63583 6F FRUS’E‘RATEGN AS IT PLE‘LA'YES
'E‘G ”ESE RUMEEE 0F- REINFGEQEB TRIALS
z‘T’E‘tEGE 1’8? FRUSTRAFEVE E‘EOHREWAEEE

finest: face fin an?“ 0:5 535. D.
IifiCfifGM‘e’ STATE UNEVEREET‘I
Deibert S. McE-‘i’emy, Jr.

{973

 

Missw

LIBRARY

Michigan State
University

This is to certify that the

thesis entitled

Odor of Frustration as it Relates to the
Number of Reinforced Trials prior

to Frustrative Nonreward
presented by

Delbert S. HcHenry, Jr.

has been accepted towards fulfillment
of the requirements for

Ph. D- degree in

 

72?
'zéfiwy/

Date 7/25/13

0-7 639

 

___r -. "—A ——..~-.

 

ABSTRACT

ODOR OF FRUSTRATION AS IT RELATES TO THE
NUMBER OF REINFORCED TRIALS PRIOR
TO FROSTRATIVE NONREHARD

By
Delbert S. HcHsnry, Jr.

While a nuaber of studies have demonstrated that rodents

secrete an odoroue substance as a function of frustrative non-

reward, little is known about the response properties of odor

emission. The purpose of the present study is four-fold:

(l)

(2)

(3)

(4)

To deternins if the concentration of odor-of-fruetration
is syeteeaticelly related to the nuaber of reinforced
trials preceding frustrative nonreward.

To deteraine if the odor concentration on the second
trial of frustrative nonreward is greater than that elit-
ted following the first exposure to frustrative nonreward.
To deternine if the paper covering the floor under the
odorant anieal acts as a depository for the odorous sub-
stance.

To determine the pherononel reaction of e aale albino

rat detecting the odor of a nonfrustratad eale conspecific.

To this end, g; froa five odorant groups were placed, in-

dividually, into the center chaebsr of a three-chanbered box

for six trials per day over nine consecutive days. The{§s of

each group received a pre-deternined nueber of nonreinforced

Delbert S. HcHenry, Jr.

trials (0, 18, 36, 48, 54), with the remainder of the 54

trials (54, 36, 18, 6, O) reinforcing approach toward the

food cup. Frustrative nonreward followed the final rein-

forced trial of day nine.

The existence of odor of frustration, and its concen-
tration, was measured in terms of the latency of a naive de-
tector'g to leave one of the end chambers and enter the odor-
ized center chamber. Latency of the detector to leave this
odorized area was also used as an index of odor concentra-
tion, since previous studies had shown that conspecifics
find odor of frustration mildly aversive.

It was found that:

(l) Odorant groups differed in latency to approach the food
cup on day nine. This was interpreted in terms of dif-
ferences in level of “food expectation”.

(2) Amount of urine excreted by the odorants following frus-
trative nonreward was directly related to the number of
reinforced trials prior to frustrative nonreward. This
was interpreted as showing differences in level of frus-
tration following frustrative nonreward.

(3) Latency of the detector g; to enter, then leave the odor-
ized center chamber was not systematically related to the
number of reinforced trials the odorant‘gs received prior
to frustrative nonreward.

(4) Rate detecting odor of a nonfrustratad male conspecific
tend to approach faster and escape slower than rats de-

tecting odor of a clean chamber.

Delbert S. HcHenry, Jr.

(5) Exhausting the odorized air of the center chamber fol-
lowing frustrative nonreward, but prior to detector
testing, yielded a non-significant tendency to approach
more slowly, and leave faster than detector g; of the
control group.

(6) Detectors were slower to enter but faster to leave an
area infused with odor-of-frustration secreted as a
function of the second trial of frustrative nonreward,
relative to detectors receiving odor-of-fruetration from
odorants receiving their first trial of frustrative non-
reward.

Discrspant findings between comparable studies were dis-
cussed in terms of procedural differences. An improved method-
ology based on the findings of the present study was proposed.
And a brief summary of phenomena related to odor-of-frustration

was given.

ODOR OF FRUSTRATION AS IT RELATES TO THE
NUMBER OF REINFORCED TRIALS PRIOR
TO FRUSTRATIVE NONREHARD

By

'- .
M1

Delbert S? HcHenry, Jr.

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Psychology

1973

LIST OF TABLES
LIST OF FIGURES

TABLE

OF CONTENTS

INTRODUCTION . o . o o o o o o o o o o . o .
HETHDD o o o o o o o o o o o o o o o o o o o
SUbjBCt. o o o o o o s o o o o o o o o o o o
Applrltus o o o o o o o s o o o o o o o o o
PICCUduro o o o o o o o o o o o o o o o o o
RESULTS o o o o o o o o o o o o o o e o o o o
DISCUSSION o o o o o o o o o o o o o o o o o
A DISCUSSION OF RELATED TOPICS o o o o
Strain Differences in Odor Sensitivity. .

Possible Secretory Glands Associated With
Oder of Frustration and the Role of Urine in
OdIr .f Fruatrlti.n Secretiln o o o o o o s
Odor of Frustration as One Instance of a
General Stress Odor . .

BIBLIOGRAPHY .

ii

Page
iii

iv

33
36
3B

Table
1.

LIST OF TABLES

Page

Day nine acquisition performance
f.r f.Ur .d.r.nt ngUpI o o o o o o o o o o o 20

Trial one means and standard deviations

of odor chamber approach, escape, and the
difference score (approach - escape) for

.ight dCtBCt.r group... o o o o o o o o o o o 23

Trial two means and standard deviations

of odor chamber approach and escape for
eight detector groups . . . . . . . . . . . . 24

iii

LIST OF FIGURES

Figure Page

1. Latency to approach the food cup as a
function of acquisition day . . . . . . . . . . 19

2. Estimated quantity of urine for five

odorant groups, with measures taken
after frustrative nonreward . . . . . . . . . . 21

iv

INTRODUCTION

A number of investigations have provided evidence for
the existence of odors secreted by an individual rodent as
a function of operations designed to produce frustration,
i.e., nonreinforcoment in a situation previously associated
with reinforcement.

In general these demonstrations have taken one of two
forms. First, it has been shown that a nonfrustratad con-
spscific makes an immediate and characteristic response
upon receipt of an odor associated with frustrative non-
reward. Hhsn an odor elicits a characteristic response
from o conspocific it is termed a pheromone (Korlson &
Luschsr, 1959). A second type of supportive evidence has
come from those studies which report that experimental ani-
mals can use the odor produced by a frustrated conspecific
as d cue for the solution of a discrimination learning prob-
lem.

For example, Harrison & Ludvigeon (1970) successfully
trained female albino rats to choose a food baited goal box
in a T-maze conditional discrimination problem when using

odor-of-frustrotion as a cue. Forty-sight So were separ-
ated into four groups of twelve g; each. The first group
(NC) received odor-of-frustrated-conspocific versus odor-of-

clsan paper. A second group (RN) received odor-of-fruotrated

1

2

conspecific versus odor-of-rewarded conspecific. And a third
group (RC) received odor-of—rewarded conepecific versus odor-
of—clean paper. The fourth group received only odor-of-clean
paper as a control for non-experimental discriminative cues
such as odor of food pellets emanating from the baited goal
box. Odor-ofwreward and odor-of—frustration were presumed
to be secreted or excreted by a food deprived odorant animal
placed at the choice point of the T-mazs. The presentation
of food to the odorant animal was intended to elicit odor-
of reward, while frustrative nonreward was intended to elicit
odor-of-frustration. The above chance performance of the NC
and RN groups was interpreted by Morrison and Ludvigeon as
showing a cue function for odor-of-frustrotion. However,
the use of compound cues in this study makes the evidence
equivocal. For example, the NC group could have learned the
T-more problem using the species-specific scent of the in-
dividual odorant animals.

Several spontaneous alternation studies have shown that
rats tend to avoid an area infused with their own scent (c.f.,
Schultz & Tapp, 1972) but to approach on area containing the
scent of other conspecifics (Reiff, 1956). Bower & Alexander
(1967) found that mice could distinguish between odors as-
sociated with two conspecifics in a Y-maze discrimination
problem. Archer (1968) showed that odor of a strange male
mouse caused an increase in aggressive behavior between male
cage mates. Taken together, these studies support the pos-

sibility that 'odor-of-frustration' might better be labelled

'odor-of-frustrated-rat' with the characteristic scent of the

3

odorant conspecific contributing to stimulus control of the
choice behavior of the indicator animal in the Harrison and
Ludvigeon study.

A similar problem arises in determining the nature of
the cue controlling the choice responses of the second in-
dicator group. Recall that this group received adar-of—a
frustrated rat versus adar-af-a-rsinforced rat. Uhils Har-
rison & Ludvigeon interpret the successful performance of
this group in terms of cue control by odor-af—frustratian
an alternative explanation is possible. Southhall & Lang
(1969) showed that rate could use the odor from a single pel-
let as a cue to solve a T-maze discrimination problem. It
seems possible that group two of the Harrison and Ludvigeon
study (the RN group) solved the conditional discrimination
problem using odor of food or food particles rather than
ador-of—frustration. In spite of the evidence cited above,
Harrison and Ludvigeon found that the RC group receiving odor-
of a reinforced conspecific versus odor-of-clesn paper did
not choose the baited goal box significantly above chance.

It is not clear whether the discrepancy in findings is due

to methodological differences in studies or an odor insensi-
tivity by the §s of the RC group. Using essentially the some
apparatus and design, HcHonry (unpublished research) was un-
able to replicate the Harrison and Ludvigeon findings when
the characteristic scent of rat was common to the two cue

values of the successive conditional discrimination, i.e..

adar-of-frustrated rot versus ador-of-a nonfrustratad rat.

In general, the use of a learned response confounds the

learning process with the relationship between the odor of
frustration and the unconditioned response to that odor.
Under certain circumstances it may be impossible to deter-
mine whethor the topography of a learned odor indicator re-
sponse is due to learning variables or detection of the odor.
For example, a number of studies have used the “double
alternation" learning paradigm in providing evidence for the
existence of an odor associated with non-receipt of an ”ex-
poctsd' reward (Ludvigson.& Sytsma, 1967: Ludvigson, 1969).
Typically this procedure involves the double alternation of
reward (R) and non-reward (N) in an RRNN pattern of successive
events in the goal box of a straight alley. 'A second char-
acteristic of these studies is that homogeneous goal box
.events are arranged such that each g’in a squad receives a
given ordinally numbered trial before any g receives the next
trial, and the goal box event of a given trial is the some
for each,§. A number of studies have shown that rats need
an external cue to learn this pattern (Bloom & Capoldi, 1961),
with learning manifested by significantly slower running speeds
on non-reinforced triale relative to running speeds on rein-
forced trials. Ludvigeon & Sytsmo (1967) have shown that when
the above conditions are met, i.e., double alternation paradigm
with homogeneous goal box events, rats show patterned running
in the area of the goal box. Seaga, Ludvigeon & Rsmley (1970)
have implicated an odor cue from the preceding conspecific
by showing that anaemic rats do not show patterned running

under the double alternation condition.

An unambiguous interpretation of the outcome of these
studies is that rats give off a substance which perseverates
after the frustrated animal is removed, and that the presence
of this substance is associated with one type of behavior by
the detecting animal, and a different type of behavior in its
absence. Unfortunately, this type of study reveals nothing
about the relationshipbetwsen odor emission, odor—of-fruetra-
tion and the response or class of responses made by a con-
specific detecting odor-of-fruetration. Both pheromonal prop-
erties and the cue prOperties of odor-of-frustration would be
expected to elicit responses incompatible with approach toward
the goal box, so that increased running time in the presence
of the odor is not an unambiguous demonstration of the aversive
properties of the odor when using this paradigm.

Rather than imposing a learned indicator response (dis-
crimination paradigm) upon the odor detector, a number of in-
vestigators have used designs showing the interaction between
learned responses and responses elicited by odor of frustra-
tion. HcHoee & Ludvigeon (1966), for example, trained on ex-
perimental groupto run in two discriminably different alleys,
one of which was associated with a small magnitude of reward
(s- alley) and the other associated with a large magnitude of
reward (s+ alley). HcHoee and Ludvigeon found that the control
group which received an intermediate but equal amount of food
in each alley ran slower in the s- alloy than in the s+ alley.
Presumably this effect was due to an odor given off by the

experimental animals in the a? alley.

A related effect has been demonstrated by Hasserman &

6

Jensen (1969) which they term the ”pseudo-extinction effect”.
Essentially this refers to the finding that “continuously re-
warded rats show a decrease in running speed on a runway
recently traversed by other rats undergoing experimental ex-
tinction” (p. 1307). This decrease in running speed was con-
fined to the goal area where odor-of-frustration would pre-
sumably be strongest. Since §a running speed did not decrease
on those trials where the preceding odorant animal was rein-
forced it is unlikely'that the effect was due to the character-
istic scent of the conspecific.

Evidence supporting the existence of an odor associated
with non-receipt of an ”expected” reward also comes from a
study carried out by Hellgren, Fouts & Hartin (1973). Hater
deprived odorant rats received twenty-four continuously rein-
forced trials in the center compartment of a three-chambered
box. They then received a series of trials of non-reinforce-
ment in the same center compartment. Hhen detector g; were
placed in the start.box (one of the and chambers) and per-
mitted entrance into the odorized center chamber their latency
to enter the odor laden center chamber was significantly longer
than detector‘gs exposed to odor of a nonfrustratad rat. La-
tency to leave the center chamber by "escaping'' to one of the
end chambers was significantly shorter for those detectors ex-
posed to odor-of-frustrotion, relative to the performance of
those detectors receiving the scent of a nonfrustratad con-
specific.

To date, the study of odor of frustration has been limited

to demonstrations of its existence. Determining the response

7

characteristics of odor emission as it relates to the oper-
ation of frustrative nonreward has been hampered by the need
to use an indicator response of a conspecific receiving odor-
of-frustration. Host responses can be measured using devices
that relate to the response in a known way, e.g.. clocks and
counters. Unfortunately, odor-af-frustration cannot yet be
measured using a mechanical device which gives a one-to-one
relation between a dial reading and some value of a response
parameter. The only "device“ which is sensitive to odor of
frustration is another rat which makes an unconditioned re-
sponse upon detection of the odor. Unlike the clocks and
counters, it is not known how changes in.the indicator re-
sponse relate to changes in some characteristic of odor emis-
sion.

An incidental finding of the previously cited study by
Seago,‘3t.{gl. (1970) points to a possible indicator response,
namely, latency to approach an area infused with odor-of-frus-
tration, which may be sensitive to graded changes in the con-
centration of the odor.. Recall that four groups of rats were
tested in a straight alley using a double alternation paradigm.
The g; of two groups were made anaemic as a consequence of
olfactory bulb removal; the rebaining two groups were tested
intact, and presumably were macrosomatic. The normal go showed
patterned running as expected while the anaemic animals didn't.
But of particular.intersetwas the additional finding that

the magnitude of patterning (difference between latency to

enter the goal box on reward versus nonreward trials) was a

function of the number of preceding g; on a given trial, i.s..

the § run seventh showed stronger patterning than ths‘g run
second. Such a finding suggests that either: (1) odor of
frustration accumulated as successive g; were tested on a
given trial and that latency to approach an area infused with
this odor is a function of the odor concentration, i.e.. a
pheromone effect, or (2) a greater concentration of the odor
provided a more‘oasily detected cue signaling nonreinforcs—
ment in the goal box area, or (3) perhaps both of those factors
were operating in additive fashion. In any case it is clear
that the indicator response was sensitive to the concentration
of the odor. This finding may permit a study of the response
properties of odor emission, especially as it relates in par-
allel fashion to the traditional response measures indicative
of a frustration effect, s.g., running speed in alloy two of
a double-alley apparatus (Amsel L Rousoell, 1952) or bar press
amplitude (Notterman & Hintr, 1965).

Amsel (1950) has proposed that the magnitude of the frust-
ration effect of “invigorating responses which follow (frus-
trative nonreward)" depends upon the strength of the antic-

ipation of reward, r - sg (Spence, 1956). The strength of

9
r - sq, in turn, is determined by such factors as magnitude

a: reward, and number of reinforced trials. Peckham & Amsol
(1967) has confirmed that the frustration effect is influenced
by reward magnitude, and a number of studies have found that
number of reinforced trials prior to frustrative nonreward is

positively related to the strength of the frustration affect
(Hug, 1970: Stimmell L Adams, 1969: Yslon, 1969).

Yelen (1969) trained three groups of rats in a double

alley apparatus constructed such that the goal box of alley
one also functioned as the start box for alley two. Each
group of rats was given 12, 36, or 60 trials with a 97 mg.
Noyes food pellet consistently available in both goal boxes.
The gs of all three groups were then shifted to a 501 rein-
forcement schedule for goal box one, with goal box two baited
on all trials. The magnitude of the frustration effect, as
manifested by significantly faster alley two running speeds
following goal box one nonreinforcemont versus goal box one
reinforcement, was directly related to the number of prior
reinforced trials. The frustration effect was largest for
.ths 60 reinforced-trials-group, second largest for the 36
roinferced-trials-group, and smallest for the 12 reinforced-
trials-group.

In summary, a number of studies have demonstrated the
existence of an odor associated with frustrative nonreward,
and the time has come to begin a study of the response char-
acteristics of odor emission. Yelen (1969) has shown that
the response of running following frustrative nonreward is
influenced by the number of reinforced trials preceding frus-
trative nonreward. Ths.purposo of the present study was to
determine the extent to which odor-ef-frustratien is similarly
influenced by this variable. Odor concentration was measured
in terms of the latency of an odor detecting‘g to approach
an area infused with odor-of-frustratien emitted by an odor-
ant animal receiving different numbers of prior reinforced
trials. Since Seaga,:gt.‘gl. (1970) showed that approach

latency is sensitive to sdor-ef—frustration concentration,

.e ,
.. -
J ".l v.
v.1-v'...

. s,
a
—. - v
a I- A-
- vo'h
. . I
s
a n.
L... ,., .
”em \
On
- .I-‘
4 s
’P
u
N ‘-
a‘
l . g.-
v'eggfil" .'.
I

’s>':4

.
low
(
' I
\a
at,
I
”A

‘4

(soul."'

'1.

(‘11

D
\ v
'. 'e
‘ "

f

‘ . r: ‘ J:
O A o
- Q. < l I g
- J ‘.
e O
. , .
. t. ' ‘

1«‘l

[ '«u

. -§‘

: tlz

' u".
f

0-: d

:70”

-. .1 - .
l 0 .
, 't
I
) r* 'r f
.,n g.

:‘.f 2'

.‘ l 7‘
5 .. :‘
”wm‘mj
. i".:‘
I rl._ "'_
Q 0-1)
.1.':l?‘[w

emu. J

O -\
| ,
;:II:

T -3E3'

VI] 111‘:

CURL"

J J
No
A. .4

.7. a ‘,
» m- ‘,
_. .. '1.
l L
.1 ‘ .
l aq\/ .11
n IT
5", :‘l.
-'
x
*«n.~' 1
1.1- .
. .i F,
.,.. . . c
I
T 7‘ "'tn,
.e [d l'
-V ..- _
D 0' Ll'.
O“ ‘f \
‘l. “l.
'Ll l( _ "
VI- r‘

v B
.4 t

10

the function relating latency of approach by odor detecting
go to number of prior reinforced trials of the odorant group
should give an indication of odor concentration as it relates

to the number of prior reinforced trials.

METHOD

Subjects:
One hundred and eight male, albino rats (Spraguo-Dawley

strain) served so So. Each‘g was experimentally naive, and

was 95 to 105 days old at the beginning of experimentation.
Upon arrival at the laboratory each rat was housed individ-
ually in an 0' X 10' metal cage, and provided with 32 libitum
food and water for a five day period. At the completion of
this period each.§ was reduced to approximately 005 of its

free feeding weight by imposing a 10 gram per-day deprivation
schedule.

Apparatus:

‘ The testing apparatus consisted of a three-chambered box,
with each chamber measuring 11' X 7%” X 0' high. Adjoining
chambers were constructed of +' clear plexiglas, and were
separated by 7%” X 0' high guillotine doors, also constructed
of clear plexiglas. Each chamber was covered by a hinged,
plexiglas lid measuring 11' X 0'. Strips of water proof butch-
er paper ovorlayed with absorbent Scott towsling covered the
floor of the apparatus. Clean paper could be pulled from paper
.rolls located at one end of the apparatus through a slot located
at the base of one and of the chamber: paper soiled with urine

and boli could be pulled from the apparatus through a slot

11

located at the other end of the apparatus. Forty-five mg.
Noyes food pellets were delivered down a i" (0.0.) rigid
plastic tube into a plastic food dish with a removable, clear
plastic top. The food dish, measuring 1' X 2', was located

in the center chamber. and was attached to one of the side
walls. Latency of response measures for the detector So

were taken by using microswitches attached to the guillotine
doors and one photo-relay located 4%" inside the center cham-
ber, with a second photo-relay 4}” inside the "goal box".

Each microswitch and photo-relay was part of the timing cir-
cuitry programmed through 20 v. electro-mechanical components.
Two clocks, capable of resolving .01 seconds provided measures
of response latency. Odor laden air was removed from the
chamber by using two 20 v. blowers. A 1 5/8“ rubber hose
connected the input port of one blower to the 1 5/0" exhaust
hole cut in the end wall of the goal chamber. A similar blower
and hose arrangement provided room air into the chamber through
a 1 5/0' hole cut in the start box and wall.

Procedure:

The 108 So were randomly divided into five groups of odor
emitters and eight groups of odor detectors. The size, treat-
ment and function of each of the thirteen groups was as follows:1
(1) A group of eight detector.rats received odor of a clean

(center) chamber in order to provide reference data for

 

1At the end of the description for each group is a code
enclosed in parentheses. This coderwill be used as a group
label, and is intended as a mnemonic device to help the reader
recall the.function and treatment assigned each group.

(2)

(3)

(4)

(5)

(6)

(7)

(8)

12

group three. (D-OCC) refers to Detector - Odor of Clean
Chamber.

A group of four nonfrustratad odorant So was placed in

the center chamber in order to provide a scent character-
istic of a nonfrustratad male rat. (0-0 R) refers to
Odsrant - O Reinforcement in the center chamber.

A group of eight detector rats received odor of a non-
frustratad conspecific so as to provide data on the
characteristic response elicited by odor of a nonfrus-
tratad male; and secondly, these gs provided “reference
data” for the detectors which received odor of a frus-
troted male rat. (D-O R) refers to Detector - receiving
odor associated with 0 Reinforcement.

A group of eight odorant animals received 40 nonreinforced
trials, followed by 6 reinforced trials, and finally frus-
trative nonreward. (0—6 R) refers to Odorant - 6 Rain-
forcements in the center chamber.

A group of eight detector rats received odor-of-frus-
tration from the So of group four. (D-6 R) refers to
Detector - receiving odor associated with 6 Reinforcements.
A group of eight odorant §s received 36 nonreinforced
trials, followed by 10 reinforced trials prior to frus-
trative nonreward. (0410 R) refers to Odorant - 10 re-
inforcements in the center chamber.

A group of sight detector rats received odor-of-frustra-
tion from thelgs of group six. (D-lB R) refers to De-
tector - receiving odor associated with 10 Reinforcements.

A group of eight odorant §s received 18 nonreinforced

(9)

(10)

(ll)

(12)

13

trials followed by 36 reinforced trials prior to frus-

trative nonreward.n (0-36 R) refers to Odorant - 36 Re-

.inforcements in the center chamber.

A group of sight detector rats received odor-of-frue—
tration from the go of group eight. (0-36 R) refers to
Detector - receiving odor associated with 36 Reinforced
trials.

A group of sixteen odorant So received 54 reinforced
trials and O nonreinforced trials prior to frustrative
nonreward. Eight go of this group received a single
trial of frustrative nonreward. (0-54 R-F1) refers to
Odorant - receiving 54 reinforced trials in the center
chamber followed by 1 trial-of Frustrative nonreward.

A second group of eight go received a second trial of
frustrative nonreward four minutes following following
the first trial of frustrative nonreward. Those So are
coded (0-54 R-Fz).

A group of eight detector rats received odor-of-frue-
tration hypothetically emanating from the paper covering
the floor at the time that eight of the So from group
ten received frustrative nonreward, i.s., the odorized
air of the center chamber was exhausted following frus-
trative nonreward, leaving only the paper as.a.eeurce of.
odor-of—frustration. (D—54 R-E) refers to Detector -
receiving 54 reinforced trials in the center chamber -
with the odorized air Exhausted following frustrative

nonreward.

A group of eight detector rats received odor-of-

14

frustration following the first nonreinforced trial of
the remaining eight rats of group ten. The center cham-
ber was not doodorized (i.s., the air wasn't exhausted)
following frustrative nonreward. (0-54 R-Fl) refers to
Detector - receiving the odor associated with 54 Rain-
forced trials in the center chamber and - one trial of
Frustrative nonreward.

(13) A second group of eight detector rats received odor of
frustration following the second nonreinforced trial of
the second set of group ten rate. This group is coded
(0-54 R-FZ).

One week following the onset of the deprivation schedule
a randomly determined odorant S was removed from the colony
room and carried to the cubicle containing the test apparatus.
Each odorantԤ_rsceived six massed trials per day. with all
nonrewarded trials (as specified above for each odorant group)
administered prior to the presentation of the reinforced trials.

This procedure allowed equivalence between groups for handling,

exposure to the apparatus, deprivation level at the time of
the nonreinforced test trial (trial seven of day nine), and
number of trials on the final test day. A

On all nonreinforced trials that preceded reinforced
trials on odorant §,was placed in the center chamber with the

. guillotine doors lowered, for a 45 second period. Food was

not presented and a clean food cup (one without the odor of

Noyes food pellets emanating from it) was covered with a

plastic lid. At the completion of the trial §,was removed

from the center chamber to the home cage for a 15 second

15

inter-trial invsrval. §,was then returned to the center cham-
ber for trial two. Trials two through six were identical to
that described for trial one.

Reinforced trials were identical to nonreinforced trials
with the obvious exception that ten 45 mg. Noyes food pellets
were delivered, all at once, down the +' tube into the plastic
food cup for Se consumption. The odorant S was placed into
the center chamber "facing away from" the food cup at a point
as far away from the food cup as possible. Using a stop watch,
measures of the time to approach the food cup were taken in
an effort to get an independent measure of the development of
”expectation" (as a function of the number of reinforced
trials).

During separate training sessions on days one through
nine the odor-detoctor'gs were placed into the start box (one
of the end chambers) for a one minute period in an effort to
make the indicator response (latency to approach the odor
laden center chamber) loss under the control of stimuli as-
sociated with the start box in subsequent test trials, and
more under the control of the independent variable (the hypo-
thetical differences in odor concentration in the center
chamber).

For the training that occurred on day nine, and for the
odor-of-frustration test trial, the 108.§s were grouped into
eight squads of fourteen Se each. Because there were only
four So providing odor-of-a-nonfrustratod male rat, group
(0-0 R), these g; were assigned to two squads. As such

each g of this group provided odor to two different detectors,

16

each in a different squad. The purpose of grouping §s into
squads was to control for temporal variations in factors which
may affect olfactory sensitivity of the detectors (e.g., humid-
ity) and short term deprivation of the odorants.

The odor of frustration test trial followed the sixth
trial of day nine. The odorant §.was removed from the cham-
ber following consumption of the pellets; the chamber was
cleaned by replacing the paper floor covering, exchanging the
food-odorized-feeding cup with a Clean one covered with the
plastic lid, and exhausting the odorized air from the chamber.
The odorant‘g was returned to the center chamber and ten 45
mg. Noyes food pellets were delivered into the closed food
cup. One implication of this procedure which should be made
explicit is that food pellets were present in the closed food
cup at the time that 211 detector gs were tested. At the end
of sixty seconds the odorant §,was removed from the center
chamber, and a count of the number and approximate size of
the urine spots on the paper covering the floor was taken.
Thirty seconds after removal of the odorant §_a naive detector
§ (i.e., one which had never recoived°food in the test chamber,
had not experienced the odor of another rat in the test cham-
ber, and had not explored any part of the test chamber except
the start box) was placed into the start box. Five seconds
later the guillotine door separating the start box from the
center chamber was raised and than lowered as the detector‘g
entered the center compartment. Simultaneous with the lowering
of the first door, the second guillotine door separating the

”goal box” from the center chamber was raised to allow the

17

detector §,to escape from the area infused with the various
odors (e.g., a clean center chamber, frustration, and the
characteristic scent of a nonfrustratad conspecific). If

the detector Sbfailed to enter the center chamber after two
minutes he was removed from the apparatus and his latency to
enter the center chamber was recorded as two minutes. Center
chamber escape latencies were similarly recorded.

Each detector §,was returned to the start box following
one of four inter-trial intervals (15 seconds, 45 seconds,
ninty seconds, and five minutes) for a re-exposure to the some
odor conditions prevailing during trial one. Trial two was
carried out in a manner identical to trial one.

Following the second trial for each detector the oppor-
atus was cleaned by ”pulling“ clean paper into the apparatus,
replacing the food cup, and wiping the inside surface of the
walls with a damp Scott towel. The odorized air of the chamber

was removed by activating the blowers for thirty seconds.
RESULTS

Latencios to approach the baited food cup by the odorant
,gs were recorded, and are summarized in Figure l for those So
which received nine days of reinforcement (Group (0-54 R-F1)
and Group (0-54 R-F2)).

The initial portion of the function shows a precipitous
drop in latency to approach the food cup between day one and
day three, with the attainment of a relatively stable asymptote
by the third block of six trials. The.accurocy of this de-

scription is supported statistically by a between-days

LATENCY TO APPROACH THE FOOD CUP

(Total no. of sec. in 6 trials)

300

250

200

150

100

50

1B

 

 

 

1 I l L 1 l I I
I 2 3 4 5 6 7 8 9

ACQUISITION DAY

Figure 1. Latency to approach the food cup as a
function of acquisition day.

BI

 

 

 

I I I I I I l l
e 8 T 8 a I5 E S I
YAO l/IOITIBIUOOA
s as qua boo? ed: dosciqqa oi yonstaJ .1 erupid

.vsb noiiieiupos is national

005

053

003

051

00!

08

(ioiol UO' or see gu 9 mole)
I'VlEl/ICA .LO VbbBOVCH .LHE LOOD Cflb

19

comparison of the performance of groups (0-54 R-F1) and (O-
54 R-Fz); (F .. 31.93, df .. 15,120, P (.001. The 93;; mg
comparison (Tukey B test) showed significant drops in latency
between days one and two and days two and three only; P<< .05.

A between-groups comparison of the day nine performance
of those odorant gs receiving one, three, six, or nine days
of reinforcement yielded an (F . 5.81, df . 3,31, P< .01.)
A‘gggt hag comparison showed significant differences in day
nine performance between group (0-6 R) and the remaining three
groups (0-18 R), (0-36 R), and (0-54 R). A non-significant
difference was obtained between group (0-54 R) and groups
(0-10 R) and (0-36 R). The means and standard deviations for
the day nine performance of the four groups are presented in
Table 1.

It is frequently assumed that urination is an emotional
consequence of the presentation of an aversive stimulus (e.g.,
Donny & Ratner, 1970). Figure 2 suggests that the amount of
urination was systematically related to the number of rain-
forced trials prior to frustrative nonreward. Since it was
impossible to measure the area of the urine spot with a ruler,
because to do so would require lifting the lid on the center
chamber and disrupting the odorized area prior to testing the
detector S, an estimate of the area of the spot was made in
terms of (or compared to) the area covered by a half-dollar
(assigned anarbitrary value of five), a quarter (4), a nickle
(3), a dime (2), and a spot smaller than a dime (1). Because
of the large number of zoro's (non-urinators) and the extreme

amount of variability of urine scores, the differences in

20

Table 1. Day nine acquisition per-
formance for four odorant
groups.

ODORANTS x(ssc.) S. D.

 

0-6 R 126.00 90.50
0-18 R 32.75 17.14
0-36 R 42.75 49.01

0-54 R 23.50 12.20

 

 

 

 

 

I!

i‘.

1

.14
I

131..”

Pl F -I I

$ O
I\L —1I

1...

INE

r
all.)

W

3.

“60301111111111?! OF UR

f

LA.-

«.0
as

3
is“
al-

03

'3
la.

er‘
~.

“'I
_/

1

04
on

I

C.

-
I

N \l
(I

50
C)

f-b

 

21

 

 

 

 

 

 

 

"ESTIMAT

o-oev 0-6:: we R‘ omen o-s4n

ODORANT GROUP

"Figurs 2. Estimated quantity of urine for
five odorant groups, with meas-

"-ures taken after frustrative
nonreward.

. .
"—- fl I
- -. - . é. ’7

IS

 

 

 

 

 

 

 

 

 

 

Fl 456-0 REE-0 R 81-0 Fl 8-0 RO-O

cIUOFIE) Tl/IAFIOCIO

In? enixu 1s yiiinaup botamlieB
-easm diiw ,equozp inszobo evil
aviiszieurl 13316 neflad eeiu
.b1swsinon

.S stupid

05

CS
08

3|
0|

EBLIWV1ED OnVI/llllA 0h new”;

22

amount of urine excreted was not significantly different be-
tween groups: (H a 5.96, df - 4, P:>».05).

By assuming that group (D-O R) represents the expected
number of nonfrustratad go which would urinate in the center
chamber in a one minute period it is possible to compare, via
the chi-square statistic, the number of Se urinating in the
nonfrustratad control group to the number of go urinating in
‘each of the remaining four frustration groups. The chi-square
value for groups (0-6 R), (0-18 R), and (0-54 R-Fl) was iden-
tical and equalled 6.25, P < .02. For group (0-36 R) the chi-
square value equalled 12.25, P<: .001.

Despite the large differences in “expectation” as mani-
fested by the day nine acquisition performance of the odorant
groups, and apparent differences in level of frustration, as
manifested by differences in magnitude of urination between
groups, the performance differences beteeen detector groups
was neither significant nor systematically related to the
number of reinforCed trials prior to frustrative nonreward of
the odorant _S_p; approach (F s 0.74, df :- 4,35, P) .05); escape
(F e 0.90, df = 4,35, P:> .05); the difference score, approach-
escape (F a 1.15, df . 4,35, P) .05). Table 2 gives the
means and standard deviations for the detector groups.

A comparison of performance differences was made between
group (D-54 R'Fl) which received the odor elicited as a func-
tion of the first frustration trial given group (0-54 R-FZ),
and group (D-54 R-Fz), which received odor-ef-fruetration
elicited as a function of the second frustration trial ad-

ministered to group (0-54 R-FZ): approach (T e 1.23, df a 14

23

Table 2. Trial one means and standard deviations
of odor chamber approach, escape, and
the difference score (approach - escape)

for eight detector groups.

 

 

 

 

 

DETECTORS MEASURE X(..c.) S. D.
approach 47.36 39.90
D-OCC escape 23.30 40.40
difference 23.99 47.61
approach 21.34 13.37
0-0 R escape 44.46 46.05
difference -23.10 41.30
approach 23.07 15.59
D-6 R escape 20.07 32.24
difference 17.75 54.07
approach 16.32 13.24
D-lB R escape 29.11 35.03
difference -12.75 30.25
approach 30.71 31.11
D-36 R escape 34.62 30.70
difference -3.91 46.06
approach 17.16 14.75
D-54 R-F1 escape 54.60 45.37
difference -22.31 30.33
approach 29.45 23.92
0-54 R-Fz ..C.P. 35.43 27o‘0
difference -5.97 14.49
approach 30.33 16.52
D-54 R-E escape 34.27 27.99
difference -3.96 20.92

 

 

24

Table 3. Trial two means and standard deviations
of odor chamber approach and escape for

eight detector groups.

 

 

 

 

 

DETECTORS MEASURE x(sec.) S. 0.
D-OCC approach 22.79 27.21
escape 43.20 51.14

D-O R approach 10.24 22.56
escape 36.60 42.10

D-6 R approach 16.44 14.95
escape 23.01 32.29

'D-10 R approach 5.90 6.70
escape 54.04 55.77

D-36 R approach 16.94 10.79
escape 45.00 45.97

D-54 R-F1 approach 16.01 16.00
escape 33.97 39.71

D-54 R-Fz approach 24.11 30.46
escape 51.99 57.16

D-54 R-E approach 20.97 25.77
. escape 52.05 56.17

 

 

 

 

 

 

 

 

 

 

IOI.II1 .1 (III- Iill'l l

m”...— ..

 

--—.--—~ -....—._

.q...._.... .—- -

25

P) .05): escape (T :- l.00, P) .05).

In order to determine if the paper covering the floor
under the frustrated rat was a source of odor-of-frustration
a comparison was made between group (D-54 R-E) and group (D-
54 R-Fl); approach (T a 1.60, df - l4, P:> .05) and escape
(T - 1.00, df . 14, P;> .05).

In order to determine the pheromonal reaction to odor
of a nonfrustratad male conspecific a comparison was made be-
tween group (D-OCC) and group (D-O R); approach (T - 1.75,
df is 14, P< .05): escaps.(T :- 0.97, .df - 14, P >.05); and
the difference score, approach - escape (T - 2.11, df - l4,
p< .05).

Essentially the same set of analyses were made an the
trial two performance of the various detector groups. Since
none of the analyses yielded statistically significant values
only the means and standard deviations are presented. These

may be seen in Table 3.
DISCUSSION

The finding that detector gs exposed to the character-
istic scent of a conspecific have a faster latency to approach
an area containing that odor, and a slower latency to leave
that area, relative to detectors exposed to the odor of a
clean chamber, supports the findings of Roiff (1956), and
Hollgren, Fouts, & Hartin (1973).

The failure to demonstrate an odor-of-frustration effect,
however, is contrary to the outcome of those studies which

report increased latencies to enter an area presumably infused

26

with odor-sf-frustration (Collerain & Ludvigson, 1972; HcHsss
A Ludvigson, 1966; Mellgren, Fouts, & Martin, 1973; Hasserman
L Jensen, 1969). Such a discrepancy in results forces one to
make a detailed examination of seemingly trivial procedural
differences which may be responsible for the different out-
comes of comparable studies.

Since the Hellgren, gt. 2;. (1973) methodology provides
the closest approximation to the methods of the present study,
it should be easiest to ferret out the relevant differences
by a comparison of these two studies. First, there are a
number of particulars which the studies have in common. Both
studies used male, albino rats of the Sprague-Dowley strain.
In both studies So were approximately 100 days old. The ap-
paratus of each study consisted of a three-chambered box, with
each chamber separated by a guillotine door, and the floor
of both apparatuses was covered by removable paper. Each
chamber of the Hellgron box was 15” X 5+" rather than the 11”
X 7}” chambers used in the present study. Most importantly,
in both studies the latency to enter the odorized chamber
was measured from the time the guillotine door was raised
until § broke a photsboam located approximately 4" into the
center chamber (4' exactly for Mollgren, 4}" in the present
study).

There are a number of procedural details which differ
for the two studies, some of which can be ignored based on
the findings of other studies. For example, Hellgren used
.5 cc of water as reinforcement, but most studies have used

Noyes pellets to obtain an odor-of-frustration effect.

27

Unlike the present study, the detectors of the Hellgren study
were exposed to the experimental apparatus on the test trial
only. It might be suggested that odor-of-frustration potent-
iates an initial fear of a novel apparatus, a fear which might
be absent in the present study since detector gs received nine
one-minute exposures to the start box. Other studies, how-
ever, havo obtained a detector aversion to the odorized area
after the detectors had received considerable pro-experimental
exposure to the to-be-odorized area (HcHose & Ludvigeon, 1966;
Hasssrman & Jensen, 1969).

There are two procedural differences which may be re-
sponsible for the discrepant findings. First, Hellgren ad-
ministered 12 trials of frustrative nonreward, at the rate
of three trials per day, with each detector receiving the
hypothetical sdor-of-frustration during each trial. Since
Hellgren combined his data over trials it is impossible to
determine if the odor affect developed after the first trial
of frustrative nonreward. If this were the case it could ac-
count for the discrepant findings. The fact that both Hellgren
and the present study obtained essentially the same pheromonal
effect when studying the detector reaction to odor of a non-
frustratad male conspecific (i.e., the odor elicits a rapid
approach and lengthy exploration of the area) suggests that
the differences in performance between the detectors receiving
odor-sf-frustration in tho twocstudies are due to differences
in the number of frustration trials given. Detector group
(D-54 R-Fz) was included in the present study to provide data
relevant to this point. Hhile a comparison of groups (D-54 R-Fl)

20

and (D-54 R-Fz) did not yield a significant T value, an in-
spection of Table 2 shows that group (D-54 R-Fz), which re-
ceived the odor-af-fruotration from the second nonreinforced
trial of group (0-54 R-Fz), had a longer latency to enter the
odorized center chamber and a shorter latency to leave this
area relative to group (D-54 R-F1) which received odor-of-
frustration from the first trial of frustrative nonreward of
group (0-54 R-Fz) odorants. Such a finding would be consist-
ent with studies using other measures of frustration. Ansel
A Roussell (1952), for example, observed an increase in run-
ning speed in alloy two of a double alley apparatus over the
first five trials of frustrative nonreward. Tortsra (1973)
found that panel pressing amplitude increased over the first
11 trials of frustrative nonreward, and then leveled off to
form a stable asymptote.

A second difference between the Hellgren, 35, 51. (1973)
study and the present investigation concerns the manner in
which an odor detector could ”escape" from the odorized cen-
ter chamber. Hellgren permitted escape either by re-ontering
the start box gr the opposite and box; the present investigat-
ion permittod entrance only into the end chamber opposite the
start box. The So could not escape back into the start box
once the guillotine door was lowered. There are two obser-
vation which were made that are relevant to this methodological
difference. First, it was observed that a large number of

detectors, once having entered the center chamber, would make

a vigorous effort to re-sntsr the start box. For example,

if the animal's tail extended back into the start box, thus

29

preventing complete closure of the guillotine door, the animal
would attempt to pry the door open with his nose. Such a strat-
egy obviously complicates an interpretation of the escape
latencies of the detectors. But more importantly, a number

of detectors were able to ”break” the photobsam located 4%”
inside the center chamber, without making a complete entry
into the center chamber, and thus preventing closure of the
guillotine door separating the start box from the center cham-
ber. When this happened the detector was able to rs-enter the
start box, and thus make an escape response, the latency of
which could not be measured. While only three So made this
form of escape, all go had a potential for this pattern of
responding, and procedural changes should be made in future
studies to control for this possibility. The use of photo-
beams as "movement sensors" also had the undesirable conse-‘
quence of contributing to the ”within-groups-variobility',
since some §s actively explored the light source while others
tended to avoid the area of increased illumination.

A second factor in the present study which appeared to
contribute to the large variability within groups was the type
of reaction elicited by the raising of the guillotine doors,
on event which the detectors did not experience during the
pro-teat exposures to the start box. For same So lifting the
guillotine door appeared to be the salient event responsible
for a short latency lungs foreword into the center chamber:
for other.§o door raising caused freezing. Either form of
reaction would undoubtedly mask any effects of odor-of-frus-

tration.

30

There are a number of potential remedies for reducing
the within groups variability and making the dependent var-
iable less under the control of stimuli associated with arbi-
trary procedural considerations and more under the control of
the independent variable. Rather than using a randomized
groups design it might be better to use a randomized blocks
design (Edwards, 1960), where each detector §Iwithin a given
group would be matched with a single g from each of the other
comparison groups on the basis of latency to enter the non-
odorized center chamber during a pro-toot trial.

A second possible improvement would be to use a hinged
or tilt floor in a two-chambered apparatus, rather than photo-
relays in a three-chambered box. Essentially a tilt floor is
a floor mounted on a fulcrum so as to allow a small amount of
vertical movement (e.g., 1/16”) at the two ends of the floor
as the weight of the rat shifts from one side of the pivot
point to the other. Deflection of the floor would be the
mechanical event responsible for closure of an electrical
relay integrated into the response timing circuitry. Such a
device has been used successfully in a number of studies of
odor effects in rodents (e.g., Doty, 1971), and would have
the advantage of eliminating the need to use photobsams. The
use of a two-chambered box would simplify an interpretation
of the escape latencies, since‘g would not have an option of
escape routes, and would not be engaged in a highly probable,
but ineffectual means of escaping, namely, attempting to es-
cape under the closed guillotine door separating the start

box from the odorized center chamber. While a tight fitting

31

guillotine door could be used as a partition to restrict the
hypothetical odor-of-frustration to the center chamber during
the twenty to thirty seconds between odor emission and testing
of the detector, the results of the present study suggest that
it would be best to raise the guillotine door just prior to
placement of the detector into the start box. A concern with
the possibility of immediate odor diffusion into the start
box is probably unwarranted since the double alternation studies
(e.g., Ludvigeon & Sytsma, 1967) and the Waessrman & Jensen
(1969) study of the pseudo-extinction effect show that the
odor is detected immediately in front of, or within the first
half of, the goal box. The odor had not diffused in detectable
strengths, into the middle portion of the alley, even with an
inter-trial interval of several seconds.

Certainly there are too many studies reporting a de-
tector aversion to odor of frustration to justify labelling
the effect a ”phantom phenomena". Implementation of the sug-
gested improvements in the methodology of the present invest-
igation may well facilitate the study of the response pro-

parties of odor emission.
A DISCUSSION OF RELATED TOPICS

There are three topics which are of indirect concern to
the present study, and which will be discussed briefly in an
effort to provide the reader with a summary statement regard-
ing: (1) strain differences in odor sensitivity, (2) the sec-
retory glands of rodents which may be responsible for the

production of the odorous material, and the role of urine as

32

a medium possibly containing the odorous substance, and (3)
other experimental operations producing aversive odors.
Strain Differences in Odor Sensitivity:

Early clinical studies of albinism in humans (e.g., Ogle,
1070) suggested that a lack of pigmentation is associated with
anaemia, or at least, a weakened sensitivity to olfactory stim-
ulation. Young (1957) has suggested a physiological basis for
this presumed relationship by claiming that the pale yellow
or dark brown pigment of the olfactory epithelium is necessary
for olfactory sensitivity. Briggs & Duncan (1961) have car-
ried the argument one step further by suggesting that caroten-
oids are responsible for the coloration of the olfactory epi-
thelium and are the chemical reactants of the olfactory recep-
tors rssponsible for the conversion of chemical energy to elec-
trical energy.

Houlton (1960) carried out an odor discrimination study
comparing the olfactory sensitivity of male black Norway rats
to male albino rats. More specifically, n-hexyl alcohol was
placed in one of two drinking bottles available to‘g at all
times, the position of the bottles was changes periodically,
with the drinking spout of one bottle connected to electric
shock. The other spout was not connected to electric shock.
Olfactory sensitivity was determined by reducing the con-
centration of tho odorant in the water bottle until there
was no significant deviation from a chance drinking score
of 50% (an equal amount taken from each bottle). The results
showed that pigmented rats were superior at 35 days, but ro-

tosting So at 160 days resulted in superior performance (i.e.,

33

a lower threshold) by the albino rats. Jennings & Keefer
(1969) have also failed to find differences in olfactory sen-
sitivity between male albino rats and headed males of the
Long-Evans variety.

More recently, Houlton (1962) has provided a critical
review of the physiological and biochemical evidence relating
the pigmentation of the olfactory epithelium to beta-carotene
and these two factors, in turn, to olfactory sensitivity.
Chromatographic evidence indicates that beta-carotene is not
part of the chemical complex responsible for the pigmentation
of the olfactory epithelium, and that the concentration of
beta-carotene is not related to olfactory sensitivity in dif-
ferent species.

Possible Secretor Glands Associated with Odor-of-Frustration

 

end the Role of Urine in Odor-sf-Frustration Secretion:

The specific secretory Gland (or glands) responsible for
the production of the olfactory material associated with frus-
trative nonreward has not been determined. There are two prob-
able reasons for this failure. First, members of the phylo-
genetic ordor Rodontia, e.g.. fig; and Rattus, possess an extra-
ordinary numbor of secretory glands and a variety of behaviors
associated with the use of those glands. Ssbacoous glands are
located over most of the surface of the body of rats (Mantegna,
1963). The preputial gland is located near the urethra, and
since both the amount and chemical composition of the secret-
ions of the preputial gland are affected by adrenal activity
(Lesher, Lorincz, & Rothman, 1954) it would be a prime candidate

for investigation. Hus also possesses secretory glands on the

34

sales of the feet (Tembrock, 1960) the output from which pro-
vides odor trails which conspecifics can detect.

The second reason so little is known about the physio-
logical basis of odor-sf-fruatration is that even the most
sophisticated methods of chemical analysis (e.g., gas-liquid
chromatography) are ineffectual in the study of the odor
molecules and their sources. Presumably this is because such
a minute amount of the substance is involved (Valonta A Rigby,
1968).

There has been considerable discussion regarding the role
of urine as it relates to odor of frustration (Schultz A Tapp,
1973; Deutsch, 1970). A number of studies (e.g.. Wasserman
A Jensen, 1969) including the present investigation have found
a strong relationship between frustrative nonreward and quan-
tity of urine output. Other studies, however, have found an
odor-of-fruetration effect in the absence of observable urine
spots (e.g.. Morrison A Ludvigson, 1970). Collerain A Lud-
vigeon (reported in Reyniorse, in press) have attempted to
resolve this inconsistency by suggesting that extremely small
amounts of urine may be excreted, which might be detected under
ultra-violet light but not under white light. Efforts assoc-
iated with the present study to replicate this finding using
a Sylvania lamp (FOTBS.1BLB) discharging ultra-violet light
were not successful. Urine spots on the Scott towsling were
no more visible under ultra-violet light than they were under
white light. The ultra-violet portion of the electromagnetic
spectrum ranges from about 0 millimicrons to 300 millimicrons,

and since Reyniorse (in press) doesn't report the discharge

35

wavelength used by Collerain A Ludvigeon it is possible that
urine does fluoresce under some ultra-violet wavelength not
used in the present study.

While the means of depositing the odorsus substance has
not been determined it is probable that the substance is
secreted on to the surface upon which the odorant animal treads.
Brill (1967) was able to transfer the paper odorized by the
presence of a nonfrustratad rat to a separate apparatus and
still get the odor effect of spontaneous alternation, i.s.,

a tendency for a nonfrustratad rat to avoid his own odor trail.
Carr, Hartorano A Krames (1970) were able to show that male
mice preferred an area containing sawdust odorized by a non-
stressed male mouse to sawdust which absorbed the odors do-
posited by a mouse recently defeated in an agonistic bout with
a dominant conspecific. The purpose of including group (D-54
R-E) was to determine if the odorized paper covering the floor
of the center chamber was sufficient to produce an odor-of-
frustration effect.. As reported in the results section a
comparison of both the approach and escape performance of
groups (D-54 R-E) and (D-54 R-Fl) yielded non-significant dif-
ferences. Surprisingly, the mean approach latency of group
(D-54 R-E) was longer than both groups (D-54 R-Fl) and (D-O R).
Speculation is possible, (e.g., exhausting the odorized air
for group (D-54 R-E) eliminated the approach-eliciting compon-
ent of the characteristic scent of the odorant, but did not
disturb the source of the odor-of-frustration located on the
odorized paper) but serious consideration of the implications

of this finding should await the demonstration of reliable

36

group differences as a function of this manipulation.
Odor of Frustration as One Instance of a General Stress Odor:

The use of the term oder-of-frustration is meant to imply
nothing more than the fact that frustrative nonreward of an
"odorant" §,is associated with behavioral changes in a detector
S placed in the area previously occupied by the frustrated
conspecific. A number of theorists have suggested that frus-
trative nonreward is an aversive event in the same way that
the presentation of shock is aversive (e.g., Wagner, 1969),
and that odor-of-frustration is really an odor-of-atress pro-
duced by a number of aversive events (Harrison A Ludvigeon,
1970). A comparison of the behavioral outcomes of exposure
to odors produced by frustrative nonreward and physical stress
tend to provide indirect support for this proposition.

Valenta A Rigby (1960) were able to demonstrate that male
albino rate could distinguish between the odor-of-shock stress
and the odor-of-an-unstressed conspecific. This ability was
manifested by an increased latency to bar press to stress
odor when that odor was associated with a bar press-punishment
contingency. Evidence was presented in the introduction that
odor-of-frustration could also provide a cue function (Hor-
rison A Ludvigeon, 1970).

The unconditioned responses to odor-of-shock stress and
odor-of-frustration are very similar. A number of studies
have demonstrated that rodents tend to avoid an area where a
conspecific has been previously stressed (Hullsr-Velten, 1966;
Rottman A Snowdon, 1972) or to have an increased latency to

enter that area (Courtney, Reid, A Wooden, 1960). Evidence

37

has been presented that odor-of-frustration produces a similar
reaction (Collerain A Ludvigson, 1972; Wasserman A Jensen,
1969; Hellgren, Fouts A Martin, 1973). While it is possible
that the operation of frustrative nonreward and physical stress
produce two qualitatively different odors, in the absence of
direct evidence, it is more parsimonious to assume that the
some mechanism and odor substance are involved in the two

aversive events.

BIBLIOGRAPHY

BIBLIOGRAPHY

Amssl, A. The role of frustrative nonreward in noncontin-

uous reward situations. Psychological Bulletin,
1950, 55, 102-119.

Amssl, A., A Roussell, J. Motivational properties of frus-
tration: I. Effect on a running response of the
addition of frustration to the motivational com-
plex. Journal of Experimental Psychology, 1952,
43, 363-360.

Archer, J. The effect of strange male odor on aggressive be-
havior in male mice. Journal of Hammalogy, 1960,
49, 572-575.

Bloom, J. H., A Capoldi, E. J. The behavior of rats in re-
lation to complex patterns of partial reinforce-
ment. Journal of Com arative and Ph siols ical

Psychology, 19 1, 4, 2 1-2

     

Bowers, J. H., A Alexander, B. K. Mics: Individual recog-
nition by olfactory cues. Science, 1967, 150,
1200-1210.

Briggs, H. H., A Duncan, R. B. Odour receptors. Nature,

Brill, H. Parameters of odor trail avoidance in the spontan-
eous alternation of the rat. (Doctoral Dissertation,
University of Cincinnati), 1967.

Carr, W. J., Martovano, R. 0., A Krames, L. Responses of mice
to adore associated with stress. Journal of Com ar-
ative and Physiological Psychology, 1970, 71, 22;-
220.

Collerain, I., A Ludvigson, H. W. Aversion of conspecific
odor of frustrative nonreward in rats. Psychonomic
ScienceI 1972, 27, 54-56.

Courtney, R. J., Reid, L. 0., A Wooden, R. E. Suppression of
running time by olfactory stimuli. Psychonomic
Science, 1960, 12, 315-316.

Denny, M. R., A Ratner, S. C. Comgarative Psychology, Home-
wood: Dorsey, 1970.

30

39

Deutsch, M. E. Olfactory stimuli and the "pseudo extinction"
effect. Science, 1970, 169, 402.

Doty, R. L. Homsspecific and heterospecific odor preferences
in sexually naive peromyecus maniculatpp Bairdi and
persmyscus leucopus noveboracensis. (Doctoral Dis-
sertation, Hichigan State University). 1971.

Edwards, A. L. Experimental Design in Psycholo ical Research.
New York: Holt, Rinehart and Winston, 1960.

 

Hug, J. J. Frustration effects after varied numbers of partial
and continuous reinforcement. Incentive differences
as a function of reinforcement percentage. Psychon-
omic Science, 1970, 21, 57-59.

Jennings, J. W., A Keefer, L. H. Olfactory learning set in

two varieties of domestic rat. Psychological Re-
E.rt!' 24’ 3-15.

Karlson, P., A Luschsr, H. 'Pheromones': A new term for a
class of biologically active substances. Natu o,
1959. 103, 55-56.

Lasher, N., Lorincz, A. L., A Rothman, S.~ Hormonal effects
on sebaceous glands in the white rat. II. The ef-
fect of the pituitary-adrenal axis. Journal of

Investigative Dermatology, 1954, 22, 2 -3l.

Ludvigson, H. W. Runway behavior of the rat as a function of
inter-subject reward contingencies and constancy of
daily reward schedule. Psychonomic Science, 1969,
15 (1), 41-43.

Ludvigson, H. W., A Sytsma, D. The sweet smell of success:
Apparent double alternation in the rat. Psychonomic
Science, 1966, 6, 203-204.

HcHose, J. H., A Ludvigson, H. W. Differential conditioning
with nondifferential reinforcement. Psychonomic
Science, 1966, 6, 405-406.

Hellgren, R. L., Fouts, R. 5., A Hartin, J. W. Approach and
escape to conspecific odors of reward and nonreward
in rats. Animal Learnin and Behavio , 1973, l,

Hontagna, W. Comparative aspects of sebaceous glands. In
W. Mantegna, R. A. Ellie, A A. F. Silver (Eds.),

Advance in Biolo of Skin, Vol. 4, New York:
Hacmillan, 1963. Pp. 32-45.

40

Harrison, R. R., A Ludvigson, H. W. Discrimination by rats of
conspecific odors of reward and nonreward. Science,

Houlton, D. G. Studies in olfactory acuity. 5. The comparaive
olfactory sensitivity of pigmented and albino rats,
Animal Behavior, 1960, 0, 129-133.

Houlton, D. G. Pigment and the olfactory mechanism. Nature,

Hullsr-Velten, H. Uber den Angstgeruch bei der Hausmaus
(Hus musculus L.). Zeitschrift fur Ver loichsndo
Physiologie, 1966, 52, 401-429.

Notterman, J. H., A Hintz, D. E. D namice of Res ones. New
York: John Wiley and Sons, 1

 
   

Ogle, E. Anaemia (or cases illustrating the physiology and
pathology of the sense of smell). Mod,-chir, Trans.,
1070, 35, 263-290.

Peckham, R. H., A Ameel, A. Within-subject demonstration of
a relationship between frustration and magnitude of
reward in a differential magnitude of reward dis-

crimination. Journal of Experimental Psychology,

Roiff, H. Untersuchungen uber naturliche und synthetische
Geruchstouffe, die bei Rotten und Hausen sine
stimulierende Wirkung auslosen. Acts Tropics,
1956, 13, 209-310.

Reyniorse, J. H. Communication elements constraining animal
learning and performance. Symposium on nonverbal
communication, U. of Toronto, 1973.

Rottman, S. J., A Snowdon, C. T. Demonstration and analysis
of an alarm pheromone in mice. Journal of Com ar-
ative and Physiological Psychology, 1972, 01, 403-
490.

Schultz, E. F., A Tapp, J. T. Olfactory control of behavior
in rodents. Psychological Bulletin, 1973, 21-44.

Seaga, J., Ludvigson, H. W., A Rsmley, N. R. Effects of ones-
mia an apparent double alternation in the rat.
Journal of Com arative and Ph eiolo ical Ps cholo

         

Southhall, P. F., A Lang, C. J. Odor cues in maze discrimin-
ation. ngchonomic Science, 1969, 16, 126-127.

 
    

Behavior Theor and Conditionin . London:
nivsreity Press, 19

Spence, K. W.
Yale

 

41

Stimmel, D. T., A Adams, P. C. The magnitude of the frus-
tration effect as a function of the number of pre-
viously reinforced trials. Psychonomic Science,
1969, 16, 31-32.

Tembrock, G. Communication in selected groups: Landeammals,.
In T. A. Sebesk (Ed.), Animal Communication, London:
Indiana University Press, 1960. Pp. 330-404.

Tortsra, D. F. The effect of incentive size on response ampli-
tude during acquisition and extinction. (dissert-
ation, Hichigan State University), 1973.

Valenta, J. 0., A Rigby, M. K. Discrimination of the odor of
stressed rate. Science, 1960, 161, 599-601.

Wagner, A. R. Frustrative nonreward: A variety of punishment.
In B. A. Campbell, A R. H. Church (Eds.), Punish-
ment and Avereive Behavior, New York: Appleton-
Century-Crofts, 1969. Pp. 157-104.

Wasserman, E. A., A Jensen, D. D. Pheromones and the “pseudo-
extinction"'effect. Science, 1969, 166, 1307-1309.

Yelen, D. Magnitude of the frustration effect and number of
training trials. fipychonomic Science, 1969, 15,

Young, J. Z. The Life of Hammals= Oxford: Clarendon Press
1957.

   

M'TITI'ITIEIITILEMIQIRIfiEIiIlflflE/ilflflfliflffim‘flfl'ES