Conditioning Situation Versus Intermittent Stimulus

This book made available by the Internet Archive.
Acknowledgments
I wish to extend deep appreciation to my colleagues, Drs. Michael Chase,

Carmine D. Clemente, Ronald M. Harper, Rebecca K. Harper, and Barry M.
Sterman, for reading parts of the manuscript and for their most helpful
comments. Special thanks are due to Dr. Irving Louis Horowitz, chairman of
the board and editorial director of Transaction Publishers for his excellent
suggestions and especially for his generous help with selecting the title of the
book. I also wish to thank the publishers of the University of Chicago Press
for permission to reproduce some of the illustrations used in this book.
I also owe very best thanks to my daughter Joanna-Veronika Warwick for her
valuable editorial assistance.
W.W
Vll
Introduction
Traditionally, in laboratory research, the process of conditioning, both

classical and instrumental type, is initiated with a single intermittent
stimulus (such as a tone, flash of light, etc.). Because of its role in evoking
conditioned behavior, the use of an intermittent stimulus has become an
indispensable part of laboratory
research on conditioned behavior.
A question arises whether the same scheme of conditioning may be applied to

behaviors occurring in real life. Let us analyze some cases from human
behavior. One example is illustrated by the sound of an approaching plane. If
this event happens in wartime, this sound can evoke the defensive reaction of
escaping to a shelter. But in peacetime, the sound of a plane usually evokes
only a slight
orienting reaction.
Another case, less dramatic than the first mentioned is an event observed
everyday in school. A bell sounding before class results in students entering
the classroom; the same sound at the end of the class results in students
leaving the room. Therefore, the same sound of a bell evokes two different
behaviors, depending on the
circumstances in which it occurs.
In each of these examples, the single stimulus (the sound of an approaching

plane or ringing of a bell) does not determine the reaction; instead, the
decisive factor here is the situational background against which the single
stimulus appears.
A question can be asked—what happens when the situational background is

suddenly changed? In a case related to me, a young married couple invited
the wife’s mother, who was still living alone in the “old” country, to live with
them in the United States. The mother came, but could not adjust to the new
life-style and became openly very unhappy. On her request the daughter took
hei back to her home in the old country. There are similar cases among
2 Conditioning
new immigrants to this country. One immigrant told me, “there is no life
here” and constantly returned to talk about his earlier times “there”
Similar observations were made on domestic pets. Some animals cannot

easily adjust to new conditions of life. I know of a case where a dog was given
to an older couple who lived about 100 kilometers from the animal’s previous
home. One week later, looking very exhausted, this dog managed to find its
way back to its former masters. The animal had probably not eaten anything
during the several days of its trip. So deeply touched were its masters, they
decided to keep the dog.
In the laboratory, reactions to situational stimuli are often disregarded, and

their occasional occurrence is sometimes treated as a phenomenon requiring
a new theory.
This book discusses various cases of situational influences on behavior and

attempts to provide a proper explanation.
Chapter 1 briefly describes some of I. P. Pavlov’s historic achievements in
creating a theory of conditioned reflex, a basis for understanding all
behavioral activities of the organism from a physiological point of view.
Chapter 2 presents results of studies by Konorski and Miller on motor

(instrumental) conditioned reflexes showing various kinds of these reflexes
and some differences from the classical Pavlov-ian conditioned reflexes.
Chapter 3 deals in full with early data on the phenomenon of so-called

“switching” obtained in studies of Konorski and Miller and in the laboratory
of Asratyan.
Chapter 4 describes the effects of changes in the experimental situation on

the value of conditioned reflexes.
Chapter 5 consists of two parts. The first part presents results of studies on
the role of the experimental situation in instrumental conditioning; the
second part describes the studies on the single (intermittent) conditioned
stimulus tested in various situations.
Chapter 6 discusses examples of complex instrumental reflexes such as

detour behavior and activities related to food intake and sleep.
Chapter 7 narrates cases of conditioning of activity of internal organs.
Introduction 3
Chapter 8 shows the therapeutic role of EEG feedback in epilepsy.
Chapter 9 contains a theoretical discussion of the data described in previous

chapters. It emphasizes the crucial role of situational background in
determining the final results of conditioning procedures.
Rounding out the book, a general summary and conclusions once again
emphasize the effect of the situation on conditioning processes.
Classical Conditioned Reflexes
Ivan Pavlov’s historic discovery of the principle of the conditioned reflex was
an enormous break in the scientific approach to the understanding of brain
functions.
One of Pavlov’s first observations in this area was made during a study on the
secretion of saliva in dogs. Namely, Pavlov found that the animals salivated
not only when they were eating food offered by a caretaker, but also when
they saw the caretaker approaching or heard his footsteps. This simple
observation indicated that the inborn reaction of salivation to food (or other
objects) in the mouth may also appear to originally neutral stimuli which
accompanied or immediately preceded the food intake.
As a physiologist, Pavlov chose a physiological interpretation of this

phenomenon (Pavlov 1926,1960). He accepted the concept of the reflex
introduced by Descartes (1764) that “an external or internal stimulus falls on
some one or another nervous receptor and gives rise to a nervous impulse;
this nervous impulse is transmitted along nerve fibers to the central nervous
system. There, due to the existence of nervous connections, it gives rise to a
fresh impulse which passes along outgoing nerve fibers to the active organ
where it excites a specific activity of the cellular structures” (Pavlov 1960,
p.7). Pavlov considered the secretion of saliva to the presence of food in the
mouth (and specifically to the tactile and taste stimuli produced by the
morsel) an inborn reflex and called it an unconditioned reflex (UCR) while
the salivation observed to the stimuli previously neutral (such as e.g., sounds
of the caretaker’s footsteps) he called a conditioned reflex (CR). The stimulus
that produced unconditioned reflex was
6 Conditioning
called an unconditioned stimulus (USC). The previously neutral stimulus that

produced conditioned reflex was called conditioned stimulus (CS) (e.g., a
definite sound or a visual object). This terminology has been widely accepted
by researchers.
Pavlov emphasized that the establishment of a conditioned reflex is

dependent on the application of the unconditioned stimulus (which is also
called the reinforcing stimulus or simply reinforcement) during or following
the action of the conditioned stimulus; it is necessary that the conditioned
stimulus starts to operate before the application of the reinforcing stimulus.
If the conditioned stimulus is not followed by the unconditioned stimulus,
the conditioned reflex cannot be established.
According to Pavlov, “conditioned reflexes are phenomena of common and
widespread occurrence: their establishment is an integral function in
everyday life. We recognize them in ourselves and in other people or animals
under such general names as ‘education’, ‘habits’ or ‘training’; all of these are
really nothing more than the results of an establishment of new nervous
connections during the post-natal existence of the organism. They are, in
actual fact, links connecting the extraneous stimuli with their definite
response reactions” (Pavlov 1960, p. 26).
In Pavlov’s laboratory a great variety of conditioned stimuli were used. These

included various sounds (called by Pavlov “stimuli of the acoustic analyzer”),
visual objects (“stimuli of the visual analyzer”), touch (“stimuli of the tactile
analyzer”), various smells (“stimuli of the olfactory analyzer”), changes in
environmental temperature (“stimuli of the thermal analyzer”), and even the
sensory feedback produced by performance of a passive movements (“stimuli
of the motor analyzer”). Practically each change in the environment could
become a conditioned stimulus on condition that it could be noticeable by the
corresponding receptor. For instance, a thermal stimulus of 45°C applied to
the dog’s skin could be made into a conditioned stimulus, whereas a stimulus
of 38-39°C, similar to the usual temperature of the dog’s skin, was ineffective
in spite of the fact that food reinforcement was offered after each application
of this stimulus.
Manipulations with a variety of conditioned and unconditioned
Classical Conditioned Reflexes 7
stimuli resulted in an accumulation of a great amount of experimental data

that served as constant enrichment of the Pavlovian theory. Let us
reminiscence with some topics of study in Pavlov’s laboratory (Pavlov 1960).
One of these topics was the process of inhibition of the previously acquired
behavior. Inhibition could be either external or internal. External inhibition
could be caused by an appearance of an extraneous stimulus, not involved in
the conditioned reflex arc. For instance, such extraneous stimulus could be a
sudden noise, not used previously, an unknown person who appeared
suddenly during the presentation of the conditioned stimulus, etc.
Internal inhibition is strictly related to the already formed conditioned reflex

and occurs inside the arc of this reflex. A simple example of internal
inhibition is a case when the reinforcing stimulus (e.g., food) has been
withheld. When this procedure occurs repeatedly in the following trials, the
conditioned reflex (e.g., salivation) gradually diminishes and finally stops
appearing to the conditioned stimulus. This phenomenon was called by
Pavlov the extinction of the conditioned reflex (Pavlov 1960, p. 49). It could
be obtained even during a single session when successive presentations of the
stimulus remained nonreinfored. However, the conditioned reflex to this
stimulus recovered as soon as the reinforcement was given again.
Internal inhibition observed in Pavlov’s laboratory included a form called the

conditioned inhibition. This type of inhibition was obtained when the
conditioned stimulus was preceded by a new stimulus and, in this case, the
reinforcement was not given. After several repetitions of this manipulation,
the conditioned stimulus, preceded by the new stimulus, never produced the
conditioned salivation whereas the conditioned stimulus applied alone did
(Pavlov 1960, Ch.V.).
Another case of internal inhibition studied in Pavlov’s laboratory was the “

inhibition of delay.” This kind of inhibition consisted ol extending the time
interval between the beginning ol the action ol the conditioned stimulus and
the reinforcement. The prolongation of the period of the delay could be
achieved gradually when (after establishment of a conditioned reflex with a
usual reinforcement
8 Conditioning
given 1-3 seconds after the beginning of the action of CS) the reinforcement
was delayed by an additional, e.g., 5 seconds, in each consecutive session,
until the delay reached several minutes. As a result of such procedure, the
conditioned salivation did not start immediately after the onset of the
conditioned stimulus, but only 30-60 seconds or longer after the beginning of
action of the CS. The same result could be achieved when after the
establishment of the conditioned reflex the delay in reinforcement was
prolonged to the desired length rapidly, without any gradual steps. However,
in this case, some complications in the dog’s behavior could appear during the
training period (Pavlov, p.89).
Finally, a form of internal inhibition was obtained when two similar

conditioned stimuli were used and one of them was always reinforced by food
whereas the other was not. As a result of this manipulation, the conditioned
salivation was present only to the CS which was reinforced each time,
whereas it was absent to the other stimulus. This form of internal inhibition
has been called differential inhibition or differentiation (Pavlov 1960, pp.l 17-
125 and following).
It should be mentioned, however, that the above described differential

inhibition does not mean a complete disappearance of the reflex. In some
circumstances such as, for instance, the entrance of a new person into the
experimental chamber, the previously inhibited reflex may be disinhibited
and the conditioned salivation may reappear (Pavlov 1960, p. 115).
Generalization of Stimuli
An important phenomenon observed mainly during the initial phase of

establishing the conditioned reflex is the tendency to react not only to one
particular stimulus but to all stimuli similar to it. For instance, when a
conditioned reflex has been established to a tone of specific frequency by
reinforcing it with food, many other tones will also produce salivation in spite
of the fact that they were never reinforced. The value of the conditioned reflex
produced by these similar stimuli depends, however, on the degree of
similarity to the original tone. The same can be said about stimuli of
Classical Conditioned Reflexes 9
other sensory systems (i.e., other analyzers). This phenomenon was called
generalization of stimuli (Pavlov 1960, p. 113). It is often observed both in the
laboratory and in natural life.
The research on conditioned reflexes not only consisted of compiling more

and more experimental material supporting Pavlov’s concept, but also
involved some new topics and experimental techniques. Studies performed in
the laboratory of Bykov (1957) may serve as an example of such extension of
research on conditioned reflexes. The results of these studies showed that it
is possible to establish a conditioned reflex not only in relation to salivation
but also to secretion of other glands of the alimentary system. It was
demonstrated in a child provided with a stomach fistula that when the food
reinforcement was always given after the sound of a trumpet, the stomach
secretion started to flow to the sound of the trumpet before the food was
given (experiments of Bogen, described by Bykov 1941).
The ability of conditioning was also observed on the gall bladder; namely, the
contractions of the gall bladder increased when food was shown to a hungry
dog (experiments of Kurtsin and Gorshkova, described by Bykov 1957).
The extension of the research on conditioned reflexes also included some new
methods. An example of such novelty were experiments conducted by
W.H.Gantt who also was one of Pavlov’s pupils. Gantt chose the functions of
the heart and the circulatory system as the object of research and
demonstrated that the rules of conditioning also extended to this area (Gantt
1944, 1960, Gantt and Hoffman 1940).
All the above described experimental results, which represent only a small
part of the pioneer achievements of Pavlov and his pupils, were obtained with
the use of single intermittent stimuli. In their research there was no problem
of the situational background. However, Pavlov was aware of the existence of
a great number of various stimuli in the environment which could influence
the experimental results. In order to avoid interference of external stimuli
during experiments with conditioned reflexes, a special building (known also
as a “tower of silence ”) was constructed (see fig. 3 at the beginning of Lecture
3, Pavlov 1960).
10 Conditioning
This building consisted of four chambers on each of three floors. Each

chamber was fully isolated from the external environment, securing stable
conditions for the experiment. This allowed the investigator to concentrate on
the effects of the intermittent stimuli and not pay attention to the situational
background against which these stimuli were applied.
This led to an almost complete neglect of the problem of the situation in

Pavlov’s laboratory and later in the laboratories of Pavlov’s followers. In fact,
by using special training methods, the problem of the environmental
background was removed from most experimental studies on conditioning.
Nevertheless, the problem of the situation exists. It will be discussed in

further chapters of this book.
Instrumental Conditioned Reflexes
As usually happens in science, new theories develop by the addition of

previously unknown facts or by different interpretations of the known facts.
Although Pavlov’s concept and theories of the conditioned reflexes seemed to
encompass the whole problem of behavior, one of its mechanisms still
remained unclear. That was the mechanism of the motor conditioned
reflexes.
According to Pavlov (1951), the general rule of the conditioned reflex also
included the motor conditioned reflexes. When a movement is performed, it
produces a sensory input of impulses from the muscle. When this input is
each time followed by the reinforcement (e.g., food), the conditioned
salivation appears to the movement (such as a flexion of the dog’s leg) before
the food has been offered. This was demonstrated by Krasnogorski (1911; cit.
by Pavlov 1951), who used a special device to produce a passive lifting of the
dog’s leg. In this case, therefore, the motor conditioned stimulus was indeed
analogical to the acoustic, visual, or tactile conditioned stimuli (see a scheme
of this conditioned reflex in fig. 2.1).
An important question remained unanswered. In order to get the motor

sensory input, the movement must be performed first. In the laboratory it
was produced through the use of a special device. But how is the movement
initiated and performed in natural conditions? This question was not
answered by Pavlov’s concept (see Windholz and Wyrwicka 1996).
The problem of the motor conditioned reflexes was experimentally analyzed

by two medical students of Warsaw (Poland) University, Jerzy Konorski and
Stefan Miller. After reading Pavlov’s
11
12 Conditioning
FIGURE 2.1
MOVEMENT ► IMPULSES ► SALIVATION
(classical CR)
A diagram of motor conditioned reflex in experiments of Krasnogortski
according to the explanation of I.P. Pavlov (see Ch. 2).
Movement, a passive movement of the dog, evoked by a special devise

(provided by the experimenter).
Impulses, nervous impulses deriving from the performed movement to the

brain sensory system.
Salivation, conditioned reflex of salivation.
As shown on the diagram, the conditioned reflex of salivation appears only

after the performance of the movement by the dog. In the laboratory, the
movement appears due to the use of a technique producing the passive
movement. However, there is no explanation of origin of the active
movement of the animal in free conditions (without provoking any passive
motor activity).
just published book on conditioned reflexes, they became enthusiasts of

Pavlov’s concept of conditioned reflexes. They found, however, that there was
an area in the behavior of the organism that was not included in Pavlov’s
concept. This area was the conditioned motor behavior (occasionally also
called by them “voluntary motor behavior”). Konorski and Miller, therefore,
started their own experiments in the provisional laboratory space offered to
them by a friendly professor of psychology.
The design of their first study was very simple. Using a mechanical device
which could be controlled by them from a distance, they produced a flexion of
the dog’s leg to an acoustic stimulus which was immediately followed by food.
After several repetitions of this procedure, the dog started to perform the leg
flexion actively (i.e., without any help of the mechanical device) to the
acoustic stimulus. Using various kinds of reinforcement (positive
reinforcement such as food, or negative reinforcement as a puff of air to the
ear, as well as the presence of the reinforcement in some cases and its
absence in other cases), they found that four
13
varieties of the motor conditioned reflexes exist. Let us briefly describe each
of these varieties.
First variety. When a specific movement (e.g., flexing the leg) is performed to
a conditioned stimulus (e.g., a tone) and is always followed by obtaining a
reward (e.g., food), then this movement will always spontaneously appear to
this stimulus.
Second variety. When a specific movement (e.g., leg flexion) appearing to a

conditioned stimulus (e.g., a tone) is never rewarded (e.g., by food), whereas
the reward is offered only to the conditioned stimulus alone (without the
movement), then the leg’s flexion stops being performed, and in its place an
antagonistic movement (usually an extension of the leg) will appear to the
conditioned stimulus.
Third variety. When a specific movement (e.g., leg flexion) appearing to a

conditioned stimulus (e.g., a tone), is always rewarded by withholding a
punishing stimulus (e.g., an airpuff to the ear), then the specific movement
will always appear to the conditioned stimulus. This variety is often referred
to as an “avoidance reaction.”
Fourth variety. When a specific movement (e.g., leg flexion) appearing to the
conditioned stimulus (e.g., a tone), is always followed by a punishing
unconditioned stimulus (e.g., an air puff to the ear), then this movement will
cease to appear and an antagonistic movement (e.g., leg extension) will be
performed instead.
Each variety was experimentally documented. The young investigators,

however, needed to hear an opinion about their research and discuss the
results. Therefore, they wrote to Pavlov informing him about their study. In
answer, Pavlov invited Konorski and Miller to his laboratory in Leningrad,
where they had an opportunity to repeat and confirm their previous
experimental results on a greater number of dogs and under better
experimental conditions (Miller and Konorski 1928, Konorski and Miller
1936, Konorski 1937, 1948, 1967).
Konorski and Miller concluded that the motor conditioned reflexes differed
from the Pavlovian reflexes in that the reinforcement in the former was
dependent on the performance (or inhibition) of a specific movement,
whereas there is no such condition in the Pavlovian reflexes. Therefore, they
called the motor condi-
14 Conditioning
tioned reflexes “conditioned reflexes type II,” while reserving the name
“conditioned reflexes type I” for Pavlovian reflexes. Later the terms “classical
conditioning” for conditioned reflexes type I, and “instrumental conditioning”
for conditioned reflexes type II, were introduced by Hilgard and Marquis
(1940, Kimble 1961) and became commonly used by the behavioral scientists.
Konorski and Miller admitted that the prototype of the conditioned reflexes
type II was studied earlier by Thorndike (1911) who called this kind of
behavior “the trial-and-error learning.”
In 1948, Konorski published a book entitled Conditioned reflexes and neuron

organization (dedicated to I.P. Pavlov and C.S. Sherrington), in which he tried
to explain the experimental results obtained in Pavlov’s laboratory according
to the newest findings of neurophysiology, based mostly on Sherrington’s
studies of the spinal cord (Sherrington 1929, 1947). Konorski’s new
interpretation included a theoretical explanation of the phenomena called by
Pavlov “irradiation,” “negative induction,” “positive induction,” “internal
inhibition,” and others. Instead of the explanation of the formation of the
conditioned reflex by “the meeting of the waves irradiated from different
points” (of the cerebral cortex) Konorski proposed “the formation of
excitatory synaptic connections between two coupled center....” (Konorski
1948, p. 251). He also proposed that the “chief feature of the conditioned
reflex” is not the excitation of the conditioned centre but, instead, the
excitation of the “unconditioned centre” (p. 251). He also interpreted the
“negative induction” as “inhibition of the conditioned reflex by an
antagonistic reflex” (p.251) and the “positive induction” as “summation of the
excitato-inhibitory conditioned reflexes with facilitation predominant”
(p.253). The process of “internal inhibition” which was interpreted by Pavlov
as an “inhibitory process occurring in the conditioned centre,” was explained
by Konorski as “Inhibition and excitation of unconditioned centre” or “Pure
inhibition of unconditioned centre” (p.252). The full list of Konorski’s new
interpretations related to the conditioned reflexes can be found in pages 251-
254 of his book (Konorski 1948).
Early observations of Konorski and Miller on conditioned reflexes type II

(Miller and Konorski 1928, Konorski and Miller
15
FIGURE 2.2
One of the First Records of Type II CRs
Upper graph: CS (lamp).
Lower graph: record of the movements of the left foreleg.
a) The movement is performed in the presence of the CS and in intervals.
b) The movement is performed only in response to the CS.
(Konorski and Miller, 1933).
A copy of an original recording of the dog’s movements (lifting of the left

foreleg).
a) The dog performs the movement not only during the action of the
conditioned stimulus (CS) but also during the intervals between the
successive applications of the CS.
b) When food is offered exclusively in the presence of the CS, and never in its
absence, the movements appear only during the action of the CS.
1936, Konorski 1967) included the fact that the trained instrumental
movement appears not only in the presence of the conditioned stimulus but
also in its absence, i.e., in response to the experimental situation. As shown in
part a of fig. 2.2 (which is a copy of an original record taken during the early
period of the training), the instrumental movement (lifting the left foreleg)
appears not only during each application of the conditioned stimulus but also
in the intervals between the successive applications of CS. When the re-
16 Conditioning
FIGURE 2.3
« mil l) miimm ii
in umumwm> urn » i h n n i. v4^
W l * --* ■ * — ul 1 1 Ml ...
Alimentary Type II CR Established to the Experimental Situation
From top to bottom: movements of the foreleg, pressing the lever, salivation.
Arrow denotes the increased portion of food.
(Konorski and Miller, 1933).
The establishment of an instrumental reflex to the situation. When in the

absence of the intermittent stimulus each lifting of the right foreleg and
placing it on a platform was followed by food, this movement became
conditioned to the situation.
inforcement (food) was given exclusively in the presence of the CS, the
instrumental movements started to be performed only during the presence of
CS and not in its absence (fig. 2.2, part b).
In another experiment, the dog was trained to lift its right foreleg and place it
on a small board; each movement was rewarded with food. No intermittent
CS was used, therefore this instrumental movement was performed to the
situation only (fig. 2.3). When food reinforcement ceased to follow each
movement, the trained movement gradually disappeared. This supported the
view that
Instrumental Conditioned Reflexes 17
the instrumental conditioned reaction was established to the situation, and

not to a specific single stimulus.
These documented observations are of great importance for the theory of

conditioning. They show clearly that the conditioned movement is connected
not only with the intermittent CS but also with the situational background
against which it appears. The movements appearing in the intervals between
the applications of the CS are usually extinguished by being nonreinforced.
Nevertheless, later observations showed that the intertrial instrumental
movements can still occasionally reappear (personal observations of the
author).
Early studies of Konorski and Miller (Miller and Konorski 1928, Konorski and
Miller 1936, Konorski 1939, 1967) included the phenomenon of “switching” or
“interchange” which consisted of the appearance of different reactions to the
same CS in specific circumstances. This problem will be discussed in the
following chapter.
Konorski’s experimental studies on conditioned reflexes conducted in the

later period comprise the problem of “plasticity” of the conditioned reflexes,
that is, the possibility of the transformation of conditioned reflexes type I into
conditioned reflexes type II (Konorski and Wyrwicka 1950), an exchange of
excitatory into inhibitory conditioned reflexes (Konorski and Szwejkowska
1952a,b) or alimentary into defensive conditioned reflexes, and vice versa

(Konorski and Szwejkowska 1956).
All the studies mentioned above, as well as most studies conducted in his
laboratory by his pupils and associates, were described and analyzed by
Konorski in his last book entitled Integrative Activity of the Brain. An
interdisciplinary approach published, in 1967.
Motor behavior was the subject of studies by several investigators with a

psychological rather than physiological approach. In 1911, Thorndike
published his book Animal Intelligence , in which he described his
experiments on learning of various motor tasks by animals such as fish,
chicks, cats, dogs, and monkeys. Fish had to find an opening in a partition
inside the aquarium in order to reach the food placed behind the partition.
Chickens had to find the correct way in a maze to the food placed at the goal.
The re-
18 Conditioning
quired motor tasks for cat and dogs included pressing a lever, pulling a cord,
turning a knob, pushing the door, or in some special experiments, licking the
fur or scratching. Monkeys had to perform more complex movements, such
as manipulations with a wooden peg, hook, metal bar, or turning a knob by
270°, to get a reward. This kind of behavior, which was called by Thorndike
“trial-and-error learning,” corresponds to Konorski and Miller’s “conditioned
reflexes type II” (as mentioned earlier).
Extensive studies on the problem of motor conditioned behavior were

conducted by Skinner (1938, Ferster and Skinner 1957). Skinner also
distinguished two types of behavior: the type corresponding to classical
conditioned reflexes was called the “respondent behavior,” and the type
corresponding to conditioned reflexes type II of Konorski and Miller was
called the “operant behavior.” The respondent type of behavior refers “to
conditioning which results from the contingency of a reinforcing stimulus
upon a stimulus ■” whereas the operant type of behavior results “from
contingency upon a response ” (Skinner 1937, pp.272-73).
Skinner developed special techniques for studying the operant behavior. His
subjects were either pigeons or rats. The operant task for pigeons was to peck
at the visual stimulus and for rats to press the lever. The movements of the
subjects were automatically recorded enabling an immediate analysis of the
obtained data. This resulted in the compiling of a great amount of
experimental data related to such problems as discrimination of responses,
generalization of stimuli, delayed responses, inhibition (extinction), chained
schedules of conditioning, and others (Ferster and Skinner 1957). Skinner
also theorized on problems such as “drive” and “drive and conditioning,” and
discussed the effects of emotions and drugs on conditioning (Skinner 1938).
However, he did not relate the behavioral results of his studies to brain
functions.
One of the most important points of Skinner’s theoretical approach was the
nature of the stimulus. Skinner emphasized the fact of the presence of great
variability of stimuli in the environment and the necessity of the
differentiation of the chosen stimulus before its conditioning. He also
stressed that a chosen motor act to be conditioned is never completely new
but it could be performed
19
occasionally earlier. Therefore, when speaking about the conditioned motor

behavior we should use a comparative measure: to prove that the number of
performances of this motor behavior is significantly greater after than before
conditioning.
Because all motor activity of the organism is based on the inborn motor
reflexes, it is possible that the movement chosen for the training is not new
but could have been performed occasionally in the history of the subject.
Being aware of this, the investigator must be sure that the chosen movement
was never performed by the subject during the preliminary control sessions.
The appearance of that movement during the training sessions can be
considered proof of the establishment of this movement as an instrumental
conditioned reflex. The problem of the absolute novelty of the movement
should never be considered.
The above discussed points are only a part of Skinner’s hypotheses and
suggestions. The interested reader is directed to Skinner’s original
publications (Skinner 1938, Ferster and Skinner 1957).
The Phenomenon of Switching
As mentioned previously in chapter 2, early research on instrumental

conditioning has shown that, in some circumstances, a definite conditioned
stimulus may not produce a motor reaction trained to it, but instead, a
different reaction that has been trained to a different stimulus before (Miller
and Konorski 1928; Konorski and Miller 1933, 1936; Konorski 1939, 1948,
1967). Here are some examples of their observations.
In studies on this phenomenon, called by Konorski and Miller (1928, 1936;

Konorski 1939, 1967) cases of “switching” (also called by them “transfer” or
“interchange”), the authors tried to find the exact conditions which
determined the appearance of this behavior.
Switching in Homogeneous Instrumental Reactions with Different Defensive

Reinforcement
In a dog, a classical (Pavlovian) conditioned reflex (CRI) had been established

when a tone was always reinforced with the introduction of a small portion of
acid solution into the dog’s mouth. The conditioned reaction consisted of
salivation and slight defensive movements. In further sessions, the sound of a
bell was always followed by a puff of air into the dog’s ear; however, when the
dog’s left foreleg had been passively lifted during the action of the bell, the
puff of air was not given; as a result, an instrumental avoidance reaction (i.e.,
active lifting ot this leg) was established. When this reaction was firmly
established, the previously used CS, the tone, was applied. Already after
several trials, the tone began to evoke both salivation (the CRI previously
trained to the
21
22 Conditioning
tone) and the lifting of the left foreleg (CRII, the reaction trained to the bell
but never to the tone). According to Konorski’s interpretation, the obtained
result demonstrated that a CS connected by training with a definite defensive
CRII, may elicit this reaction to another CS related to a different defensive
reinforcement. Konorski also hypothesized that similar cases of switching
may also occur in conditioned reflexes with approach type of reinforcement
(Konorski 1939, p. 20). This supposition was verified by Fonberg (1961, 1967),
as described later in this chapter.
Switching in Heterogeneous Instrumental Reactions
This type of experiment was performed by Konorski and Miller during their
visit to Pavlov’s laboratory (Konorski and Miller 1936, Konorski 1939). First,
an alimentary instrumental reaction of lifting the right hindleg to a specific
noise and a defensive (avoidance) reaction of lifting the right foreleg to the
sound of bubbling water were established in a dog. During each session a
band was fastened on the trained leg for recording purposes. It was found that
when the band was fastened on one leg only, the dog lifted this leg both to
alimentary and defensive stimuli. Only when the bands were fastened on both
trained legs, were the reactions correct, i.e., the alimentary CS produced the
alimentary CRII, and the defensive CS produced the defensive (avoidance)
CRII. However, when the alimentary CS was applied after a number of trials
with the defensive CS, the dog first performed a defensive CRII and then only
the correct alimentary CRII. And, vice versa, when the defensive CS was
preceded by a number of successive applications of alimentary CS, the
alimentary CRII appeared first and then only the correct defensive CRII was
performed. This case of switching (regarded by Konorski and Miller as a
result of the “inertia of excitation”) was not always observed in all dogs.
Switching Produced by Factors Other Than Inertia of Excitation
An alimentary CRI was established separately to the metronome,
The Phenomenon of Switching 23
to a tactile stimulus, and to the lighting of a lamp; in addition, a defensive

CRII consisting of lifting the left foreleg to avoid a puff of air into the ear, was
established to the sound of a bell. Then a process of extinction of the CRI to
the metronome was performed. During a number of sessions the metronome
was applied 2-10 times per session without the food reinforcement, whereas
the tactile stimulus and the lamp were always followed by food. The band (for
recording purposes) was always fastened on the left foreleg. During the first
17 extinction sessions salivation to the metronome beats was only partly
decreased. When, starting with the 18th session, the metronome was applied
many times in a row in each session without food reinforcement, the salivary
CRI rapidly diminished and the avoidance lifting of the left foreleg
(established to the bell but never used with the metronome), began to appear
both during the action of the metronome and during the intertrial intervals.
Thus, a deep extinction of on alimentary CRI resulted in the performance of a
defensive CRII to the alimentary CS.
A question was raised as to whether the alimentary instrumental reaction

may appear to a defensive CS in conditions different than the inertia of
defensive excitation. In order to answer this question, an alimentary classical
type conditioned reaction (CRI) was established in a dog to the tactile
stimulus, the lamp, and beating of the metronome (120/min). In the same
dog, an alimentary instrumental reaction (CRII) of lifting the right foreleg to
the sound of a bell, and a defensive instrumental reaction of lifting the right
hindleg to the tone to avoid an air puff to the ear was established. Then a new
defensive CS, the sound of bubbling water, reinforced by acid-to-mouth, was
introduced. It was found that already the very first application of the
unconditioned stimulus, acid-to-mouth, produced the defensive movement of
the right hindleg. In the following trial, the same movement appeared to the
bubbling. In the third trial, both the lifting of the foreleg (an alimentary CRII)
and the lifting of the right hindleg (a defensive CRII) were performed to the
bubbling of water. A similar effect was observed when a new stimulus, the
sound of a rattle, was followed by a puff of air. Already in the third trial the
rattle evoked lifting the right hindleg (a defensive CRII, previously
established to the tone) and then
24 Conditioning
lifting the foreleg (an alimentary CRII, previously established to the bell).
These experiments showed that the alimentary CRII may appear to a
defensive CS in conditions others than inertia of preceding excitatory
processes.
The obtained results were interpreted by Konorski and Miller (1933, 1936;
Konorski 1939, 1967) in the following way. There are two kinds of stimuli: the
determining and the releasing stimuli. The determining stimuli, such as a
band on the dog’s leg, or overall experimental situation, do not elicit the
conditioned reaction but they determine which kind of reaction is to be
performed; the alimentary or defensive background produced by preceding
application of unconditioned stimulus could also be considered a determining
stimulus.
The releasing stimuli are usual intermittent stimuli that elicit the reaction. As
previously described, a partly inhibitory (differential) alimentary CRI could
elicit a defensive CRII trained to other CS; also, a deep extinction of an
alimentary CRI could result in the appearance of a defensive CRII established
to a different CS. An application of an aversive unconditioned stimulus could
also release a previously established alimentary CRII.
The boundary between the determining and the releasing stimuli is not
absolute, however; for instance, a continuous determining stimulus may be,
at the same time, a releasing stimulus, and vice versa, a sporadic stimulus
may both determine and release the reaction (Konorski 1967, p. 389).
As already mentioned earlier in this chapter, Konorski and Miller observed a

transfer between two CRIIs, each of which was trained with a different
aversive unconditioned stimulus.
Later studies by Fonberg (1961, 1969) showed that the transfer of

instrumental reactions is possible both in avoidance and in alimentary CRIIs,
as described below.
Transfer in Avoidance Behavior
Two different avoidance CRIIs were trained in four dogs in the usual
chamber. The conditioned stimuli were the whistle, the bell, the tone, and the
light. The training was conducted in separate
series to each CS. Each animal was trained to lift the right hindleg to the CS 1
to avoid a nociceptive reinforcing stimulus which was either an electric shock
or a puff-to-ear in various dogs. Then, in another series of sessions, the dog
was trained to place its left foreleg on a platform to stimulus CS2 to avoid a
nociceptive stimulus. After the establishment of the second CRII, a test was
performed in which CS1 and CS2 were applied in the same session. It was
found that both reflexes (lifting the right hindleg to CS 1 and placing the left
foreleg on the platform to avoid nociceptive stimulus) were usually
performed correctly and they did not interfere with each other when the
animal was calm. However, when the animal was excited, both avoidance
reactions appeared in the same trial (to either CS1 or CS2), often
simultaneously or almost simultaneously; these movements also were
observed in the intertrial intervals. This kind of switching was observed in
two dogs.
According to Fonberg’s interpretation, these results support a hypothesis that

there exist specific centers in the brain, corresponding to each nociceptive
stimulus; these centers are all connected to a common defensive center. In
the state of excitement (such as fear) the common center may activate the
specific defense centers, resulting in the appearance of all trained avoidance
responses to each CS.
Approach Behavior
In another study (Fonberg 1967), two dogs were trained to press a lever to the
tone to obtain food. Then chronic electrodes were implanted in these dogs’
lateral hypothalamic “feeding” area (Anand and Brobeck 1951). Electrical
stimulation of the hypothalamic points corresponding to the tips of the
electrodes, produced eating in satiated animals. When the hypothalamic
stimulation was substituted for food reinforcement, the dogs began to
perform the previously trained alimentary CRII (lever pressing).
In further sessions, the instrumental response was reinforced either with

food or with the electrical stimulation in the lateral hypothalamus. It was
observed that the hypothalamic stimulus did
26 Conditioning
not produce an increase of hunger but rather a state equivalent to food

reward. In this experiment, therefore, a specific kind of switching took place
between two different reinforcements of the approach type, the food and the
electrical stimulation in the lateral hypothalamus. According to Fonberg’s
interpretation, the hypothalamic stimulation could produce sensations
similar to those provided by food intake, leading to the performance of the
response that was originally trained for food reinforcement.
The Studies on Switching Conducted in Asratyan’s Laboratory
The phenomena of switching were extensively studied on dogs in Asratyan’s

laboratory with the use of the classical conditioned reflexes (Asratyan 1951,
1961, 1965). In the first study, in sessions conducted in the morning by
Shitov, the CS of 120/min beats of the metronome was always reinforced with
food. In the same dogs, in sessions conducted in the afternoon by Yakovleva,
the same stimulus (metronome beats of 120/min) was reinforced by an
electric shock. As a result, in the morning session the beats of metronome
elicited an alimentary reaction (salivation and approach to the feeder), and in
the afternoon session the same CS produced a defensive reaction of lifting the
leg that was to receive the electric shock.
Even more complex experiments on switching were conducted by Struchkov

(Asratyan 1961). First, two conditioned reflexes were established in each dog,
an alimentary CR to the sound of a buzzer (reinforced with food) and a
defensive reflex to the tactile skin stimulation reinforced with an electric
shock to the skin. These reflexes were trained in the same experimental
compartment. Then, a parallel series of sessions began in another
experimental compartment in which the previous CSs were used but the
reinforcements were reversed. That is, the sound of buzzer was now
reinforced with electric shock, and the tactile stimulus was reinforced with
food. After some training each of these CSs could evoke two different
conditioned reactions, an alimentary CR in one compartment and a defensive
CR in the other compartment.
In experiments conducted by Pressman (Asratyan 1951) stimulating

electrodes with a registering device were attached to the dog’s left hindleg and
a defensive conditioned reflex was established to the tactile stimulation of
that leg. Then the stimulating electrodes and the registering device were
attached to the dog’s right hindleg (which was never used in experiments
before), and the usual experimental procedure started. This time, the CS
evoked lifting the right hindleg and not the left hindleg which was previously
used in the training. When the electrodes were attached to both hindlegs, the
CS evoked lifting either the left or right hindleg, in irregular manner.
However, after application of an electric shock to one of these legs, the dog
began to lift only that particular leg in further trials. These experiments are
similar to those performed by Konorski and Miller who obtained the same
results when the band was fastened on both hindlegs of the dog.
In other experiments performed by Shitov and Zamyatina (Asratyan

1961,1965), each experimenter conducted two sessions daily, one in the
morning and another in the afternoon. In the morning session, the acoustic
CSs were reinforced with food, and in the afternoon they were not reinforced
at all. As a result, the same acoustic CS elicited an alimentary (salivary) CR in
the morning session and an inhibitory (no salivation) CR in the afternoon
session. This experiment showed that switching may occur not only between
two excitatory reactions but also between the excitatory and inhibitory states.
According to Asratyan (1965), the phenomena of switching are a result of

action of permanent situational stimuli called by him tonic stimuli. In some
cases, the tonic stimulus was a different experimenter, and, in other cases, a
different time of the day. A different kind of reinforcement in the afternoon
session than that in the morning session could also be a factor in switching,
playing the role of a tonic stimulus. The tonic stimuli do not evoke the
conditioned reaction but they may facilitate the action of phasic stimuli , i.e.,
the intermittent stimuli (such as a tone).
The effects of tonic stimuli were demonstrated by Sakhiulina (1955, cit. by

Asratyan 1965, pp. 139-140) who recorded the electroencephalographic
responses (EEG) from various points of
28 Conditioning
the dog’s cerebral cortex during the experiments on switching. In the morning
session, an auditory stimulus was paired with electric shock to the dog’s left
hindleg; in the afternoon session, the same auditory stimulus was paired with
an electric shock to the right hindleg. This training resulted in lifting the left
hindleg to the CS in the morning session and lifting the right hindleg to the
same CS in the afternoon session. At the same time, the EEG record showed
the presence of high amplitude waves in the leads from the boundary area
between the motor and parietal areas, ipsilateral to the trained legs, i.e., from
the left hemisphere in the morning sessions and from the right hemisphere
in the afternoon sessions. This EEG pattern appeared at the beginning of the
session, before the application of the CS, and persisted throughout the
session. This kind of EEG activity could reflect the action of the tonic
stimulus.
Other Studies on Switching
More recently, the problem of switching was studied on human subjects by a

number of investigators (Kimmel and Ray 1978, Murrin and Kimmel 1986,
Lachnit 1986, Kimmel and Lachnit 1990, and others). In their studies, a
strictly defined stimulus was selected, often referred to as a “context” or
“contextual stimulus,” or, according to the terminology used by Asratyan, a
“tonic” stimulus against which a “phasic” (i.e., intermittent) CS was acting. In
their experiments, the tonic stimulus preceded and lasted through the end of
duration of the phasic stimulus. The reinforcement was an electrodermal
shock to the finger. In the case when the tonic stimulus preceded CS, the
reinforcement was withheld. The results of these experiments showed that in
cases of the application of the tonic stimulus the conditioned response was
absent. This effect was considered by the experimenters a case of switching.
These experiments were similar to studies of Pavlov’s laboratory on

“conditioned inhibition” in which the reinforcement was withheld each time a
novel stimulus preceded the action of the CS (Pavlov 1960, Ch.5). The
difference between the “context” ex-
periments and Pavlovian studies was that the context studies were conducted
on human subjects whereas Pavlovian studies were conducted on dogs.
Studies on the effect of the “context” were also conducted on pigeons by

Rescorla (1984). In his experiments the contextual stimulus was the design of
lining of the walls of the experimental compartment combined with a visual
stimulus (blue or yellow light); some of these combinations were reinforced,
whereas other were not reinforced. The results of these experiments led the
author to conclude that CSs become associated with the context in which they
occur, and that manipulations of CS (by reinforcing or not reinforcing it)
produce changes in the associations between the context and the CS which, in
turn, influence the performance to the other CSs used in these experiments.
An especially striking result was that related to the extinction experiment. It
showed that the extinction process related to one CS strongly influenced the
subsequent behavior related to the other CS.
The experimental data described in this chapter are strictly related to the
influence of the situational background on behavior. They also support the
hypothesis that conditioning of the situation does take place, and its effect on
behavior dominates over the effect of the single conditioned stimulus. This
problem will be further discussed later in this book.
Cases of Dependence of Conditioned Behavior on Situational Factors
Observations similar to those reported by Konorski and Miller (1933, 1936;

Konorski 1939, 1967) were made in the laboratory of Kupalov (Denisov and
Kupalov 1933, Kupalov 1961). In a study on dogs, two kinds of illumination of
the experimental chamber were used, full illumination in some sessions and
poor illumination in other sessions. It was observed that when the
conditioned reflex was trained in full light and then tested in poor light, the
value of the reflex remained at a high level for several following sessions, and
then only decreased and stabilized at a low level. And, vice versa, when after
several weeks of training in poor light full light was used again, the
conditioned reflex remained at low value for the next several days.
Kupalov (1961) explained this phenomenon as an effect of conditioning of the

experimental situation. According to him, there are two kinds of brain
processes involved in conditioning. One process is related to the intermittent
conditioned stimulus that produces the conditioned reaction. The other
process is related to the experimental situation (such as the kind of
illumination) which does not produce the reaction but may change the tonus
of certain parts of the brain. Kupalov called this second process a “shortened”
or “truncated’ conditioned reflex. According to Kupalov, therefore, the role of
the experimental situation is limited: it can only modify the tonus of the
excitatory process but cannot play a decisive role in the performance of the
reflex. Kupalov stressed, however, that the strength of these “shortened”
31
32 Conditioning
reflexes can be unexpectedly high. This interpretation will be discussed in

chapter 9, together with other theoretical problems.
A specific case of situational influence on conditioned reflexes was observed

by Doty and Giurgea (1961; Giurgea 1989). In a dog, they paired moderate
electrical stimulation of the left occipital cortex with similar stimulation of
the left motor cortex. As a result, stimulation of the left occipital cortex
regularly produced a flexion of the right forepaw and turning of the head to
the right (as if they stimulated the left motor cortex). When conditioning was
well established, electrical stimulation of the right homotypical cortical
regions including the right ectosylvian area was applied. It turned out that
this time, stimulation of either the right occipital cortex or the right motor
cortex produced regularly the previous response, i.e., a flexion of the right
forepaw and turning head to the right. Therefore, the motor reaction
remained the same, as previously trained, no matter which, left or right,
cortical area was stimulated. The authors called this phenomenon a
“conditioned transfer.”
However, this transfer was observed only when the general experimental
situation was unchanged. After placing the animal on the experimental table
and then rotating the table by 180°, the previously observed effect was no
longer observed. That is, stimulation of the left occipital cortex or left motor
cortex produced normal contralateral movements, whereas stimulation of the
right occipital cortex remained ineffective. On the other hand, stimulation of
the right motor cortex elicited contralateral reactions, i.e., head rotation to the
left and flexion of the left forepaw. As commented by Giurgea (1989) “even
such as ancient, phylogenetically and ontogenetically rooted physiological
response as motor contralaterality following efficient motor cortex
stimulation might be critically modulated by shortened conditional reflex in
the experimental situation.”
The influence of the experimental situation in conditioning was earlier

demonstrated in studies conducted in Konorski’s laboratory. The topic of the
first of these studies (Konorski and Szwejkowska 1950) was the chronic
extinction of conditioned reflexes against an excitatory background. In two
dogs, several
Conditioned Behavior on Situational Factors 33
alimentary classical conditioned reflexes were established to various stimuli.

One of these CSs was chosen for chronic extinction. This stimulus was
applied, without being reinforced with food, only once during each session,
among other conditioned stimuli which were reinforced. After a number of
sessions, the conditioned salivation stopped appearing to the non-reinforced
stimulus which became an inhibitory stimulus. Then the extinguished
conditioned reflex to this stimulus was restored by offering food again after
each application of this stimulus. The same procedure was then used with
each of the remaining stimuli. It appeared that chronic extinction was
dependent on the strength of each stimulus and a degree of its fixation; the
acoustic CSs (which produced strong salivation) required 21 to 62 sessions to
become extinguished whereas the visual CS (which produced a weak
salivation) required only 15 sessions.
In a further series of experiments, the chronic extinction against an inhibitory

background was studied on two other dogs by Szwejkowska (1950). This time
the conditioned stimulus used for extinction was applied in separate
experimental sessions during which the food was not given at all. Such
sessions were interspersed with sessions in which the other stimuli were
applied with the usual reinforcement. It was found that the chronic extinction
of a conditioned reflex against an inhibitory background occurred more
rapidly than the chronic extinction against the excitatory background,
requiring only 16 sessions in the case of using a strong acoustic stimulus, a
bell.
The results of both studies (Konorski and Szwejkowska 1950, Szwejkowska

1950) suggest that the differences in the results were caused by the use of two
different situations in the process of extinction. This supports the hypothesis
that the experimental situation plays an important role in the conditioning
processes.
The role of the situation in the extinction of the instrumental conditioned

reflexes was also studied by Wyrwicka (1952). In these experiments,
conducted on three dogs, the course of extinction was studied against either
an excitatory or inhibitory background. In the experiments on the excitatory
background, only one conditioned stimulus, the sound of bubbling water, was
used. This stimu-
34 Conditioning
lus was applied several times in each experimental sessions at intervals of 3-5
minutes and was not reinforced with food. However, food was given during
the intervals, without any relation to the intermittent stimulus. Already on
the first session of this study, the latency of the instrumental reaction to the
CS, which was 4 seconds in the first trial, increased to 12 seconds in the 7th
trial. On the other hand, the salivary CR which was 35 points on the
manometer scale, in the first trial, decreased to 18 in the 7th trial although
the salivation during eating remained about the same up to the end of the
session. During the following sessions both salivary and instrumental
conditioned reactions to the CS gradually diminished and after 6-8 sessions
they completely disappeared. However, as soon as the reinforcing stimulus,
food, was given again to the extinguished stimulus (bubbling), both salivary
and instrumental conditioned reflexes gradually recovered.
Several months later, another extinction series of sessions was conducted on

the same dogs. This time, however, a different conditioned stimulus, a
whistle, was used, and no food was offered either to CS or during the intervals
between the successive trials. This was, therefore, an extinction against the
inhibitory background. When the reflex to the whistle was entirely
extinguished, the reflexes to other stimuli were tested. It appeared that both
instrumental and salivary reactions to these stimuli (which were not used
during the process of extinction) were strongly diminished. Similarly to the
extinction against the excitatory background, both salivary and instrumental
CRs gradually recovered as soon as food reinforcement was offered again.
The changes in the experimental procedure also appeared an important

situational factor in behavior. In experiments on dogs by Zbrozyna (1953) an
acoustic or visual (light) stimulus was applied when the dog was eating and a
few seconds later the food was withdrawn. After several repetitions of this
procedure, the dog stopped eating and turned away from the feeder as soon as
the withdrawal signal was given. However, when the same stimulus was
applied before food was offered it evoked salivation and approach to the
feeder. Therefore, the same stimulus could elicit two different reactions
depending on the procedural situation, in the
presence or absence of food. In the presence of food the conditioned stimulus

evoked a negative reaction, i.e., cessation of eating and turning away from the
feeder; and in the absence of food, the same stimulus produced a positive
feeding reaction. The author interpreted these results as a phenomenon of
non-identification of a stimulus appearing against a different situational
background.
Cases of Dependence of Behavior on the Experimental Situation in

Psychopharmacological Studies
In a study of Giurgea (1970) on dogs, two classic conditioned reflexes were

established in each animal. One of these reflexes was trained in the usual
Pavlovian room, and another reflex was trained in Kupalov’s “free-behavior”
room, to the same conditioned stimulus. The two rooms were separated by a
long (of about 30 m) corridor. In the middle of the corridor there was a small
room in which injections were given. It was found that chlorpromazine (1
mg/kg, intravenously) strongly reduced the value of conditioned reaction
when the dog was tested in the Pavlovian room 10 minutes later.
On the other hand, when the dog was tested on another day, in the free-
behavior room, the same amount of chlorpromazine did not have any effect
on conditioned reaction. The same dog, after injection of this drug, taken just
in front of the Pavlovian room, showed a full neuroleptic effect, whereas
when placed in front of the free-behavior room, this dog appeared almost
normal as if he did not receive any drug. When taken back in front of the
Pavlovian room, this dog again showed the neuroleptic behavior whereas no
such behavior was observed when placed in front of the free-behavior room.
Such alternation in behavior could be induced 10-12 times during a period of
30-45 minutes, in various dogs.
This finding supports earlier observations of Kupalov (1961) that dogs were
more resistant to neurogenic procedures in the free-behavior situation than
in the Pavlovian room. The situational effects on behavior were also
demonstrated in studies of Zuckermann and Buffy (1960), and more recently
by Khananashvili (1983).
Poulos and Hinson (1982), in a study on rats, also observed a

36 Conditioning
strong dependence of the effects of the neuroleptic drug halo-peridol on the

environmental background. They found that about 20 daily injections of this
drug resulted in the development of a highly significant tolerance to the
cataleptic effects. Then, a group of rats was injected twice daily, once in room
A with haloperi-dol, and once in room B with saline, in random alternation
from day to day (but always with haloperidol in room A and saline in room B).
After about 20 days, tolerance to haloperidol was present only in room A
where the rats received haloperidol. No tolerance was present after
haloperidol injections in room B (where the rats usually received saline);
instead, an almost total catalepsy was seen.
In studies of Corson (1967), and Corson and O’Leary-Corson (1968), dogs

with salivary fistula and urinary bladder fistula were tested for
electroencephalic and respiratory activity in two different rooms. One of these
rooms was an experimental compartment in which the dogs received painful
shock to a hindpaw to a tone. Another room was a control compartment in
which the dogs never got a painful shock. When tested in the control room
during a 3-hour session, the dogs urinated a normal amount of low-density
urine; they also showed no spontaneous salivation and their heart rate
lowered over time. However, when tested in the experimental room, where
they obtained shocks following a tone, the dogs urinated much less with a
high density urine, produced some salivation, and showed elevated heart rate.
When they were tested again in the experimental room, without receiving any
tones or shocks, the dogs still showed the previously observed reactions when
the tones and shocks were used. This suggested that the experimental
compartment alone was able to signalize, through a “shortened” conditioned
reflex of the situation, a “danger” related to the previous experience in the
experimental room (cit. by Giurgea 1989).
In another study, conducted on morphine-reacting rats, Eikelboom and

Stewart (1979) demonstrated that saline may evoke responses similar to the
hyperthermic response usually produced by morphine, but only when the
saline injection was given in the environment in which the rats usually
obtained morphine. No hy-

pothermic response was seen in the rats’ home cage where they usually did
not receive morphine injections.
Eikelboom and Stewart (1979) also observed the environmental effects of

daily injections of morphine on body temperature in 30 male rats. The
animals were divided into four groups, one group received injections of 5 mg,
the second group 25 mg and the third group an increasing dose of up to 200
mg of morphine. The fourth group received only saline injections. In each rat
the rectal temperature was measured five times a day in three different
environments: the home cage (a neutral environment), the pre-injection
environment where the rats were placed before the daily injection, and the
injection environment where the animals remained after the injection. The
injections were followed by a period of abstinence.
The tests for conditioning showed that rats in the morphine groups,
compared to the saline group, showed an anticipatory hypothermia in the pre-
injection environment; this was just opposite of the unconditioned
hyperthermia to morphine. On the other hand, the rats in the morphine
groups showed a conditioned hyperthermia when tested in the injection
environment after a period of abstinence. The results indicate that the
changes of body temperature depended on the environment which was
related to the morphine injection. In other words, it can be said that
conditioning to the environment took place.
Circling behavior produced by apomorphine in rats, and its dependence on

the environment were reported by various investigators (Ungerstedt 1976,
Silverman and Ho 1981). It was found that even simple placing of the rats in
the cage where they previously obtained apomorphine injections resulted in
the appearance of circling. This effect was even stronger when the rats
obtained saline injection before the test.
The environmental effects were demonstrated by Melchior (1988) in a study

on tolerance to ethanol in male mice. It was observed that
intracerebroventricular (ICV) injection of a very small amount of ethanol
produced a brief but substantial hypothermic response. However, after eight
ICV injections of 2 mg of ethanol at 2-hour intervals, four injections per day
during two days, an
38 Conditioning
environmental tolerance to the hypothermic effect of ethanol developed. The
environment-dependent tolerance was also seen when these mice were tested
with intraperitoneal injections of ethanol.
A study on gastric secretion conducted on cats by Wyrwicka and Garcia (1979)

also showed the environmental influences. The experiments were conducted
on two groups of cats. In one group of 6 cats basal gastric secretion, and in
another group of 7 cats pentagastrin-induced gastric secretion, was collected
during a 2-hour session, two or three times a week. There were two kinds of
sessions. In one session, the cat’s movements were restricted by placing the
animal in a harness, and in the other session, the cat was unrestricted and
could move freely in a cage. It was found that in 4 of 6 cats in the basal
secretion group and in 5 of 7 cats in the pentagastrin secretion group, gastric
acid output was significantly higher in sessions in harness than in sessions in
cage. These differences in acid output were due to acidity rather than volume
of secretion. It was hypothesized that the restriction of the movements in the
harness produced a “reflex of freedom” which could not be accomplished
because of confinement. This led to neural disturbances in the autonomic
system, resulting in changes in gastric acid secretion.
Special observations on the situational effects on behavior of human subjects

were reported by Lhermitte (1986). Although his observations were made in
patients after frontal lobectomy, nevertheless they provided important
material concerning the role of definite environmental situations in behavior.
Two patients, a man 51 (patient 1) and a woman 52 years of age (patient 2),
were taken to a doctor’s office, a lecture room, a car, an apartment, and a gift
shop. It was found that the patients were excessively dependent on
environmental cues. Here are the excerpts of observations made in various
situations.
When in the doctor’s office some medical instruments were put on the desk,
the female patient immediately picked up the blood pressure gauge and
measured the doctor’s blood pressure. Then she took the tongue depressor
and examined the doctor’s throat.
Both patients were taken by the doctor to a lecture room with a buffet of food
and beverages. Patient 1 helped himself to the food
and the orange juice and behaved like a guest. In contrast, patient 2 behaved
like a hostess. When she saw some stacks of chairs, she proceeded to set the
chairs side by side. Then she offered some foods on various plates and a glass
of orange juice to the doctor. She did not serve herself.
The doctor took patient 1 and his female friend to his apartment and said that
it was a “museum.” Patient 1 immediately started to examine the paintings as
if he were in a real museum. His behavior was appropriate for any usual
museum visitor. Moreover, when he saw that one painting was down on the
floor, he took the hammer and nails (which had been left next to the
painting), hammered a nail into the wall and hung the painting. Then the
doctor and patient walked around the apartment and also entered the
bedroom. When the patient saw the bed without the bedspread, he
immediately started to undress and then got into bed and prepared to sleep.
The behavior of patient 2 in the bedroom was slightly different. She did not
undress but lay down immediately. Then she accompanied the doctor to a
table with various items used for intramuscular injections. When the doctor
showed her a syringe, she took it and gave the injection to the doctor.
When discussing the patients’ behavior, Lhermitte proposed defining it by the

term environmental dependency syndrome.
All studies described in this chapter show the decisive role of the
environmental situation on the behavior. From the point of view of
conditioning, the behavior became associated (conditioned) to the situation
due to the repetitive application of the reinforcing stimulus in this situation.
This problem will be discussed from the theoretical point of view in a later
chapter.
Studies on the Role of the Situation and the Intermittent Stimulus in
Conditioning
As described in chapter 1, all Pavlov’s experiments were conducted using

single, intermittent stimuli appearing against a permanent and unchangeable
environment. The problem of situation in which the conditioned stimulus
appears was practically not discussed. However, later observations by Vatsuro
(1948) revealed that changes in experimental setting could result in a
decrease of conditioned reflexes in dogs and monkeys. Stroganov (1948)
observed inhibition of conditioned reflexes after transferring experiments
from one experimental chamber to another. Similarly, Beritov (1948) found
that extraneous factors introduced into the usual situation might inhibit or
completely change the established conditioned motor behavior.
A systematic study on the role of the experimental situation on the

conditioned reflexes type II (instrumental conditioning) was undertaken by
Wyrwicka (1956, 1958, 1993). The study consisted of several series of
experimental sessions conducted daily for a two- week period, separately for
each of 2-5 dogs in each series. The sessions were conducted always about the
same time of the day. The reinforcement was food (bread) in all experimental
series.
First Series: One Experimental Situation
In a usual Pavlovian chamber, each dog was trained to perform two reactions:
one was lifting the right hindleg (CR1) to an acous-
41
42 Conditioning
tic stimulus, the sound of a rattle (CS1), and the other was placing the right
forepaw on the feeder’s platform (CR2) to a visual stimulus, a rotating disc
(CS2). Each reaction was trained in a separate series of sessions. A two- week
series of daily sessions with CS1-CR1 alternated with a fortnight series of
sessions with CS2- CR2. It was observed that at the beginning of the next
series each CS evoked a “wrong” reaction, i.e., the reaction trained during the
preceding series (for instance CR2 to CS1, or CR1 to CS2). Only when the
wrong reaction was not reinforced by food did the correct reaction appear. It
was also found that during further sessions of a series, the dogs often
performed both reactions to each CS (i.e., both CR1 and CR2 to either CS1 or
CS2).
Second Series: Two Similar Experimental Situations
A separate group of dogs was trained in two experimental chambers, A and B.

The chambers were similar to each other, but were not identical. The stimuli
and reactions were the same as those used in the first series. Reaction CR1 to
stimulus CS1 was trained in chamber A, and reaction CR2 to stimulus CS2
was trained in chamber 2 in fortnight series of daily sessions. The results of
this experimental series were quite similar to the results obtained in the first
experimental series conducted in one situation. That is, at the beginning of
the new fortnight series, each stimulus evoked the reaction trained in the
preceding series (CR2 instead CR1, and vice versa) and during further
experimental sessions each stimulus often evoked both reactions. In addition,
during an acute extinction of CRI or CR2 (by not reinforcing them with food),
limited to one session only, both CRI and CR2 appeared to each stimulus, CS1
or CS2.
Third Series: Two Different Experimental Situations
In this experimental series, each dog was trained in two situations different
from each other. One of these situations was a usual Pavlovian chamber
(situation A) in which the dog was trained to perform reaction CR2 (placing
the right forepaw on the platform)
Intermittent Stimulus in Conditioning 43
to a visual stimulus CS2. Another situation was a large empty pen (without
any stand or a platform) arranged on the floor of a room located in a separate
building (situation C). In this situation the dogs were trained to perform
another instrumental reaction CR3 (lifting the left foreleg in one dog, and
assuming a “begging” posture in the other dog) to a stimulus called “noise.”
Then, a new acoustic stimulus, the beating of a metronome, mCS, was

introduced to situation A and reaction CR2 was easily established to it. After
several days of training, the metronome device was transferred to situation C
and stimulus mCS was applied. The reaction of one dog was the performance
of reaction CR3 instead of reaction CR2 which was recently trained to the
same stimulus (mCS) in situation A. The same result was obtained when a
tactile stimulus (produced by a device attached to the dog’s back) was used as
a new stimulus A and then tested in situation C.
The above described experiments suggest that in the process of conditioning,

the instrumental reaction creates associations not only with the single
intermittent stimulus but also with the permanent complex of stimuli of the
environmental background against which the stimulus is applied. Moreover,
the above experiments suggest that the role of the situation may be dominant
over that of the intermittent stimulus.
This supposition is supported by additional facts obtained earlier (Wyrwicka

1952; also cited by Konorski 1967, p.409). Namely, it was found that when an
instrumental conditioned reaction is already established to one conditioned
stimulus, (e.g., a whistle), the same instrumental reaction will appear to all
other new stimuli, not only acoustic, but also visual and tactile stimuli, used
in the same situation (including the same experimental procedure) for the
first time. This fact was explained as a result of establishment of the
conditioned connections not only with the intermittent stimulus but also
with the situation as a whole. When speaking about a conditioned
instrumental reaction it should be remembered that it is evoked by a complex
of stimuli: the intermittent stimulus (CS) plus the situation, and not only by
the single CS.
In relation to the above described experiments, it would be interesting to

quote some special experiments by Soltysik (1975). In
44 Conditioning
his study, dogs were trained to perform a definite instrumental reaction to get
a food reinforcement. However, when the experimental procedure was
changed, namely the food reinforcement was given immediately after the
start of application of a definite conditioned stimulus, the dog began to eat at
once, without first performing the instrumental reaction. After training (in
which the delay in offering food was gradually extended) it was possible to
obtain a period up to 10 seconds of waiting for food, without the instrumental
reaction. In this case, the situation included a procedure different from the
standard procedure used with other conditioned stimuli; such a change in the
experimental procedure was also a change in the situation and this resulted in
obtaining a different kind of behavior. This study further supports the idea
that changes in the situation may result in changes in the conditioned
reaction.
Studies on changeability (“plasticity”) of the conditioned stimulus by Kozak et

al. (1961) and Kozak and Westerman (1966) also showed that situational
factors, such as altering the temperature in some parts of the animal’s body,
presence or absence of food, and others, could change the reaction.
The Role of the Intermittent Stimulus
Considering the obtained data, a question may be asked— what is the role of a
single stimulus in the process of instrumental conditioning? Can it retain its
acquired property of evoking the instrumental reaction when it is used in an
environment different than that in which it was originally trained? In order to
answer this question, another study was undertaken, described as follows.
In a special study (Wyrwicka 1958, 1993), four experimental series of

sessions, two of them with alimentary reinforcement (“alimentary group”)
and two others with defensive reinforcement (“defensive group”), were
conducted on a total of 14 dogs. Three to five dogs were used for each series.
The training was conducted separately for each dog of the experimental
series. First, in a usual Pavlovian chamber (uE), each dog in the alimentary
group was trained to lift the right foreleg and place it on the feeder platform
(aCR) to stimulus aCS (the sound of a flute) for food reinforcement. In the
same chamber, each dog in the defensive group was trained to lift the right
hindleg (dCR) to stimulus dCS (rhythmic sound of a whistle) to avoid an
electric shock to skin. Each dog of either alimentary or defensive group was
trained separately. The training was conducted daily with 8-10 trials per
session.
After 4-5 weeks of training, each intermittent conditioned stimulus used in

the training was tested in a different situation. In order to extinguish the
orienting reaction (such as sniffing and examining the environment) in the
new situation, the test was always preceded by several habituation sessions in
the test situation. There were four test situations.
The Neutral Test Situation
This test situation was a large empty room with no furniture except for a
small table and a chair for the experimenter. The dog was allowed to move
around freely. Neither conditioned nor unconditioned stimuli had been used
here before.
In the alimentary group of 5 dogs, first a control neutral stimulus (nS), the
sound of a rattle (never used before with these dogs) was applied for 15
seconds. This stimulus evoked only a slight orienting reaction. A few minutes
later, the alimentary test stimulus (taCS), the sound of a flute (to which the
alimentary instrumental CR was established in the uE before) was applied for
15 seconds. In all dogs, the taCS evoked only a motor reaction such as
approaching the flute and sniffing it, or coming to the experimenter and
placing the forepaw on her knees. However, the instrumental reaction
previously trained to this stimulus in the uE was not performed.
In the defensive group of dogs, first a control nS, the sound of a rattle (never
used with these dogs before), was applied tor 15 seconds, evoking only a slight
orienting reaction. Then the test CS (sound of whistle in two dogs, and
flashing light in two other dogs) to which the defensive CR was established in
the usual situation before, was applied for another 15 seconds. It was found
that the test stimulus produced general anxious movements in-
46 Conditioning
eluding shaking head (in case when electric shock to the ear was used as
reinforcement in the uE). Nevertheless, the previously trained avoidance
reaction (dCR) was not performed.
In both groups of dogs, therefore, neither alimentary nor defensive CS tested

in the new neutral situation were able to evoke the previously trained
instrumental CR, but still elicited a general motor reaction (classical
conditioned reflex) related to the type of reinforcement used previously.
The Homogeneous Test Situation I
In this series, the test situation was the same large empty room that was used
in the neutral test situation. However, the reinforcing stimuli (food for the
alimentary group, and electric shock for the defensive group), were given in
this situation previously.
In the alimentary group, the test was carried out on the five dogs previously
used in the neutral test situation. In the first two sessions each dog was
allowed to move freely and received a piece of meat every 20-30 seconds.
Food was either thrown on the floor (for 2 dogs) or was offered in a bowl
placed on a wooden beam lying along the wall of the room (for the other 3
dogs). Then during the third session, a new control stimulus, a certain type of
noise, was applied for 15 seconds, producing no distinct reaction. One minute
later, a conditioned stimulus, the sound of a flute (previously used in the
training in the usual situation), was applied for 15 seconds. It turned out that
the instrumental reaction previously trained in the uE was not performed by
either of the two dogs which obtained food on the floor; one of them came to
the experimenter and performed a “begging” movement. On the other hand,
the three other dogs that obtained food in the bowl placed on a wooden beam
did perform the aCR (placing their forepaw on the beam), previously trained
in the uE.
In four dogs of the defensive group, first the control stimulus (the sound of a
rattle) was applied for 15 seconds producing no distinct reaction. Then, one
and a half minutes later, the test conditioned stimulus (flashing light for two
dogs, and the sound of a whistle for two other dogs) was applied for 15
seconds. The avoid-
ance reaction, previously trained in the uE, was not performed by any of the
dogs, but shaking the head and anxious movements were observed. Then, an
electric shock was given and the procedure was repeated. As before, there was
not a clear reaction to the control stimulus (the rattle), but the application of
the test stimulus elicited touching the ear (to which the shock was previously
given in the uE) in one dog. only when, in addition, a band had been fastened
on his leg, did this dog perform the previously trained movement of lifting
this leg to the test stimulus. No previously trained reaction was performed by
another dog, even when the band was fastened to its leg. However, two other
dogs did perform the previously trained reaction when the test was preceded
by an electric shock.
These experiments showed that the previously used conditioned stimulus
elicited only a general defensive reaction but not the instrumental reaction
previously trained in theusual situation. Only when some facilitating
elements were present, such as a beam (for the alimentary group) or an
electric shock (for the defensive group) would the previously trained CR
appear.
The Homogeneous Test Situation II
In two dogs the alimentary reaction of lifting the right hindleg to the beating
of the metronome was trained in the usual situation (uE). On the same days,
another alimentary reaction of lifting the left foreleg in one of these dogs, and
begging movement in the other dog, to the tactile stimulus, were trained in
the test situation. When the metronome beating was applied in the test
situation, none of these dogs performed the reaction previously trained in the
usual situation, i.e., lifting the right hindleg. Instead, both dogs performed the
movement trained in the test situation (i.e., lifting the right foreleg in one dog
and begging movement in the other dog).
Two dogs of the defensive group were trained to lift their right hindleg to the
flashing light in the usual situation, and to stand on hindlegs to the sound of
a buzzer in the test situation; both movements secured the animal against the
electric shock applied to the
48 Conditioning
left ear. Then each conditioned stimulus was tested in the other situation. It
was found that neither stimulus evoked the instrumental reaction which had
been trained to it before in the uE. Instead, it elicited the reaction which had
been trained to a different stimulus in the test situation, i.e., standing on
hindlegs to the sound of a buzzer in one dog, and lifting the hindleg to the
flashing light in the other.
The experimental results obtained in the homogeneous situation II in both

alimentary and defensive groups of dogs, showed that the conditioned
stimulus, previously used in the training of a definite instrumental reaction,
does not evoke this reaction when it is applied in a different test situation.
Instead, it evokes the instrumental movement which had been trained in the
test situation to a different conditioned stimulus. In other words, the
conditioned stimulus tested in a different situation evokes the reaction
related to this new experimental situation but not to the conditioned stimulus
used in the previous training.
The Heterogeneous Test Situation
In three dogs, the basic training was conducted in two different situations,
one with alimentary reinforcement (situation A) and the other with defensive
reinforcement (situation D). Situation A was a usual Pavlovian chamber.
Situation D was arranged on the floor in an empty part of a room located in
another building; the experimental part of situation D was separated from the
observation section with a wooden partition. In both situations, the recording
bands were fastened on all four legs of the dogs.
In situation A, the sound of a metronome for two dogs and a moving black
disc for the third dog were used as alimentary conditioned stimuli; the
instrumental reaction was placing the right foreleg on the feeder platform
(aCR). In situation D, the sound of a whistle was used as a defensive CS; the
instrumental reaction was lifting the right hindleg to avoid a shock to the
same leg (dCR). The training was conducted in alternate series each lasting
for several days in each situation. Then the test was performed in which the
defensive CS was applied in the alimentary situation, and the
TABLE 5.1
Results of Training of Two Various Instrumental Conditioned Reflexes, CS1-

CR1 and CS2-CR2, in Various Situations
the CS to which it was trained in the given situation
TABLE 5.2
Testing an Instrumental Conditioned Reflex, CSl-CRlin Various Situations
alimentary CS was applied in the defensive situation. The results of this

procedure are described below.
The Application of the Defensive Conditioned Stimulus

in the Alimentary Situation
When the defensive CS was applied in the alimentary situation for the first
time, one dog performed a slight but distinct lifting ol
50 Conditioning
the right foreleg (defensive CR). Another dog reacted to the defensive CS with
a slight lifting of the right foreleg (partial aCR), followed by shivering,
bristling of hair, and anxious behavior. No clear reaction to defensive CS was
seen in the third dog. When the defensive CS was repeated in the next trial,
all three dogs performed the lifting of the right hindleg (dCR). When a control
stimulus, the sound of bubbling water (never used with these dogs before)
was applied, a slight orienting reaction only was observed in one dog, while
two other dogs performed an alimentary CR. When the control stimulus was
repeated, all three dogs performed the alimentary CR.
The Application of the Alimentary Conditioned Stimulus in the Defensive

Situation
The first application of the alimentary CS in the defensive situation did not
produce any motor reaction in two dogs, but the third dog performed 3 slight
movements of the right hindleg (i.e., the defensive movements). The
repetition of the alimentary stimulus in this situation again did not produce
any effect in the first two dogs. The third dog, however, first lifted its right
hindleg and then its right foreleg and held it in this position for several
seconds; this resembled the aCR trained in the alimentary situation.
Summarizing the results obtained in heterogeneous situations it can be said

that an alimentary CS applied in the defensive situation, and a defensive CS
applied in an alimentary situation evoked only a general conditioned reaction
connected with each stimulus or did not evoke any reaction at all; only in a
few cases an adequate instrumental reaction (or one similar to it) was
observed. (See Table 5.1 and Table 5.2).
Examples of Complex Conditioned Reflexes to the Situation
Although the famous experiments of Pavlov’s laboratory were based strictly

on a single external stimulus and a single conditioned reaction, other
investigators used complex stimuli and complex reactions in their studies on
conditioning and learning.
The most eminent cases of complex conditioning were provided by Thorndike

(1911). As already mentioned in chapter 2, Thorndike conducted his
experiments on various animals such as fish, chicks, cats, dogs, and monkeys.
His studies provided many examples of complex stimuli as well as complex
motor behaviors. In spite of the fact that Thorndike called the topic of his
experiments “trial-and-error learning,” the behaviors he studied can be
regarded as cases of the complex instrumental conditioning.
Detour Behavior
More recent studies of complex instrumental conditioning include a topic

such as detour behavior. A “detour” or a roundabout behavior,” the ability to
go around an obstacle on the way to a reward, was considered a result of 4
insight by Kohler, author of the famous book The Mentality of Apes (Kohler
1925). A similar view was expressed by other investigators who conducted
theii observations on rats (Higginnson 1926, Hsiao 1929, Tolman and Honzik
1930); birds (Teyrovski 1930, Lorenz 1932, 1939); reptiles (Fisher 1933); and
fish (von Schiller 1942). However, Thorpe (1956) emphasized that before
accepting an explanation of detour
51
52 Conditioning
behavior as a result of insight, one must first determine whether this

behavior appears in an untrained animal as its first response.
The problem of detour behavior was also studied by Beritashvili and his
associates (Beritashvili 1941, Beritashvili and Akhmeteli 1941) who conducted
their observations on dogs, rabbits, hens, and pigeons. The results of their
studies led them to a conclusion that the animal performs the detour
behavior due to a psycho-neural process which reproduces a picture of the
external world in the animal’s mind thus guiding its movements towards the
food. Later, Roginski and Tikh (1956) studied detour behavior on chickens,
rats, and monkeys, and found that monkeys learned to make the detour
fastest, and chickens most slowly; they concluded, therefore, that the detour
reaction depends on the phylogenetic level of the animal. Pavlov (1949),
however, expressed the opinion that the detour behavior has an acquired,
conditioned reflex characteristic.
In order to gather more data concerning the mechanisms involved in the

performance of detour behavior, systematic studies were conducted by
Wyrwicka (1959).
In the preliminary testing period of this study it was found that adult dogs
easily made a detour when food was placed behind a visible obstacle. A
question arose, therefore, as to whether this reaction was inborn or acquired
by experience. In order to answer this question, a series of experiments was
performed on puppies that had been kept in a controlled environment from
birth and never had any occasion to make a detour before.
The study was conducted on 28 puppies, male and female, 12-131 day old, of 8
different litters from 5 female and 8 male dogs. Mother and puppies of each
litter were housed in a large pen, 2x3.5 m at the bottom and about 1.5 m in
height. Their home compartment was empty and the pups had no occasion to
go around any object. They were first nursed by the mother, and later they
were fed solid food (mixture of cooked meat and cereal) with the mother.
The tests were carried out separately for each pup, in a large, empty room
(different from the pen). The experimental space was partially divided by a
wire-net partition as schematically shown in fig. 6.1.
Examples of Complex Conditioned Reflexes to the Situation 53
FIGURE 6.1
Samples of detour behavior of puppies (1-4 months old) that were controlled
from birth and never had occasion to make the detour before. (Photocopies of
sketches of detour behavior of single puppies, made by the experimenter
during the test session).
Straight line, wood partition; wavy line, wire-net partition; square at the
bottom of each sketch, starting place; empty circle at the front of wire-net,
site of the bowl and milk during preparatory sessions; circle with a cross, site
of the bowl behind the wire-net partition during the test session; the broken
line with small arrows traces of detour behavior.
In the preliminary series of 5 sessions (each lasting 15 minutes), each of 12

puppies of the first group was allowed to walk in both parts of the
experimental compartment (i.e., before and behind of the partition); 11 other
pups of the second group were allowed to remain only in front of the
partition. The third group of 5 pups was trained in a space arranged in the
garden with a wire-
54 Conditioning
net partition (similar to that used in the room) where they were allowed to
run both in front of and behind the wire-net.
Then the testing series of experimental sessions was performed, separately

for each pup. At the beginning of each session, the pup was offered food in a
bowl placed in front of the wire-net partition. After approximately 20 daily
sessions, the bowl with food was placed behind the partition, just across from
the previous location of the bowl. The food was clearly visible to the pup, but
it could not be reached from the front of the partition. In order to get the
food, the pup had to go first to the end of the partition (i.e., to go away from
the site where the food was placed behind the fence) and then make a U-turn
toward the food (see sketches in fig. 6.1).
It turned out that most of the 12 pups of the group that had been allowed to
walk in both parts of the experimental compartment before the test easily
found the way to the food in seconds; only a few pups in this group needed
more time to make a detour. Similar results were obtained on 5 pups that
were allowed to run both in front of and behind the partition in their
preliminary training in the garden.
On the other hand, none of the 11 pups that had been allowed to remain only
in front of the partition before the test could find the way to the food.
Independent of age, they only repetitively scratched the partition at the site
where the food was visible behind the wire-netting. This behavior was
observed for several consecutive sessions until finally, earlier or later, each
pup started to walk back and forth alongside the net, and accidentally found
the passage to the bowl (see a schematic pathway of a pup in fig. 6.1). On an
average, 4-6 sessions were needed for each pup to learn the detour. It is
interesting that age did not influence the response. When the test was
conducted on 4-month-old pups, their learning of the detour took even longer
than learning in the 2-month-old pups.
Once the pup learned to make a detour in a room, it was able to make it also
in a different surrounding, namely in the garden with the use of the same (or
similar) partition and the same procedure. This means that the situational
stimuli were recognized by the pup as the same as those used in the room
before, and thus, the detour reaction was performed.
In order to gather more information concerning the nature of this reaction,

an extinction test was performed separately on each of 3 pups. After the
establishment of the detour behavior, the experimental sessions were
continued but the food was withheld , i.e., the bowl behind the partition was
empty. There were 3 trials per session. The pups first continued to run to the
bowl behind the partition, but after 10 sessions in 2 pups and 30 sessions in
the third pup, they stopped going around to the bowl. After the extinction of
the detour reaction in the room, this behavior was also absent in the garden.
It was concluded, therefore, that the detour behavior is not a result of

“insight” Neither is it an inborn unconditioned reaction. Instead it is a
behavior, independent of age, acquired through individual experience, i.e., a
complex instrumental conditioned reflex.
In this case, both the stimulus amid the reaction are complex. The stimulus is
composed of permanent situational elements such as visual stimuli of the
wire-net partition, the bowl, and the passage. The motor reaction involves
both excitatory and inhibitory processes. First, the initial motor reaction to
the bowl (scratching the partition in front of the animal in an attempt to
reach the food behind the partition) had to be fully extinguished, i.e.,
inhibited by not obtaining the reinforcement (food). Then only the correct
motor reaction, a U-turn, rewarded by food, can gradually develop
(fig6 1 )- . u u ,
Of course, there are more instances of behavior where both stimuli and
reactions are even more complex. The manipulatory reactions of animals of
various species described by Thorndike (1911) and chain reflexes analyzed by
Skinner (Skinner 1938, Ferster and Skinner 1957) may serve as examples of
complex conditioned behavior. The directional instrumental reactions studied
by Lawicka (1959) and Konorski and Lawicka (1959) also belong
to this category.
Food Reinforcement as a Conditioned Stimulus
Let us try to analyze the meaning of food as an important reinforcing

stimulus in the conditioning process. Usually, little atten
56 Conditioning
tion is paid to food which is so commonly used in a variety of conditioning

techniques. The attention of the experimenter is usually concentrated on the
animal’s activity preceding the food reinforcement.
However, some studies in our laboratory showed that the act of eating can be
considered an instrumental conditioned reflex. In experiments of Wyrwicka
and Chase (1972), cats with electrodes implanted in the lateral hypothalamus
were offered a milk and broth mixture in two identical containers. Drinking
from one container was rewarded by desirable electrical stimulation in the
hypothalamus (the desirability of obtaining the hypothalamic stimulation
was first tested by the method of self-stimulation arranged so that the cat
pressed a lever to get that stimulation). Drinking from another container was
not reinforced by pleasurable hypothalamic stimulation. It appeared that cats
concentrated on drinking milk-broth mixture only from the container with
stimulation and ignored the same mixture in the other container. In other
experiments, cats learned to eat slices of bananas or tasteless and odorless
jellied agar when this was rewarded by hypothalamic stimulation (Wyrwicka
1978, 1981, 1988).
Feeding behavior, therefore, should be considered a complex conditioned act.

It is usually preceded by a number of classical and instrumental conditioned
reactions which prepare the animal for obtaining food. The classical type
reactions include secretion not only from salivary glands but also from
various internal organs (see Bykov 1957) as well as changes in respiration and
circulation (Gantt 1960, Gantt and Hoffmann 1940) and some unconditioned
motor activities such as orienting and investigatory reactions. Instrumental
preparatory behavior depends both on the previous experience with food and
on the present circumstances. In the laboratory, the preparatory instrumental
reaction is simply a trained movement, easy to observe and record. In nature,
the preparatory activity may be more complex, such as seeking food, hunting
or fighting for food. In various circumstances, this behavior may start a long
time before eating. For instance, in human life, the preparatory feeding
behavior may include all activities related to obtaining food such as shopping
for food, preparation of a meal, etc.
FIGURE 6.2
PREFEEDING ► ACT OF EATING
Instrumen. CRs Complex or Oral and

(preparation of meal or Other Instrumen. CRs
going to restaurant, etc.) (masticatory and other CRs
related to the kind of food)
A self-explanatory diagram of activities occurring before and during the act of

eating.
Let us now analyze the instrumental type behavior related to alimentary

behavior. This behavior is composed of two parts as shown in a diagram in fig.
6.2. the first part of this diagram shows the preparatory instrumental type
activity preceding eating such as performance of the previously trained
movement (in the laboratory) or a complex of motor reactions such as
seeking food or hunting, etc. (in nature).
The second part of the feeding behavior is the act of eating itself. A closer look
at this act shows that it is quite complex and dependent on the kind of food to
be consumed. In mammals, the first feeding behavior is sucking. The fact that
it appears early after birth indicates that sucking behavior is inborn (although
it is not excluded that it can occur even before birth in the womb by sucking
the surrounding fluids). Later in life, the young start to follow the mother or
other adults of their colony in seeking food and eating that food (Galef 1977,
1978; Galef and Clark 1972). This suggests that the motor ability to eat
various kinds of food (other than mother’s milk) is acquired by experience
later in life. This ability is based on a number of inborn reflexes, such as
mastication, i.e., the movements of opening (digastric reflex) and closing
(masseteric reflex) the mouth (Clemente et al. 1966; Chase and McGinty 1970
a,b; Chase and Babb 1973) as well as the movements of the tongue and other
parts of the mouth cavity. The motor performance may vary depending on the
kind of food to be eaten. The motor (instrumental) reaction of eating a piece
of meat is different than the motor reaction of eating creamy pudding or
drinking milk. Due to the repetitive experience with various kinds of food, the
act of eating gradually becomes more and more com-
58 Conditioning
plex. Even later in life some further adjustments in the act of eating may take
place when the unknown kinds of food are being offered. Therefore, the food,
which is commonly called an “unconditioned stimulus,” turns out to be a
specific conditioned stimulus , and the act of eating becomes a conditioned
instrumental reaction. In other words, the alimentary instrumental
conditioned reaction is reinforced not by food as an unconditioned stimulus
but by food as a conditioned stimulus when it is seen or sniffed by the
subject. In this case, only the final effect of eating, the sensory satisfaction
resulting from food consumption, can be considered the “reward” or the
“reinforcement” (Wyrwicka 1975).
A question can be asked whether or not the act of eating can depend on the
visual or olfactory stimuli related to a particular portion of food,
independently of the situation where this food has been offered. It seems that
environmental factors may still influence the reaction of eating. One of these
factors is the state of satiation. In this case, the act of eating may not occur. In
another case, the unsatisfactory environment such as a dirty tablecloth or the
presence of a constant loud noise, etc., may inhibit the consumption. There
are many such factors restraining eating.
On the other hand, there are also factors that facilitate eating. An example is
the already quoted study by Wyrwicka and Chase (1972) in which cats were
offered food in two containers; eating from one of these containers was
rewarded by desirable hypothalamic stimulation, while eating from the other
container was not. The cats chose to eat only from the container where they
obtained the pleasant stimulation.
Experiments with completely satiated goats which refused to eat any more
can serve as another example. These animals, however, started to eat again
when they obtained a pleasant light electrical stimulation in the “feeding
center” in the lateral hypothalamus each time they ate, even when this was
leading to overeating (Wyrwicka et al. 1959, 1960, 1982, 1988).
Presleep Behavior
A case of conditioning to the situation was observed in cats in
relation to sleep. As it was demonstrated by Sterman and Clemente (1962),

Clemente et al. (1963), Yamaguchi et al. (1963), Clemente (1968), McGinty
and Sterman 1968, and others, electrical stimulation of the basal forebrain
area (BFA) produced sleep accompanied by specific changes in the
electroencephalic activity (EEG) of the brain. In a more recent study by
Wyrwicka and Chase (1994), experiments were conducted on three chronic,
unanesthetized, undrugged cats, bearing monopolar electrodes implanted in
the basal forebrain area. Light electrical stimulation (2-4 V, 50-100 Hz, 1 ms
dur/imp, 0.5 s trains at 1/s frequency, for 2-5 minutes each time, followed by
a nonstimulation period of the same duration) resulted in producing
synchronized EEG spindles which were accompanied by the cat assuming the
posture typical for presleep behavior such as crouching with the head down,
and the eyes partly or fully closed. After several sessions with the use of BFA
stimulation, two of the cats started to demonstrate presleep behavior almost
immediately after entering the experimental compartment, ignoring the food
(which was always present in the experimental compartment). The third cat
usually ate some food and afterwards began to show presleep behavior. When
stimulation was withheld in the extinction procedure, the cats still continued
to exhibit the presleep behavior in the absence of the stimulation during
several consecutive sessions. Thereafter their usual behavior (of the pre-
stimulation period) gradually returned, including consumption of the offered
food. The authors concluded that repeated BFA stimulation produced
conditioning of the presleep behavior to the complex of environmental
stimuli of the situation.
Conditioning of the Activity of the
Internal Organs
Intensive studies were undertaken by N.E. Miller and his associates on the
possibility of producing instrumental conditioning related to internal organs
innervated by the autonomic nervous system. Miller and DiCara (1967)
demonstrated that rats immobilized by an injection of curara and artificially
respirated learned either to increase or decrease their heart rate (dependent
of the experimental design) to obtain the rewarding electrical stimulation in
the basal forebrain bundle (cf. Olds and Milner 1954; Olds 1962). In other
experiments, the curarized rats learned to increase their heart rate (or
decrease it, dependent on the chosen experimental procedure) to avoid an
electric shock to the skin (DiCara and Miller 1968a). Similar results were
reported by Throwhill (1967).
Using the same method, it was possible to change the vasomotor activity, i.e.,
to obtain either vasoconstriction or vasodilation, when the required activity
was reinforced by desired electrical stimulation in the basal forebrain bundle
(DiCara and Miller 1968b). Another study showed that thirsty, noncurarized
dogs learned to increase salivation when rewarded with access to water; it was
observed that salivation in these dogs increased with the training (Miller and
Carmona 1967). It was also possible to increase the rate of urine formation in
one group and decrease it in another group of curarized rats when electrical
stimulation in the basal forebrain bundle was offered as a reward (Miller and
DiCara (1968).
A question arose as to whether the obtained instrumental autonomic

responses can be entirely independent of the central ner-
61
62 Conditioning
vous system. In order to answer this question, a high dose of curara (4.8.
mg/kg over 3 hours) was used, resulting in full stoppage of muscular activity;
in spite of complete immobilization of the animal, rats still learned to change
their heart rate (DiCara and Miller 1968c).
However, different results were obtained in experiments on dogs by Black

(1967). This investigator studied the instrumental responses of the heart
rewarded by withholding of shock, under various degrees of curarization.
When a higher degree of curarization was used (4.1 mg/kg over 8 hours),
resulting in a complete suppression of muscular activity, the possibility of
obtaining instrumental conditioning of the heart rate strongly diminished.
These results led to a conclusion that instrumental conditioning of heart rate
may be mediated by some central processes which are in charge of motor
activities (Black 1967).
Instrumental Conditioning of the Electrical Activity of the Brain
Several studies have shown that electrical activity of the brain (EEG) can be
conditioned, i.e., it becomes an instrumental conditioned reaction. Attempts
of conditioning of the electroencephalic function of the brain were made by
various investigators. One such study was that of Olds and Olds (1961), who
recorded unitary discharges from paleocortical structures of the rat’s brain.
Each time the number of discharges from the unit increased, the rewarding
electrical stimulation of the medial forebrain bundle was applied. As a result,
the number of discharges became considerably greater.
In another group of experiments conducted on cats, Izquierdo et al. (1965)

studied the possibility of establishing a conditioned EEG rhythm to a tone
applied while the animal was sleeping. The tone was repeatedly applied until
EEG arousal was no longer elicited by the tone, i.e., until full habituation had
been achieved. Next, the same tone was applied for 4 seconds; two seconds
after its discontinuation, a weak electric shock to the cat’s skin was applied.
After a number of such trials the tone alone began to evoke
Conditioning of the Activity of the Internal Organs 63
EEG arousal. Each time the EEG reaction occurred to the tone, the shock was
withheld. After a number of repetitions of this procedure, EEG arousal began
to appear in each trial while the cat remained lying down and apparently
asleep as shown by electromyographic recording (EMG) from the animal’s
neck muscles. In this experiment desynchronization of the EEG pattern
became an instrumental avoidance reaction.
The problem of conditioning of EEG waves was also studied by Carmona in

the laboratory of Miller (Miller 1969). In one group of cats, the appearance of
high amplitude slow EEG was each time rewarded with the electrical
stimulation of the medial forebrain bundle, whereas in another group a low
amplitude EEG activity was rewarded in the same way. As a result, the cats of
each group learned to produce more of the particular rewarded rhythm.
Similar studies were also conducted on human subjects by Kamiya

(1962,1968). In his experiments, appearance of the alpha rhythm from
occipital cortex leads was each time reinforced by a small monetary reward.
This procedure resulted in a significant increase of the alpha rhythm. Other
studies include those of Nowliss and Kamiya (1970) as well as experiments of
Rosenfeld et al. 1969) and Brown (1970).
An attempt was also undertaken to condition, in cats, the alpha-rhythm (high-

amplitude waves, 8-12 cycles per second) appearing in parieto- occipital
cortical leads. Each time such a burst appeared, a small food reward (piece of
meat) was given to the cat. After a number of such trials alpha activity began
to appear each time the tone was applied (Wyrwicka 1964).
These first observations on conditioning of EEG activity should be considered

as preliminary results of studies undertaken later with the use of precise
techniques which included both better experimental control and detailed
analysis of the results. Namely, in a study conducted on cats, an EEG activity
called “sensorimotor rhythm” (SMR) of high amplitude waves of 12-14 cycles
per second from the coronal gyrus of the cortex (Sterman and Wyrwicka 1967,
Wyrwicka and Sterman 1968, Sterman et al. 1969) was used. Each time a
burst of SMR appeared an automatic switch delivered a small portion of milk
to the cat. After a number of trials, the
64 Conditioning
SMR burst appeared more often and at regular intervals (Wyrwicka and
Sterman 1968). When a milk reward was withheld during the extinction
procedure, the frequency of the SMR bursts temporarily increased at the
beginning of extinction. The finding of a temporary increase of the
instrumental reaction, which is usually observed at the beginning of the
extinction process (see Konorski 1948,p. 217; 1967, p. 361), supports the view
that the sensorimotor rhythm became conditioned as an instrumental type
reaction.
Further research focused on precise localization of the brain region where the
sensorimotor rhythm is initiated. Research of Sterman and Clemente (1962),
Clemente and Sterman (1963, 1967), Clemente et al. (1963), Sterman and
Wyrwicka (1967), Sterman et al. 1969) revealed that this region is situated in
the basal forebrain area (BFA). According to the stereotaxic atlas of the cat’s
brain of Jasper and Ajmone- Marsan (1960), the coordinates for this area are
A 13.0-15.0, L 2.5-3.0, H—4.0, although some differences from cat to cat may
exist. Bilateral lesions in the basal forebrain area in cats resulted in
suppression of sleep (McGinty and Sterman 1968).
Electrical stimulation of the basal forebrain area leads to the production of

presleep behavior such as cessation of eating and assuming a sitting or lying
position with the head lowered and eyes half or fully closed. Usually one or
two minutes later, spindle bursts of sensorimotor activity (SMR) appear on
the EEG recording. Both the behavioral and EEG responses resemble those
observed during natural presleep behavior.
Repeated stimulation of the BFA leads to conditioning of the SMR as shown

in studies of Wyrwicka et al. (1962), and Clemente et al. (1963). In
experiments on cats, the BFA stimulation period was each time preceded by a
tone (Tl). After several repetitions of this procedure, Tl alone started to evoke
the spindle bursts of SMR. A control stimulation in the thalamus preceded by
another tone (T2) in the same animal, did not evoke SMR; instead, a high
frequency EEG pattern was observed. Repetitive stimulation of this thalamic
area led to an appearance of high frequency EEG activity to T2, without any
appearance of SMR. These ex-
Conditioning of the Activity of the Internal Organs 65
periments further supported the observation that electrically elicited EEG

bursts typical of presleep stage can be conditioned.
Research conducted on human subjects suffering from insomnia showed that

presleep behavior can also be conditioned. In a study of Poser et al. (1965),
beats of a metronome were paired with a sleep- inducing barbiturate
(methohexitone) once a day, for 16 days. Toward the end of this period, the
subjects reported relaxation and soporific tendencies during metronome
beats.
In a treatment used by Evans and Bond (1969), administration of a sleep-

inducing barbiturate (methohexital sodium) was paired with self- produced
counting to 28 as a conditioned stimulus. After eight weekly sessions with
this procedure, an increase in the amount of sleep, from an initial 2 hours to a
final 6 hours a night, was observed.
In a more recent study on human subjects by Caruso et al. (1988), classical

conditioning of sleep onset was produced using a sleep- promoting hypnotic
drug, triazolam, that served as an unconditioned stimulus. It was paired with
pill ingestion, and a tone, together with the environmental stimuli, served as
the conditioned stimuli. When, under the same conditions, lactose (a
placebo) was substituted for triazolam, sleep latency remained shorter than
the previously determined baseline.
Earlier studies on animals revealed the presence of a EEG slow-wave pattern

from the sensorimotor cortex during behavioral inactivity or trained
inhibition (Gastaut 1958, Kogan 1960, Anokhin 1961, John etal. 1961,
Rowland 1961, Donhoffer and Lissak 1962, Roth et al. 1967, Sterman and
Wyrwicka 1967, Wyrwicka and Sterman 1968). A systematic study on
somatomotor suppression and excitation by stimulation of the orbital gyrus
in unanesthetized freely moving cats was conducted by Chase and McGinty
(1970 a, b). They found that orbital cortical stimulation resulted in the
suppression of the masseteric reflex and the digastric muscle contraction;
however, a gradual reduction of the effectiveness of the cortical stimulation
was observed as the animal passed from the alert state into quiet sleep.
In experiments by Babb and Chase (1974), cats were trained to produce the
sensorimotor rhythm (SMR) while the masseteric and
66 Conditioning
digastric reflexes were monitored throughout the session. It was observed

that during SMR small amplitude masseteric reflexes were depressed,
whereas large amplitude masseteric reflexes were not. According to the
authors’ interpretation, the suppression of the masseteric reflex is related to a
decrease of afferent excitatory input produced by reduced cortical discharge;
on the other hand, strong reflex evocation would compensate for the reduced
input, masking the inhibitory effect.
In a special study of Chase and Harper (1971), cats were trained to produce a
synchronized EEG pattern of 12-14 cps of sensorimotor activity to get a milk
reward. During the presence of SMR, somatic motor activity was inhibited; at
the same time, heart rate decreased, and the respiratory pattern became more
regular than during control periods. In a later study, neurons in the thalamus
were found to discharge in a “burst- pause” manner during SMR, similar to
that found in quiet sleep (Harper 1973, Harper and Sterman 1972). These data
suggest that the sensorimotor rhythm is related to motor inhibition.
In all experiments described in this chapter a process of instrumental

conditioning took place. In each case the subject was required to produce a
definite reaction such as an increase or a decrease in heart rate, obtain
vasoconstriction or vasodilation, increase or decrease the amount of
salivation or urine formation, etc. In some other experiments an increase of
neuronal discharges or an appearance of the EEG waves of a definite
frequency was a conditioned reaction. Only after the performance of the
required task was the reward given. In each case, the process of instrumental
conditioning was performed on the internal organs of the body. In other
words, this process took place in the internal environment of the body. This
suggests that there exists not only an external situation in which the
conditioning usually occurs, but also an internal situation which plays an
important role in the life of the organism. This problem will be further
discussed in the following chapter.
Therapeutic Role of the EEG Feedback in Epilepsy
The elicitation of specific changes in functions of various internal organs such

as heart rate, increase or decrease of salivation, and others described in the
previous chapter, suggest that these changes have been acquired by
conditioning. For instance, according to observations of Miller and Carmona
(1967) thirsty dogs increased salivation with the training. Similarly, heart rate
increased when the animal was rewarded by desired electrical stimulation in
the basal forebrain bundle (DiCara and Miller 1968). These and other
examples of changes in functions of the internal organs suggest that these
events take place accordingly with the basic mechanisms of instrumental
conditioning.
A question can be asked as to what instrumental method has to be used by

the organism to achieve the required functional change. For instance, an
increase in respiration can result in an increase of heart rate; similarly, the
movements of the tongue or the jaws may increase salivation. In these cases
the required instrumental reaction has been performed with the use of
another reaction (respiration, movements of jaws, etc.).
However, these instrumental reactions occurred also in immobilized

(curarized) animals. In such cases, the motor activity of one system could not
be used to produce the reaction in anothei system. In spite of this, the animal
somehow finds the means leading to the performance of the required
behavior.
Especially intriguing is the achievement of a required pattern of electrical

activity of the brain. Some such cases as described
67
68 Conditioning
in the previous chapter demonstrate that with the use of a specific reward, it
is possible to obtain the required EEG pattern both in animal and human
subjects (Olds and Olds 1961, Miller 1969 / experiments of Carmona/, Kamiya
1962, Wyrwicka 1964, and others).
The development of studies on conditioning of electrical activity of the brain

led to the question as to whether it would be possible to use some of the
evoked EEG patterns in a therapy of certain neural disorders. One such
disorder is epilepsy.
Treatment of Epilepsy with the Use of Sensorimotor Training
Epilepsy can be described as a state of disregulation of the function of the

motor system, characterized by convulsions (seizures), often with a
temporary loss of consciousness. The EEG recording in epileptics showed a
completely disordered picture of the electrical activity, which was replaced by
slow regular waves. These slow waves were seen in epileptics only in the
absence of seizures and were observed mostly in the sensorimotor cortical
area. These waves were described in the previous chapter as a “sensorimotor
rhythm” being related to motor inhibition.
The finding of the inhibitory character of the sensorimotor rhythm led to the
question of whether it would be possible to use the SMR as a natural
treatment in conditions involving disregulation of the motor system in
epilepsy in humans.
In order to answer this question, the following study was conducted on an

epileptic by Sterman and Friar (1972). The subject was a 21-year old female
with a history of convulsive disorder which began when she was 16 years old.
The onset was insidious and consisted of nocturnal, generalized major
seizures. There was no family history of epilepsy. Neurologic examinations
failed to demonstrate any localized lesions. The EEG recordings taken three
years before and just prior to the study showed a generalized spike-wave
activity in the left fronto-parietal area of the brain. The seizures were
preceded by a non-specific aura, but there were no localizing features of any
precipitating factors. Daytime inci-
dents were limited only to some small movements such as wrinkling of the
brow associated with the left lateral deviation of the eyes, crossing of the right
arm to the left knee, and falling to the left. The majority of seizure incidents
were currently nocturnal, occurring in the later portion of a night s sleep.
The frequency of incidents varied from 2 to 4 per month at the onset to 1 per
3 months one year later. During the 12-month period immediately preceding
the treatment, seizures were present irregularly at an average rate of about 2
per month. The subject had variable success in using a combination of drugs
such as Dilantin, Mysoline, Peganone, Dianox, and Ritalin. In the preceding
12 months she also daily used Dilantin and Mebarol.
The training techniques included the EEG recording from the scalp
sensorimotor and occipital areas, using needle electrodes. Placement of
electrodes was standardized by reference to the international (10-20) system.
The electrocardiogram (EKG) and electromyogram (EMG) were also
recorded. During the sessions the patient was sitting comfortably on a
reclining chair. The instructions were delivered through standard tape
recording. The device recording the SMR activity (feedback) consisted of a
rectangular unit suspended from the ceiling several feet above and in front of
the patient. The unit contained two rows of ten small lamps covered with
transparent colored Plexiglas. When the EEG activity of appropriate
amplitude and duration appeared, the lamps in the top row were successively
lighted; each advance was accompanied by the sounding of a single chime
which was used as a reward. With the eleventh successful EEG response the
lights of the top row were extinguished and the lamps of the bottom row were
activated; each time this was accompanied by a double chime reward. Usually,
200 to 300 rewards were obtained during a single session lasting from 30
minutes to one hour. The sessions were conducted once a week during the
first month and then twice during the following three months.
The results of this treatment were dramatic. Initially, the patient showed a
moderately low level of SMR feedback. Later, however, after adjustment of
the reward from 12-14 cycles/sec (c/sec) to 11-13 c/sec, a gradual increase in
the production of SMR feed-
70 Conditioning
back was observed. After three sessions there were no seizures for the
following period of 3 months, except one mild nocturnal seizure. At the same
time, some changes occurred in the personality of this patient. Being
previously a quiet and rather timid, non-assertive individual, the patient
progressively became more outgoing, with personal, confidence and even an
enhanced interest in her appearance. She also reported a shorted latency to
sleep onset, and a more rapid awakening in the morning. Moreover, this
patient voluntarily ceased her previous practice of supplementing
anticonvulsant medication while experiencing an aura, without seizure, and
reported a reduction in such occurrences. In a later phase of the study she
obtained an additional small monetary reward for producing SMR; this
resulted in the highest rate of performance she had ever showed.
A reduction of epileptic seizures following the training of 12-15 c/sec SMR

feedback was confirmed in further studies of Sterman and his colleagues
(Sterman et al. 1974, Sterman and MacDonald 1978, Sterman and Shouse
1980) as well as in the studies of Finley et al. (1975), and Seifert and Lubar
(1975). However, several other authors who had confirmed seizure reduction
challenged the concept of a specifically therapeutic 12-15 c/sec EEG rhythm.
Some investigators reported that the enhancement of this EEG pattern was
not always accompanied by a decrease in seizures (Kaplan 1975, Wyler et al.
1976, Kuhlman and Allison 1977, Kuhlman 1978, Cottetal. 1979).
In order to obtain more data, the following study was undertaken on 8 poorly
controlled seizure patients by Sterman and MacDonald (1978). Four of these
patients were rewarded alternately for facilitation and suppression of 12-15
c/sec SMR activity. All these patients showed a seizure reduction only during
positive reward tor the 12—15 c/sec pattern. Four other patients were
rewarded for the facilitation and then suppression of 18-23 c/sec rhythm.
Three patients of this group showed significant seizure reduction.
A further study by Sterman and Shouse (1980) was conducted on another

group of 8 epileptic patients who were not adequately controlled by
anticonvulsant medications. The patients ranged in
age from 18 to 35 years, and their history included diagnosed epilepsy from 8
to 24 years. The techniques used for this group included not only the training
with the use of various frequencies of waves but also sleep EEG spectra,
clinical EEG and anticonvulsant blood levels. The training was conducted
mainly in the patient’s home and lasted 3 months.
The investigators found a compliance with training instructions and an

acquisition of the responses. Overall anticonvulsant blood levels were low
and unrelated to the EEG or seizure changes. The clinical EEG corresponded
to sleep EEG and seizure rate outcomes. Power spectral analysis of samples of
“non-rapid eye movement (non-REM) sleep from all-night EEG recordings in
the first subgroup of patients corresponded to the pattern of seizure rate
changes in this group. EEG changes were also limited to sensorimotor cortex
in the second subgroup, but were linear, and paralleled a progressive decrease
in seizure rate. Both groups showed the same pattern of EEG changes with
seizure reductions. Low and high frequency waves were reduced and
intermediate frequencies were increased. The correlation analysis confirmed
this relationship. The pattern of these changes suggested a normalization of
sensorimotor EEG substrates related to the EEG feedback training.
The results of studies described above, confirmed by other investigators
(Finley et al. 1975, Kaplan 1975, Wyler et al. 1976, Ellertsen and Klove 1976,
Kuhlman 1978, Kuhlman and Kaplan 1979, Wyler et al. 1979). Cott et al.
(1979), emphasized the dependence of the results on the procedure while
Lubar et al. (1981) stressed that the severity of epilepsy symptoms may
influence the EEG feedback therapy.
A study by Lantz and Sterman (1992) confirmed the observation of Lubar et

al. that the effect of feedback training was indeed dependent on the severity
of symptoms which could include a variety of cognitive, motor, and
psychosocial functions. An improvement in these functions could occur only
in subjects who were successful in learning the trained response and reducing
seizures.
EEG feedback activity is a unique case of conditioning which occurs entirely

inside of the internal situation. This kind of conditioning is produced by an
appearance in the brain of a complex
72 Conditioning
pattern of motor excitation produced by a specific physiological state which

normally leads to epileptic seizures. The occurrence of this pattern of
excitation is a conditioned stimulus which evokes a specific conditioned
reaction. The conditioned reaction in this case consists in low-frequency, high
amplitude EEG waves (i.e., the sensorimotor rhythm, SMR). The appearance
of this rhythm indicates that the state of motor inhibition has taken place
leading to the disappearance or prevention of the seizures.
The acquisition of this reflex which should be considered a defensive

instrumental reaction fully depends on the subject. The patient must learn to
recognize a change in sensations preceding the occurrence of epileptic
seizures and immediately initiate an inhibitory action (by producing SMR)
thus preventing the attack. If the patient acquires this reaction by practice,
the seizures may be avoided each time. The clinical records described above
show that such control of epilepsy is possible to achieve.
Theoretical Comments
According to experimental results, the relationship between the intermittent

stimulus (CS) and the situation in which it is applied can be described as
follows.
When two instrumental reactions to two different intermittent conditioned

stimuli (one acoustic and the other visual), respectively, were trained in one
situation , each CS often elicited the reaction related to the other CS, or both
instrumental reactions. The same results were obtained when the training
was conducted in two similar situations. However, when one of these
reactions was trained in one situation and the other in another situation,
completely different from the other one, only one instrumental reaction (CR)
was obtained in each situation (that which was trained in this situation) (see
the summary in Table 5.1).
When each intermittent CS was separately applied in various situations, the

following results were obtained.
An application of the CS in the neutral situation (in which neither an

unconditioned [US] nor a conditioned stimulus was even applied) did not
evoke a specific CR, but only a general reaction which contained some
elements of alimentary or defensive reactions connected with this stimulus in
the previous situation.
An application of the CS in the homogeneous situation (in which no

instrumental CR was ever trained but the same unconditioned stimulus, food
or shock, was used in the previous situation), evoked general alimentary or
defensive behaviors. In some cases a truncated or delayed instrumental
reaction was performed, especially when some element of a previous
situation, such as a beam resembling the platform, was present, or when an
unconditioned stimulus (electric shock) was given.
73
74 Conditioning
An application of the CS in another homogeneous situation (in which an

other instrumental reaction had been previously trained to a different CS)
evoked this other instrumental reaction.
An application of the CS in the heterogeneous situation in which an

alimentary CS was tested in a defensive situation (where a defensive reflex
was previously trained), and vice versa, a defensive CS was tested in an
alimentary situation (where an alimentary reflex was previously trained),
evoked a general conditioned reaction, and only in a few cases (especially
when the test stimulus was repeated), a correct instrumental reaction
appeared to the tested stimulus.
These results are summarized in Table 5.2.
The experimental data show that a single conditioned stimulus produces the
definite instrumental CR only in the situation in which it was trained. Once
the situation has been changed, the conditioned stimulus is not able to
produce the correct instrumental conditioned reaction. Sometimes, a
diminished and modified reaction may be present, hardly resembling the
correct CR.
On the other hand, the conditioned stimulus tested in various situations

always elicited general activity with some features related to the
unconditioned stimulus previously used in the training. In the case of a
defensive CS, it was a general defensive reaction, e.g., crouching, shaking the
head, touching the ear (with paw) where the electrodes were attached before,
or escaping; in the case of alimentary CS, the dog came up to the
experimenter and assumed a begging posture. The correct instrumental CR
previously established to the test CS appeared mostly in the presence of fa-
cilitatory components related to the training.
The performance of the general behavior related to the previously used

procedures, indicates that the intermittent CS retains its classical (Pavlovian)
connections with the unconditioned stimulus, even when it is tested in a
different situation. However, it does not retain the instrumental-type
connections with the CS and UCS when it is applied in a different situation ,
A question can be asked, therefore, about the role of the environmental

situation in the process of conditioning. The previous observations made by
Konorski and Miller (1933, p. 114; Konorski
Theoretical Comments 75
1967, pp. 359-360) showed that it is possible to establish an instrumental

reaction to the situation alone. Later, conditioning to the situation alone was
used by other investigators (Wyrwicka et al. 1959,1960, cit. by Konorski 1967,
p. 405). A more recent study by Wyrwicka and Chase (1994) on the
conditioning of the presleep state to the experimental situation in cats is a
further case of this kind of conditioning. The process of conditioning,
therefore, is not limited to the single stimulus, but can occur also to a
complex of permanent environmental stimuli.
The above described data support a view that the process of conditioning
involves both the intermittent conditioned stimulus and the situation in
which this stimulus appears. However, due to the experimental procedure
which consists of giving the reinforcement only to the intermittent CS and
never to the permanent complex of situational stimuli, the reaction to the
situation alone ceases to appear. This is a partial inhibition of the conditioned
reflex to the situation. In other words, a differentiation takes place between
the intermittent CS+situation and the situation alone.
This does not mean that the CR to the situation alone completely disappears.
It may still reappear in various circumstances, such as a sudden disturbance
of the usual experimental procedure, etc. Moreover, the previously described
experiments showed that the conditioned stimulus retains its acquired
property of producing the CR only in the situation in which the training took
place. When the CS was applied in a different situation, the CR was absent.
This was especially evident in the case of instrumental conditioned reflexes.
Let us now discuss various authors’ interpretations of the role of the

situation, already mentioned in previous chapters. Konorski and Miller (1933,
1936, Konorski 1939, 1967) divided all stimuli into two groups. The
permanent stimuli of the situation were called by them the “determining”
stimuli, whereas the intermittent stimuli were called the “releasing” stimuli.
The determining stimuli decide which reaction has to appear, while the
releasing stimuli decide when it will happen. When studying their papers, the
reader may draw a conclusion that Konorski and Miller did not regard the
situation as a separate factor in conditioning. Rather they tried
76 Conditioning
to eliminate the problem of the role of situation in conditioning, and in their

usual procedure with conditioned reflexes, they quickly removed the intertrial
appearances of the CR to the situation (by not reinforcing them) to
concentrate only on the intermittent CS (cf. Konorski 1967, p. 359).
However, the problem of the situation persisted and in many cases seemed to
be the decisive factor in behavior. One such instance is the study of Denisov
and Kupalov (1933, Kupalov 1961), described in chapter 4. Briefly, in
experiments on dogs, two kinds of illumination of the experimental chamber
were used, full illumination in some sessions, and dim illumination in other
sessions. When the conditioned reflex was trained in full light and then tested
in dim light, its value remained at a high level during several successive
sessions and only then decreased. Vice versa, after training in dim light with
reflex testing in full light, its value remained low for the next several sessions.
Kupalov (1961) suggested that this was an effect of conditioning to the
experimental situation. According to Kupalov, however, the situation is not
able to evoke the conditionwed reaction but can change the tonus of some
parts of the brain. Therefore, he called the reflex to situation a shortened or
truncated conditioned reflex.
However, this explanation cannot be accepted when confronted with facts

such as the appearance of instrumental conditioned reactions to the situation
only (cf. Konorski and Miller 1933, Konorski 1967, p.360; Wyrwicka et al.
1959, Wyrwicka and Chase 1994), or the absence of this reaction to the CS
applied in a situation completely different from the situation in which this
reaction was trained (cf. experiments described in chapter 5). However,
through the use of special experimental procedure, the conditioned reflex to
the situation became differentially inhibited. It was, therefore, considered a
“shortened” or “truncated” conditioned reflex (Kupalov 1961).
Nevertheless, these terms cannot change the finding that in instrumental

conditioning, the conditioned stimulus retains its power only in the situation
in which it was trained (Wyrwicka 1958). When the CS is applied in a
different situation, it does not produce the previously established
instrumental reaction. Instead, it pro-
duces another instrumental CR, that which had been trained in this new
situation. This phenomenon leads to the conclusion that, in fact, the
situational complex of stimuli dominates over the intermittent stimulus.
As described in chapter 3, it was possible to obtain two different conditioned

reactions when the same intermittent CS was applied in two different
situations. Most data on this phenomenon (called “switching ”) were obtained
in Asratyan’s laboratory, mainly as a result of training. According to
Asratyan’s interpretation, switching results from the action of a specific
permanent stimulus, e.g., the person conducting the experiment or the time
of the day. The permanent stimulus was called by Asratyan the tonic stimulus
which prepares the organism for a response to the phasic stimulus (i.e.,
intermittent CS). For instance, it was possible to obtain an alimentary
conditioned reaction to a CS in the morning session, and a defensive
conditioned reaction to the same CS in the afternoon session. Asratyan (1965)
hypothesized that the switching phenomena result from the ability of the
brain to adapt to environmental changes.
Although Asratyan’s interpretation seems to be reasonable, the fact is that it

does not take into account that the phenomenon of switching may be a result
of conditioning to the permanent situational factor (such as the experimenter
or the time of the day). All other interpretations can only add to this basic
fact.
Similarly, even such a convincing interpretation as Konorski’s theory of two

kinds of stimuli, those that determine the kind of reaction and those which
release the reaction (Konorski 1967), does not clearly speak about
conditioning to the situation. It should be said that the determining stimulus
(e.g., a cuff on the dog’s leg), is simply another conditioned stimulus.
Moreover, this situational conditioned stimulus dominates over the usual
inteimit tent stimulus which only retains the role ol an anonymous lactoi
evoking the reaction.
The role of the environment in behavior was demonstrated in clinical studies

of Lhermitte (1983, 1986 1 and II). As described in chapter 4, his frontal
lobectomy patients showed an extreme dependence on the situation, leading
to an exaggerated imitative
78 Conditioning
behavior (Lhermitte 1986,1) as well as to a kind of facilitation in responding

related to each situation in which the patients found themselves (Lhermitte
1986,11). This environmental dependence syndrome” (a term introduced by
Lhermitte 1986,11) could result from the frontal lobectomy. Nevertheless, the
change in patients’ behavior revealed one of the most important mechanisms
of behavior normally obscured by other events.
It would also be interesting to quote here a behavior theory by Beritashvili

(1971). His concept concentrates not on the reflexes but on memory.
According to him, there are three kinds of memory: image memory,
emotional memory, and conditioned-reflex memory.
Image memory is the highest form of memory found in higher vertebrates.

According to Beritashvili, even a single presentation of a food object results in
the formation of the image of this object in the brain; this leads to “image-
driven” behavior each time the animal enters the room where the original
presentation of food took place. Emotional memory originates in the fear
related to a noxious agent. Conditioned-reflex memory (of a food object)
presumably depends on the structural development of synaptic mechanisms.
Beritashvili concludes his theoretical considerations concerning memory in

vertebrates by hypothesizing that each of the three forms of memory (image,
emotional, and conditioned-reflex) gradually develop phylogenetically from
fish to monkeys. All three kinds of memory are also present in man. The
memory of man, however, differs specifically in that it includes “verbal-logical
memory.”
Although Beritashvili’s concept of behavior is so different from the

theoretical interpretations of other authors, some of his views are close to the
problem of the role of situation discussed in this book. Beritashvili stresses
the role of image memory as a reinforcing agent and its connections with the
environment where the presentation of reinforcement occurs. In that way,
the importance of the situation gains further support even from a distant
point of view.
Special attention should be given to conditioning of the brain electrical

activity (EEG) which occurs inside the body, i.e., in the
internal situation. This type of conditioning was described in chapters 7 and

8. Examples of such conditioning of the electrical activity of the brain focused
on self-producing the sensorimotor rhythm (SMR) related to the inhibition of
epileptic seizures. The active production of SMR by epileptic patients
indicates that this kind of behavior can result from learning, i.e., from the
acquisition of an instrumental conditioned reflex. This reflex is evoked by the
appearance in the brain of a complex pattern of motor excitation produced by
specific physiological conditions (which normally lead to an epileptic seizure).
The occurrence of this pattern of excitation is a conditioned stimulus which
evokes a specific instrumental conditioned reaction. The conditioned reaction
in this case is a self-produced specific physiological state of the brain which is
expressed by an appearance of low-frequency, high amplitude EEG waves
(SMR). The appearance of SMR shows that a state of motor inhibition has
taken place including the inhibition of epileptic seizures.
This inhibitory reaction can be understood as an avoidance reaction, or, using

the conditioned-reflex terminology, a defensive instrumental conditioned
reaction. A question arises as to what is the reinforcement for this reaction.
The answer is quite simple: the reinforcement is a decrease of the motor
excitation which prevents an epileptic attack.
Elere the whole conditioned act—stimulus, reaction, reinforcement—occurs

totally inside the brain. Ot course, the external environmental elements such
as the room, recording devices, time of day, etc., also exist but their role in the
process appears minimal. The main environment for the occurrence of this
process is the internal situation in the brain in which the conditioned stimu
lus, the avoidance reaction, and the reinforcement, occur. It may be said that,
in this case, instrumental conditioning to the internal
situation takes place.
Let us add a few words of explanation of the term internal situation. It can be
understood as a complex of memories of previous experiences of the
organism. These memories may serve as an internal background for the final
reaction. However, which of these memories are used as the main factors for
the reaction has to be
decided by the subject.
'
Summary and Conclusions

Chapter 1 summarizes the famous work of I.P. Pavlov. Using a single
intermittent stimulus in the same unchangeable situation, isolated from
external influences, Pavlov created the concept of the conditioned reflex as
the fundamental unit of the activity of the brain. A conditioned reflex is an
acquired function based on an unconditioned (inborn) reflex such as
salivation to an object (including food) placed in the mouth, or lifting the leg,
to pinching the skin of that leg. When an originally neutral stimulus, for
example a tone, repeatedly precedes the unconditioned stimulus (also called
the reinforcement) salivation or a defensive movement begins to appear to
the tone and becomes a conditioned reaction.
Conditioned reactions, later called classical or Pavlovian conditioned reflexes,

can be established practically to any of the previously neutral stimuli, such as
various tones, visual objects, tactile and even some taste stimuli. The
abundant experimental data obtained in Pavlov’s laboratory led him to
construct a theory of conditioned reflexes which included such functions as
differentiation between stimuli, various forms of inhibition as well as mutual
influences of some reflexes on the other reflexes, etc.
Chapter 2 deals with the motor conditioned behavior, which was not included
in Pavlov’s theory. According to Pavlov, the performance of a movement
produces sensory impulses analogous to those produced by visual or acoustic
stimuli. When these impulses are followed by an unconditioned stimulus
(e.g., food in the mouth) they become a conditioned stimulus (CS). Pavlov,
however, did
not discuss the origin of the movement itself.
The mechanisms of the motor conditioned reflexes were first experimentally

analyzed by two young scientists, J. Konorski and
81
82 Conditioning
S. Miller. They distinguished four varieties of these reflexes, two of them

based on a positive unconditioned stimulus (such as food in the mouth) and
two others on a negative unconditioned stimulus (such as pinching the skin).
Konorski and Miller concluded that the motor reflexes differ from the
Pavlovian classical reflexes in that the reinforcement depends on the
performance (or inhibition) of a specific movement, whereas there is no such
condition in the Pavlovian reflexes. In their terminology, the motor
conditioned reflexes were called conditioned reflexes type II (CRII), while
Pavlovian reflexes were given the term classical conditioned reflexes type I
(CRI) (Miller and Konorski 1928, Konorski and Miller 1936, Konorski 1939,
1948,1967). Later, conditioned reflexes type I were called classical conditioned
reflexes, while conditioned reflexes type II were called instrumental
conditioned reflexes (Hilgard and Marquis 1940, Kimble 1961).
The observations of Konorski and Miller include the fact that the trained CRII
appears not only to the intermittent conditioned stimulus (CS), but also to
the situation in which it is applied, in the absence of the CS.
Chapter 2 also summarizes some results of studies in instrumental

conditioning by other investigators such as Skinner (1938), Ferster and
Skinner (1957), and others.
Chapter 3 deals with the phenomenon called “switching” which consists of an

appearance of a conditioned reaction different than that originally trained to a
definite CS. The first observations were made by Miller and Konorski (1928)
who found that the instrumental reaction was dependent on fastening a band
on the animal’s leg or a change in the reinforcement rather than a change of
the CS. Later studies of Fonberg (1967) showed that switching is possible not
only between various reflexes with the same reinforcement, but also between
the CRII with alimentary or defensive reinforcement, as well as between the
CRII reinforced by either food or desirable electrical stimulation in the lateral
hypothalamus (cf. Olds and Milner 1954, Olds 1962).
Chapter 3 also describes the results of extensive studies of switching by

Asratyan and his associates (Asratyan 1951, 1961, 1965). In their studies the
appearance of a different CR depended
Summary and Conclusions 83
on factors such as different time of the session, different reinforcement, or

using a band on the leg. Switching was also possible between the excitatory
and the inhibitory CSs. Each of these stimuli represented a different
permanent factor of the situation which had been connected through training
with a different CR to the same intermittent CS.
Studies on switching included more recent experiments on human subjects

with the defensive type of reinforcement; in these studies an additional CS,
called a “contextual stimulus” (or “context”), preceding the usual CS, was
applied (Kimmel and Ray 1978, Kimmel and Lachnit 1990, and others). The
effects of the contextual stimulus were also studied on pigeons by Rescorla
(1984). The studies on “context” resemble earlier experiments of Pavlov on a
form of conditioning called by him “conditioned inhibition” (Pavlov 1960, Ch.
5).
Chapter 4 describes various studies on rats in which the situation turned out
to be a potent factor producing CRs to some situational elements, such as the
neuroleptic effects in the front of the room where injections of
chlorpromazine were previously given (but not in the front of other rooms).
Similarly, a specific conditioned circling was observed in the environment
where apomor-phine injections were previously given. Chapter 4 also
includes a description of the situational effects on the behavior of human
patients after frontal lobectomy.
Chapter 5 describes the systematic research on the role of the situational

background and the intermittent stimulus in conditioning. When two
instrumental CRs, with two different motor effects, one to an acoustic
stimulus and the other to a visual stimulus, were trained in the same
situation, each CS often evoked both CRs in the same trial or a CR trained to
other CS. However, when each CR was trained in a situation completely
different from the other situation, the correct CR was always obtained (see
Table 5.1). When the CS was tested in the same or similar situation, it
produced the correct CR. But when the CS was tested in a situation
completely different from the original one, it did not evoke the correct CR but
instead the CR which had been trained in that different situation to a
different CS.
84 Conditioning
This includes the case of heterogeneous CRs. When the CS, originally trained
to produce a defensive CR, was tested in the alimentary situation (in which
alimentary CRs were trained to different CSs) it evoked some general
defensive behavior (such as escape) but not the previously trained
instrumental CR. The same was observed when an alimentary CS was tested
in a defensive situation: it evoked a general alimentary posture such as
begging, but not the previously acquired CR. However, an introduction of
elements from the original situation (such as an object resembling the
feeder) led to the performance of the correct CR. These observations show
that the intermittent stimulus may retain its ability to evoke the acquired
instrumental CR only in the situation in which it was originally trained (see
Table 5.2).
Chapter 6 analyzes three examples of complex conditioning: detour, feeding,

and presleep behaviors. The experimental analysis showed that each is a case
of an instrumental conditioned reflex, occurring in a definite environment
and acquired during a conditioning training, without the use of any
intermittent conditioned stimulus. Each of these cases is complex and
consists of a few simple motor activities. The detour behavior requires a bowl
with food visible across an obstacle (fence). The act of eating also requires
some preliminary learning, related to the kind of food to be consumed.
Presleep behavior consists of a number of motor actions such as selecting a
convenient place to lie down, assuming a specific position, closing the eyes,
etc. All these motor actions depend on the situation in which they are
supposed to occur.
Chapter 7 describes the conditioning of functions of various internal organs

such as heart rate, increase or decrease salivation, vasoconstriction or
vasodilation, and others.
Chapter 8 focuses on conditioning of electrical activity of the brain (EEG) in

both animal and human subjects. Clinical studies showed that a certain EEG
(decribed as the “sensorimotor rhythm” or the “EEG feedback”) reduced or
even prevented the epileptic seizures. This EEG pattern can be conditioned,
i.e., produced by the patient, after a conditioning training. The EEG feedback
conditioned reflex is evoked by a complex of stimuli deriving from
Summary and Conclusions 85
the internal situation of the brain and is fully accomplished by an acquired

EEG activity occurring in the brain. It is, therefore, an example of
conditioning to the internal brain environment. An analysis of the clinical
data has shown the beneficial effect of this specific EEG therapy.
Chapter 9 discusses the experimental and observational data described in

previous chapters from a theoretical point of view.
Conclusions
The use of intermittent stimuli as the basis of the theory of conditioned

reflexes by Pavlov was a basic achievement in explaining the function of the
brain related to behavior.
Pavlov’s theory, however, has left some questions unanswered, one of which
is the role of the environmental background or situation in which the
conditioned reflex (CR) occurs.
A number of observations show that the intermittent CS can retain its

acquired properties only when it is applied in the situation where it had been
established by the training. This is especially true for the instrumental type of
conditioning. The instrumental type of conditioning cannot be evoked in a
new situation that is completely different from the original one. Only when
some elements related to the original situation are present, may the
instrumental reaction appear.
However, in the case of very strong defensive unconditioned stimulus (for

example, a painful electric shock to the skin), the related CS, used in the test
situation, can evoke a general defensive reaction of classical type.
Each change in the situation can alter the performance of the CR.
The CR can be established not only to an intermittent stimulus but also to the
complex of permanent situational stimuli. An example of CR to the situation
is the appearance of the intertrial CRs, observed often in instrumental CR
studies. These CRs to the situation are usually eliminated by non-reintorcing
them.
The phenomenon of “switching” can be explained as a consequence of

establishing the CR to each situation, which predominates over the effect of
the intermittent stimulus.
86 Conditioning
The “truncated” conditioned reflex is nothing else than a conditioned reflex to
the situation, partly inhibited by non-reinforcing it.
The facts of conditioning observed in pharmacological studies can also be

explained by conditioning to the situation.
A special instrumental conditioning occurs in the therapy of epilepsy. It has

been possible to establish a specific EEG pattern related to the suppression or
prevention of seizures in epileptic patients. This fact can be understood as a
case of establishing a defensive instrumental conditioned reflex to the
internal situation.
The examples described in this book are only a few selected cases showing the
crucial role of the situation in behavior. More studies on this topic are
needed.
References
Anand, B.K. and Brobeck, J.R. (1951). Hypothalamic control of food intake in
rats and cats. Yale J. Biol. & Med. 24:123-140.
Anokhin, P.K. (1961). Electroencephalographic analysis of cortico-subcorti-cal

relations in positive and negative conditioned reactions. Ann. N.Y.Acad. Sci.
92: 899-983.
Asratyan, E.A. (1951). Principle of trans-switching in conditioned reflex

activity. J. Higher Nerv. Activity 1: 47—54 (Russian).
Asratyan, E.A. (1961). The initiation and localization of cortical inhibition in

the conditioned reflex arc. Ann. N.Y. Acad. Sci., 92: 1141-1159.
Asratyan, E.A. (1965). Compensatory Adaptations, Reflex Activity and the

Brain (S.A. Corson, Transl.). Oxford: Pergamon Press.
Babb, M. and Chase, M.H. (1974). Masseteric and digastric reflex activity
during conditioned sensorimotor rhythm. EEG clin. Neurophysiol. 36: 357-
365.
Babkin, B.P. (1949, 1971). Pavlov, A Biography. Chicago and London: The
University of Chicago Press.
Beritashvili, J.S. (1941). The comparative study of individual behavior in dogs,
rabbits and hens. Transactions of the Beritashvili Physiological Institute,
4:229.
Beritov (Beritashvili), J.S. (1948). Fickleness of individual reflexes under

influence of external stimuli. In General Physiology of Muscular and Nervous
System. Ch. 11. Acad. Nauk USSR, Moscow.
Beritashvili, J.S. (1971). Vertebrate Memory Characteristics and Origin. (John

S. Barlow, Transl.). New York, Plenum Press.
Beritashvili, J.S. and Akhmeteli, M. (1941). On the behavior of pigeons in

overcoming an obstacle. Transactions of the J. Beritashvili Physiological
Institute, 4: 248.
Black, A.H. (1967). Operant conditioning of heart rate under curare. Techn.
Rep. No. 12, October 1967. Dept, of Psychology, McMaster University,
Hamilton, Ontario, Canada.
Brown, B.B. (1970). Recognition of aspects of consciousness through

association with EEG alpha activity represented by a light signal.
Psychophysiology, 6: 442-452.
Bykov, K.M. (1957). The Cerebral Cortex and the Internal Organs. Transl. and
edited by W.H. Gantt. New York: Chemical Publ.
Caruso, L., Zozula, R., Davis, E.G. et al. (1988). Classical conditioning ol sleep
onset in humans. Sleep Res. 17:82.
88 Conditioning
Chase, M.H. and McGinty, D.J. (1970a). Modulation of spontaneous and

reflex activity of the jaw musculature by orbital cortical stimulation in the
freely moving cat. Brain Res. 19: 117-126.
Chase, M.H. and McGinty, D.J. (1970b). Somatomotor inhibition and

excitation by forebrain stimulation during sleep and wakefulness: orbital
cortex Brain Res. 19:127-136.
Chase, M.H. and Babb, M. (1973). Masseteric reflex response to reticular
stimulation reverses during active sleep compared with wakefulness or quiet
sleep Brain Res.59:42 1-426. F '
Chase, M.H. and Harper, R.M. (1971). Somatomotor and visceromotor

correlates of operantly conditioned 12-14 c/s sensorimotor cortical activity.
EEG Clin. Neurophysiol. 31:85-92.
Clemente, C.D. (1968). Forebrain mechanisms related to internal inhibition

and sleep. Conditioned Reflex, 3:145-174.
MH *’ KnaUSS ’ TK ’ Sauerland > E.K., and Sterman, M.B. (1966) Inhibition of

a monosynaptic reflex by electrical stimulation of the basal forebrain or the
orbital gyrus in the cats. Experientia, 22: 844-
C emente, C.D. and Sterman, M.B. (1963). Cortical synchronization and sleep
patterns in acute restrained and chronic cats induced by basal forebrain
stimulation. EEG. clin. Neurophysiol. Suppl. 24172-187.
Clemente C.D and Sterman. M.B. (1967). Basal forebrain mechanisms for
internal inhibition and sleep. In: Sleep and Altered States and Consciousness.
Ch. V 111 , pp. 127-147.
Clemente, C.D., Sterman, M.B. and Wyrwicka, W. (1963). Forebrain inhibitory

mechanisms; conditioning of basal forebrain induced EEG synchroni-zation
and sleep. Exp. Neurol. 7:404-417.
Corson, S.A. (1967). Cerebrovisceral theory: A physiological basis for psy-

cnosomatic medicine. Internat. J. Psychol. 4: 234-241.
Corson, S A. and O’Leary-Corson, E. (1968). The effect of psychotropic drugs

on conditioning of water and electrolyte excretion: Experimental researech
and clinical implications. In: Psychotropic Drugs in Internal Medicine
txcerpta Medica International Congress Series, 182* 147-164
Cott, A., Pavloski, R.P. and Black, A.H. (1979). Reducing epileptic seizures
lough operant conditioning of central nervous system activity procedural
variables. Science 203: 73-75.
Denisov, P.K. and Kupalov, P.S. (1933). The magnitude of conditioned reflexes
m bright and obscure rooms. Arch. Biol. Sci. 35: 689. (Russian). Cit by Giurgia
1989.
^b^CIerseher Parish L Homme ' Et un ,raite de la formation du foetus. Ed.
D, Cara, L. V. and Miller, N.E. (1968a). Long term retention of instrumentally
learned heart-rale changes in the curanzed rat. Comm. Behav. Biol. 2 (Part
•A.). 19—23.
D* Cara ’ and Miller, N.E. (1968b). Instrumental learning of vasomotor l79 PO

| n 485y |486 ' earnmg ‘° reSp0nd diff erentially in the two ears. Science
References 89
Di Cara, L.V. and Miller, N.E. (1968c). Changes in heart rate instrumental^
learned by curarized rats as avoidance responses. J. Comp. Physiol. Psychol.
Donhoffer,^H. and Lissak, K. (1962). EEC changes associated with the

elaboration of conditioned reflexes. Ann. physiol. Acad. Sci. Hungarica 21.249
255. Doty R.W. and Giurgea, C. (1961). Conditioned reflexes established by
coupling electrical excitation of two cortical areas. In: Brain Mechanisms and
Learning (A. Fessard, R.W. Gerard and J. Konorski, eds.). Blackwell Sclent.
Publ., pp. 133-151.
Eikelboom, R. and Stewart, J. (1979). Condit.oned temperature effects using

morphine and the unconditioned stimulus. Psychopharmacology 61: 31-38.
Ellertsen, B. and Klove, H. (1976). Clinical application of biofeedback training
in epilepsy. Scand. J. behav. Therapy. 5: 133-144.
Evans, D.R. and Bond, I.K. (1969). Reciprocal inhibition therapy and
conditioning in the treatment of insomnia. Behav. Res. Ther. 7a: 323 323.
Ferster, CV.B. and Skinner, B.F. (1957). Schedules of Reinforcement. New
York: Appleton-Century-Crofts.
Finley W W Smith, H.A. and Etherton, M.D. (1975). Reduction of seizures and
normalization of the EEG in a severe epileptic following sensorimotor
feedback training. Preliminary Study. Biol. Psychol. 2: 189-203.
Fishel, W. (1933). Uber bewahrende und wirkende Gedachtmsleistung. Biol.

Zbl 53'349.
Fonberg, E. (1961). On the transfer of two different defensive conditioned
reflexes type II. Bull. Pol. Acad. Sci. Cl.II, 9.47 -49.
Fonberg, E. (1967). The motivational role of the hypothalamus in animal

behavior. Acta Biol. Exper. {Warsaw) 27: 303-318. .
Fonberg, E. (1969). The role of the hypothalamus and amygdala in lood

intake, alimentary motivation and emotional reactions. Acta Biol. Exper.
(Warsaw) 29: 335-358. . r
Galef, B.G.,Jr. (1978). Differences in affiliative behavior of weanling ra -
lecting eating and drinking sites. J. Comp. Physiol. Psycho . 92: 431-437. Galef
B.G.,Jr. and Clark, M.M. (1972). Mother’s milk and adult presence, two factors
determining initial dietary selection by weanling rats. J. Comp. Physiol.
Psychol. 78: 220-225.
Galef B G Jr (1977) Social transmission of food preferences: an adaptation for

weaning m rats. 7. Comp. Physiol. Psychol. 91: 1136-1140.
Gantt, W.H. (1944). Experimental Basis for Neurotic Behavior. New York.
Garm WH (1960) Cardiovascular component of the conditioned reflex to pain

food and other stimuli. Physiol. Rev. 40, suppl. 4: 266-291.
Gantt WH. and Hoffmann, W.C. (1940). Conditioned cardio-respiratory

changes accompanying conditioned food reflexes. Am. J. Physiol. (Proc.).
Gastaut, 3 fM1958). Some aspects of the neurophysiological basis of

conditioned reflexes and behavior. In: Neurological Basis of Behavior (G b.
Wolstenholme and C.M. O’Connor, eds.), pp. 255-276. Boston. Little,
Brown and Co.
90 Conditioning
Giurgea, C.E. (1970). Central excitability tonus and its modulatory influence
upon a neuroleptic activity. In: Modern Problems of Pharmacopsychiatry. The
Neuroleptics. Vol. 5: 91-94. Basel, Karger.
Giurgea, C.E. (1989). Kupalov’s concept of shortened conditional; reflexes:

psychophysiological and psychopharmacological implications. Pavl. J. Biol Sci
24: 81-89.
Harper, R.M. (1973). Relationship of neuronal activity to EEG waves during

sleep and wakefulness. In: M.I.Phillips (ed.) Brain Unit Activity during
Behavior, pp. 130-154. Springfield, IL.: C.C. Thomas.
Harper, R.M. and Sterman, M.B. (1972). Subcortical unit activity during
conditioned 12-14 Hz sensorimotor rhythm in cats. Federation Proceedings
31:404.
Higginson, G.D. (1926). Cit. by Thorpe 1956.
Hilgard, E.R. and Marquis, D.G. (1940). Conditioning and Learning. New
Hsiao, H.H. (1929). Cit. by Thorpe 1956.
Izquierdo, I., Wyrwicka, W., Sierra, G. and Segundo, J.P. (1965).

Etablissement d un reflexe de trace pendant le sommeil natural chez le chat.
Actual. Neurophysiol. 6: 277-296.
Jasper, H.H. and Ajmone-Marsan, C. (1960). A Stereotaxic Atlas of the

Diencephalon of the Cat. Ottawa: The National Research Council of Canada.
John, E.R., Leiman, A.L. and Sachs, E. (1961). An exploration of the functional
relationship between electroencephalographic potentials and differential
inhibition. Ann. N.Y. Acad. Sci. 92: 1160-1182.
Kamiya, J. (1962). Conditioned discrimination of the EEG alpha rhythm in

humans. Paper presented at the Meeting of the Western Physiological
Aassociation. San Francisco, April 1962.
Kamiya, J. (1968). Conscious control of brain waves. Psychology Today 1:57-

60.
Kaplan, B.J. (1975). Biofeedback in epileptics: equivocal relationship of
reinforced EEG frequency to seizure reduction. Epilepsia 16: 477-485.
Khananashvili, M.M. (1983). The Pathology of the Higher Nervous Activity

(Russian). Moscow: Edit. Medizina.
Kimble, G.A. (1961). Hilgard and Marquis Conditioning and Learning. New
Kimmel, H.D. and Ray, R.L. (1978). Transswitching: Conditioning with tonic
and phasic stimuli. J. Exper. Psychol.: General 107: 187-205.
Kogan, A.B. (1960). The manifestations of processes of higher nervous

activity in the electrical potentials of the cortex during free behavior of
animals. EEG clin. Neurophysiol. Suppl. 13: 51-64.
Kohler, W. (1948, orig.1925). The Mentality of Apes. London.
Konorski, J. (1939). Principles of cortical switching. Przeglad Fiziol. Ruchu 9:

191-242. (Polish; a French summary).
Konorski, J. (1948). Conditioned Reflexes and Neuron Organization. London:

Cambridge University Press.
Konorski, J. (1967). Integrative Activity of the Brain. An Interdisciplinary

Approach. Chicago: University of Chicago Press.
References 91
Konorski, J. and Lawicka, W. (1959). Physiological mechanisms of delayed

reactions. Acta Biol. Exp. (Warsaw ) 19: 175-197.
Konorski J. and Miller S. Conditioned reflexes of the motor analyzer. Trudy
Fisiol. lab. Pavlova. 6: 119-298 (Russian).
Konorski, J. and Szwejkowska, G. (1950). Chronic extinction and restoration

of conditioned reflexes. 1. Extinction against the excitatory background.
Acta Biol. Exp. (Warsaw ) 15: 135-170.

Konorski, J. and Wyrwicka, W. (1950). Transformation of conditioned
reflexes of the first type into conditioned reflexes of the second type. Acta
Biol. Exp. (Warsaw ) 15: 193-204.
Kozak, W., Macfarlane, W.V. and Westerman, R. (1962). Long-lasting

reversible changes in the reflex responses of chronic spinal cats to touch, heat
and cold. Nature (Lond .) 193: 171-173.
Kozak, W. and Westerman, R. (1966). Basic patterns of plastic change in the

mammalian nervous system. In: Nervous & Hormonal Mechanisms of
Integration, XXth Symposium of the Society for Experimental Biology. 20:
509-544.
Krasnogorsky, N.I. Cit. by Pavlov 1951.
Kuhlman, W.N. (1978). EEG feedback training: enhancement of sensorimotor

cortical activity. EEG clin. Neurophysiol. 45: 290-294.
Kuhlman, W.N. and Allison, T. (1977). EEG feedback training in the treatment
of epilepsy: some questions and some answers. Pavlovian J. Biol. Sci. 12 : 112
122 .
Kuhlman, W.N. and Kaplan, B.J. (1979). Clinical applications of EEG feedback
training. In R.J. Gatchel and K.P. Price (eds). Clinical Applications of
Biofeedback: Appraisal and Status. New York: Pergamon.
Kupalov, PS. (1961). Some normal and pathological properties of nervous

processes in the brain. Ann. N.Y. Acad. Sci. 92: 1046-1053.
Lachnit, H. (1986). Transswitching and contextual conditioning. Relevant

aspect of time. Pavlovian J. Biol. Sci. 21: 160-172.
Lachnit, H. and Kimmel, H.D. (1990). Contextual conditioning. A comparison

of Eastern and Western views. Pavlovian J. Biol. Sci. 25: 174-179. Lantz, D.
and Sterman, M.B. (1992). Neuropsychological prediction and outcome
measures in relation to EEG feedback training for the treatment of epilepsy.
In: Thomas L. Bennett (Ed.) The Neuropsychology of Epilepsy. New York:
Plenum Press.
Lawicka, W. (1959). Physiological mechanism of delayed reactions. II.
Delayed reactions in dog and cats to directional stimuli. Acta Biol. Exp.
(Warsaw) 19:199-219.
Lhermitte, F. (1986). Human autonomy and the frontal lobes. Part II. Patient
behavior in corn-lex and social situations. The “environmental dependency
syndrome.” Ann. Neurol. 11: 335—343.
Lorenz, K. (1932). Betrachtungen uber das Erkennen der arteigenen

Triebhandlungen der Vogel. J. Ornithol. 80:50.
Lorenz, K. (1939). Vergleichende Verhaltungsforschung. Zool. Anz. 12,
Supplementband, p.60.
Lubar, J.F., Shabsin, H. Natelson, S.E., Holder, G.S., Whitsett, S.F., Pamplin,
92 Conditioning
W.E. and Krulikowski, D. (1981). EEG operant conditioning in intractable

epilepsy. Archives Neurol. 38: 700-704.
McGinty, D.J. and Sterman, M.B. (1968). Sleep suppression after basal
forebrain lesions in the cat. Science 160: 1253-1255.
Melchior, Chr.L. (1988). Environment-dependent tolerance to ethanol

produced by intracerebroventricular injections in mice. Psychopharmacology
96: 258-261.
Miller, N.E. (1969). Learning of visceral and glandular responses. Science

163:434-445.
Miller, N.E. and Carmona, A. (1967). Modification of visceral response,

salivation in thirsty dogs, by instrumental training with water reward. J.
Comp. Physiol. Psychol. 63: 1-6.
Miller, N.E. and Di Cara, L.V. (1967). Instrumental learning of heart-rate

changes in curarized rats: shaping and specificity to discriminative stimulus.
J. Comp.Physiol. Psychol. 63: 12-19.
Miller, N.E. and Di Cara, L.V. (1968). Homeostasis and reward: T-maze
learning induced by manipulating antidiuretic hormone. Am. J. Physiol. 215*
684-686.
Miller, S. and Konorski, J. (1928). Sur une form particuliere des reflexes
conditionnels. C.R. Soc. Biol. (Paris) 99: 1155-1157.
Murrin, M.R. and Kimmel, H.D. (1986). Determinants of tonic and phasic
reactions in transswitching. Pavlovian J. Behav. Sci. 21: 117-123.
Newmann, T. (1978).
Nowliss, D.P. and Kamiya, J. (1970). The control of electroencephalographic

alpha rhythms through auditory feedback and the associated mental activity.
Psychophysiology 6: 476-484.
Olds, J. (1962). Hypothalamic substrates of reward. Physiol. Rev. 42* 554-

604.
Olds, J. and Milner, P. (1954). Positive reinforcement produced by electrical

stimulation of septal area and other regions of rat brain. J. Comp. Physiol.
Psychol. 47: 419-427
Olds, J. and Olds, M.E. (1961). Interference and learning in paleocortical

systems. In Brain Mechanisms and Learning (A. Fessard, R.W. Gerard, J.
Konorski and J.F. Delafresnaye, eds.), pp. 153-187.
Pavlov, I.P. (1949). Pavlovskie Sredy 1949, p. 574. (Russian).
Pavlov, I.P. (1951). Fiziologicheskii mechanism tak nazyvanych proizvolnykh

dvizenii. In: I.P.Pavlov’s Polnoe Sobranie Sochinenii, Vol. 3, Book 2, 2nd
edition, enlarged, pp. 315-319. Moscow and Leningrad: Academia Nauk SSSR.
Pavlov, I.P. (1960, first printed 1927). Conditioned Reflexes. New York: Dover
Publications.
Poser, E.G., Fenton, G.W. and Scotton, L. (1965). The classical conditioning of
sleep and wakefulness. Behav. Res. Ther. 3: 259-264.
Poulos, C.X. and Hinson, R. (1982). Pavlovian conditional tolerance to halo-

peridol catalepsy: Evidence of dynamic adaptation in the dopaminergic
system. Science 218: 491-492.
Rescorla, R.A. (1984). Associations between Pavlovian CSs and context. J.

Exper. Psychol.: Animal Behavior Processes. 10: 195-204.
References 93
Roginsky G.Z. and Tikh (1956). Roundabout ways in animals (Summary).

Problems of the modern physiology of the nervous and muscle systems.
Tbilisi, p. 384.
Rosenfeld, J.P., Rudell, A.P., and Fox, S.S. (1969). Operant control of neural
events in humans. Science 165: 821-823.
Roth, S.R., Sterman, M.B. and Clemente, C.D. (1967). Comparison of EEG
correlates of reinforcement, internal inhibition and sleep. EEG clin.
Neurophysiol. 23: 509-520.
Sakhiulina, G.T. Cit. by Asratyan 1965.
Schiller, R von (1942). Unwegversuche an Elritzen. Z. Thierpsychol. 5:101.

Seifert, A.R. and Lubar, J.F. (1975). Reduction of epileptic seizures through
EEG feedback training. Biol. Physiol 3: 157-184.
Sherrington, C.S. (1929). Sherrington’s Femer Lecture. Proc. Roy. Soc. B, Vol.
CV, p.332. Cit. by Konorski 1948, p. 72.
Sherrington, C.S. (1947). Integrative Action of the Nervous System. Cit. by
Konorski 1948, p.72.
Skinner, B.F. (1938). The Behavior of Organisms: an Experimental Analysis.

New York: Appleton-Century.
Soltysik, S. (1975). Post-consummatory arousal of drive as a mechanism of

incentive motivation. Acta Neurobiol. Exp. 35: 447-474.
Sterman, M.B. (1981). Power spectral analysis of EEG characteristics during

sleep in epileptics. Epilepsia 22 : 9 5 -10 6 .
Sterman, M.B. (1986). Epilepsy and its treatment with EEG feedback therapy.
Ann. Behav. Med. 8: 21—25.
Sterman, M.B. and Clemente, C.D. (1962). Forebrain inhibitory mechanisms:

Sleep patterns induced by basal forebrain stimulation in the behaving cat.
Exp. Neurol. 6: 103-117.
Sterman, M.B. and Friar, L. (1972). Suppression of seizures in an epileptic

following sensorimotor EEG feedback training. EEG clin. Neurophysiol.
33. g9_95
Sterman, M.B. and MacDonald, L.R. (1978). Effects of central cortical EEG
feedback training on incidence of poorly controlled seizures. Epilepsia 19. 207
—222. Sterman, M.B., MacDonald, L.R. and Stone, R.K. (1974). Biofeedback
training of the sensorimotor EEG rhythm: Effect on epilepsy. Epilepsia 15.
395-417. Sterman, M.B. and Shouse, M.M. (1980). Quantitative analysis of
training, sleep EEG, and clinical response to EEG operant conditioning in
epileptics. EEG clin. Neurophysiol. 49: 558-576.
Sterman, M.B. and Wyrwicka, W. (1967). EEG correlates of sleep. Evidence
for separate forebrain substrates. Brain Res 6. 143—163.
Sterman, M.B., Wyrwicka, W. and Roth,S.R. (1969). Electrophysiological

correlates and neural substrates of alimentary behavior in the cat. Ann. N. Y.
Acad.Sci. 157:723-739. . L u . .
Stroganov, V. V. (1948). The effect of a change of the situation upon the

higher
nervous activity of the dog. Trudy Eisiol. Lab. Pavlova 13: 128-149.
Szwejkowska. G. (1950). The chronic extinction and restoration of
conditioned reflexes. 2. The extinction against an inhibitory background. Acta
Biol. Exp. (Warsaw ) 15: 171-184.
94 Conditioning
Teyrovsky, V. (1930). A study of ideational behaviour in the Garden Warbler.

Pub. Fac. Sci. Univ. Masaryk, No. 122.
Thorndike, E.L. (1898). Animal Intelligence. An experimental study of the

associative processes in animals. Psychol Monogr. 2, No. 8.
Thorpe, W.H. (1956). Learning and Instinct in Animals. London: Methuen.
Throwhill, J.A. (1967). Instrumental conditioning of the heart rate in the

curarized rat. J. Comp. Physiol. Psychol. 63: 7-11.
Tolman, E.C., and Honzik, C.H. (1930). Cit. by Thorpe 1956.
Ungerstedt, U. (1976). 6-Hydroxydopamine induced degeneration of the

migrostriatal dopamine pathway: The turning sundrome. Pharmacol. Therap.
(B). 2:37.
Vatsuro, E.G. (1948). The conditioned reflex situation and its effect on
learning of conditioned reflexes. Trudy Fisiol. Lab. Pavlova 13: 21-127.
(Russian).
Windholz, G. and Wyrwicka, W. (1996). Pavlov’s position toward Konorski

and Miller’s distinction between Pavlovian and motor conditioning paradigm.
Integr. Physiol. Behav. Sci 31: 338349.
Wyler, A.R., Lockard, J.S., Ward, A.A. and Finch, C.A. (1976). Conditioned
EEG desynchronization and seizure occurrence in patients. EEG clin.
Neurophytsiol. 41: 501-512.
Wyler, A.R., Robins, C.A. and Dodrill, C.B. (1979). EEG operant conditioning
for control of epilpsy. Epilepsia 20: 279-286.
Wyrwicka, W. (1952). On the mechanism of the motor conditioned reaction.

Acta Biol. Exp. (Warsaw) 16: 131-137.
Wyrwicka, W. (1956). On the effect of experimental situation upon the course

of motor conditioned reflexes. Acta Biol. Exper. (Warsaw) 17: 189-203.
Wyrwicka, W. (1958). Studies on the effects of the conditioned stimulus
applied against various experimental backgrounds. Acta Biol. Exper.
(Warsaw) 18:175-193.
Wyrwicka, W. (1959). Studies on detour behavior. Behaviour 14: 240-264
Wyrwicka, W. (1964). Unpublished experiments.
Wyrwicka, W. (1975). The sensory nature of reward in instrumental behavior.

Pavlovian J. Biol. Sci. 10: 23-51.
Wyrwicka, W. (1982). Some aspects of the origins of food preferences. Brain

Res. Inst. Bull. 6: 3-4, 10-11.
Wyrwicka, W. (1988). Brain and Feeding Behavior. Springfield IL ■

C.C.Thomas.
Wyrwicka, W. (1993). The problem of switching in conditional behavior

Integr Physiol. Behav. Sci. 28: 239-257.
Wyrwicka, W. and Chase, M.H. (1972). The act of eating as an instrumental

reaction rewarded by electrical stimulation of the brain. Physiol Behav 9-717-
720.
Wyrwicka, W. and Chase, M.H. (1994). Conditioning of brain stimulation-

induced presleep behavior. Physiol. Behav. 56: 883-889.
Wyrwicka, W„ Dobrzecka, C. and Tarnecki, R. (1959). On the instrumental

reaction evoked by electrical stimulation of the hypothalamus Science 130*
336-337.
References 95
Wyrwicka, W., Dobrzecka C. and Tarnecki, R. (1960). The effect of electrical

stimulation of the hypothalamus on the conditioned reflexes, type II. Acta
Biol. Exper. (Warsaw ) 20: 121-136.
Wyrwicka, W. and Garcia, R. (1979). The effect of electrical stimulation of the

dorsal nucleus of n. vagus on gastric acid secretion in cats. Exp. Neurol. 65:
315-325.
Wyrwicka, W., Sterman, M.B. and Clemente, C.D. (1962). Conditioning of
induced EEG sleep pattern. Science 137: 616-618.
Wyrwicka, W. and Sterman, M.B. (1968). Instrumental conditioning of

sensorimotor cortex EEG spindles in the waking cat. Physiol. Behav. 3: 703-
707.
Zbrozyna, A. (1953). A phenomenon of non-identification of a stimulus

operating against different physiological background in dogs. Soc. Sci. Lodz ,
Sec. Ill, No. 26 (Polish with an English summary).
Zuckermann, E. and Buffy, A. (1960). Protein-bounded acetylcholine output

in cortical efferent blood after conditioned reflexes. Exp. Neurol. 2: 423-428.
Name Index
Akhmeteli, M., 52 Allison, T., 70 Anand, B.K., 25 Anokhin, P.K., 65 Asratyan,

E.A., 2, 77
Babb, M., 57, 65 Beritashvili, J.S., 41, 52, 78 Black, A.H., 62, 70 Bogen, K.M., 9
Bond, I.K., 65 Brobeck, J.R., 25 Brown, B.B., 63 Buffy, A., 37 Bykov, K.M., 9, 56
Carmona, A., 61, 63, 67, 68 Caruso, L. 65
Chase, M.H., 57, 58, 65, 66. 75, 76 Clark, M.M., 57 Clemente, C.D., 57, 59, 64
Corson, S.A., 36 Cott, A., 70
Davis, E.G., 65 Descartes, R., 5 Di Cara, L.V., 61, 67 Donhoffer, H., 65 Doty,
R.W., 32 Eikelboom, 36 Ellertsen, B., 71 Evans, D.R., 65
Ferster, C.B., 18, 19, 55 Finch, C.A., 70 Finley, W.W., 70, 71 Fisher, W., 51
Fonberg, E., 24 Fox, S.S., 63 Friar, L., 68
Galef, B.G. Jr., 57
Gantt, W.H., 9, 56 Garcia, R., 38 Gastaut, H., 65 Giurgea, C.E., 35, 36

Gorshkova, cit. by Bykov, 10 Harper, R.M., 66 Higginson, G.D., 51 Hilgard,
E.R, 14 Hinson, R., 35 Ho, 37
Hoffman, W.C., 9, 56 Hsiao, H.H., 51 Honzik, C.H., 51
Izquierdo, I., 62
John, E.R., 65
Kamiya, J., 63, 68 Kaplan, B.J., 70, 71 Kimble, G.A., 14 Kimmel, H.D. 28
Khananashvili, M.M., 35 Klove, H., 71 Knauss, T.K., 57
Konorski, J., 2, 6. 10, 11, 12, 13, 14, 15, 17, 18, 21, 22, 31, 32, 33, 36, 43,55,64,78
Kozak, W., 43-44 Kohler, W., 51 Krasnogorski, N.I., 11, 15 Krulikowski, D., 91
(first author L.F.Lubar)
Kuhlman,W.N., 70 Kupalov, P.S., 31, 37 Kurtsin, L.K., 9

Lachnit, H., 28 Lantz, D., 71 Lawicka, W., 55 Leiman, A.L., 65 Lhermitte, F.,
38, 39, 78, 79
98 Conditioning
Lissak, K., 65 Lorenz, K., 51 Lubar, J.F., 70, 71
MacDonald, L.R., 70 Marquis, D.G., 14 McFairlane, W.V., 43-44 McGinty, D.J.,

57, 59, 65 Melchior, C.HR-L, 37 Miller, N.E., 61, 63, 67, 68 Miller, S„ 2, 11, 12,
13, 14, 16, 17, 18, 22
Murrin, M.R., 28
Natelson, S.E., 71 Nowlis, D.R, 63
Olds J. and Olds M.E., 61, 68 O’Leary-Corson, E., 36
Pavlov, I.P., 2, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,41,52 Poser, E.G., 65 Poulos, C.X.,
35 Pressman, cit. by Asratyan, 28
Ray, R.L., 28 Rescorla, R.A., 29 Roginski, G.Z., 52 Rosenfeld, J.P., 63 Roth,

S.R., 65, 67 Rowland, 65 Rudell, A.P, 63
Sachs, 65
Sakhiulina, G.T., 27 Sauerland, E.K., 57
Schiller, von, R., 51 Seifert, H.R., 70 Sherrington, C.S., 14 Shitov, 27 Shouse,

M.M., 70 Skinner, B.F., 18, 19, 55 Soltysik, S., 43-44
Sterman, M.B., 13, 57, 65, 66, 68, 70 Stewart, J., 36 Stroganov, E.G., 41
Struchkov, M.I., cit. by Asratyan Szwejkowska, G., 33, 35
Thorndike, E.L., 17, 18, 51, 55 Throwhill, J.A., 61 Teyrowski, V., 51 Tikh, N.A.,
52 Thorpe, W.H., 51-52 Tolman, E.C., 51,
Ungerstedt, U., 37
Ward, A.A., 70 Westerman, R.,44 Windholz, G., 11 Whittsett, S.F. 71 Wyler,

A.R., 70
Wyrwicka, W., 11, 17, 33, 41, 43, 44, 52, 56, 58, 59, 63, 64, 65, 68
Yakovleva, cit. by Asratyan, 26 Yamaguchi, 59
Zamyatina, 29
Zbrozyna, A., 34, 65
Zozula, R., 65 (first author Caruso, L.)
Zuckerman, E., 35, 37
Subject Index
Basal forebrain area, 59
Chain reflexes, 55 Conditioned reflexes classical (type I), 2, 5 instrumental

(type II), 2, 5 four variations of instrumental conditioned reflexes, 12-13 to
the situation, 51 Conditioned inhibition, 7 transfer, 32
Conditioning of the sensorimotor (SMR) rhythm for milk in cats, 63-64;

paired with barbiturate increases amount of sleep in humans suffering from
insomnia, 65
Dependence of behavior on situationed cues, 31
in patients after frontal lobectomy, 38 Determining stimuli, 24 Detour

behavior, 51-55 Differential inhibition, 8 EEG responses between left-right
location as a result of switching, 28
Drugs and conditioning, 18
Effect of excitatory versus inhibitory background on conditioning, 32-33

“Environmental dependence syndrome” (term introduced by Lhermitte), 39
Establishment of an instrumental coditioned reflex to the situation, 15-16
Experiments of context conducted on pigeon, 29
Extinction of the conditioned reflex, 7, 18, 23
Feeding as a complex conditioned act, 56, 58a;

preparatory feeding reactions as con ditioned behavior, 56 factors inhibiting
or facilitating con sumption by brain stimulation, 58 Generalization of
stimuli, 8, 9
Hypothalamus, 56
Lateral hypothalamic “feeding center,” 58
Pleasurable lateral hypothalamic stimulation, 56
Intermittent stimulus, 1,2 Inhibition of delay, 7 Internal inhibition, 7

“Insight,” 55
Instrumental conditioning of electrical activity from the paleocortical area in

rat’s brain, 62
Konorski’s interpretation of phenomena introduced by Pavlov such as

irradiation, negative induction and positive induction, 14
Kupalov’s view of situation as shortened (truncated) conditioned stimulus,

31,76
Learning by curarized rats to increase or decrease heart rate to obtain

rewarding brain stimulation or avoiding electric shock in the skin, 61
Learning by noncurarized thirsty dogs to increase salivation when rewarded
with access to water, 61 Learning to make a detour by puppies controlled from
birth, 52 Learning to desynchronize EEG pattern during sleep in cats and in
human subjects, 63
Lesions in basal forebrain area result in suppression of sleep in cats, 64
100 Conditioning
Method of producing a passive movements in dog, 11
Neutral test for
homogeneous situation I, 46 homogeneous situation II, 47 heterogeneous

situation, 48
defensive stimulus tested in alimentary situation, 49 alimentary stimulus

tested in defensive situation, 50
Non-identification by dogs of a stimulus produced by a permanent change in

situation, 34-35
Pavlov’s concept of motor conditioned reflexes, 11
Phasic stimuli as intermittent conditioned stimuli, 27-28
Prevention of epileptic seizures by
self-producing SMR (inhibitory EEG rhythm) by the subject, 79
Releasing stimuli, 24
Reversal of reinforcements, 26-27
Role of situation with the use of morphine, 36-37 apomorphine, 36-37

ethanol, 37
Sensorimotor rhythm (SMR) as related to motor inhibition, 66
Studies on value of two different reactions trained in
one situation, 41-42; two similar situations, 42; two different situations, 42-
43
Seizures reduction following training of 12-15 c/sec SMR feedback, 70
seizure reduction may be dependent of severity of symptoms, 71 Secretion

from salivary glands and other internal organs should be considered a
conditioned behavior, 56 Situation in Pavlovian (classical) type of
conditioning, 3
Skinner’s theoretical approach to conditioning, 18
operant behavior, 18 respondent behavior, 18 Substitution of food

reinforcement by electrical stimulation in the lateral hypothalamus, 25-26
Switching (transfer, interchange) definition homogeneous switching, 21-22
heterogeneous switching, 22-23 occurring between excitatory and inhibitory
states or other heterogeneous instrumental reactions, 27 Switching produced
by extinction, 23-24, or by excitation, 25
Therapeutic role of EEG feedback in epilepsy, 67
Thorndike’s experiments, 17, 18 “Tower of silence” of Pavlov, 10,11 Trial and

error learning”, 1 Temporary increase of EEG bursts at the beginning of the
extinction process (as a proof of conditioning), 64 Theoretical comments on
the role of situation in behavior, 73-79 Theory of Beritashvili, 78
conditioned-reflex memory 78 emotional memory, 78 image theory, 78
Unconditioned reflex, 5 Unconditioned stimulus, 6

'
V
BOSTON
PUB
BRARY
3 9999 03793 703 2
iL#r
No longer tty* property of the
Boston
ubrary.
Sale of this material benefits the Library
Boston Public Library
COPLEY SQUARE
- : "AT"
The Date Due Card in the pocket Indicates the date on or before which this
book should be returned to the Library,
Please do not remove cards front this pocket
(Continued from front flap)
the functions of various internal organs, and the conditioning of electrical

brain activity leading to inhibition of epileptic seizures. In her concluding
chapter, Wyrwicka discusses theoretically the data mentioned previously.
Conditioning opens up rich possibilities for continued exploration. This

revealing work will interest scientists specializing in behavioral sciences,
psychologists, neuroscientists, educators, as well as students of biology.
About the Author
Library of Congress: 99-16172 Printed in the U.S.A.
Jacket design by Lynn E. McPhearson
Wanda Wyrwicka is a research neurobiologist at the University of California

at the Los Angeles School of Medicine. Her books include The Mechanics of
Conditioned Behavior, The Development of Food Preferences, Brain and
Feeding Behavior, and from Transaction, Imitation in Human and Animal
Behavior.

Conditioning Situation Versus Intermittent Stimulus

Uploaded by

Copyright:

Available Formats

Conditioning Situation Versus Intermittent Stimulus

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Conditioning Situation Versus Intermittent Stimulus

Uploaded by

Copyright:

Available Formats

This book made available by the Internet Archive.

I wish to extend deep appreciation to my colleagues, Drs. Michael Chase,

Traditionally, in laboratory research, the process of conditioning, both

research on conditioned behavior.

A question arises whether the same scheme of conditioning may be applied to

circumstances in which it occurs.

In each of these examples, the single stimulus (the sound of an approaching

A question can be asked—what happens when the situational background is

Similar observations were made on domestic pets. Some animals cannot

In the laboratory, reactions to situational stimuli are often disregarded, and

This book discusses various cases of situational influences on behavior and

Chapter 2 presents results of studies by Konorski and Miller on motor

Chapter 3 deals in full with early data on the phenomenon of so-called

Chapter 4 describes the effects of changes in the experimental situation on

Chapter 6 discusses examples of complex instrumental reflexes such as

Chapter 7 narrates cases of conditioning of activity of internal organs.

Chapter 8 shows the therapeutic role of EEG feedback in epilepsy.

Chapter 9 contains a theoretical discussion of the data described in previous

As a physiologist, Pavlov chose a physiological interpretation of this

called an unconditioned stimulus (USC). The previously neutral stimulus that

Pavlov emphasized that the establishment of a conditioned reflex is

In Pavlov’s laboratory a great variety of conditioned stimuli were used. These

Manipulations with a variety of conditioned and unconditioned

Classical Conditioned Reflexes 7

stimuli resulted in an accumulation of a great amount of experimental data

Internal inhibition is strictly related to the already formed conditioned reflex

Internal inhibition observed in Pavlov’s laboratory included a form called the

Another case of internal inhibition studied in Pavlov’s laboratory was the “

Finally, a form of internal inhibition was obtained when two similar

It should be mentioned, however, that the above described differential

An important phenomenon observed mainly during the initial phase of

Classical Conditioned Reflexes 9

The research on conditioned reflexes not only consisted of compiling more

This building consisted of four chambers on each of three floors. Each

This led to an almost complete neglect of the problem of the situation in

Nevertheless, the problem of the situation exists. It will be discussed in

As usually happens in science, new theories develop by the addition of

An important question remained unanswered. In order to get the motor

The problem of the motor conditioned reflexes was experimentally analyzed

MOVEMENT ► IMPULSES ► SALIVATION

Movement, a passive movement of the dog, evoked by a special devise

Impulses, nervous impulses deriving from the performed movement to the

Salivation, conditioned reflex of salivation.

As shown on the diagram, the conditioned reflex of salivation appears only

just published book on conditioned reflexes, they became enthusiasts of

Instrumental Conditioned Reflexes

Second variety. When a specific movement (e.g., leg flexion) appearing to a

Third variety. When a specific movement (e.g., leg flexion) appearing to a

Each variety was experimentally documented. The young investigators,

In 1948, Konorski published a book entitled Conditioned reflexes and neuron

Early observations of Konorski and Miller on conditioned reflexes type II

Instrumental Conditioned Reflexes

One of the First Records of Type II CRs

Upper graph: CS (lamp).

Lower graph: record of the movements of the left foreleg.

a) The movement is performed in the presence of the CS and in intervals.

b) The movement is performed only in response to the CS.

(Konorski and Miller, 1933).