimental psychology. Hull
has stated his assumptions about drive in terms of individual

aninals. We find him using in the Principleg 2; Behavior

 

such phrases as "the organism", "an organism", ”the individ-
ual" and "the individual or the Species”. It is obvious
that Hull intended his speculations to apply to the animal
as an individual.

On the other hand, the conventional methodology of
experimental psychology is not conceived in terms of the
individual aninal. Groups of animals are used in psycho-
logical experiments, and conclusions are drawn in terms of
the behavior of these groups of animals. Furthermore, all

of the tests of Hullian notions have been in terms of group

“...——

1h

behavior. Consideration of individual animals results in
failure of the theory (7).

‘What is demanded here is some sort of an individual
which has its genesis in a group, or in a class of individuals.
Leonard and Goodman have provided such a conception in their
calculus of individuals (22). This is the fusion-individual,
or sum-individual, of a class of individuals. The fusion-
individual has the same logical type as the individuals
which are members of the chiss, but derives from the class
itself. The notion fusion is conceived as a heterogeneous
relation between an individual -- the fusion-individual --
and a class. An individual is said to stand in that relation
to a class, when everything that is discrete from the
individual is discrete from every member of the class, and
conversely. Their first postulate assumes that every class
which has members has a fusion-individual, 1.9., has a sum.

Therefore, in view of the above considerations, we will
understand by the term organism, a sum-individual of a class
of individuals. When we use the term animal we will under-
stand the common-sense usage of the term, e.g., g rat is an
animal, 5 dog is an animal, 2 human being is an animal.

The theory of drive, as an empirical theory, is inti-
mately related to measurement. We will want to speak about
the amount of drive Operative in a given situation. Or, to
speak more precisely, we will want to speak about the magni-

tude of the drive of a given organism. The use of the phrase

C
9.
I
I
r
.- V‘ " ”—I—Q
.- m.—
\
...
e p —
t 3 0 v
I
c
E
e

7"

,‘

 

the magnitude 23 drive in equations that represent empirical

 

situations, implies that we have at hand some method of
determining the magnitude of drive. The theory of drive

is, in fact, an application of the general theory of measure-
ment to a concrete situation, 1.6., every measurement pro-
cedure is a model of the general theory of measurement.

How this is accomplished can best be explained by showing

the parallels between the notation of the general theory of
measurement (22) and the notation of the theory of drive.
Some discussion of the general theory of measurement will
have to come first.

By the term measurement, we will understand the assign-
ment of numbers as names of the preperties possessed by
objects. This definition of measurement differs from that
commonly found in the writings of psychologists, in that it
is more restrictive. The definition excludes, from the class
of measurement procedures, the assignment of numbers as
names of objects. Thus, the use of numbers in a nominal
sense is not measurement. It is to be emphasized also, that
measurement is an empirical procedure. Measurement is some-
thing that a person does, through the actual manipulation of
events in the environment, in accordance, we would assert,
with the above definition.

A theory of measurement has as its purpose, the elabor-
ation of the nature of the relationships which must exist

between: (1) the objects of measurement; (2) the class of

l6

preperties with respect to which the objects are measured;
(3) the number signs which are used as names of the magnitudes
of these preperties. The use of the phrase the glggg g;
pgoperties in (2), above, is to be noted. A measurement
procedure, that is, a valid measurement procedure, is con-
cerned with only a single class of preperties. Examples
might be heights, weights, numbers of moles on the body,
Stanford-Binet I.Q.'s, etc. Comparison of any two such
classes of preperties, say for example, Stanford-Binet I.Q.
and number of moles on the body, would require additional
empirical knowledge, and would involve two applications of
the theory of measurement.

The general theory of measurement (22) assumes three
primitive ideas. These are: a class, K, of objects of
measurement, e.g., buildings; a class, L, of preperties
which the objects of measurement possess, e.g., heights;
and a relation, R, which takes members of K as arguments,
e.g., smaller than. According to the examples, aRb would
be interpreted as building a is smaller than building b.
Later theorems show that R has £952 of the preperties of the
less-than relation which holds between numbers.

Given the three primitives, three notions are defined.
These are: a relation, S, taking members of K as arguments;
a relation, Q, taking members of L as arguments; and a
notation for the expression £22 magnitude 2; £22 prepertz,

.it20: the member gf the property-class, possessed 21:33

'l

O
,
O
t
t e o -
K
D

g r
_‘ '
I

 

17

object, i.e., ”mag(a)". Later theorems show that S has
.3922 of the preperties of the relation identity, which holds
between numbers; Q is an ordering relation for mag(a), and
corresponds to the relation less-than which holds between
numbers.

Three primitive sentences, or postulates, are required.
Postulate l asserts: if two objects, a and b, have the
preperties M and N, respectively, then M is identical with
N if and only if a stands in the relation S to b. Postun
late 2 asserts: R is transitive. Postulate 3 asserts:
there is some preperty, M, which a possesses. Between them,
postulates l and 3 assert that each member of K possesses
one and only one member of L.

Theorems are then deduced which present interesting and
important preperties of the relations R, S and Q. Three of
these theorems are particularly important because of the
formal parallelism between them and the ordering axioms for
the real number system. These three theorems show that Q
is an.ordering relation for mag(a). It is through these
three theorems that the relationship between R and S and the
number system is established.

The present theory of drive requires five distinct
applications of the theory of measurement. That is to say,
we will be concerned with five different classes of preperties
attributable to organisms, and also to animals. The animals

considered will, of course, be rats; white, male, albino rats.

 

18

The five classes of preperties will be: 1, privation with
respect to food; 2, privation with respect to water;

3, general activity; A, drive; and 5, reaction potential.
To illustrate the parallelism between the general theory of
measurement and the theory of drive we list in Table 1h,
below, the corresponding notation for the theory of measure-
ment and one of our classes of preperties. We chose priva-

tion with respect to food.

TABLE 1h

PARALLELS BETWEEN THE NOTATION OF THE GENERAL THEORY
OF MEASUREMENT, AND THE PRESENT THEORY OF DRIVE,
IN THE CASE OF FOOD PRIVATION

 

 

Theory of measurement Theory of drive
"an "a"

"mag( )" "priv( ,x)"

"mag(a)" "priv(a,x)"
A class of qualities M, N, A class of privations M1,
0, OtCe M2, M3, etc.
mag(a). a DI,.(7M)(Ma) priv(a,x). = DI,.(7M)(Mx,a)
Expressions like "mag(a)", Expressions like "priv(a,x)",
"mag(b)" may be substituted "priv(b,x)" may be substi-
for expressions like "M", "N” tuted for expressions like
etc. "M1", "M2", Etc.

 

Tables similar to 1h could be constructed to show the
identical parallel relations between the notation of the
theory of measurement and each of the other classes of

preperties used in the theory of drive.

19

Four primitive ideas are required for the theory of
drive. These are need, general activity, reaction potential
and drive. Of these four, the notion of need is most
interesting. Need is a notion which Hull borrowed from
common-sense for incorporation in his theory. Hull speaks
of actual and potential needs (16, p. 57). We will consider
that needs are dispositional preperties of animals (6).

That is to say, needs are constant. An animal has a con-
stant need for such and such an amount of food per day. It
makes no difference whether the animal has eaten its fill
only ten minutes ago, he still needs a given amount of food
per day. Similarly, in the case of water, an animal has a
need for a constant amount of water per day; and for Optimal
conditions it needs this amount of water regardless of the
state of its thirst at any given time. It will thus be seen,
that we chose the notion of potential need, not the notion
of actual need. Instead of speaking of the actual need of
an animal for a given substance, we will Speak of privation
with respect to that substance. Thus, we hepe to make the
notion 2323 a little less ambiguous. Need will be conceived
as a relation between an environmental support and an animal.
The expression "xNa" is to be interpreted as x is a need

of animal a.

Adequate treatment of the concept of need requires
another symbol concerned with need, besides the relational

symbol. We shall wish to speak of privation with reSpect to

20

a need. For this purpose the expression "priv(a,x)" is
used, and is to be interpreted as the magnitude of the
privation of a with respect to the need x.

General activity is likewise conceived as a relation
between the activity of an animal and that animal. The
expression "yAa" is to be interpreted as y is the activity
of a. ‘When we use the term general activity, or when we
speak of the magnitude of the activity of an animal, it will
be understood that we refer only to activity as measured in
the present situation. Any other use of the term activity
is a use which is not included in the present system, and
the present system makes no statement about any other use
of the term activity.

We shall wish also to Speak of the magnitude of the
activity of an animal. For this, we will use the expression
"act(a)", to be interpreted as the magnitude of the activity
of a.

The notion of reaction potential was conceived, by Hull,
as a relation between a stimulus and a response. He used
it in the sense of a functor, i.e., a symbol taking number
signs as values. Hull's use of the term reaction potential
corresponded to the notation "mag( )" in the thesry of
measurement. Unfortunately Hull was not consistent in his
usage (16, pp. 3hh-3h5L We are here interested in reaction
potential only in the sense of the magnitude of reaction

potential. We will be concerned with a notation for the

magnitude of reaction potential only, and with only those
other considerations which Hull's general system forces

upon us. As was explained in the introduction, it will be
necessary to assume that all of the variables, in the Hullian
system, which determine reaction potential, except drive and
the magnitude of reward, are constant over all of the eXperi-
mental groups. Thus, the measures of drive which will be
obtained are measures of relative drive. Differences which
we will seek are relative differences.

The expression "ef(a)" is to be interpreted as the
magnitude of the reaction potential of a. Likewise, the
expression "dr(a)" is to be interpreted as the magnitude
of the drive of a.

Three postulates are required for the theory of drive.
These three postulates correSpond to dummy equations (1),

(2) and (S) in the introduction. The postulates are set up
in the form of Carnap's bilateral reduction sentence (6),

ices,
Q1 03 OQZ 3 Q3

In each case Q1 corresponds to the assumptions that are
necessary for the assertion of Q2 5,Q3. In each case

Q2 §,Q3 corresponds to the dummy equations of the intro-
duction. We have made one modification of Carnap's usage.
For Carnap,'Q3 represents a dispositional predicate. For
the present usage, we have modified this interpretation.
Q2 and Q3 are the same sort of notion, predicates taking

definite descriptions as arguments. It is necessary to

22

assume that the descriptions exist in each case. For example,
in the case of priv(a,x) it is necessary to assume that the
prOperty M, such that, a has M, exists. This assumption is
made.

The first postulate is:
Ni:.= Nib.y6N7a:.(x):xeN?a.x.# y. >.priv(a,x) z pP1V(b,X):.S7.

. priv(a,y)<.priv(b,y)33dr(a)éidr(b)
Postulate 1 says, in affect, that'if two organisms, a and b,
have identical needs, one of which is y, and if they have
been equally deprived with respect to each of their needs
except y, then the deprivation of a with respect to y is
less-than that of b if and only if the drive of a is less-
thanthat of b. Of course, y is the test need in this
assertion, and it is necessary to assume that all of the
organisms have all of their needs in common, as well as to
assume that their privations with reapect to all of their
needs, except y, are equal.

The second postulate is:

(Hu):.(v):ia : be.veA-?a3u '—= V:.~?e
act(a)<iact(b) E'dr(a)4:dr(b)

Postulate 2 says, in affect, that if two organisms, a and
b, have all of their activities in common, and that they all
do the same thing, implies that the magnitude of the
activity of a is less-than that of b if and only if the
drive of a is less-than that of b.

23

The third postulate is:
ef(a)4_ef(b) E dr(a)¢:dr(b)

Postulate 3 says, in affect, that the reaction potential of
organism a is less-than that of b if and only if the drive
of a is less-than that of b. This postulate comes directly
from Hull, and it should be obvious that it is only true in
the limited context of the present theory, where drive is
the only variable which effects reaction potential.

The reader will note that in the consequences of the
postulates we have used the relation less-than. The relation
less-than is more powerful than the relation identity. We
could have set the postulates up in terms of identity. Had
we done this, we would be unable to assert any statement
involving the less-than relation. The holding of identity
can be derived from the holding of less-than, but not con-
versely. Since we wish to assert both kinds of statements,
we must begin with the relation less-than. This consider-
ation points out another parallel between the general theory
of measurement and the present theory of drive. The theory
of measurement was begun with the relation, R, which corres-
ponds to the less-than relation. It will be remembered that
the relation less-than is an ordering relation. Less-than
orders the number domain which is its field. Identity, which

is, never-the-less, a very useful relation, is not an ordering

relation.

The arguments to the relation less-than, in the postu-
lates will not be numbers. These are expressions like "the
magnitude of the drive of", etc. EXpressions cf the form

used are called functors. Functors are functions which take

 

numbers as values. We Speak of £23 magnitude of the privation
of a with reapect to x, when we have in mind some unique

value of the magnitude of the privation of a with respect to
x. Functors are a form of definite description, as the
definition of priv(a,x) shown in Table 1h indicates. The
general theory of singular descriptions is dealt with in
Principia Mathematics in elk and *30. We may make similar

remarks about act(a), dr(a) and ef(a).

Theory
Let: "x", "y" and "z" be variables denoting possible needs
"u" and "v" be variables denoting acts
"a", "b", "c" and "d" be variables denoting organisms
"xNa" denote x is a need of organism a
"uAa" denote u is an activity of organism a
"priv(a,x)" denote the expression the magnitude of
the privation of organism a with
respect to the need x
"act(a)" denote the expression the magnitude of
the activity of organism a

"dr(a)" denote the eXpression the magnitude of

the drive of organism a

JJ

,‘

 

 

5.-

25

"ef(a)" denote the expression the magnitude of
the reaction potential of organism a.

An expression of the form "xNa" will be meaningful if
and only if it takes a variable, or a constant, denoting a
possible need as a member of the domain of N, and a variable,
or a constant, denoting an organism as a member of the
converse domain of N.

An expression of the form "uAa" will be meaningful if
and only if it takes a variable, or a constant, denoting an
act as a member of the domain of A, and a variable, or a
constant, denoting an organism as a member of the converse
domain of A.

An expression of the form "priv(a,x)" will be meaningful
if and only if it takes in first argument position a variable,
or a constant, denoting an organism, and in second argument
position a variable, or a constant, denoting a need.

An eXpression of the form "act(a)" will be meaningful
if and only if it takes as argument a variable, or a constant,
denoting an organism.

An expression of the form "dr(a)" will be meaningful if
and only if it takes in argument position a variable, or a
constant, denoting an organism.

An expression of the form "ef(a)" will be meaningful if
and only if it takes in argument position, a variable, or a
constant, denoting an organism.

We will assume as logical vehicles: (1) Principia

 

(a

 

Mathematigg of Whitehead and Russell (AD); and (2) The

 

general theory of measurement of Leonard (21).
We will use the mathematical relations less-than (41),
identity (=), and diversity ($).

The primitive sentences of the theory are:

-e —~ .—+
P 1. N'a = N'b.y¢N7a:.(x):x¢N'a.x4=y.D .priv(a,x) == priv(b,x):.D .

priv(a,y) < priv(b,y) E dr(a) < dr(b)
P 2. (5 u):.(v):B-?a = Fbwefaau = v:.:> .
act(a) (act(b) E dr(a) ( dr(b)
P 3. ef(a)( ef(b) .-__-_ dr(a) (dr(b)
The above prepositions are consistent with a universe
of: (1) two distinct sum-individuals, a and b; and (2)
four distinct elements, c, d, e, and f.
Following are certain theorems derived from the primi-
tive sentences, specifically for the purposes of testing in

the present thesis:

Verbal statement of the theorems
On the hypothesis that two organisms have all of their
needs in common, and that their degrees of privation with
respect to all of their needs, except the test need, are
equal, then:
Th. 1. If their degrees of privation with reSpect to the
test need are equal, then their drives are equal,

and conversely.

Th. u. If the degree of privation with respect to the test

need of organism a is less-than that of organism b,
then the reaction potential of a is less-than the
reaction potential of b, and conversely.

Th. 5. If their degrees of privation with respect to the
test need are equal, then their reaction potentials
are equal, and conversely.

On the hypothesis that two organisms engage in the same
activity, then:

Th. 2. If their activity is equal, then their drives are
equal, and conversely.

Th. 8. If the activity of organism a is less-than the
activity of organism b, then the reaction potential
of a is less-than the reaction potential of b, and
conversely.

Th. 9. If their activities are equal, then their reaction
potentials are equal, and conversely.

On the assumption that all of the determiners of reaction
potential, with the exception of drive, are constant for two
organisms, then:

Th. 3. If their reaction potentials are equal, then their
drives are equal, and conversely.

On the hypothesis that two organisms have all of their
needs and activities in common, and that their degrees of
privation with respect to all of their needs, except the test
need, are equal, then:

Th. 6. If the degree of privation of organism a with respect

to the test need is less-than the degree of privation
of organism b with reSpect to the test need, then

the activity of a is less-than the activity of b,

and conversely.

Th. 7. If their degrees of privation with respect to the
test need are equal, then their activities are equal,
and conversely.

On the hypothesis that two organisms have all of their
needs and activities in common, and that their degrees of
privation with respect to all of their needs, except the
test needs, are equal, then:

Th. 10. If the activity level of organism a is less-than
the activity level of organism b, then the reaction
potential of a is less-than the reaction potential
of b, and conversely.

Th. 11. If the activity level of organism a is equal to the
activity level of organism b, then the reaction
potential of a is equal to the reaction potential
of b, and conversely.

Theorems 10 and 11 are interesting, because the test
needs can differ for a and b. For example, theorem ll asserts
that if two organisms are deprived with respect to different
needs, and if their activities are equal, then their reaction

potentials are equal.

29

Theorems of the present system are designated "Th".
Theorems from the theory of measurement are designated

'T", after leonardb practice.

5

311.1. 117a ‘N'b. 4,1”: IJRa:.(x):xeu'a.x¢y.).
priv(a,x) ‘ priv(b,x):.).
priv(a,y) ‘ priv(b,y) 5 dr(a) ‘ dr(b)

Dem.‘
:l.*4.ll Hp(N,x,y,a,b).).
'~[pr1v(a.3) < priv(b,y)] 3 ~[dr(a)< dr(b)] (1)
(l)[b/a,a/b] Hp(N,x,y,b,a).).

~[priv(b,y)<priv(a,y)3 ~[dr(b)<dr(a)3 (2)

(l).(2).*4.38 Hp(N,x,y,a,b).):
~[priv(a,y)<'priv(b,y)].~[pr1v(b,y)‘cpriv(a,3)].
5 . ~[dr(a)<:dr(b)].~[dr(b)< dr(a)] (3)

(3).T6.64.‘4.11.Dual.DN 23D

Th.2. (Ju):.(V):A'9I " Avbnr £478. 3 uava).
act(a) ' act(b) 2 dr(a) ' dr(b)
Dem.

Same proof as Th.l., starting with 12.

Th.3. ef(g) ' ef(b) 2 dr(a) ‘ dr(b)
Dem.

Jame proof as Th.1., starting with r3.

30

—~ " l‘ ,

fh.4. N's 3 Nth. ,y=:Z]e;.(x):x'iflda.x4-y.).
priv(a,x) 3 priv(b,x):.).
priv(a,y)< priv(b,y) E ef(a)<ref(b)
Dem.

fl and is ) -ED

4 a a --.
Th.5. ;;'a ‘ N'b._,y¢LJ'a:.(x):x"={Ham-5131.).

priv(a,x) 3 priv(b,x):.).

priv(a,y) 3 priv(b,y) ef(a) 3 ef(b)

Home
Same proof as fh.2., starting with £h.4.

4 «v ., -..
Th.6. h'a 3 N'b.,,y€ ii'a:.(x):x¢N'a.x-_r 3.).
priv(a,x) 3 priv(b,x):.(§u):.(v);ara ‘ ATb.
Vaikfa ) u=v:.).priv(a,y)< priv(b,y) 5 act(a) < act(b)
Dem. V

.l.r2.*ll.02.31.‘3.47.*4.21.)z:
{re 3 II'—’b._ (,3; 6117's: .(x):xe [item 741.).
priv(a,x) 3 priv(b,x):.(§u):.(v):A;h 3 Ash.
Véa?& ) u‘v:.):
priv(a,y): priv(b,y) 5 dr(a)<:dr(b):
dr(a)< dr(b) 5 act(a) < act(b) (1)

(1).*4.22 ) TED

31

-‘.

«4 4 ,’
Th.7. N'a 3 N'b. ,ye H'a:

.(x):xeN'a.x#-y.).
priv(a,x) 3 priv(h,x):.
(Ju):.(v):a7a 3 Afbae Ava ) u 3 v:.).
priv(a,y) 3 priv(b,y) 2 act(a) 3 act(b)
Dem.

Same proof as Th.1., starting with Th.6.

4 do 1.

1311.8. (Qu):.(v):a'a 3 A'b.Y€A'8 ) u'v:.).
act(a) < act(b) = ef(a)< ef(b)
Dem.

£2 and. P3 ) lED

Th.9. (Ju):.(v):A'a 3 A'b.ve A78 ) u‘v:.3.

act(a) 3 act(b) 5 ef(a) 3 ef(b)

00111 e

Th.2. and Th.3. ) ZED

Th.10. 137’s 3 Il'b.j,zc1%,:8:.(x):x€117a.x3-y.xf—=z. 3.
priv(a,x) 3 priv(b,x):.
a a» .5
(4u):.(v):A'a " A'bn'é s'e ) u‘vz.).
act(a)< act(b) 3 ef(a)< ef(b)
DO’me
- 4 '4’ 4’
"3.31[Hu):(v):A" 3 wt.“ A“! D 1137/9.
act(a)< act(b) 2 ef(a)< ef(b)/r,
“’ 3’ I» —-;
N's 'N'b.j,zgid'azdxhxeN'mx-iy.x--~1.).

priv(a,x) 3 priv(b,x)/q]::.)::.

32

(Ju):.(v):s'a 3 A'b.v€rs—'.a ) u3vz.).

act(a)<act(b) 2 ef(a)<ef(b)::)::

-~ 4 - «9

N's 3 N'b.i,z EH'&:.(x):x& N'a.xeiy.xs- 2.).
priv(a,x) 3 priv(b,x):.

(3n):.(v):A7h 3 A'b.ve;A?h ) u3vz.).

act(a)<act(b) 2 ef(a)<ef(b) (1)

rs.8.(1).*3.22.35 3 ZED

Th.ll.

1578 3 N'lib.f,z etélyezdx):xe-N-?a.xa*y.x—7Lz.).
priv(a,x) 3 priv(b,x):.
a --% -~>
(Ju):.(v):A'a 3 A'b.ve A's ) u3vz.).
act(a) 3 act(b) 3 ef(e) 3 ef(b)
Dflme

Sane proof as Th.10., starting with Th.9.

STATEMENT OF THE PROBLEM

The problem for the empirical part of this thesis was
to put to experimental test certain of the theorems derived
in the section on theory. Using water privation Theorems
h, S, 6 and 7 were tested. Using food privation Theorems h,
S, 6 and 7 were tested. Finally Theorem 11 was tested.

In combination with the above, two magnitudes of food
reward and two magnitudes of water reward were used in
training the animals, to determine whether there was any
interaction between the amount of reward and the degree of

privation.

 

PREDICTIONS

Using Theorem h we have:

Hyp. 1e

Hyp. 30

HYp. “0

With food privation, the greater the level of
privation, the greater the force of reSponse

(Test data).

With food privation, the greater the level of pri-
vation, the greater the number of trials to
extinction.

With water privation, the greater the level of pri-
vation, the greater the force of reSponse (Test data).
With water privation, the greater the level of pri-
vation, the greater the number of trials to

extinction.

Using Theorem 5 in conjunction with Hull's postulate on

the magnitude of reward, we have:

Hyp. Se

HYPO 60

‘With food privation, the force of response of the
small reward group will be less than the force of
response of the large reward group (Training data).
With water privation, the force of the reSponse of
small reward group will be less than the force of

reSponse of the large reward group (Training data).

Using Theorem 6 we have:

Hyp. 70

With food privation, the greater the level of pri-

vation, the greater the activity level (Test data).

e e
t

e e
C

. g Q
i

e e

e e

O I

,‘9

 

Hyp e

Hyp.

Hyp e

HYp e

Hyp.

35

8. With food privation, the greater the level of pri-
vation, the greater the activity on extinction days.

9. With water privation, the greater the level of pri-
vation, the greater the activity level (Test data).

10. With water privation, the greater the level of pri-
vation, the greater the activity on extinction days.

Using Theorem 7 we have:

11. There will be no differences in activity among the
four sub-groups in the training series.

Using Theorem 11 we have:

12. For a given level of privation of food or water,
where the activity level of a food privation animal
is equal to the activity level of a water privation
animal, the force of response of the food privation
animal is equal to the force of reSponse of the water

privation animal.

SUBJECTS

The subjects used in the present study were RB experi-
mentally naive male albino rats from the colony maintained
by the Department of Psychology of the Michigan State College.
The ages of the animals at the beginning of their use as
subjects ranged from 100 to 120 days.

APPARATUS

The apparatus used in the present problem was especially
constructed for the series of problems of which this thesis
is the third. Davis (8) and Smith (3h) used the apparatus
in their studies, and with one minor modification the appar-
atus is the same as they reported. The apparatus itself
consists of a combination activity chamber and panel-pushing
device, so constructed that one may obtain from it a measure
of the activity of the eXperimental animal, as well as
measures of reSponse latency and the force with which the
animal reSponds. It consists of a 1/2 inch plywood box
with overall dimensions of 20 x 16 x 11 inches.

Figure 1 presents a cross-section of the apparatus.

The bottom of the activity chamber was a false floor which
was supported by three springs near the edge, and by a rubber
ball at its center. At the four corners of the false floor,
small, attached, rubber balls served as steps, preventing the
floor from tipping more than 1/1, inch. ‘

A guillotine door at one end of the activity chamber,
when raised, gave access to a hinged h inch by 2 inch panel.
This panel was constructed of rectangular piece of plywood,
1/16 inch thick. At the upper end of the panel a small piece
of 1/2 inch plywood, 2-1/2 inches long was attached. This

formed the base for hinge.

38

28.:—

Hausa
no»:

mbedmdmmd .mo wagdun 4<ZOHsommImmono .H mquHm

\\\\\\\\\\§\\\\§\\\\\‘V\\N\N\\N\\\\\ .\\ \\\\\\\\\\\\\\\\N\\\\\\N\\\\\\\\\\\\

 

 

 

Hausa

H025
on com H3

doe—5
own-qua

'5

 

 

 

 

.83
8:05.30

 

 

 

. §SE§S§Sassessisss..,

\a

 

\\\\\\..\\\\ 7 l

a\\\\\\\\\&

 
   

 

€38 9...:
at.“ no

 

- AF
\\\\\\\\\\\\.
C

80:

 

sea-.—

.31..

 

\\‘ .\\\\\\ ‘\\\\\\\\\\\\\\\\\\\\\\\ \\\\\\\‘ \\\‘ \\\\\\

(III: ‘ \\\\s .'

Bananas

€38 «L3

 

IOhOE

cussed-s5

The flat grey interior was illuminated by a 7-watt
bulb, located at the end of the box Opposite the guillotine
door. The bulb was covered by a piece of Opal-flashed glass.

The box was entered from the tOp, through a lO-3/h
square inch hinged door. In the center of this door was a
large clear-glass observation window.

Activity level was measured by a device consisting of a
GE 2-36KRL mercury switch, suSpended vertically beneath the
false floor and connected in series with a Gorrell and Gorrell
115 volt electric counter. The mercury switch was situated
beneath the floor in such a manner that movement by the
animal caused the floor to tip, causing the mercury in the
tube of the switch to flow, momentarily making and breaking
the circuit in accordance with the vigor and frequency of
the movements of the animal.

A thin metal rod, hinged at the tOp of the panel was
twisted so as to extend to the back of the panel in one
direction, and to the tOp of the box in the other. The rod
was so placed that it "rode” back on the panel when the
latter was pushed Open, and at the same time the upper half
of the rod came forward towards the activity chamber. By
means of this rod, the force applied to the door was trans-
mitted to a slender stick of wood which was glued to a
light plastic wheel (a child's toy roulette wheel) mounted
on a plastic axel. The force of the response, applied to the

panel, was thus transmitted into rotary movement of the wheel.

1"

hO

The extent of movement of the wheel was read in degrees of
angle. Because the wheel offered very little resistance to
the metal rod, the initial movement of the panel caused it

to turn out of the range of further movements of the rod.

The extent of rotation of the wheel depended upon the initial
force with which the panel was struck, not on the distance the
panel was moved. Thus, force of reSponse was read in degrees
of rotation of the wheel.

Response latency is not considered in the present report
because the experimenter feels that his own latency is con-
founded with the animal's response latency;

Single reward pellets were placed on a tray located l/h
inch below the lower jamb of the panel. Metal walls were
built up on either side of the tray to discourage exploratory
behavior. The corners of these walls were bent toward the
panel, forming stOps to prevent the animal from forcing the
panel and breaking it. On the tray itself, a small well of
solder was constructed to hold the food pellet in place.

Water reward was administered through a curved glass
tube attached by a rubber stapper to a burette. The
burette was clamped to a standard ring stand, and the whole
assembly was retractable. The tube was so adjusted that the
end was located in the same position in which food pellets

were placed.

PROCEDURE

That portion of the procedure which was common to all
of the sub-groups will be presented first. We will then

discuss the individual groups separately.
Habituation

All animals were handled for a period of 20 minutes on
each of three days prior to their first experience in the
apparatus. This handling consisted of stroking the animals,
and allowing them to run about on tOp of a large table. On
the fourth day each animal was placed in the activity box
for a period of six minutes, during which time its activity
level was recorded. These six-minute habituation trials
continued for five days, or through the eighth day of the
experimental series. This six-minute activity measure is
standard for the entire experimental series. The activity
measures obtained during the time period day A through day 8,
inclusive, represent the initial activity levels of the

animals under conditions of ad libitum feeding.
Training

On the ninth day each animal was given the regular six
minutes in the box, during which time activity level was

recorded. At the end of six minutes the guillotine door was

I‘I

 

raised, allowing the animal access to the hinged panel. The
panel was Open to its fullest extent (approximately two
inches) making the reward easily accessible to the animal.
After the animal had taken the reward, the panel was closed,
and the guillotine door was lowered. After a period of 30
seconds had elapsed, the guillotine door was again raised,
presenting the animal with the Open panel and the reward.
This procedure continued for eight trials. On the ninth
trial the panel was Open only approximately l/2 inch, so
that the animal had to Open the panel the rest of the way
to obtain the reward. Another trial was given in the same
manner, making a total of'lO rewarded responses for the day.
The following day the entire procedure was repeated with the
sole exception that for the first two trials the panel was
Open approximately l/2 inch, and on the last eight it was
closed all of the way. On the remaining six days of training
each animal made 10 responses to the panel fully closed. At
the end of training each animal had received 80 rewarded
trials in the apparatus.

The sixeminute activity level, before the trials of
each training day, was recorded.

The force of the reSponse for each trial was recorded
on the last three training days.

All animals were trained on 22-1/2 hours of privation

of the reward substance.

(1

 

Testing

On the 17th day of the experimental series,testing was
begun. The animals were randomly assigned to the particular
privation level at which they would be tested on a given day.
The apprOpriate privation manipulations were carried out
(details to be Specified below) and the animals were intro-
duced into the apparatus. The six-minute activity level was
recorded. Then the animals were given four rewarded trials
on the panel, in exactly the same manner as on the last six
days of training. The force of response was recorded for
each of the four test trials. If a subject refused to
respond for a period of six minutes it was removed from the
apparatus, and the expression "NR" (no reSponse) was recorded
for that day's trials. Each animal was tested at each pri-
vation level considered, and the order of presentation of
privation levels was randomized for each subject. Six pri-
vation levels were studied and there were seven testing days.
The reason for the extra testing day is that one of the

privation levels was MB hours without food (water).
Extinction

On the 2hth day of the series each animal was again
assigned at random to one of the six privation levels. The
animals were placed in the apparatus, the activity level was
recorded, and extinction was begun. An extinction trial

was conducted as was an ordinary testing trial, with the

I.

 

 

exception that no reward was given the animal. Extinction
trials were continued until an animal had refused to make a
response for three consecutive minutes. An animal was con-
sidered to have responded if and only if it Opened the panel
and placed its nose in the position that it would have had

to assume to get a reward. The force of each response was
recorded, as well as the number of responses to criterion.

A summary of the procedure common to all groups follows:

Days 1 - 3 Handling, 20 minutes per day.

Days A - 8 Habituation in apparatus, sixdminute activity
level recorded.

Days 9 - 16 Training, 10 trials per day. Activity recorded
on all days. Force of response recorded on
days 1h, 15 and 16.

Days 17 - 23 Testing, h trials per day under the apprOpri-
ate privation conditions. Activity level and
force of reSponse recorded each day.

Day 2h Extinction, to a three minute criterion.
Activity level and force of response recorded.

All animals were weighed at the beginning and end of
the experimental series, and at least one other time during
the series.

The main division of the present experiment is with
respect to the privation substance. Half of the animals
were deprived with reSpect to food and half were deprived
with respect to water. .

Of the 2h animals which were placed on food privation,
12 received a large reward pellet and 12 received a small

reward pellet. The large pellets weighed, on the average,

MS

0.32 gram. The small reward pellets weighed, on the average,
0.08 gram.

Of the 2h animals placed on water privation, 12 received
a large water reward and 12 received a small water reward.
The large water reward was, on the average, 0.20 cc. of
water, and the small water reward was, on the averaga.0.12 cc.
of‘water.

On days 1 through 8 all animals were on an ad libitum
feeding schedule. During the training series, the animals
were on 22-1/2 hours food (water) privation. During the
testing series, the animals were deprived for various numbers
of hours, depending upon the particular random sequence of
privations for any given animal. The privation levels
studied were 1, 2, 6, 12, 22-1/2 and h8 hours. During the
extinction series an animal was extinguished at only one
of these six privation levels.

It was felt that any cumulative effect of privation
would seriously impair the validity of the experimental
results, so an attempt was made to keep the animals' weights
as close to ad libitum weight as possible. To this end, one-
half hour after each animal had completed his performance
for a given day, it was allowed free access to a wet mash
mixture and to water for a period of hS minutes. During the
testing series,animals were also allowed.free access to the

mash and water for us minutes just prior to the beginning of

 

 

us

their privation period. An exception to the general pro-
cedure is the case of 22-1/2 hours privation, where the
single feeding time was lengthened to one hour.

Those animals deprived with respect to food (water)

had water (food) available in their home cages at all times.

RESULTS

Activity level

The habituation activity levels of all MS animals were
broken out in terms of the sub-groups into which the animals
would fall, and analysis of variance was done to determine
whether there were differences in activity level with
reSpect to the groups-tO-be. The results of the analysis
are presented in Table 1. There are no significant differ-
ences between groups-tO-be. There are significant differ-
ences between the five habituation days. The general tend-
ency is for activity level to decrease over the five-day
habituation period (see Figure 2). The days-times-groups
interaction term is not significant, indicating that all
four groups-tO-be behaved in the same way over the five
habituation days.

Table 2 presents the results of an analysis of variance
of the difference in activity level between the mean of the
last two habituation days, and the first day of training.
This analysis should indicate whether the change in priva-
tion, as such, is instrumental in bringing about an increase
in activity. There were no differences between the four
groups-to-be. The interaction term groups-times-days is not
significant, indicating that all of the groups-to-be behaved

in the same way. There is a significant difference between

I“

 

TABLE 1

ANALYSIS OF VARIANCE OF THE HABITUATION

ACTIVITY FOR THE FOUR GROUPS-TO-BE

ua

 

 

Source of variance

 

d.f. Sum of squares F p
Total 239 5,506,71u.40
Between animals 47 1,855,949.60
Between groups 3 1u0,311.uo 1.20 ns*
Between animals
within groups uh 1,715,638.20
Within animals 192 3,650,76h.80
Days A 1,559,h30.02 35.86 .01
Days x Groups 12 177,837.81 1.36 ns

Pooled animals x
days interaction

176 1.913,u96.97

 

*ns denotes not

significant

1+9

.hmo
meacumau panda emu so Una .mhwo dodgesuanmn
e>au on» no nose :0 He>ea mua>ape< .N easmam

N49

 

A

4

 

OOH

com

com

8:

TEAS? XLIAILOV

ANALYSIS OF VARIANCE OF THE ACTIVITY DATA COMPARING

TABLE 2

THE FIRST DAY OF TRAINING WITH THE MEAN OF

THE LAST TWO HABITUATION DAYS

50

 

Source of variance d.f. Sum of squares F p
Total 95 1,232,86u.99
Between subjects A? 705,139.u9
Between groups 3 39,178.h9 .4]_ ns*
Between subjects
within groups uh 665,961.12
Within subjects h8 527,725.50
Days 1 217,6h6.26 33.70 .01
Days x Groups 3 25,910.11 1.3h ns
Error uh 28h,l69.13

 

*
ns denotes not

significant

(0

,Q

I,

 

51

days, indicating that the change from ad libitum feeding and
watering conditions to 22-1/2 hours privation did signifi-
cantly increase activity level.

The analysis of variance of activity on the training
series is given in Table 3 (see also Figure 3). There are
differences significant at the five percent level of confi-
dence between: (1) the four sub-groups, (2) large and small
reward, and (3) successive training days. The difference
between sub-groups can be attributed to the difference between
large and small reward, since the differences between food
and water, and the interaction term food-water—times-large—
small-reward, were not significant. Thus, hypothesis 11 is
not denied by these results. There was a general tendency
for activity level to increase with successive days of train-
ing, in the case of“the large food reward group only.

The analysis of the activity levels on the testing
series is given in Table A. Inasmuch as there were differ-
ences in the training data, introduced by the use of two
levels of reward, the analysis of the testing activity is
a covariance analysis.

Table A shows significant differences between privation
levels. The F-value for this factor was significant beyond
the one percent level of confidence. More interesting than
this result, however, is the finding of significant F-values

for the two interaction terms food-water-times-privation

1“

 

TABLE 3

ANALYSIS OF VARIANCE OF THE ACTIVITY DATA
FOR THE TRAINING SERIES

52

 

 

 

Source of variance d.f. Sum of squares F p
Total 383 13,310,719.ho
Between animals h? 7,689,586.10
Between groups 3 1,h65,872.60 3.37 .05
Food vs water 1 323,292.16 2.23 ns*
Large vs small 1 930,23h.hh 6.h2 .05
Food vs water x
Large vs small 1 212,3h6.00 1.h7 ns
Between animals
within groups A3 6,223,713.50
Within animals 336 5,621,133.30
Days 7 231,818.50 2.09 0.5
Days x Groups 21 503,905.50 1.51 ns
Days x food-water 7 176,519.7h 1.59 ns
Days x large-small 7 170,911.5h 1.54 ns
Days x food-water
x large-small 7 156,h7h.22 1.hl ns
Pooled animals x
days interaction 308 h,885,h09.30

 

* ,
ns denotes not significant

(a

.‘

 

53

.oazoaw Heuausaaogwo ndou new

omen madmawhu unwao on» we game no Ho>oa hua>apo< .m oadmam

H4O

 

hops! Hamfim
nous: omawu
coca Hausa
600% owned

 

03

com

com

co:

oom

TEAS? ELIAILDV

TABLE h

ANALYSIS OF CO-VARIANCE OF ACTIVITY LEVEL
ON THE TESTING-SERIES

 

 

 

Source of variance d.f. Sum of squares F p

Total

Deprivation 5 692,590.h0 12.97 .01

Animals in 2,815,729.99 5.61 .01
Groups 3 179,291.98 1.50 ns*
Foodpwater 1 53,2h3.28 1.33 ns
Large-small 1 85,56h.26 2.1h ns

Food-water x
Large-small l 21,321.37 <11 ns

Between animals
within groups A3 1,715,892.77

Animals x Deprivation
Groups x Deprivation 15 604,557.10 3.77 .05

Food-water x
Deprivation 5 29l,h07.12 5.h6 .01

Large-small x
Deprivation 5 182,989.30 30u3 001

Fooddwater x
Large-small x

 

Deprivation 5 126,962.09 2.38 .05
Error 219 2,339,737.83
*

ns denotes not significant

I.

level and large-small reward-times-privation levels. The
significance of the food-water-times-privation level inter-
action term indicates that the food groups did not behave

in the same way over the privation levels as did the water
groups. Reference to Figure A and to Table 5 indicates that
only the large food group deviated from the trend of the
water groups. These data do not support either hypotheses
7or 9.

The large-small-reward-times-privation level interaction
term was significant for the same reason that the food-water-
times-privation level interaction term was; i.e., the large
food group deviates from the trend of the other groups. It
is noteworthy that the large food group deviates from the
trend of the other groups only with respect to the early
privation levels. Table 5 is included to point up these
differences in trend.

A remark on the reliability of measurement of activity
is in order at this point. Johnson (19, p. 136) recommends
the use of the error term and the individual mean square in
estimating the reliability of measurement. We obtain an
estimate of the reliability coefficient if we subtract from
unity the ratio of the error mean square to the individual
mean square. Using this procedure, with the:resuh:s of the
analysis of variance of the activity data for the training
series, we have as an estimate of the reliability coefficient,

Ilt‘xx = 00910

56

.mddopm Hencesaaoaxo ado.“ no.“

soape>apa no mao>oa Mam me Some co Ho>ea huaeapo< .: madman

ZOHB<>Hzm mo mtDOm

 

m: m.mm ma 0 NH
Ill]. H0963 HH gm“
.lll. hopes owned
..I||.. doom Hagm |...|:|.ll.l.
.....III coo.“ owes \..||.il.l|.|u I
\\
\\\ .
\,
\ /
\\\\\. .\\
‘\ \\\\\\ /// \
\\\\\ \
\\\|\ I
\\ .
\
\
\
\\\

 

OOH

com

com

8:

cm:

'IEAE'I IIIAILOV

57

 

: u n u m.mm
. u - 1 n - u - NH
Ho. Ho. so. u - u u a a n - u 0
Ho. Ho. Ho. . u u u u - u a n n u . u a m
Ho. Ho. Ho. . s u u s u u u u s u a u u u u u a

 

we m.~m NH 0 m m: m.mm ma 0 m m: m.mm ma o m a: m.mm ma e m
coape>aaa mo medom coape>aaa we madom coape>aaa no masom soapmpana mo manom
owwwq HHeEm omamq HHmEm

 

 

 

 

 

 

poem aopmz

 

mmDomu mDOh mme mom MBH>HBU< OzHBmwB mme mom
.MOZMQHRZOO ho qm>mq Bzmommm mzo mme B< BZ<UHmHZOHm .mMozmmHmmHQ mo zmmae¢m

m mqm<a

 

.I 8 I ' | 1:. I
II. I ...: I II I! I I‘
' I II II I .I’ II In...

' (I I .- ' ll" IIIIII ‘1'? ‘ull‘l

'7 .0 ’ .II I
. I
l I 8' T I. .I
I T
I
V I

III..."

I l I. (I .

Table 6 summarizes the results of analyses of variance
of activity on the extinction day. It is seen that none of
the F-values is significant. Hypotheses 8 and 10 are not

supported.

Force of response

For the training series, force Of reSponse was recorded
only for the last three training days. The analyses of these
data, using the mean force of response for each animal for
each day, are given in Tables 7 and 8.

Table 7 presents the results of the analysisfbr the
food groups. None of the F-values is significant, indicating
no differences in force of response attributable to size of
reward, and no differences in force of response on the last
three days of training. The lack of significance in the
interaction term groups-times-days, indicates that both
groups behaved similarly on the last three days (see Figure
5). These results, with respect to size of reward, do not
support hypothesis 5. These results, with reSpect to differ-
ences between days, indicate that the habit for the panel
pushing response was at a maximum by the end of training.
It was hOped.thatthis would be the case for valid results
on the testing data.

Table 8 presents the results of the analysis of variance
of the force Of reSponse on the training series for the water

reward groups. The total degrees of freedom here are hi

(Q

59

TABLE 6

SUMMARY OF THE ANALYSES OF VARIANCE OF
ACTIVITY AT EXTINCTION

 

 

 

Source of variance d.f. F p
Water-food 1 (.1 ns*
Large-small 1 4.1 ns
Groups 3 1.16 ns
Deprivation 5 4 1 ns

 

fins denotes not significant

TABLE 7

ANALYSIS OF VARIANCE OF FORCE OF RESPONSE ON THE
LAST THREE DAYS OF TRAINING FOR THE FOOD GROUPS

 

 

 

Source of variance d.f. Sum of squares F p

Total 71 3.15h.089

Between animals 23 2,054.842 451 ns*
Groups (Large-small) 1 3.167 4,1 ns
Between animals
within groups 22 2,051.675

Within animals h8 1,099.2h7 .
Days 2 37.h06 211‘ ns
Days x Groups 2 ’ 52.795 1.15 ns

Pooled animals x
trials interaction uh 1,099.0h6

 

*ns denotes not significant

 

 

_ — . . -__ 2 - . _
Q
..7 i. —- , -..—ha- a ..- .a _ - , ..
—
x o
—
- .. . ~—. _, - —..- A -f . , , . _ - . . -
. i - . - i _ .4 - o - » ’
I U
- ~ -m- — .— .—-—..- - ._—- . . . - ..——-..—~ - . a ,
‘ i
n
\ xi
.-
-

 

60

TABLE 8

ANALYSIS OF VARIANCE OF FORCE OF RESPONSE ON THE
LAST THREE DAYS OF TRAINING FOR THE WATER GROUPS

 

 

 

Source of variance ' d.f. Sum of squares F p
Total H1 3.AA5.736
Between animals 13 1,821.169
Groups (Large—small) 1 1,010.381 1h.95 .01
Between animals
within groups , 12 810.788
Within animals 28 1,62h.567
Days 2 78.832 411 ns*
Days x Groups 2 350.917 3.52 0.5

Pooled animals x
trials interaction 2h 1,194.818

 

*ns denotes not significant

61

masonm Hausgapoawo .38 new @5593 we when

0095 anma on» so omcoamoa mo oeaom .m madman

 

N49
m m a
..I/
I
I,
[ll
"ll'l
\I/
\ I
\ I
\ I,
\\ ll
.\\

.Il... popes .395
.III have! 0mg
. ..III. coo.“ Hanan
IIII. coon owns

 

OH

ON

on

0:

SISNOJSHH .50 30 31.0.21

instead of 71, as in Table 7. This is due to the fact that
data for 10 animals were not available. The difference for
size of reward is significant beyond the one percent level
of confidence. This result supports hypothesis 6, since it
is the small reward group which showed the smaller mean force
of reSponse. There were no significant differences over the
last three days of training. The significance of the inter-
action term size of reward-times-days reflects the fact that
the small reward group shows a steady decline in force of
response over the last three days of training, while the
large reward group does not (see Figure 5).

Estimates of reliability of measurement of force of
response give rix_= 0.95 for the food groups, and rkx = 0.91
for the water groups.

Table 9 presents the results of the analysis Of variance
of the force of response on the training series. The
original data were not homogeneous, and hence a square root
transformation was used. The data for this analysis were
obtained by averaging the four test trials for each pri-
vation level for each animal.

In Table 9 the F-value for privation levels was signifi-
cant beyond the one percent level of confidence, indicating
an increased force of response with increased privation. The
only other significant F-value was the interaction term size

of reward-times-privation level. This interaction term was

significant because the large food reward group showed an

63

TABLE 9

ANALYSIS OF VARIANCE OF FORCE OF RESPONSE
ON THE TESTING DATA

 

 

 

Source of variance d.f. Sum of squares F p
Total 287 30104347
Deprivation S 37.14118 15016 .01
Animals 47 144.097 6.95 .01
Groups 3 5.203 <31 ns*
Food-water 1 2.151 <11 ns
Large-small 1 2.411 4-1 ns
Food-water x -
Large-small 1 0.641 ((1 ns
Between animals
within groups 44 133.691
Animals x Deprivation 235 119.889
Groups x Deprivation 15 11.230 1.52 ns
Food-water x
Deprivation 5 3.502 1.42 ns
Large-small x
Deprivation 5 6.323 2.56 .01
Food-water x
Large-small x
Deprivation 5 1.405 <1 1

Error (pooled animals
x deprivation) 220 108.659

 

*ns denotes not significant

 

1‘

6L;

increase in force of response with increased privation, over
the early privation levels, while the other groups did not.
See Figure 6 and also Table 10. These results do not support
either hypotheses 1 or 3.

Table 11 summarizes the results of analyses of variance
of the number of trials to extinction. None of the F-values
is significant. Hypotheses 2 and 4 are not supported.

According to hypothesis 12 there should be a strong
positive Pearson product-moment correlation coefficient
between the force of reSponse of animals in the water groups
and animals in the food groups, when the animals are matched
according to activity level. Comparison of the activity
levels of the food and water group animals disclosed 55
cases when the hypothesis of Theorem 11 was satisfied.

Since the correlation of 0.12 is less than its standard

error (0.13), hypothesis 12 is not supported.

The validity of the theory of drive

We are here concerned with empirical validity. The
reader will recall that the consequences of the three primi-
tive sentences were set up in terms of the relation less-
than. This relation is an ordering relation, and implies
linearity. In terms of our present eXperiment this means
that there should be a linear relation between hours of
privation and activity, and between hours of privation and

force of response. Since significant differences were

65

.mmsoam HmOCcEaLeaNm anew new

soapwpfiaa no mao>ea Kan me home so omsoamoa no ooaom .0 ocsmfim
zoqa<>asm m3 sizes

m: m.mm ma e ma
7 1 om

 

 

 

O
.3

 

YIIit nous: Hausa
lill. pope; owned
. . Ooom Hamsn
llllu OOOM OthH

TDUOH

HSNOJQBW HO

66

 

.. .. - .. ms...
U U U U U U U U NH
. u u u u . Ho. . . Ho. . u e
8. 8. .8. .. .. .. u .. 8. .. .. .. 8. n .. .. m
HO. HO. HO. U U U U U U U HO. U U U U HO. .l U U U H

 

 

we m.mm NH e m we m.mm ma e m we m.mm ma e m we m.mm ma 0 m
coapm>aaa mo masom soapm>aaa we made: soauw>aaa no madom soapw>aaa mo mndom

 

 

 

 

owned HHeEm owned HHeEm

 

 

pooh sope3

 

 

mmbomo Adezmszmme mbom Ema mom mmzommflm @O momom mo OZHBmmB Ema mom
.mozmmHmzoo ho Am>mq Bzmommm mzo Ema Ed BE¢OHEHZOHm .mmOzmmmmmHQ ho zmmee<m

OH mqm<8

 

I l b I

' III. I i

O O c. I
In! U
U U
It. I
II. x
U U I l.

inlol|

..OI-t' 'll.

TABLE 11

67

SUMMARY OF ANALYSES OF VARIANCE OF THE

NUMBER OF TRIALS TO EXTINCTION

 

 

P

 

Source of variance d.f. F
Water-food l .4.1 ns*
Large-small l 4.1 ns
Groups 3 2.12 ns
Deprivation 5 1.15 ns

 

*ns denotes not significant

0\
LL)

demonstrated between both of the test variables with respect
to hours of privation, we are in a position to test the
implications of the use of the relation less-than.

Table 12 presents the analysis of variance of deviations
from linearity for the activity data of the testing series.
It is seen that the F for deviations from linearity is not
significant. Hence, the implications of the use of the
less-than relation are supported for the activity data.

Table 13 presents the analysis of variance of the
deviations from linearity for the force of reSponse data of
the testing series. It is seen that the F for deviations
from linearity is not significant. Hence, the implications
of the use of the less-than relation are supported for the
force of response data.

In consequence, we may say that the present experiment
is an adequate model for the theory of drive, at least
within the range of privations used.

The least squares regression of the activity data on
hours of privation is

a.=-2.8446h + 202.1159

The least squares regression of force of reSponse on

hours of privation is

f = 0.021811 4' 5e1889

 

C

LN
\0

TABLE 12

ANALYSIS OF VARIANCE OF DEVIATIONS FROM LINEARITY
NG

 

 

 

 

 

 

 

FOR ACTIVITY ON THE TESTI SERIES
Source of variance d.f. Sum of squares F p
Deprivation 5 694.074.230
Linearity l 627,183.095
Deviat. fm. linear. 4 66,891.135 1.57 ns*
Error 220 2,340,056.82
*
ns denotes not significant
TABLE 13
ANALYSIS OF VARIANCE OF DEVIATIONS FROM LINEARITY
FOR.THE FORCE OF RESPONSE ON THE TESTING SERIES
Source of variance d.f. Sum of squares F p
Deprivation 5 37.4484
Linearity 1 32.9334
.Deviat. fm. linear. 4 4.5150 2.29 ns*
Error 220 108.6587

 

*
ns denotes not

significant

 

IN

DISCUSSION

The discussion of the results of the present study will
be divided into three sections. We will consider first the
implications of the study for activity as a measure of drive.
Secondly, we will consider the results of the study with
reference to the concept drive. Finally, the notion of

need will be considered.

Activity as a measure of drive

 

The fact that activity level drOps during the five
habituation days supports the findings of Smith (34), in
the same situation. It seems to take approximately five
days to establish a consistent level of activity, from which
we may work. This reduction is certainly not the result of
a reduction in privation, since the animals were on an ad
libitum feeding schedule on all five days, but seems more
due to the reduction in the amount of exploratory behavior
in which the animals engage.

The finding of a significant increase in activity from
the last two days of the habituation series to the first day
of the training series, would seem to indicate that the
increase in privation was reSponsible for the increased
activity. There is, however, an alternative eXplanation.

It is possible that the change in privation conditions,

.ii ....

- ...-.. __ _ - . —
.

I U
_ ‘_ \

 

71

independently of the fact that the change was an increase,
resulted in increased activity. There are some studies
which seem to suggest this (8, 9, 4), although their results
are not a sufficient test. Some unpublished work from the
Michigan State College laboratory bears more directly on
this point*. Water privation was utilized, and activity was
recorded over a ten-day period. Precautions were taken to
see that the animals had no Opportunity to associate water
reward with the activity chamber. Four groups, of four
animals each, were run. Two of the groups were on the same
conditions of privation throughout the lO-day period, one
group deprived for 3-5 hours, and the other for 18-20 hours.
Two groups were switched on privation kavels half way through
the lO-day period; one group being switched from 3-5 to 18-20
hours privation, and the other being switched from 18-20 to
3-5 hours privation. Both of the groups which were switched
on privation levels showed an increase in activity level
from the first 5 days to the last 5 days. Neither group
which remained at the same privation level showed any in-
crease in activity from the first to the last 5 days.
Furthermore, the difference-between 3-5 and 18-20 hours of
privation was not significant. These results support the
assumption that the increase in activity level from the last

days of habituation to the first day of training, in the

 

“Behan, R. A. and Campbell, F. Activity under two
levels of water privation, in the rat. Unpublished.

H“..”

72

present study, was due to the change in privation conditions,
and not to the increase in privation, as such.

The results of Smith (34) and Davis (8) pointed to the
existence of an anticipatory learning factor as influencing
activity. In the present study, the fact that the large
food reward group showed an increase in activity during the
eight days of training, points to the same conclusion. Also,
the existence of significant interaction terms between food-
water and privation level, and between large and small
reward and privation level, points to the same conclusion.
It might be thought that such a result would be the conse-
quence of the particular apparatus used in the present study.
That is, due to the fact that the activity measures were
taken in the same apparatus in which the animals learned
the panel-pushing response. This is doubtful, however,
since the other three sub-groups did not show a similar
increase in activity during the training series. If associ-
ation with reward, as such, were a sufficient condition for
an increase in activity, one would expect that these sub-
(groups would have shown an increase in activity parallel
‘to the increase of the large food reward group. It would
zappear, from the present results, that some minimal amount
<3f reward is necessary.

With respect to the results with activity on the testing
Berries, it is obvious that activity is not determined by

hcrurs of privation of food or water. Here again, activity

73

level seems more dependent upon the amount of reward given

in the situation, and the change of privation conditions.
Only the large food group shows the increase in activity
with increased hours of privation, and this only with reSpect
to the lower levels of privation as contrasted with the
higher levels.

It is well known that activity is a complexly determined
aspect of the animal's behavior (10, 13, 26, 27, 28, 29, 30,
34, 39). The implication of the present series of studies,
thgt the amount of reward that animals are given in a par-
ticular situation influences the amount Of activity that
the animals diSplay in that situation, merely adds one more
factor to be considered, if one attempts to use activity as
a measure of drive. Drive is conceived as a unitary con-
struct. To attempt to measure a unitary thing by reference
to a multiply-determined aSpect of behavior like activity,
seems a little ridiculous.

There is also a question of precision of measurement.
The results of the testing series suggest that there must be
some substantial change in the level of privation, since
differences occur only after 12 hours of privation. If
drive is supposed to increase with each hour of privation,
activity level does not reflect this fact. Further, it is
believed that this insensitivity of activity to changes in
drive is not due to the apparatus used. The rat could make

the counter run very swiftly by merely sitting and scratching.

.nr ..

 

It is noteworthy that the writer could find no studies of
activity as a function of privation where privation inter-
vals were less than 12 hours.

The implications of the present study are that activity
level would not be a satisfactory indicator of the magnitude

of drive.

The theory Of drive

The theory of drive presented in the second section of
the present thesis has not stood the eXperimental test. Of
the 12 hypotheses tested, 10 were not confirmed. Four
hypotheses were derived from Theorem 4. Theorem 4 is to
the effect that reaction potential will vary with hours of
privation. Two hypotheses were concerned with privation
with reSpect to food. Of these, hypothesis 1 used force of
reSponse as a measure of reaction potential, and hypothesis
used the number of trials to extinction as an indicator of
reaction potential. Hypothesis 1 failed in the case of the
small food reward group, but was supported in the case of
the large food reward group. Hence hypothesis 1 fails.
Hypothesis 2 failed in the case of both reward groups.

Two of the hypotheses derived from Theorem 4 were con-
cerned with water privation. Of these, hypothesis 3 used
the force of reSponse as a measure of reaction potential,
and hypothesis 4 used the number of trials to extinction as
a measure of reaction potential. Both of these hypotheses

.failed for the large and the small water reward groups.

.l’ 1.1 -1.

Theorem 4 was not supported.

Two hypotheses were derived from Theorem 5 in conjunction
with Hull's postulate on the magnitude of reward. Hypothesis
5: with food privation, using the training data, the force
of reSponse of the small reward group will be less than the
force of reSponse of the large reward group. Hypothesis 5
was not confirmed. Hypothesis 6: with water privation,
using the training data, the force of response of the small
reward group will be less than the force of response of the
large reward group. Hypothesis 6 was confirmed. The question
arises, was the failure of Hypothesis 5 due to failure of
the drive theory, or failure of Hull's postulate on the size
of reward. The result with Hypothesis 5 is exactly what
would have been expected if we had not used two sizes of
reward in the two groups, hence thecnnclusion is that the
failure of Hypothesis 5 is not due to the failure of the
drive theory.

Theorem 5 was supported.

Four hypotheses were derived from Theorem 6. Two of
these are concerned with food privation. Hypothesis 7:
using the test series activity, the greater the privation
the greater the activity. Hypothesis 7 was not supported
in the case of the small food reward group. Hypothesis 8:
using the extinction series activity, the greater the pri-
vation the greater the activity. Hypothesis 8 was not

supported. Two of the hypotheses derived from Theorem 6

76

were concerned with water privation. Hypothesis 9: using
the test series activity, the greater the privation the
greater the activity. Hypothesis 9 was not supported.
Hypothesis 10: using the extinction series activity, the
greater the privation the greater the activity. Hypothesis
10 was not supported.

Theorem 6 was not supported.

Theorum 7 was tested with only one hypothesis, number 11.
Hypothesis 11 asserted that there would be no differences
among the sub-groups, with respect to activity, on the train-
ing series. The differences among the sub-groups that appear
are due to the use of two sizes of food reward, hence
Hypothesis 11 is not disconfirmed. In this connection see
the discussion of Hypothesis 5, above.

Theorem 7 was supported.

A single hypothesis, number 12, tests Theorem 11.
Hypothesis 12 failed.

Theorem 11 was not supported.

There is no consistent pattern of support and failure
for the experimental hypotheses. If Theorem 11 had been
supported, we could have said that these hypotheses involving
the relation less-than failed, and those involving the
relation identity were supported. However, Theorem 11

failed. We must conclude that there is something wrong with

the theory of drive.

 

)PI I|1

77

The theory of drive has three primitive sentences. The
first of these reduces drive in terms of privation with
respect to a need. The second reduces drive in terms of
general activity. The third relates drive to the rest of
the Hullian system. Postulates 2 and 3 are necessary for
empirical test of the theory. We have already discussed
activity, and noticed two draw-backs. First, activity is
multiply-determined; and second, activity is not a sufficiently
precise measure of drive.

We will not here concern ourselves with postulate 3,
since it is forced upon us by external considerations.

Before we examine postulate 1, it will be necessary to
consider some different usages of the term drive. There are
three usages of the term drive which are common in present
day psychology. The first of these is to consider that drive
is a stimulus (12, 14, 24, 3). This is perhaps the oldest,
and just beginning to come back into the literature. Before
this notion of drive will become useful, it will be necessary
to clarify the notion stimulus. Very little has been done
along this line, and this usage of the term drive is outside
the scepe of the present thesis, hence we will go on to
other uses of the word.

The remaining two usages of the term drive differ, not
as to the function which drive plays in the explanation of
behavior, but rather in the degree of abstractness with which

the term is considered. Both usages depend upon the prior

78

assumption of need; assuming that it is possible to withhold
some environmental support, and thus deprive the animal with
reSpect to that environmental support, i.e., with reSpect to
that need. The first notion of drive: drive(l), abstracts
away completely from the specificity of need (16, 23, 37).
As Munn puts it "... drives, as such are usually blind.

They activate, it is true, but most of them do not, until
learning has occurred, turn activity into apprOpriate
channels." (26, p. 84). The second usage of the term drive:
drive(2), is much less abstract than the first, and some
individuals seem to use the term almost synonymously with
the term need (33, 39). It differs from the concept need

in that the drive activates the animal, but the activity is
more or less Specific, depending upon the need with respect
to which privation occurs. For example, privation of water
(food) results in water (food) seeking behavior on the part
of the animal. We will confine our remarks to the latter
two notions of drive. I

The present thesis has its origin in a test of drive(l).

However, as was pointed out, both drive(l) and drive(2)
assume that privation with respect to a need results in the
animal's being activated. In this sense the present thesis
is a test of both notions. If Hypothesis 12 had been the
Ionly hypothesis denied by the present study, it would have
kneen possible to fall back on drive(2). However, the denial

of‘Hypotheses 7, 8, 9 and 10, also, makes this impossible.

.2

,‘I

79

Reference to Table 5 will make it plain that privation with
reSpect to a need, per so, does not result in increased
activity. Indeed, within the limits of the present study,
it would seem that a learned anticipation and the amount

of reward were the most important factors. Hence, the
present experimental results legislate against both of the
notions drive(l) and drive(2).

 

The_category of need

This section will constitute an attempt at a speculative
account of the reason for the failure of the theory of drive.
The notion of drive rests upon the prior notion of need. The
kinds of needs which psychologists assume are taken from the
common-sense mythology of everyday life. In this reSpect
psychology does not differ from the other sciences. The
scientific enterprise is characterized by the fact that the
scientist begins with common-sense. The contributions of
science consist in the modifications of that with which they
start.

With this observation in mind we may look to the notion
of need as one of the sources of the present failure of
drive. If the basic notion is faulty the derivative notion
must also be faulty. In his treatment of the notion of need,
Hyll may be taken as typical of psychologists in general.
.Hull considers the needs of animals as independent of each

other. Animals have needs for food, water, air, avoidance

80

of tissue injury, maintenance of an Optimum temperature,
defecation, urination, rest, sleep and activity. These of
course are only the primary needs. Other psychologists,

for example Lewin (23), would include a large number of
other kinds, including some that are learned. Each of these
needs acts independently, and then the results of the indi-
vidual action is summated to give an over-all effect.

The thesis of the present section is that this notion
of need is singularly simple-minded, and thus faflacious. It
will be argued that the different needs listed above, as well
as the learned needs that an individual may acquire, do not
act independently of each other. The mode of action of
needs is rather to be considered as an interaction. Further-
more, the conception that equal hours of privation with
respect to a need results in equal drives, is likewise an
over simplification. If needs interact with one another,
then it is meaningless to Speak of equal hours of privation
with reSpect to a need. In other words, the denial of the
independent action of needs necessitates a denial of the
assumption that equal hours of privation with reSpect to a
need result in equal effects.

The evidence for the above position is sketchy, and
incomplete. However, there is some. The studies of Ver-
planck and Hayes (38), Finger and Reid (10), and Calvin and
Behan (5) all indicate that there is an interaction effect

between food and water privation. If animals are deprived

f‘

81

of food they automatically cut down their water intake, and
also, if animals are deprived of water they automatically
cut down their food intake. It is conceivable that there
is no need for food or water, but that there is a need for
an Optimal food-water balance, and that the animal adjusts
its intake of the one substance when deprived of the other,
in such a way as to maintain this balance.

This interaction effect is conceivable, with respect to
all sorts of nutritive substances. It is well known, now,
that the prOper utilization of organic food substances
requires the presence of certain minerals in.minimal amounts.
Furthermore, it is not inconceivable that many kinds of sub-
stances might interact with each other 23 constitute a_gggg.

It is thus seen, that the objection here is not to the
notion of need, but rather to what constitutes a need, and
how needs are classified. It should, however, be noted that
a simple reclassification of needs will not allow one to go
back to the rather simple minded assumption that equal hours
of privation with reSpect to a need have equal effect on the
animal, regardless of the need. For example, in the present
study, the large food group showed no increase in activity
until 12 hours of privation had elapsed. Animals deprived of
water showed no increase in activity over the entire testing
range. The force of response data yielded similar results,
with the exception of the hB hour privation level. On the

other hand, animals deprived of air, become active immediately,

 

f\

82

but only for short periods of time. Similar behavior
patterns are seen in the case of certain learned needs.

For example, the anxiety that arises when certain individuals
are deprived with reSpect to learned needs may be the same
kind of phenomenon.

The present notion would not be hard to test, in a
preliminary way, at least with food and water. If there is
an Optimal food-water balance for the rat, and presumably
for other animals also, then the ratio of food intake to
water intake should be a constant, or should vary in a
systematic way, under different conditions of privation.

A number of studies have shown a consistent food-water intake
ratio when animals are on ad libitum feeding schedules.

They have also shown constant reductions in either food or
water intake, when privation is with respect to water or
food (5, 10, ll, 28, 31, 32, 35, 38). In other words,
present evidence encourages the view, but is insufficient

to maintain it systematically.

SUMMARY AND CONCLUSIONS

The present thesis was concerned with a test of the
Hullian theory Of drive, using privation with reSpect to
food and with respect to water. The position was formalized,
and consequent theorems were tested.

A combined activity box and panel—pushing device was
used in the present study. The panel-pushing device was so
arranged to provide a measure of force of response, and of
the number of trials to extinction.

Forty-eight male albino rats were used as subjects in
the study, and were divided into two major groups:

1. twenty-four animals deprived of water;

a. twelve animals given a large reward (0.20 cc.)
b. twelve animals given a small reward (0.12 cc.)

2. twenty-four animals deprived of food;

a. twelve animals given a large reward (0.32 gm.)
b. twelve animals given a small reward (0.08 gm.)

All animals were given habituation training in the
amoparatus during the first five days Of the eXperiment,
wtiile on an ad libitum feeding schedule. Activity level was
recorded.

On days 6 through 13 all animals were trained on the
panel-pushing task, under 22-1/2 hours of apprOpriate pri-

vation and reward. Each animal was given ten trials per day.

85

The implications Of the study for activity as a measure
of drive, and for drive as an eXplanatory construct, were
discussed.

The category Of need was discussed, and an alternative

interpretation Of need, as interaction, was suggested.

81+

Activity level was recorded on all 8 days, and force of
response was recorded on the last 3 days.

At the completion of the training series the animals
went immediately to the testing series, where they were
tested after different numbers of hours Of privation, 1.6.,
on 1, 2, 6, 12, 22-1/2 and h8 hours of apprOpriate privation.
The order Of presentation of privation levels was randomized
for each animal. Each animal was given h trials per day.
Activity level and force of reSponse were recorded each day.

I At the conclusion of the testing series each animal was
extinguished on one of the above mentioned hours of privation.

The results are as follows:

1. Activity is not a reflection Of privation, 233.23,
but is rather dependent upon the amount and kind of reward
in interaction with privation, and.0n a learned anticipation
of reward.

2. A general concept Of drive is not tenable, because
the correlation between the food and water groups, when
actiVity is constant, was not significant.

3. A drive concept is not tenable, because the water
groups did not show the increase in behavior measures which
the large food group showed.

h. There is a significant interaction between the kind
of reward substance and the privation level.

5. There is a significant interaction between the

amount of reward substance and the privation level.

 

1.
2.
3.
u.
S.

6.

7.

8.

1C).

11..

REFERENCES

Anderson, E. E. The interrelationship of drives in
the maleéalbino rat. Comp. Psychol. Monogr., 1938,
ILL, n0. 0

Canpbell, B. A. Design and reliability of a new activity-

recording device. J. comp. physiol. Psychol., l95h,
H7. 90‘920 -

Campbell, B. A. and Kraeling, D. ReSponse strength as
a function of drive level and amount Of drive reduction.

(lo 2520 £21282l-: 1953. 54, 97-101.

Campbell, B. A. and Sheffield, F. D. Relation of
random activity to food de rivation. J. comp.

physiol. Psychol., 1953, a , 320-322.‘*

Calvin, A. D. and Behan, R. A. Water intake as a
function Of food privation in the rat. Brit. J.

Psychol., In press, 1954-

Carnap, R. Testability and meaning. Phil. Sci.,
1936: 3: u20-M7lo

Cotton, J. W. Running time as a function of the
amount Of food deprivation. ‘i' exp. Psychol.,
1953. he, 188-198.

Davis, R. H. The quantification of drive I. Incentive
values of food and water. Ph.D. thesis on file in
the Michigan State College Library, 1953.

Deese, J. and Carpenter, J. A. Drive level and rein-
forcement. ‘J. exn. Psychol., 1951, h2, 236-238.

Finger, F. W. and Reid, L. S. The effect Of water
deprivation and subsequent satiation upon general

activity in the rat. ‘J. come. physiol. Psychol..
19529 #5. 368-3720

Gregerson, M. I. Studies on the regulation Of water
intake. II. Conditions affecting the daily water
intake of dogs as registered continuously b a
potometer. Amer. J. Physiol., 1932, 102. 3 '3u9.

[1

ll

12.

13.

1h.

15.

16.
17.
18.

19.

20.

21.

252.

3:3.

2L;,

25..

87

Guthrie, E. R. The psychology 23 learning. New York:
Harper, 1935.

Hall, J. F., Smith, K., Schnitzer, S. B. and Hanford,
P. V. Elevation of activity level in the rat following
transition from ad libitum to restricted feeding.

in assa- pflainl- 221222.. 1953. L16. #294133.

Holt, E. B. Animal drive and the learning process.
New York: Holt, 1931.

Herenstein, B. R. Performance of conditioned responses
as a function of strength Of hunger drive. .l- comp.

physiol. Psychol., 1951, uh, 210-22h.

Hull, C. L. Principles Of behavior. New York:
D. Appleton-Century, 19H3.

 

Hull, C. L. A behavior system. New Haven: Yale Univ.
Press, 1952.

Johnson, P. 0. Statistical methods i2 research.
New York: Prentice-Hall, 1989.

 

Kimble, G. A. Behavior strength as a function Of the
strength of the hunger drive. 1. exp. Psychol.,

1951. kl. 3&1-3u8.

Kleitman, N. The effect of starvation on the daily
consum tion of water by the dog. Amer. i. Physiol.,
1927. 81. 336-3&0.

Leonard, H. S. A. eneral theo of measurement.
Abstract on file In the versity of Michigan
Library, Ann Arbor, Michigan.

 

Leonard, H. S. and Goodman, N. The calculus of
individuals and its uses. 1. gym.lLogic, 1940, 5,

Ll-S"SSO

Lewin, K. Contributions pg psychological thee . The
conceptual re resentation and the measuremeng pf
ppyéhological forces. ‘Durfiam: Duke’Hniv. Press, 1938.

 

 

 

Miller, N. E. Learnable drives and rewards. In S. S.
Stevens (Ed.), Handbook pf ex erimental s cholo .
New York: John wIIey & Sons, 1951, H55-E7E.

Montgomery, K. C. The effect of the hunger and thirst

drives upon exploratory behavior. J. comp. physiol.
Psychol., 1953, us, 315-319. "

{i

O
I
O
I
....y_

K v
f- v
_ t
e
O
l
' I
I

'- e
O .
r. v E. . ' .
0 e
. I . .
. . Q o.
t .
e . ‘2 .
(‘7 t .
. ' '1 e
I .-
’ o
‘ O
' ' 2 o
' t
D g f .
O Q ‘ .
. ...\ ( '
x; . . _ .
.Q'
. -.. Q t _
. O I '
’ a
O . t .
-
e
' ( o
O
U o o .1
Q I

“III

 

 

26.

27.

28.

29.

30.

31.

32.

33.

3h.

35.

L36u

31?.

353.

39..

88

Munn, N. L. Handbook 23 psychological research 23
the rat. New York: HoughtOn-Mifflfn, I950.

 

Reed, J. D. Spontaneous activity of animals. Psychol.
Bull.» 19h7. “A: 393-h12.

Richter, C. P. Animal behavior and internal drives.

Quar. 333. 3101051, 1927, 2, 307-3u3.

Shirley, M. Spontaneous activity. Psychol. Bull.,
1929, 269 3&1'3650

Siegel, P. S. and Steinberg, M. Activity level as a
function of hunger drive. J. come. physiol. Psychol.,

1949. #29 #13-hl6.

Siegel, P. S. and Stuckey, H. L. An examination of
some factors relating to the voluntary water intake
Of the rat. .£° comp. physiol. Psychol., l9h7, hO,
271-27u.

Siegel, P. S. and Talantis, B. S. Water intake as a
function of privation level when food is withheld.

.1: come. physiol. Psychol., 1950, A3, 62-65.

Skinner, B. F. The behavior Of organisms. New York:
Appleton-Century-Crofts. I933.

 

Smith, 0. A. The quantification Of drive II. Two
methods of food privation. Ph.D. thesis on file in
the Michigan State College Library, 1953.

Strominger, J. L. The relation between water intake
and food intake in normal rats and rats with hypo-
thalmic hyperphagia. Yale J. biol. Med., 19u7. 19,
279-288. _‘ —"' “""

Thompson, W."R. Exploratory behavior as a function Of
hunger in bright and "dull" rats. J. comp. physiol.
Psychol., 1953, as, 323-326. +

Tolman, E. C. Pur osive behavior ip animals and men.
Berkeley: Univ. of California Press, 1932.

Verplanck, W. S. and Hayes, J. R. Eating and drinking
as a function of maintenance schedule. J. comn.

21123101. PSZChOlo, 1953! “-6, 327-3330 -

Warden, C. J. Animal motivation studies. New York:
Columbia Univ. Press, 1931.

F- is»

O\

_ \

89

no. Whitehead, A. N. and Russell, E. Princi ia Mathgmatica.
Cambridge: Cambridge Univ. Press, (ZnE ed.), 1925.

hl. Yamaguchi, H. 0. Drive (D) as a function of hours of
hunger. ‘02:. 0x2. P820h01o, 1951, LL23 108-1170

APPENDIX

91

 

 

 

 

TABLE 1
ACTIVITY LEVELS ON THE FIVE HABITUATION DAYS
FOR FOUR GROUPS—TO-BE
Da Animal
5' 1 2 3 4 5 6 7 8 9 10 11 12
Small Food Group
1 82 218 387 473 169 330 148 254 161 363 270 149
2 31 77 508 314 60 222 141 116 69 114 174 177
3 78 206 535 420 269 214 9a 115 41 232 2i 38
4 75 149 157 118 96 40 41 1 0 19 16 34
5 151 186 573 339 136 115 M9 7 1h 29 192 #3
Large Food Group
1 783 391 705 541 380 465 570 568 196 11 11) 84
2 31 257 380 90 118 188 533 263 150 23 12 25
3 261 455 182 253 115 64 205 187 51 90 277 129
4 153 202 81 1%2 29 130 103 63 86 247 66 226
5 218 168 359 2 8 153 102 67 97 84 54 340 81
Small Water Group
1 170 626 289 229 347 252 242 300 438 739 423 491
2 113 326 97 139 392 319 199 99 253 44 129 50
3 50 352 116 277 417 180 218 50 28 230 161 47
4 106 137 185 177 289 181 250 13 289 284 42 15
5 113 390 205 130 432 159 317 72 290 70 30 108
Large Water Group
1 325 35 322 877 415 829 417 599 397 171 356 157
2 207 400 30 104 222 93 110 215 81 154 90 201
3 10 318 43 240 200 70 184 292 110 193 136 53
4 18 123 132 97 247 98 121 57 11 161 121 103
5 63 259 104 90 366 96 83 28 3 5h 41 95

 

 

92

 

 

 

no me me sen mam see mmm saw one emu ea we: ma
eam oem ow sea __ new me: mam mos em see now NH
one eem mma sea mm: mom awn em see me 03m wee HH
emm emm and pea eon map mmm me: :3: emm see mme oa
dam sea son Ham we: Hem Hem :mm mea Ham eme noe e
me mma no es oem amp eom mam ems mem papa am: a
Hmm am sad mm mom on: a» amm mm: mam me: sea a
sea ee oma omH emm 0mm me mma m.m em: Ham and o
aeoaw poem emaem
eem Hes ems em mm eeH em 00: mme and en: mmm ma
ops em Ham e mom new ee mam mom mm mam mmm NH
can am com me :oa pea mem oem eem pom maa an: as
com am am on mma me ems 0mm emm new now me as
emm mam mem m 60H «mm 0mm emm em: mam mam :om e
as we emm on em ooa mes mam ms: co: oam sea a
sea as aom as sea on ema ems em: Hma :om pad a
as Has eem om em we oom Hem eem pen emm om e
unoac poem HHaEm

NH as ea e a a. e m in m m H

Hesac< hem

 

 

mmbomw Qoom ORB awe mom deQ GZHZHde mma

N mqmfia

zo mqm>mu MBH>HBO¢

93

3‘4“...” M... I... .II . .0. ma. . . $.ng

 

 

 

 

 

 

 

 

 

eeH OON OOH OON ,mN ONO HO: Omm OOe ONO eHO OmN OH
OHN mHH. OON meN eOH Oem mHm em ONO NON ON e N NH
HON OOO OON oem OHN Nee Om: Oem emN NOO OO OOH HH
OOH OON OmH OO eNN HHH eNm mON OOH OOH OOH :e OH
mHN Ne OeH OeH eON Hmm Hmm NON HHN OeN NON HON e
eNN OmH OOH HOH OOH eeN mmm ON OmN meN OH: Nmm O
HNH OOH OOH OO: eds ONO mmm e: eHH ONH me OHN e
OeH Oem eeN NNN om me eO meH OOO eOm mmm HOO O
9.5.5 ..Heuez owned

OO eH HeH OeH OOH HeH HOO NH: NO OOO eNN mm OH
eeH HN OH HOH OO mOH Hem OOe OOH O: NO: ONH NH
HO 3O eON NNN mOH OOH OOO OeN me. m.N OO: mHm HH
OOH meH OON OeH Os Nmm NOO men ONN HeN OH: Oe OH
e O: OOH OOH OHH OOH OH: eNe mam HON OOO eem e
mHH e: OOH emN HN aeN ON: MOO OON ONH OON NON m
HmH NH OOH OOH ON eNN mm: we: mON me mOO meN

eHH OmH OHm OeN ee OeH amN Hem OeH me OH: OON O

@395 amps? Hamsm
NH HH OH e O e O m e m N H see
HOEHQ¢
OOOOOO Ome<2 O29 ems mom OH<O OszH<ma ems zo OememH eeHeHeo<

m mem<e

 

 

 

IIIIIIIIIHHH Illllllll!m H

 

Nemwngaaas:mume,

h.

0/
OOO HON OOO NOO OOO” eON NOO He: OOO OOO OOO NOO O:
ONH OO NOH OHO OON eON NON ONN NOO Oee OHO OHN O.NN
OON ONN HH: ONN eeO OOO NON ON: OOO es: OOO OOO NH
eNN Os OOH HOH OOO OeH HOH OOO eNN NNH OHO OO: O
ONH He NHH HOH OOO ONH OO OON eNH NN ONO :o: N
OO HO NN OO eeH OON OO OOH OOH OOH eON OOO H

@5090 poem emaeg

 

 

eN O: OHN N eOO OOH OOH eO: Oee NON ONO OOO O:

oe NNH eON O: HOH OHN ONH OON OeO NHN OHN Ne O.NN

NO Os HOH ON eON OOH NNN OO: so: NON eON ON NH

NHH HNH OOO mm eOH OOH eOH ON: eON NON OOH ONH O

HO NHH ON Oe eON eOH OHN OOO OON OON OOH N

OO OO ONH OO ON ONH eO eON eN eO eHH OO H

@5090 poem HHOEm

NH OH 0H m m N o m 3 . m N H Goape>aaaen

HOEHO< mo manom

 

mmbomo Qoom GEE Ema mom mama mzHBmma Ema zo mqm>mq MBH>HBU<
d mqmda

95

 

 

 

 

 

HOO Hes Oe OeO NOO NNO NeO Os: OON OOO ONO NNO O:

eHH eeO ON eON ONN ONN ONO ONH eOH mo: NHO oeH O.NN

NHH HON OOH ON: OO: OON NHO OOH OOO NNH OHO NO NH

ON NOH NNH ON: eNO eeO OOO Oe OHN e HON OeN O

OHH OOO OeH NOO OO HO: OOO ONN OOO NH OHO OO N

NeH ONH OON eOO HOO NH: eeO HON ONO OOH NOO OO H
98.5 Loud: owned

OON eON NNO eON Ne NOH OHO on HHO OON OeN eNN O:

NNN eO OOH ONN ON eNH ONN OeN HNN OOH HON OO O.NN

OO ON OON ONH e OOH NOO OOO OOH OOH eOO NH NH

O OO HN OON N eO OOO ONO eN eO OON Oe O

O eN eO OO HH eNO OO HO: NNN eNN ONH eOO N

eO OO ONN OOH OH OeH OeH HOO OOH eO OHO NH H
macaw peed; HHOSm

NH HO ca 0 w w o m j m m H coauw>aamep

HdEHCd MO Meadow

mmbomw mama? 03M. awe mo.m mwed 6255mm; WEB zo mgmu EEHBU<

O memee

(IV

 

 

(‘11 l!

7 (ill! . l'lullfllull

96

r. Erlv

q]... .1..-
‘ o.ui-ll.o§.'u'ix-*tll’xlv.u!!. .K

 

0
(“1‘0
MNM

r4 H 11“

O.eO 0.00 O.HO O.NO e.ON OH

H.OO N.OO e.OO O.OO 0.0N NH

Nee: O.OO e.OO N.ON O.Oe HH
Odoac nepwz emaeq

O.ON O.HO O.OO O.eN O.Oe e.ON O.HN N.ON 0.0N 0.00 OH

0.0N N.HO O.ON N.NN e.OO O.NN e.ON O.NN N.OO O.NO NH

e.ON e.ee N.Oe 0.0N N.Ne 0.00 N.OO 0.0N O.ON e.OO HH
deoao neeO3.HHOEm

O.HO O.NO O.OO e.HO 0.00 e.NN O.HO O.eN e.eH 0.0N OH

O.ON N.ON O.OO N.OO O.OO N.eN H.OO 0.0N O.eO O.NN NH

0.00 O.NO N.ON H.OO e.NO 0.0N 0.0N O.eH e.NO H.ON HH
Ozopo pooh emawq

N.HO O.NN e.eH N.NN O.ON H.NO 0.00 O.eO O.ON H.NO OH

N.OO O.ON O.OH N.NN O.eN O.HO O.eO 0.00 e.OO N.OO NH

0.00 N.ON O.OH O.ON O.NO N.HN 0.00 0.0N O.NN O.HO HH

Odopu poem HHeEm

 

OH e O N O O s O N H OOO
HesHe<

 

 

 

 

UZHZHmmB mo deQ mmmme Bm<g
fine So mmbomc mbom mmB ho Q¢2H2< moam mom mmm mmmm mo womom z<m2

Q mqm<a

97

.. uh,‘ ..v .u’li
hr). IIMI h (llz/

 

 

 

 

.O NN.O ON.O NO.N O0.0 O0.0 eN.O O0.0 N0.0 NO. eN.O Oe.: O:
.O eN.O O0.0 N0.0 NO.N e0.0 O0.0 N0.0 N0.0 ON.O O0.0 O0.0 O.NN
.: H0.0 ON.O O0.0 H0.0 NO.N H0.0 OH.O OH.O ON.O N0.0 O0.0 NH
.O ee.O H0.0 ON.O HO.N e0.0 N0.0 NH.O O0.0 O0.0 O0.0 N0.0 O
.O OO.N ON.O O0.0 OO.N O0.0 NO.O OO.O H0.0 OH.O. NH. N .H N
.e OO.N O0.0 O0.0 O0.0 O0.0 .OO.: Ne.: N0.0 N0.0 NN.m Om.: H
Odoam poem Owned
OH.O HO.O OO.O OO.O H0.0 O0.0 N0.0 O0.0 O0.0 ON.O H0.0 OH.O O:
OO.O OH.O OH.O O0.0 O0.0 NH.O NN.O ON.O OH.O ON.O N0.0 ON.O O.NN
ON.O O0.0 OH.O ON.O ON.O :e.e O0.0 O0.0 O0.0 OH.O O0.0 ON.O NH
O0.0 ON.O H0.0 O0.0 H:.: O .: OH.O ON.O N0.0 O0.0 OO.: ON.O O.
.O H0.0 NH.O ON.O O0.0 Om.: H0.0 e0.0 ON.O H0.0 ON.O O0.0 N
NH.O OO.O NN.O OO.O NN.O OO.O NO.O HO.O eN.O OO.O ON.O NO.O H
Onoau.noom Haegm
NH HH OH O O N O O s O N H eOHpeeHeeoe
- HOEHO< no manom

 

 

 

 

mmHmmm Emma mmB zo mmbomw moon 039 Ema m0 AdEHZd modm mom mmzommmm mo Homom 24m:
N mqm<9

98

 

ON.O ma.© mm.> $0.0 pm.> OH.m mo.m NH.m

 

 

O0.0 ON.O eN.N O0.0 O:
OO.m NH.O Ne.O O0.0 ON.O O0.0 N0.0. e0.0 O0.0 .He.O H0.0 OH.O O.NN
OO.u O0.0 O0.0 ON.O NH.O oe.e O0.0 Ne.: ON.O ON.O H0.0 e0.0 NH
H0.0 O0.0 O0.0 N0.0 ON.O O0.0 OH.O O0.0 O0.0 O0.0 .O0.0 NN.O O
Ne.O HO.O OH.O OH.O ON.O Oe.O NH.O OO.N O0.0 O0.0 OO.O OO.O N
OO.O ON.O OO.O OO.N ON.O He.O OO.O ON.O HN.: OO.O NH.N eO.: H
Adena aeuwz emamq
O0.0 O0.0 eN.O Oe.O OO.N O0.0 He.O ON.O N0.0 ON.O O0.0 NN.O O:
H0.0 N0.0 ON.O OO.N ON.O OO.N N0.0 N0.0 O0.0 O0.0 N0.0 eO.: O.NN
e0.0 ON.O N0.0 HN.N O0.0 O0.0 e0.0 O0.0 OH.O O0.0 H0.0 O0.0 NH
NH.m NN.O HO.O eN.N OO.O OO.N NH.O ON.O NO.O OH.O OO.O OO.O O
O0.0 OO.N e0.0 OO.N O0.0 O0.0 Oe.O ON.O eO. O0.0 N0.0 OO.N N
e0.0 e0.0 OO.O HO.O OO.O OO.N O0.0 NH.O NO.O OO.:_ HO.O oe.: H
Odopm nepN3.HHOEm
NH HO 0H m w N o m d m N H coapmbfiamen
HOEHO< . ho mHSOm

 

 

 

 

 

 

 

OmHOmO ewes OON 2O OOOOOO Ome<e O39 was aO Hesta OOem OOO mOzONOOO eO mOmOm zemz
O mqmee

99

IA 1
F. q’fi'e . ‘1’! .‘i u «1": .v . .F

 

NH

 

HO.O HO.O NO.O HN.O ON.O OO.O OO.H

N0.0 Oe.O NN.O eO.O NO.O Ne.H OO.N NH

HN.O OO.H OO.O O0.0 OO.O OO.O ON.O HH
edema OOPOS ewama

OO.OH HN.O ON.H ON.O HN.O OO.O OO.O Oe.O OO.O ON.H OO.O HO.O OH

ON.O NN.O OO.O OH.O ON.O He.H NO.O HH.H OO.H OO.H OO.H O0.0 NH

ON.H ee.H NO.O eO.O OH.O OO.H HO.O He.O eO.O He.O eO.O O0.0 HH
Osoaw hepOB HHOEm

OO.H Oe.O ,O:.O OO.H O0.0 ON.O Oe.O NN.O e0.0 OO.H NO.N OH.H OH

Nu.O eO.H eO.O OO.O ON.O OO.O HO.O eO.O OO.O OO.O OO.H ON.N NH

ON.H ON.H N0.0 NN.O OO.H OH.H ON.H OO.O HN.O OO.H OO.O OO.H HH
Odoac poem emawq

HO.O OH.O N0.0 OO.e ON.O O0.0 NH.O N0.0 e0.0 O0.0 NO. OH.O OH

OO.O OO.O HO.H NN.e: NH.OH OO.O eO.N O0.0 eO.O OO.O ON.H eO.O NH

e0.0 OO.O NO.O OO.N Oe.O HO.O NN.O OO.H eO.H NO.H ON.H OO.O HH
macaw poem HHOEm

NH HH OH e O N O O s O N H NOO

 

HOEHsd

 

Ema zo mmbomu mbom mmB &O Q<ZH?<

GZHZHde ho mwdm mmmme Bmmq

0 mqmde

modm mom mmZQOmmm mo NUZMqu Z<HQME

100

 

uepddafi m cdnu nonmeam me: hoceuea Omanefi O.HNEOGN one uenu uepmeapna I$

 

 

 

OO. OO.OH N0.0 NN.O OH.O OH.N ON.O ON.O ON.O OO.O OO.O O0.0 O:
OO.O HO.O OO.O OO.O Oe.O OO.O HO.O He.O ON.O NH.O HN.O HN.O O.NN
O0.0 Oe.N: O0.0 O0.0 Oe.O OO.N OO.O eN.O NN.O NN.O Oe.O OH.O NH
NO.O HO.O eO.N N0.0 Oe.O ON.O H0.0 e0.0 Oe.O HN.O Oe.O HO.O O,
Oe.O u OH.H eO.H ON.H OO.NNH Oe.O He.H NH.O HN.O ON.O eO.O N
O0.0 . NO.H a- ON.OH OO.H O0.0 ON.O ON.O Oe.O OO.H NN.O H
@5093 week emadq
OO.O OH.O HO.O NO.O wN.O OO.O eO.O O.O eN.O O0.0 ON.O OO.O O:
ON.O ON.O N0.0 ON.N N.O NH.O NN.O 0.0 ON.O ON.O NO.H O0.0 O.NN
ON.O OH.O O0.0 ON.O e0.0 eN.O NO.O NO.O HO.O OH.H eO.O eN.O NH
N.N OH.H NN.O O0.0 ON.O eN.O He.O e0.0 NO.O NO.O NO.HH OH.O O
O.N NN.O ON.O NO.N e0.0 OO.O OO.H NN.O eH.O ON.O Oe.e eN.O N
OO.e NH.O eO.H eO.O OO.ON OO.O OH.O eO.O OH.O eO.NH OH.e ON.O H
meoac pooh HHeEm
NH HH OH e O N O O O O N H eOHpeeHeeee
HOEHO< no nesom

 

 

mmHmmm Emma mm& 20 mmpomu mock 039 mmB mo 442H2¢ modm mom mmzommmm mo Hozma<n Z¢Hamz
OH mqm¢9

\u!‘ .53!

 

 

 

l
0.
l
mepdcfie m can» «Hepdeam we: anaconda canoes 3.38.23 OH.H» pen» mopmeHOGH Ia.
OH.N ON.O eH.O OO.O N0.0 ON.H N0.0 H0.0 ON.H NO.O eH.O eH.O O:
HO.O ON.O ON.O NO.: . u . OO.: HO.H eO.O ON.H HN.O O.NN
ON.O ON.e N0.0 eN.O HN.O NH.O e0.0 O0.0NH OH.H OH.O OO.H H0.0H NH
OH.H . OO.H . N0.0 ON.NO eN.OH OO.N e0.0N OO.O OO.: HO.O O
e0.0 OO.O ON.O eO.: ON.NOH . mm.O . eO.O NO.O ON.H eO.NO N
Oe.N NN.O NO.O . NN.NN . .e NH.ON ON.O OO.O NN.O e0.0 H
@595 .3825 emadq

eO.N NO.O OO.O O0.0 OH.O Oe.O HO.O Oe.O NO.O O0.0 O0.0 OO.N O:
NH.ON OO.O NO.O OH.O NH.O u OO.O HO.H OO.O OO.H NN.O NH.OH O.NN

: eH.OH eO.O eH.O NO.H u O0.0 ON.O O0.0 eO.H O0.0 eO.H NH
OO.OO OH.NN NO.O NO.H Oe.e . OO.H ON.O OO.OH HO.N HN.H NN.H O

t u . ON.O HN.O OO.H OO.H x HO.O . NO.H OO.O N

u . eN.O ON.O OO.N a- HO.O eO.OO OO.O OO.eN OH.O N0.0H H

95.8 .333 .355
NH HH OH O O N O O O O N H eeHOeeHeeee
Owens; .Ho mgom

 

 

mMHmmm Emma. mmB zo mmbomo

mme¢3 OO.OH. mma .mo Adzflmd Sofia morn...” mmzommma mo NOZMBJNHH 24¢ng

as. an (25:11

 

Nov 5 ’56
(501/ 19 ’55

 

 

"I7'1111111117111113les