Content uploaded by Erica N Feuerbacher
All content in this area was uploaded by Erica N Feuerbacher on Mar 14, 2014
Content may be subject to copyright.
History of Dogs 46
A History of Dogs as Subjects in North American
Experimental Psychological Research
Erica N. Feuerbacher
University of Florida
C. D. L. Wynne
University of Florida and Kyung Hee University
The modern resurgence in psychological experiments involving dogs follows a long and rich tradition of using dogs as
experimental subjects in psychology. Except for a few exceptions (e.g., Pavlov, and Scott and Fuller), much of this research
is often overlooked. We trace the history of dogs as experimental psychological subjects: The work of Darwin and Pavlov
sets the stage for our focus on research emanating from North American laboratories. We end our review with the advent of
the modern renaissance of dog research. This account tracks the history of psychology as a science, providing insight into
psychological processes and theoretical corollaries of these processes generally, and shedding light on the behavior of dogs
specically. A rediscovery of this literature can only aid research being conducted today, including rejuvenating old ques-
tions, suggesting new ones, and highlighting useful methods for current issues. We hope through this endeavor that those
working with dogs will see themselves as part of this rich tradition and that a historical perspective will help integrate dog
research into a eld greater than the sum of its parts.
Keywords: history of psychology, dogs, Pavlovian conditioning, operant conditioning, animal cognition.
Dogs have much to recommend them as psychological
subjects. They are as suitable for the study of basic behavioral
processes as any typical laboratory animal but also have
a number of unique attributes that encourage their study
specically. First, dogs are a highly social species. This
allows for investigations into the mechanisms that produce
and maintain conspecic social behavior, as well as the
cognitive and behavioral by-products of those mechanisms.
Dogs’ high degree of sociality has led to them being studied
as an analog of human social behavior.
Moreover, some researchers have suggested that dogs’
close evolutionary relationship with humans might have
produced unique cognitive skills and behavioral repertoires
in dogs (e.g., Hare & Tomasello, 2005), including the ability
to respond to complex cues from a heterospecic individual,
and closer approximations to the abilities of humans than any
other nonhuman species. The uniquely intimate relationship
between dogs and humans offers researchers a unique social
system unavailable in more traditional lab animals. Finally,
the behavioral and morphological variability between
breeds of dogs offers an unrivaled system for evaluating
the interactions between phylogenetic and ontogenetic
selection in the production of behavioral traits (e.g., Spady
& Ostrander, 2008). Thus, dogs offer researchers a rich
spectrum of psychological research opportunities, some of
which are found in few other species. Recognition of dogs’
unique research prospects has been part of the impetus for
Erica N. Feuerbacher, Department of Psychology, University of
Florida; Clive D. L. Wynne, Department of Psychology, Uni-
versity of Florida and Department of Biomedical Engineering,
Kyung Hee University, Seoul, Korea.
We thank Donald Dewsbury for providing useful feedback on an
earlier draft of this manuscript.
Correspondence concerning this article should be addressed to
Erica Feuerbacher, Department of Psychology, University of
Florida, P.O. Box 112250, Gainesville, Florida 32611. Email:
firstname.lastname@example.org, Phone: (940) 390 9830, Fax: (352) 392-
Volume 6: 46 - 712011
ISSN: 1911-4745 doi: 10.3819/ccbr.2011.60001 © Erica Feuerbacher 2011
History of Dogs 47
the recent renaissance in comparative dog behavior and
cognition (e.g., Kubinyi, Virányi, & Miklósi, 2007).
The modern eld of comparative dog cognition and
behavior research can be dated to a handful of papers
published in 1997 and 1998. Adam Miklósi and colleagues
at the Eotvos Lorand University in Budapest, Hungary
reported a correlation between dogs’ problem solving
abilities, their role in human lives (pet vs. working dog),
and the owners’ perceptions of their dogs (Topál, Miklósi,
& Csányi, 1997). In 1998, the same group demonstrated
that pet dogs could locate a hidden food item by following a
human pointing gesture (Miklósi, Polgárdi, Topál, & Csányi,
1998). Independently, in the United States, Brian Hare and
colleagues reported a very similar result (Hare & Tomasello,
1998). The study of dog behavior and cognition has grown
rapidly in the subsequent years. One measure of the growth
of the eld is that at the most recent annual international
conference on comparative cognition (Melbourne, FL,
2010) over 10% of presentations concerned dogs and their
relatives, whereas at the rst such meeting in 1994, there
was not a single paper on this subject group.
One thing that has arguably failed to develop during this
period of increasing interest in dog behavior is an awareness
of the century or more of research on dogs that preceded
Miklósi’s and Hare’s groups’ papers in 1997 and 1998. It is
a common complaint that modern researchers are unaware
of the history of their elds (Goodwin, 2004), but in the
present case this phenomenon is particularly pronounced.
The research that sprang up in the late 1990s had no direct
connections to the previous scientic work. These earlier
research traditions had largely died out by the 1980s and
thus the modern eld of dog cognition and behavior is
Most classes in introductory psychology mention Pavlov
and his dogs, and most classes on animal learning mention
that the original studies that led to the development of the
concept of “learned helplessness” were carried out on dogs
in the 1950s (Solomon & Wynne, 1953). Aside from these
two contributions the only other widely-cited study on dogs
is the classic monograph, Genetics and the Social Behavior
of the Dog by John Paul Scott and John Fuller (1965). We
show here, however, that dogs have been very popular
subjects in studies on animal behavior and cognition dating
back to Charles Darwin himself. The use of dogs as a subject
species peaked in the 1960s, but declined over the rest of
the twentieth century, before starting to increase in the rst
decade of the twenty-rst century.
We have identied three broad categories of reasons for
using dogs as psychological subjects: (a) they might be a
convenient animal for the study of basic processes that are
not unique to dogs; (b) they have useful, but not necessarily
unique, behavioral or cognitive traits that can be used to
study a more general phenomenon (e.g., the morphological
and behavioral diversity between breeds can be studied to
examine whether the factors that affect morphology have a
concomitant effect on behavior); and (c) they have unique
behavioral or cognitive traits that are themselves the focus
An interesting difference between much of the earlier
work reviewed here and the current resurgence of research
on dogs is a change in the reasons for using dogs as
experimental subjects. Much current research using dogs as
psychological subjects is directed at questions about dogs
themselves, and most closely follows the third category of
reasons for studying dogs listed above. Most of the earlier
research was aimed at the rst two categories of reasons for
using dogs as subjects. The changing rationale for studying
dog psychology might be part of the reason why older papers
are seldom cited today.
Nevertheless, changing rationales for studying dogs do
not detract from the earlier studies’ potential to contribute to
current and future research. Studies that used dogs to examine
more ubiquitous psychological processes might become
the foundation for contemporary comparative research to
further tease apart the unique and nonunique behavioral and
cognitive traits and mechanisms in dogs. Similarly, when
dogs were used because of their behavioral resemblance to
humans (their social behavior, in particular), the results have
informed not only subsequent work on humans, but can also
now shed light on current questions that are being pursued
for the sake of learning about dogs themselves.
Sometimes earlier work in a eld is of limited value as, for
example, when technological advances render earlier studies
irrelevant to modern researchers. In the area of animal
behavior and cognition, however, technological advances,
while important, have not had as great an impact as in many
areas of science. Probably the only pieces of equipment
commonly used today in the study of dog behavior and
cognition that would not be familiar to researchers from
the mid-twentieth century would be the video camera and
computer touch screen (neither of which is essential to
contribute to the peer-reviewed literature on dog psychology
today). Of course, there have been advances in experimental
design and statistical analysis, but absence of these features
is hardly reason enough to ignore a study.
Of course, many earlier studies appear misguided to us
today. In some cases their errors might be obvious and render
a study uninformative. Other problems in a study might,
however, involve issues that could not become apparent until
the research was attempted. In this way, even awed studies
History of Dogs 48
can be of great use to contemporary researchers by showing
them directions of research that, although possessing face
validity, are unlikely to be productive.
Furthermore, much research carried out investigating
the cognition and behavior of dogs prior to 1998 was by
any measures competent or even excellent science. That
contemporary researchers do not incorporate this work
into their discussions is at least remiss and perhaps at times
In the current paper, we review the major lineages of
research in experimental psychology on dogs emanating
from North American laboratories (Figure 1). We start our
review by contextualizing North American comparative
psychology with Darwin and Pavlov, and end with the
advent of the modern renaissance of research using dogs
as experimental subjects in 1997. In dening our scope, we
recognize the omission of much important work from outside
North America; we have limited the paper in this way to
prevent it from becoming unwieldy to the reader and possibly
incomplete with regard to work that is unavailable in English,
and therefore, largely inaccessible to the current authors. We
hope that this paper will allow current researchers to reopen
research on past questions with a new perspective, inform
and improve current research through the addition of other
relevant data, open new lines of research suggested from
the older research with the benet of historical perspective,
and perhaps motivate others to delve into the research not
covered in this paper.
W. Horsley Gantt Howard S. Liddell
William T. James
B. F. Skinner
Marian Breland Bailey
Richard L. Solomon
Harry FrankMartin Shapiro
John P. Scott &
John L. Fuller
Michael W. Fox
James J. Lynch
François Y. Doré
Donald O. Hebb
W. Robert Thompson
Figure 1. Flow chart showing the intellectual relationships between major researchers discussed in the text. Solid arrows
(—) connect to boxes containing students, postdoctoral researchers, or laboratory associates of the primary investigator.
Dashed lines (- - -) show direct scientic collaborations between researchers. For example, W. Robert Thompson from
Hebb’s lab worked with Scott and Fuller for a summer. The diagram does not show the more subtle inuences that certainly
exist between these researchers, such from papers or interactions at conferences. Boxes contain researchers who worked
closely together (for example, the Arkansas researchers, Roscoe Dykman, Oddist Murphree, Joseph Newton, and John
History of Dogs 49
Darwin and His Advocates
It is customary to date the beginnings of Comparative
Psychology to the works of Charles Darwin (1809 - 1882). In
the Origin of Species by Means of Natural Selection (1859)
Darwin outlined how all species are related by descent to
common ancestors. This provided an intellectual framework
within which questions about the degree of psychological
similarity between humans and other species became of
central interest and inspired the rst attempts to study the
behavior of nonhumans (see Browne, 2003). Darwin theorized
in more detail about the cognitive similarities between
humans and other animals in the Descent of Man (1871).
In chapters three and four, Darwin offered a “Comparison
of the mental powers of man and the lower animals.” At
several points in that discussion Darwin referred to dogs, as
when, for example, he suggested that the love of a dog for its
master resembles the human belief in a god. Darwin listed
the emotions seen in dogs; other psychological abilities such
as imitation, memory, dreaming, and reason; moral qualities,
the concept of property, and of self. Darwin’s method in the
Descent was based on his own informal observations and
on reports offered to him by others. Darwin’s behavioral
observations would not be publishable in a modern scientic
journal; nonetheless, his ideas have often proved correct.
The nal work in which Darwin wrote extensively on
animal psychology was The Expression of Emotions in
Man and Animals (1872). Darwin’s aim in this work was to
demonstrate the universality of the expression of emotions
across different species, including humans. In that volume,
many of the illustrations of emotional expression are of a
dog, specically Darwin’s own terrier (Townshend, 2009).
Darwin was not a professor and thus had no students in the
formal sense. However, he inspired many contemporaries to
study questions that grew out of his theory of evolution by
natural selection. One of Darwin’s intellectual advocates was
George Romanes (1848−1894) (Wynne, 2007). Romanes’
book, Animal Intelligence (1883), may be counted as the
rst post-Darwinian book-length treatment of its subject,
and Romanes’ afliation with Darwin may be gauged from
the fact that this book, published only months after Darwin’s
death, included a glowing eulogy to Darwin.
Chapter 16 of Romanes’ Animal Intelligence was dedicated
to dogs. Romanes added little to what Darwin had written
a decade earlier in the Descent about dog psychological
qualities. Romanes’ anecdotes of dog “sagacity” were
generally more credulous than Darwin’s, but his approach to
animal intelligence paralleled Darwin’s. Romanes added an
interesting nurturist note to Darwin’s generally more nativist
thinking on dog intelligence. Thus Romanes argued that the
ner emotions (“Pride, sense of dignity, and self-respect”
among others, p. 439) were only present in “well-treated”
“high-life” dogs, and not in “Curs of low degree” (p. 439).
Although Romanes was mainly an anecdotalist, he was
also the author of one of the rst experimental studies of dog
behavior. Romanes (1887) tested the ability of a setter-bitch,
with which he had hunted for eight years prior to testing, to
follow his scent under a range of circumstances. As a result
of 16 tests in which he and his friends and servants walked
around a large park wearing their own boots, each others’
boots, or no boots at all, Romanes deduced that the dog
identied his scent trail “by the peculiar smell of my boots...,
and not by the peculiar smell of my feet” (p. 274). He further
concluded that had he been “accustomed to shoot without
boots or stockings, she would have learnt to associate with
me a trail made by my bare feet” (p. 274) — though he did
not carry out any experiments to test this hypothesis.
Figure 2. Charles Darwin. Father of evolutionary theory
and early researcher in dog behavior and cognition. Image
from Wikipedia in public domain.
History of Dogs 50
Sir John Lubbock
A second individual directly inuenced by Darwin to
carry out an experiment on dog behavior was Darwin’s
neighbor at Downe, Kent, Sir John Lubbock (1834−1913).
Lubbock (1884a, 1884b, 1889) gave his poodle, Van, the
opportunity to convey its ideas to him by providing it with
a set of pieces of stout cardboard, each with a word written
legibly on it. The words used included “Food,” “Bone,”
“Water,” and “Out.” Lubbock started by putting food in a
saucer and placing the “Food” card over the top of it. Next
to this, he placed an empty saucer with a blank card on top.
Then Lubbock trained the dog to bring him the “Food” card
before he gave the dog any food. In “about a month” (1884a,
p. 276), the dog brought the food card in preference to the
blank card (which did not lead to any food) on nine out of
ten occasions. Lubbock also reported that the dog would
bring him a card with “Out” on it when it wanted to go for a
walk, and a card with “Water” written on it when it wanted
to drink. It is not entirely clear how Lubbock knew that the
dog was really requesting these outcomes. He commented:
“No one who has seen him look along a row of cards and
select the right one can, I think, doubt that in bringing a card
he feels that he is making a request.” (1884a, p. 548).
Conwy Lloyd Morgan
The third early originator of the study of animal
psychology to have been directly inuenced by Darwin
and who published some observations on dog behavior
was Conwy Lloyd Morgan (1852−1936). Morgan spent
most of his career at University College Bristol (today’s
University of Bristol) and is best known for his “canon”
according to which the behavior of animals should not be
ascribed to a “higher mental faculty” if it can be accounted
for in terms of the “exercise of one which stands lower in the
psychological scale” (1894, p. 53). Morgan’s contribution to
the discussion of dog intelligence consisted in what can be
viewed as counterexamples to Romanes’ anecdotes. Morgan
described apparently ingenious behavior of dogs, which
could nevertheless be accounted for through trial-and-error
learning. The most commonly cited example is Morgan’s
dog, Tony, whose ability to open the latch to a garden gate
might by a casual observer have been considered insightful.
Morgan, however, recounted how careful observation over a
period of time demonstrated that this behavior was in fact the
Figure 4. Conwy Lloyd Morgan, originator of Morgan’s
Canon and early researcher on dog behavior and cognition.
Morgan took a more behavioral view of complex dog behav-
ior. Image copyright the University of Bristol. Reproduced
Figure 3. George Romanes, follower of Darwin who contin-
ued Darwin’s interest in dog behavior and cognition. Image
from Wikipedia in public domain.
History of Dogs 51
outcome of an extended process of trial-and-error learning.
Similarly, Morgan described how this same dog failed over
repeated days of testing to ever turn a stick so that it would
pass through the railings of a fence (Morgan, 1894).
Pavlovian Conditioning Tradition
Pavlov and Bekhterev
The British tradition inspired directly by Darwin failed to
develop into an experimental science of dog behavior, but
did set the stage for comparative lines of questioning in North
America. To nd the origins of the experimental study of
dogs in psychology, we need to briey discuss two Russian
scientists whose work guided much of the experimental
behavioral research on dogs in the early twentieth century
and whose techniques are still used.
Ivan Pavlov (1849−1936) is the name most widely
associated with psychological research in dogs, having
elucidated conditioned reexes using stulated dogs to
measure their salivary responses (thorough accounts of
Pavlov’s work can be found elsewhere: e.g., Todes, 2000,
2001). Although Pavlov’s name is widely associated
with these procedures, it is his rival, Vladimir Bekhterev
(1857−1927), whose methods were most frequently used in
North American experiments on conditioned reexes.
Bekhterev focused on motor rather than glandular
reexes. He applied shock to elicit paw exion and changes
in respiration. The reasons for the adoption of Bekhterev’s
methods over Pavlov’s are multifaceted, including issues
of historical timing and some pragmatic preferences for his
experimental techniques. Watson, in his 1915 American
Psychological Association Presidential Address, mentioned
Pavlov’s techniques but focused on and expressed his
preference for Bekhterev’s (Ruiz, Sánchez, & De la Casa,
2003; Watson, 1916), as did others (Brogden, 1962).
Pavlov’s work nonetheless grew in inuence in the United
States in the 1920s through lectures by G. V. Anrep, a student
of Pavlov’s (Liddell, 1958), and Anrep’s publication of a
complete translation of a set of Pavlov’s lectures (Pavlov,
1927). At the same time, the problem of gaining access to
Bekhterev’s Russian experimental papers, which were never
translated into English, probably mitigated his continued
Together, Pavlov’s and Bekhterev’s laboratories provided
a wealth of research into the conditioned reex in dogs and
gave the world a powerful new methodology for objectively
studying behavior. The conditioned reex was these
laboratories’ primary research interest, not dog behavior,
per se; however, the utility of their techniques and their
results formed the basis for much of the experimental work
involving dogs in North America, including some of our rst
understandings of the remarkable sensory capacities of dogs
(Razran & Warden, 1929).
Pavlovians in America: W. Horsley Gantt and Students
The entrée of Pavlov into America came through his
students, in particular W. Horsley Gantt (1892−1980). (A
second Pavlov student, Howard S. Liddell and his student,
William T. James, are discussed in the Combining Traditions
Gantt met Pavlov in 1922 and worked as a researcher in
Pavlov’s lab for nearly seven years before returning to Johns
Hopkins University (Lynch, n.d.), where he founded the
Pavlovian Laboratory in 1929. He later founded a second
lab at the Veterans Administration Hospital in Perry Point,
Maryland. In both labs, dogs were the primary experimental
subjects. Though Gantt’s initial choice of dogs was likely
a result of working with Pavlov rather than an interest in
dogs per se, as his research developed, Gantt focused on
phenomena in dogs that are potentially relevant to many
social species, including humans. As Gantt noted, “The dog
has also a special advantage 1) stemming from his long and
intimate association with the human being, and 2) because of
his very responsive and easily inuenced cardio-respiratory
system” (Gantt, Newton, Royer, & Stephens, 1966, p. 156).
Gantt identied ve elds in which he made major
contributions: (a) the relationship of the intensity of the
stimulus to the magnitude of the response (Harvey, 1995); (b)
Figure 5. Lloyd Morgan’s dog, Tony, lifting the garden gate
latch with its head. Lloyd Morgan attributed the dog’s be-
havior to gradual trial-and-error learning rather then in-
sight. Reproduced from http://www.psy.tufts.edu/psych26/
morgan.htm with permission of Robert Cook.
History of Dogs 52
psychopharmacology; (c) the role of the nervous system in
conditioning; (d) cardiac conditioning; and (e) experimental
neuroses (Reese, Peters, & Dykman, 1987). Gantt typically
utilized Bekhterev’s motor reex preparation, although
he sometimes used the salivary reex. He also attempted
to condition other physiological systems, including the
cardiac, respiratory, and renal systems (Ruiz et al., 2003).
This multifaceted approach allowed for comparisons of the
conditionability of different systems, which Gantt found
were often differentially conditionable and not in accord.
That is, the conditioned reex, or CR, in one system might
have extinguished, but the CR of another system to the
same conditioned stimulus, or CS, continued, a status he
termed schizokinesis (see Harvey, 1995; and Ruiz et al.,
2003, for more complete descriptions of Gantt’s conceptual
framework). Gantt considered his work on cardiac
conditioning the most interesting aspect of his research
(Reese et al., 1987), and Harvey (1995) suggested that this
work might have been his principal addition to Pavlov’s
The cardiovascular conditioning procedure also allowed
Gantt to elucidate more clearly a phenomenon Pavlov had
termed the social reex (Lynch & Gantt, 1968a, 1968b),
and which Gantt called the effect of person (Lynch, 1987).
This was the effect of one animal, whether conspecic or
heterospecic, on another (Gantt, 1973). In this paradigm,
Lynch and Gantt (1968b) used Pavlovian conditioning to
measure the effect of humans on dogs’ heart rates noting
that, “the cardiovascular system may be a valuable index
in studying the psychophysiology of [the] socialization
process” (p. 69). They found that a dog’s heart rate would
spike (i.e., tachycardia) when a human walked in the room,
but would decrease substantially (i.e., bradycardia) when the
human petted the dog (Gantt et al., 1966).
Gantt et al. (1966) also noted that dogs’ effect of person
response often showed specicity to a particular person.
The effects, both cardiac and motor, were larger and more
variable to people with whom the dog had a “special
relationship” (p. 152), compared to the dog’s reactions to
strangers. The exact history of the dog with the person who
elicited those amplied responses was not described, such
that the requisite contingencies to produce those effects
remain to be explored.
Gantt’s work also demonstrated various conditioning
effects. For example, when a bell was a CS for petting,
the bell elicited bradycardia (Gantt et al., 1966). However,
the dogs’ heart rate response to a tone associated with the
entry of a passive human into the room did not show any
conditioning effect (Newton, Teitelbaum, & Gantt, 1968).
The “effect of person” was also apparent when overlaid
on other experimental contingencies. For example, when
unsignaled shocks were administered to dogs, the typical
tachycardia elicited by the shock was considerably reduced
when a human was petting the dog during the shock delivery
(Anderson & Gantt, 1966). James J. Lynch pursued more
detailed analyses of the effect and found that the tachycardiac
response to the CS and US were both reduced when a human
petted the dog during the CS and US. The motor response
(paw exion) also diminished in the petting phase (Lynch
& McCarthy, 1967). This study pointed to the potential
potency of tactile stimulation to reduce aversive conditioned
In a follow up paper, Lynch and McCarthy (1969)
investigated whether the effect of person elicited by a person
entering the experimental room would vary if the dog had a
differential history with the person. They found that heart
rate increased the most when the human was associated
with shock and increased the least when the human was
associated with petting. These results further substantiated
Gantt’s view that petting might be a US for dogs. The effect
of person was even more pronounced in two neurotic dogs
(Gantt et al., 1966).
The effect of person is undoubtedly an interesting
phenomenon and worthy of rediscovery, especially with the
current interest in the intimate relationship enjoyed by dogs
with humans. In their 1966 paper, Gantt et al. concluded
that much more work on the effect of person was warranted.
They suggested that there was a large amount of conditioning
that went into producing the effect, as evidenced by its
specicity to certain people. It remained unclear, however,
whether there was also an underlying unconditioned stimulus
property responsible for the behavioral relation, especially
in the tactile component, and whether this property might be
a product of phylogeny or early socialization.
Figure 6. W. Horsley Gantt feeding dog, Gundy. Gantt was
an American student of Pavlov and researched the “Effect of
Person” in dogs. Image courtesy of the Alan Mason Chesney
Medical Archives of the Johns Hopkins Medical Institutions.
History of Dogs 53
Gantt also ventured into the domain of personality; his
interest in neuroses lent itself to research on individual
differences. He investigated erratic heart rate and respiration,
refusal to eat in the experimental chamber, loss of CRs to
food, and variable sexual reexes (Gantt, 1944). Gantt
viewed putting a dog under experimental strain as a useful
way to identify individual differences that might not have
been apparent under normal circumstances. As Gantt (1944)
noted, in this line of research he followed Pavlov, who
had made mention of similar differentiation of types (e.g.,
inhibitory and excitatory) identied by subjecting the dogs
to experimental strain and measuring the predominance of
different conditioned responses.
Gantt advocated addressing individual differences by
collecting many behavioral measures from different systems
(e.g., motor, cardiovascular, and respiratory) and measuring
their correlations. This suggestion presaged much of the
work being done today in the eld of dog temperament and
personality (e.g., Sinn, Gosling, & Hilliard, 2010; Svartberg
& Forkman, 2002), in which statistical analyses are used to
identify underlying behavioral structure that might inuence
a suite of behaviors. For those interested in this eld, a
rediscovery of Pavlov’s work (e.g., Pavlov, 1927), as well as
that of Gantt, and especially Gantt’s theoretical deliberations
(Gantt, 1944) on this topic, would be valuable.
Gantt had several other students and collaborators,
who made unique and substantial contributions to our
understanding of dog behavior. Roscoe Dykman, Oddist
Murphree, Joseph Newton, and John Peters, among others,
developed and studied what became known as the Arkansas
nervous pointer dog line. Their work is discussed below in
the Combining Traditions section.
Gantt clearly made substantial contributions to psychology
in general, and to our understanding of dog learning and
behavior in particular. He published over 700 articles (see
Horvath, 1987, for a full list of Gantt’s publications) and
was nominated for a Nobel Prize in 1970 (Harvey, 1995). At
the time that Gantt was working, neo-Behaviorism was at its
peak and Gantt’s work was largely overlooked (see Reese
et al., 1987; and Ruiz et al., 2003 for discussion of possible
contributing factors). Gantt’s attention to experimental
integrity in the tradition of Pavlov, and his use of single-
subject designs to give very complete analyses of individual
differences and their persistence over an individual’s
lifetime, make his work worthy of rediscovery. It might
shed light on contemporary issues as well as possibly reopen
worthwhile research tracks that have been largely forgotten.
The Operant Tradition:
From Thorndike to Skinner, and Beyond
While the Pavlovian tradition was emerging in Russia
and nding its way to North America, another tradition was
emerging in America: what would become known as operant
Edward Thorndike (1874−1949) reported on puzzle box
experiments with dogs in his 1898 book, Animal Intelligence,
offering the rst experimental work that could be classied
as operant. More typically associated with his work on
puzzle box learning with cats, Thorndike investigated
dogs in similar boxes. Although Thorndike makes various
references to dogs and their behavior in his book (1898), in
his 1899 article, he focused mainly on the behavior of his
cats; in one place noting only that, “the dogs presented no
difference in behavior that would modify our conclusions”
(p. 291). Much like the subsequent researchers discussed
in this section, the choice of dogs as experimental subjects
seems to have been one of convenience.
Following Thorndike’s monograph, other experimental
reports on dogs using operant methods emerged, but did
not represent a cohesive body of research. One of the more
prominent examples was W. T. Shepherd, a professor of
psychology and philosophy at Waynesburg College, who
published work comparing the “adaptive intelligence” of
dogs, cats, and rhesus monkeys (1915) and the auditory
discrimination ability of dogs (1919). Shepherd used
experimental procedures much like those of Thorndike’s.
Operant work on dogs did not begin to accelerate until
the late 1940s and early 1950s. In the operant tradition,
Clark Hull published one paper comparing the reinforcing
effectiveness of regular feeding, sham feeding, in which the
food did not reach the dog’s stomach due to an esophageal
stula, and extinction (Hull, Livingston, Rouse, & Barker,
1951). More work came from Skinner (1904−1990) and his
students. The dog research of some of these students was
one way in which Skinner exerted his inuence.
Skinner’s Students: Marian Breland Bailey and Ogden
One of Skinner’s rst graduate students was Marian Breland
(later Marian Breland Bailey, having remarried after Keller
Breland’s death) (1920−2001). While the Brelands, and
later the Baileys, did not publish their experimental research
on dogs in peer-reviewed journals, they did relate their
experiences training animals in their book, Animal Behavior
(Breland & Breland, 1966). This work and their inuence
on the elds of comparative psychology and the world of
animal training, and dog training in particular, warrant their
History of Dogs 54
mention. The Brelands started an animal training business in
1943 call Animal Behavior Enterprises (ABE) (see Bailey
& Bailey, 2006, for a more complete history). In their
business, the Breland-Baileys were particularly interested
in long-range control of behavior and worked on training
eld and utility dogs, including bomb-snifng dogs. They
constructed operant chambers for dogs, in which many
response topographies could be measured and a variety of
reinforcers delivered. Schedules of reinforcement were
evaluated, largely to improve ABE’s ability to maintain
sustained performances in their dog performers (Bailey,
personal communication, March 17, 2010). Their expertise
in constructing operant chambers was put to use by other
researchers in the eld (e.g., Murphree, Dykman, & Peters,
The Breland-Baileys spearheaded the movement for
humane training, which has particularly taken hold in the
dog training world (e.g., Pryor, 1999); Bob Bailey stated
their perspective: “Patience and preparedness are better then
brute force” (Bailey & Bailey, 2006).
Skinner had another graduate student, Ogden Lindsley
(1922−2004), who elected to study dogs for his dissertation
in large part because of their sociality, which he thought
made them a better model species for human behavior
(Lindsley, 2004). For this project, Lindsley built one of
the earliest operant chambers for dogs (Lindsley, 1956).
Lindsley investigated the effects of high doses of radiation
on beagles’ performance on discrimination tasks, calibrating
the dogs’ performance against their performance after
being administered pharmacological agents such as alcohol,
benzedrine, and nembutal (Lindsley, 2002). Lindsley (2002)
reported that with the irradiation only food consumption
was disrupted as the dogs sickened, but that the visual
discrimination and the sound-conditioned fear suppression
responses were sustained until the dog died. Several other
lines of research using dogs as subjects followed the
Skinnerian tradition, although the authors were not students
Beyond Skinner: Roger W. McIntire
One researcher working within an operant tradition but
outside the direct inuence of Skinner was Roger McIntire.
McIntire made a unique contribution to research on dogs
while working under contract for the United States Army.
At the University of Maryland, McIntire started the Canine
Behavior Lab, in which he investigated training techniques
for eldwork with Army dogs, a research program that sent
an estimated 2500 working dogs to Vietnam (McIntire,
personal communication, August 27, 2010). He published
two defense reports from this work: one that focused on
the utility of free-ranging dogs and dogs working within a
foot patrol for personnel reconnaissance (McIntire, 1965),
and another that investigated the role of early experience
and selective breeding, and the utility of reinforcement
(McIntire, 1967). Initially, the researchers used telemetered
brain stimulation as a reinforcer for holding a pointing
posture when a human scent was detected (McIntire,
personal communication, August 27, 2010). However, this
proved ineffective and McIntire soon switched to social
reinforcement from the handler (e.g., McIntire & Colley,
1967). From this work, he found no effects of the age at
which training was initiated, nor large effects of selective
breeding. He did, however, report that training was
facilitated through nding the most powerful reinforcers
and limiting the use of punishers (McIntire, 1968). McIntire
also published an experimental paper investigating the
effectiveness of human social reinforcement in training dogs
and maintaining performance in working dogs (McIntire &
Colley, 1967). Together, these papers represent some of the
few investigations into the role of human social interaction
in maintaining responding in a dog. McIntire and Colley
(1967) found that tactile reinforcement from the handler
(petting) was effective in maintaining responding, whereas
verbal praise alone was not. Although inconclusive, these
results suggested that there might be primary and secondary
reinforcing components to human interaction, a result that
hearkens back to the effect of person reported by Gantt (e.g.,
Gantt et al., 1966).
Figure 7. Hunter, one of Ogden Lindsley’s research subjects,
presses a panel in an operant chamber designed for dogs.
Hunter survived an LD50 does of 300 roentgen units. The
image is thought to be a personal picture from Lindsley’s.
Image originally published in the Journal of Experimen-
tal Analysis of Behavior, 77(3), p. 386. Image courtesy of
the Society for the Experimental Analysis of Behavior, Inc.,
which holds the copyright.
History of Dogs 55
John Paul Scott and John L. Fuller
In the 1950s, a new and highly inuential research line
was emerging that was “developmentally and socially
oriented” (Scott, 1978). John Paul Scott (1909−2000) and
John L. Fuller (1910−1992) began investigating the role
of inheritance in dog behavior, helping to launch the eld
of behavioral genetics (Dewsbury, 2009). Their work was
inuenced by that of anatomist Charles Stockard, whose
research will be briey described below in the section
As Scott and Fuller (1965) noted, the inuence of heredity
on behavior had been largely ignored during the rise of
behaviorism in the 1930s and 1940s. A new program at
Jackson Laboratory was designed to remedy this by assessing
the hereditary inuence on behavior in mammals.
The project that became known as “Genetics and the
Social Behavior of Mammals” (Scott & Fuller, 1965) had
as its backdrop the fundamental goal of understanding
human behavior; nonhuman, mammalian models were seen
as convenient ways of addressing fundamental problems of
human behavior. Dogs were chosen for the project in part
because of Scott’s perception that they exhibited greater
sociality than more common lab species, which potentially
could provide greater insights into human behavior
(Dewsbury, 2009). The second reason for using dogs was
the assumed range of genetic variability between the various
dog breeds (Scott & Fuller, 1965).
Because the work of Scott and Fuller and the history of
Jackson Laboratory have been assembled in one tome, their
1965 masterpiece, Genetics and the Social Behavior of the
Dog, as well as having been insightfully and thoroughly
chronicled in other historical reports (Dewsbury, 2009, in
press-a, in press-b), we will only give an overview of their
immense project here. Scott and Fuller (1965) identied
as one of their primary questions, “whether heredity could
produce an important effect upon behavior in a higher
animal or whether it simply set the stage for behavioral
activity” (p. 3). The dog project spanned 23 years at Jackson
Laboratory and involved the breeding, rearing, and study
of over 300 dogs. Although the project started with more
dog breeds, Scott and Fuller eventually narrowed the scope
to focus on ve dog breeds of similar morphology. This
reduced the likelihood of physical characteristics producing
any observed behavioral differences. They studied purebred
dogs of all ve breeds, as well as carefully crossed hybrids
of two of the breeds (basenjis and cocker spaniels). The dogs
were assessed on a battery of tests from birth until dogs were
1 year old.
Tests were broken into four classes: (a) performance tests,
(b) emotional reactivity tests, (c) relationship tests, and (d)
physical and physiological observations (Staff, 1950). We
will focus on the rst three. Performance tests were those
tests in which the dogs had to complete a relatively simple
task, but for which specialized training was often required
before the assessment could be made. Such tests included:
(a) leash control climbing tests, in which the dog’s ability to
learn to walk on a leash and navigate stairs was measured;
(b) trailing tests, in which the dog was trained to locate an
object; (c) barrier tests, in which the dog was placed in a
novel environment and it had to gure out how to navigate
to a hidden food bowl, with successive tests being more
and more challenging; (d) discrimination tests, in which the
dog had to make a correct choice in a T-maze based on a
conditional discriminatory stimulus; (e) manipulation tests,
in which the dog was presented with an inaccessible food
bowl (the level of inaccessibility increased over trials) and
had to learn to eventually manipulate a rope to access the
food bowl; and (f) maze tests, in which the dog had to learn
to navigate a complicated maze to access a food reward. The
emotional reactivity tests were those in which the dog was
subjected to mild but short-term stressors such as restraint,
small cages, startling stimuli, pain (e.g., shock), threatening
displays by the experimenter, or loud noises. Relationship
tests investigated patterns of social behavior in the presence
of conspecics or humans. The tests with conspecics
included dominance tests, which involved presenting two
dogs with a bone and measuring which dog had possession
of the bone and for how long, as well as observations of
social behavior including maternal and sexual behavior.
Tests involving humans included handling tests, in which
the dog’s response to human approach and petting were
recorded, and motivation tests, in which dogs were trained
to follow the experimenter into the laboratory and the
amount of time the dog was close to the experimenter during
following was measured.
These tests did not seem to trace their origins to any
specic research tradition, but Pavlov, Gantt and W. T.
James, among others, inuenced the program at Jackson
Laboratory (Fuller, 1948). Scott and Fuller (1965) also cited
Köhler as the source of their “umweg” barrier test. Köhler, in
his book The Mentality of Apes (1925/1999), described the
behavior of a single dog in a similar barrier test: the food was
visually accessible, but the only way to physically access it
was a circuitous route around the chain link fence separating
dog and food. Finally, Scott and Fuller referenced Robert
Tryon’s work using mazes to identify individual differences
in rats (Tryon, 1940).
Despite calling the compendium of their work Genetics
and the Social Behavior of the Dog, which might seem to
History of Dogs 56
imply that the research found strong implications for the
role of heredity in behavior, the work at Jackson Laboratory
found only moderate to low heritabilities—an average of
27% (Dewsbury, in press-a) — for behavioral traits. As
Dewsbury astutely pointed out, however, this result only
indicated that in those specic test conditions, there was a
large inuence of the environment. Heritability values do
not actually indicate the degree of genetic inuence, or lack
thereof, on the behavior of interest, nor are heritabilities able
to be extrapolated to other studies. Scott and Fuller (1965)
found breed differences on different tasks, but correlations
were low, suggesting that there was not a unied single
factor underlying the behavioral results (Dewsbury, in
press-a). As Scott noted, “genetics does not put behavior
in a straitjacket” (quoted in Dewsbury, in press-a, p. 21).
As his work progressed, Scott changed positions from one
that asserted a strong genetic inuence, to one emphasizing
the “plasticity and the exibility of behavior” (Dewsbury, in
One concept that emerged from the Jackson Laboratory
work that has attained widespread recognition is that
of critical periods (e.g., Scott, 1958), an idea that Scott
developed from embryology and for which he may be best
known (Dewsbury, in press-a). The concept of critical
periods suggests that there are certain juvenile phases
during which the presence or absence of certain stimuli
and experiences will have long-lasting effects in the adult,
and any missed opportunities during the critical period are
largely irremediable once that window of opportunity has
closed. Based on the Jackson Laboratory work with puppy
development, Scott and Marston (1950) identied four
different stages in puppy development. Of the greatest
practical implication was the socialization period, in which
exposure to stimuli, especially social stimuli such as humans
and conspecics, was deemed essential if the puppy was to
grow up to exhibit friendly social behavior to humans and
In 1968, the dog laboratory at Jackson Laboratory closed.
John Paul Scott had departed three years earlier, taking a
position at Bowling Green State University. At Bowling
Green, he established a colony of dogs and continued to
publish on dog behavior, especially focusing on the critical
period hypothesis and the role of experience during critical
periods in affecting attachment (e.g., Compton & Scott,
1971; Gurski, Davis, & Scott, 1980).
The inuence of Scott and Fuller’s work cannot be
overstated for research that focuses on dog behavior,
development, or genetics. It is unlikely that such a large
endeavor, including the careful breeding and rearing of many
dogs, will be undertaken again in the foreseeable future.
Their 1965 treatise is an oft-cited data source, especially
in the elds of development and behavioral genetics. Many
researchers passed through the Jackson Dog Laboratory,
particularly through the summer research program that
Scott arranged. Thus, although the research conducted in
this program often does not bear Scott or Fuller’s name, a
quick glance at the acknowledgements or grants that funded
the work point to Jackson Laboratory at Bar Harbor. For
example, Walter C. Stanley at the National Institute of
Mental Health (NIMH) published four fascinating papers on
humans and human interaction as reinforcers for normal and
deprived dogs (Bacon & Stanley, 1963, 1966; Stanley, 1966;
Stanley, Morris, & Trattner, 1965). The oeuvre emanating
from Scott and Fuller’s endeavor is much greater than
indicated by a search for either of their names.
One collaborator who stands out as having made
contributions to the psychological study of dogs beyond
his time at Jackson Laboratory is Michael W. Fox. After
working as a fellow at Jackson Laboratory, he published
an impressive number and breadth of research articles on
canids, balancing between psychology and ethology. His
research included work on development of behavior in dogs,
with attention to critical periods (e.g., Fox, 1964, 1969; Fox
& Stelzner, 1966a, 1966b), on neurological development
and neurochemical substrates of behavior (Agrawal, Fox, &
Himwich, 1967; Fox, 1968), and on the behavior and ecology
of urban dogs (Fox, Beck, & Blackman, 1975). He also
published many books, including Integrative Development
in Brain and Behavior in the Dog (Fox, 1971) and The
Wild Canids: Their Systematics, Behavioral Ecology and
Evolution (Fox, 1975/2009).
Scott and Fuller’s inuence is also seen in other research
lines, such as the work of D. O. Hebb and colleagues at
McGill University, and the cognitive tradition that would
follow. Before discussing those, we make passing mention of
Jackson Laboratory’s collaboration with Guide Dogs for the
Blind of San Rafael, CA. Clarence Pfaffenberger, a member
of the Board of Directors for Guide Dogs for the Blind, and
Scott together published one scientic paper (Pfaffenberger
& Scott, 1959), as well as a volume entitled Guide Dogs
for the Blind, Their Selection, Development, and Training
(Pfaffenberger, Scott, Fuller, Ginsburg, & Bielfelt, 1976).
Pfaffenberger also went on to publish his own book, The
New Knowledge of Dog Behavior (1963/2002), based on his
work with Guide Dogs for the Blind. Based on Scott and
Fuller’s tests, Pfaffenberger developed puppy tests (Weiss &
Greenberg, 1997) to be used as selection tools for breeders,
owners, and trainers. Similar tests in use today to assess
adoptability of shelter dogs or utility of working dogs have
their roots in Pfaffenberger’s work.
History of Dogs 57
Donald O. Hebb and Students
A more direct line of inuence from the Jackson
Laboratory to other dog research comes through the work
of Donald O. Hebb (1904−1985) at McGill University who
attended a conference at Jackson Laboratory and started
experiments on social isolation using Scottish terriers from
Jackson Laboratory. As Scott pointed out (Scott, 1978),
these experiments predated Harry Harlow’s work with social
isolation in primates. Later, a different lab would conduct a
direct replication of Harlow and Zimmerman’s classic study
(1959) of wire vs. cloth surrogate mothers with dogs (Igel &
Hebb and his students and colleagues at McGill produced
several studies investigating the effects of restricted early
experience on later behavior. Puppies were reared to 7½
months of age in different states of deprivation. Conditions
included being reared in a home environment, reared in the
lab but with normal socialization, or reared in the lab but with
no human contact except for routine cleaning and feeding.
Within this last category, the puppies could be more or less
sensory deprived, from living in a wire cage with visual access
to their environment to living in a completely enclosed cage
in which they could not see anything else. The behavioral
effects of these early experiences were investigated using
problem solving behavior (Clarke, Heron, Fetherstonhaugh,
Forgays, & Hebb, 1951; Thompson & Heron, 1954b) and
emotional responding (Melzack, 1954). The methods were
reminiscent of Scott and Fuller’s (1965) work and the
research was largely longitudinal; tests were administered
as the dogs matured, and after certain amounts of time had
elapsed, since the isolated dogs had been placed in normal
environments. The study conrmed the importance of early
experience in later development, a view also championed by
Scott and Fuller: Isolated dogs showed aberrant reactions
to humans and novel environments (freezing), lower rates
of conspecic social behavior, and poorer performance
on the mazes. These differences persisted even after the
experimental dogs had been moved to normal housing. Later
studies expanded and generally supported these ndings
by augmenting the battery of tests to which the dogs were
subjected (Thompson & Heron, 1954b), and the variety of
natural behaviors measured, such as exploration (Thompson
& Heron, 1954a), social behavior (Melzack & Thompson,
1956), and aberrant behavior, such as “whirling,” seen in
the sensory restricted dogs (Thompson, Melzack, & Scott,
During his time in Hebb’s lab, Ronald Melzack expanded
the investigation into social and sensory deprivation in
dogs by investigating emotionality, fear, and pain in
environmentally-restricted and normal dogs, as well as their
ability to learn a shock-avoidance response (Melzack &
Scott, 1957). Not all of the environmentally-restricted dogs
responded to all of the noxious stimuli, and they were also
largely incapable of learning an avoidance response, which
the authors suggested represented an inability to respond
adaptively as a result of their early experience. Additionally,
the authors mentioned other abnormal behavior of the
restricted dogs, including seemingly aimless movement
such as banging into water pipes repeatedly.
To identify whether the behavioral differences in the
isolated dogs were the result of physiological rather than
psychological decits, Melzack (1972) used a visual
discrimination task to investigate differences in deprived
and control dogs. The two groups performed differently
on some tasks, which Melzack attributed to a difference in
central nervous system arousal level, and not physiological
decits directly affecting the ability to discriminate (e.g.,
visual decits). Deprived dogs were highly aroused which
might have interfered with the discrimination performance.
Finally, Melzack and Burns (1965) published a suggestive
study in which electroencephalograms (EEG) and cortical
responses supported the hypothesis that deprived dogs were
unable to lter out irrelevant information in the same way as
control dogs, paralleling later work on dog behavior from an
information processing theory perspective (see subsection
below on Harry Frank).
By the 1950s Pavlovian and Skinnerian research traditions
were well established in North America, and the work of
Scott and Fuller at Jackson Laboratory was under way.
Thus, it is not surprising that research emerged investigating
both conditioning procedures simultaneously, and often
incorporating a developmental approach. In the work of these
scientists, we nd dogs playing a role in the development
of psychology as a scientic eld: Not only were dogs the
subjects of several seminal papers on avoidance conditioning
(e.g., Solomon, Kamin, & Wynne, 1953; Solomon &
Wynne, 1953) and learned helplessness (Overmier &
Seligman, 1967), but the research done with dogs helped
contribute to important theoretical developments of the
time (e.g., Rescorla, 1966; Solomon & Wynne, 1954). We
rst discuss William T. James, a student of the Pavlovian
researcher, Howard S. Liddell, then turn to several students
and collaborators of W. Horsley Gantt, before introducing
new names in the eld: Richard L. Solomon and his students,
and Martin Shapiro.
William T. James
Whereas most early researchers working with dogs
selected dogs for practical reasons, or because they had
characteristics that were not unique to them, William T.
History of Dogs 58
James (1903−1998) was more interested in dogs for their
own sake (Liddell, 1938). His research attempted not only
to answer questions of general psychological interest,
but also questions specic to the understanding of dogs.
Of the researchers discussed in this paper, it can readily
be argued that James’ research provided one of the most
integrative programs, similar in scope to that of the Arkansas
researchers, to be discussed in the following section. James’
research spanned a considerable range of psychological
domains, addressed entirely new questions in regard to dog
behavior, transcended any particular methodological school,
and continued for one of the longest periods of time (per
Psycinfo, his rst paper on dogs was published in 1933 and
his last in 1971), creating one of the few cohesive research
tracks working specically on dogs.
James began working with dogs when he joined an
enormous project of Charles Rupert Stockard (1879−1939),
an anatomist interested in the endocrine effects on
morphology and behavior (Shepard, 2001). Stockard ran a
research “farm” located at Shrub Oak in New York which
housed 500 mostly purebred dogs (James, 1988; Nonidez,
1941) and the products of carefully designed crosses.
James conducted behavioral tests for Stockard using the
conditioned reex techniques he had learned from Liddell.
From the research conducted at the Farm, Stockard wrote
and edited an immense treatise, The Genetic and Endocrinic
Basis for Differences in Form and Behavior (1941). One of
the chapters detailed James’ work at the Farm on conditioned
salivary and avoidance (paw exion due to shock) responses.
Stockard hypothesized that breed differences, even in
behavior, were due to differences in the ductless glands.
Scott and Fuller (1965) critiqued Stockard’s work as
promising for its time but ultimately misguided, given the
inability of his experimental designs to test a Mendelian
inheritance hypothesis. While working at the Farm, James
began research on behavioral differences between various
breeds of dogs, including basset hounds, beagles, English
bulldogs, German shepherds, salukis, St. Bernards, and
terriers, as well as carefully planned crosses of these breeds.
Given his tutelage by Liddell, much of James’ early work
focused on differences between dogs in their response to
Pavlovian conditioning procedures. Based on their reactions
to salivary and motor conditioning experiments, James
identied different typologies of dogs, similar to those
developed by Pavlov (1927).
James identied German shepherds, salukis, and terriers
as highly active (or “excited”) breeds. Conditioned salivary
reexes in these dogs typically had a short delay between
stimulus onset and response, but the response itself was
not large in magnitude. Once a conditioned response was
acquired, however, excited dogs readily generalized,
often making discrimination training challenging. Their
responses, especially to painful stimuli, were also slow to
extinguish (James, 1941a, 1953b). One excited dog emitted
600 paw exions to the CS after the shock contingency had
been turned off and the response still had not extinguished
Basset hounds, beagles, and St. Bernards represented the
other extreme: the highly inactive category. Inactive dogs
readily formed conditioned salivary responses of large
magnitude, but these soon decreased and were delayed
from the onset of the CS. These dogs would often lean on
the harness and would sometimes even fall asleep in the
experimental chamber until the stimulus was introduced
Additionally, excited dogs exed their paws with more
force than did inhibited dogs (James, 1941a), and excited
and inhibited dogs showed different patterns of activity
in a 24-hour cycle, with excited dogs being more active
(James, 1953b). James also reported that the only way he
could modify an excited dog to an inhibited dog or vice
versa was pharmacologically (James & Ginsburg, 1949). A
nal difference he noted between the excited and inhibited
dogs was their response to inescapable shock of increasing
intensities (James, 1943). James raised the question of
whether different types’ had different thresholds for pain or
tolerance of frustration.
James also crossed excited and inhibited dogs. The F1
generation fell as intermediates between the two extreme
parental breeds on the conditioning experiments. The F2
generation, however, showed a greater range, with some dogs
Figure 8. German shepherd in harness used by William T.
James’ in his work at the Cornell Dog Farm. James investi-
gated breed differences in classical conditioning paradigms.
The identity of the man with the dog is unknown. Courtesy
of Hargrett Rare Book and Manuscript Library/University of
History of Dogs 59
performing more like one of the extremes of the purebred
grandparents, or as intermediates like the F1 generation
James’ work on how different dog breeds and individuals
respond to conditioning procedures has implications for
research and practice in applied animal behavior, as well as
suggesting useful ways for assessing behavioral differences
for researchers interested in temperament and personality.
Researchers interested in “trainability” might nd the
differences in generalization and resistance to extinction
particularly useful. Additionally, his work investigating
differences between breeds provides a signicant parallel to
the work we discuss in the following section by the Arkansas
researchers who investigated differences between two strains
of one dog breed.
Moving from a strictly Pavlovian approach, James
incorporated operant research into his work after noting that
“hyperexcitable” types of dogs were sometimes impossible
to train to tolerate the harness and experimental apparatus
(James, 1950). The utility of the operant approach, as
James saw it, was that it allowed the animal freedom of
movement, alleviating the stimuli that elicited resistance
in the hyperexcitable dogs. James’ enthusiasm for such an
approach was apparent in his two publications on how to
construct operant chambers for dogs (James, 1950, 1953a).
He later published a paper on constructing an apparatus for
studying the reinforcing effectiveness of sham vs. direct
stomach feeding on operant responses and reported that only
actual consummatory behavior, and not stomach injection
of food, maintained lever pressing (James, 1963). He also
investigated dogs’ response patterns in an operant chamber
(James, 1954) and found that the presence of the experimenter
or another dog would evoke more responding than when the
dog was alone, again pointing to interesting social relations
that might affect dogs’ performances in experiments.
James also investigated social interactions between
puppies during feeding. He couched these studies in terms
of dominance relations and social facilitation. Using a bowl
from which only one puppy could eat at a time, and two mixed
litters of beagles and terriers, James (1951) found that the
terriers were always the dominant animals in this paradigm,
even when they were smaller than beagles. He further
explored dominance relations using a conditioned salivary
procedure in which two dogs were tested simultaneously
while standing next to each other. The relative status of the
dog determined whether or not it would emit a conditioned
salivary reex (James, 1936).
James also found that food-satiated puppies would eat
signicantly more food when presented with the food again
in the presence of littermates, compared to when the food
was presented again in the absence of other puppies (James,
1953b). However, puppies that were reared alone after
weaning did not show this social facilitation of feeding when
rst fed in a group (James & Gilbert, 1955).
James conducted a separate study on social facilitation
that did not involve feeding behavior and demonstrated a
clever and fascinating research line worthy of more thorough
follow-up (James, 1971). He trained one dog to avoid a US
(spray of water) in the start box of a three-compartment
runway maze and found that a naïve dog would learn the
escape response simply through social facilitation. James
could then extinguish the escape response also through
social facilitation. The decoy dog did not run from the
compartment and the target dog also stopped running,
although this procedure did not fully tease apart extinction
processes from social facilitation. This procedure warrants
further investigation as a tool for studying social learning
Finally, James also investigated the development of
sensory systems and geotropic responses in puppies (James,
1952a, 1952b; 1956), and the effects of different levels of
social deprivation on operant behavior in dogs for which the
reinforcer was visual access to a conspecic (Angermeier &
James, 1961). Here his work overlapped with the work of
Hebb and his students, as well as Scott and Fuller.
James’ expansive work has given us a greater
understanding of individual differences in dogs, as well
as possible breed differences and techniques for further
investigating these differences. On a theoretical level, he
Figure 9. Saluki dogs and a Basset hound in their kennels
at the Cornell Dog Farm, May, 1941. Anatomist Charles
Stockard founded the Cornell Dog Farm and investigated
morphological differences between breeds. William T. James
investigated behavioral differences of these same dogs,
largely using Pavlovian conditioning techniques. Courtesy
of Hargrett Rare Book and Manuscript Library/University
of Georgia Libraries.
History of Dogs 60
expanded Pavlov’s view of individual nervous differences
to differences that extend to the entire animal; for example,
dogs deemed “excitable” through a conditioning procedure
also had higher overall metabolic rates (James, 1941b).
Although James often discussed individual differences
with more of an essentialist slant than might be warranted,
he clearly understood the interaction of the animal and the
environment, and this interaction points to the utility of
studying individual differences in both steady and transitional
states during conditioning. This brief overview indicates the
breadth of James’ work and his substantial contribution to
our understanding of dog behavior and learning.
Pavlovians in Arkansas
A unique research line emerged at the University of
Arkansas in the 1960s and 1970s from a group of Pavlovians
with various ties to Gantt (Reese, 1985; Reese et al., 1987).
Together, Roscoe Dykman, Oddist Murphree, John Newton,
and John Peters established and studied the Arkansas Line
of Nervous Pointer Dogs (nervous dogs) and the normal
counterpart line (normal dogs). The dogs were derived from
two pairs of carefully selected working dogs—one pair of
dogs that performed well in the eld, and one pair of dogs
that exhibited nervous behaviors, including freezing and
cowering in the presence of humans. Such nervous pointers
would occasionally crop up in a litter, but were usually culled
by the breeder (Uhde, personal communication, August 6,
The two lines differentiated rapidly through inbreeding,
with both strains breeding almost completely true (Dykman,
Murphree, & Peters, 1969). For over 18 years the group
developed and investigated differences between the pointer
lines in such a diverse array of approaches that this research
program represents one of the most comprehensive and
integrative research programs on dogs. Their research ranged
from Pavlovian conditioning, operant conditioning, and
behavioral tests from Scott and Fuller (1965) to crossbreeding
the lines to assess possible genetic differences inuencing
the behavioral differences, and to using pharmacological and
behavioral interventions to address the therapeutic potential
for ameliorating anxiety disorders.
Much of the research from this group on the pointers
followed Gantt’s investigations into the effect of person, and
provides an interesting extension of the different types of
dogs investigated by Pavlov (1927) and William T. James
(e.g., James, 1953b). In an unrestrained environment, both
lines of pointer showed the typical decrease in heart rate to
petting and a subsequent increase in heart rate when petting
ended. However, when restrained, the bradycardia to petting
occurred in the normal line but was absent in the nervous
dogs (Thomas, Murphree, & Newton, 1972). Whether these
differences were driven entirely or in part by differences in
amount of motor movement is unknown.
Performance on operant tasks was also evaluated in the
two dog lines. In his 1974 paper, Murphree described in
detail the procedures, apparatus, and special techniques
for the nervous dogs. The lab enlisted Marian and Keller
Breland to train their staff and construct an operant chamber
with a treadle that could be pressed by the dogs, producing
automated delivery of a piece of meat (Murphree, 1974;
Murphree, Dykman & Peters, 1967b). On this apparatus,
many nervous dogs failed to perform at all, and those that
did exhibited erratic, low rates of responding, whereas the
normal dogs showed sustained responding with increasing
rates over the rst several sessions. The authors suggested
that the failure of the nervous dogs was due to their
“overwhelming anxiety or fear in any novel situation”
(Murphree et al., 1967b, p. 416). This conclusion was
supported by the improvement in bar-pressing acquisition
when nervous dogs were administered chlordiazepoxide,
a tranquilizer and anxiolytic drug (Angel, Murphree, &
DeLuca, 1974; Murphree, DeLuca, & Angel, 1974). With
a strong enough motivating operation in place (two to four
days of water deprivation), however, nervous dogs did
learn the response (Newton & Chapin, 1978). Nevertheless,
response rates of the nervous dogs were still lower than that
of normal dogs, and correlated with independent measures
of the degree of nervousness of each dog.
The research group also assessed the two lines of dogs on
tests used by Scott and Fuller (1965). Nervous dogs showed
less exploratory behavior in an empty room, had more
subjects that froze in response to a loud horn, and a greater
percentage of individuals avoided humans (Murphree,
Dykman, & Peters, 1967a). These differences appeared at
2 months of age and persisted through to the last test at 18
months of age. Some differences remained stable, and some
became amplied with increasing age.
The Arkansas researchers also investigated the behavior
of crosses between the two lines. They conducted reciprocal
crosses and tested the crossbred puppies for the effect of
person and on the behavioral tests taken from Scott and
Fuller (1965). Between offspring of reciprocal crosses, there
was no apparent difference. Both crosses behaved nearly
identically to the nervous dogs when assessed for effect of
person (Murphree, Peters, & Dykman, 1967), as well as on
the behavioral tests (Murphree & Newton, 1971; Murphree,
Peters, & Dykman, 1969).
McBryde and Murphree (1974) undertook an investigation
of whether it was possible to change a nervous pointer into
a normal one—a question with both theoretical and practical
importance. They noted that the nervous pointers still
History of Dogs 61
retained the normal behavior of “pointing” to the scent of
quail. Using this kernel of normal behavior, the reinforcer
of access to a quail wing, and sometimes a normal dog
for social facilitation of responses, the authors trained ve
nervous pointers to hunt successfully, including retrieval,
which necessarily involved approaching a human.
This behavioral improvement did not, however, generalize
to the standard behavioral tests administered to the dogs.
The authors broached a question relevant to all studies in
which behavioral or temperament assessments are made:
“How well do these behavioral tests represent the dog?”
(McBryde & Murphree, 1974, p. 83) and noted that the dogs
appeared so normal when reviewing videos taken of their
performances hunting that, “one easily gets the feeling that
the rehabilitation has been very basic and that measures of
animal behavior in an articial setting may be inappropriate”
Work on the pointers continued when some of the dogs
from the two strains were passed on to Thomas Uhde, then
at the NIMH, who had become interested in the dogs as a
potential model for human panic disorder (Uhde, personal
communication, August 6, 2010). Research by Uhde and
his collaborators focused on the physiological (e.g., Klein,
Tomai, & Uhde, 1990) and neurophysiological differences
(e.g., Gurguis, Klein, Mefford, & Uhde, 1990) that underlay
the behavioral differences between the normal and nervous
lines. The NIMH group also included further behavioral
measurements (e.g., Klein & Uhde, 1988), and discovered
that many of the nervous dogs were deaf, but demonstrated
that this deafness was not a factor in their behavioral
differences from normal dogs (Klein, Steinberg, Weiss,
Matthews, & Uhde, 1987).
The Arkansas pointer lines formed a unique research
paradigm that contributed richly to our understanding of the
genetic and environmental bases of behavioral difference
between dogs, including the development of potential
therapies for shy or nervous dogs. The researchers involved
presented a very integrative research program, systematically
investigating the dogs’ behavior using a variety of
techniques. When combined with the work of William T.
James, a fascinating picture of individual, strain, and breed
differences begins to emerge, and both research programs
have provided a multitude of ways to usefully investigate
these differences. The pointer lines, unfortunately, were
terminated by the 1990s (Uhde, personal communication,
August 6, 2010).
Richard L. Solomon and Students
In the 1950s Richard L. Solomon initiated the study
of avoidance behavior in dogs. At that time, avoidance
behavior was an area of active controversy between
behaviorist and cognitive researchers (Sidman, 1954; and
see Herrnstein, 1969, for a history of the eld). Solomon
was a student of Harold Schlosberg who worked on white
rats (e.g., Schlosberg, 1934). Why Solomon switched to
dogs is unclear, but dogs were probably convenient subjects,
and readily emitted observable escape responses that were
well suited to shuttleboxes. Solomon’s research using dogs
would have a signicant impact on the eld of psychology
(Rescorla, 1995). Solomon helped propose the two-process
theory of conditioning (Rescorla & Solomon, 1967), learned
helplessness (Overmier, 1996), and the opponent-process
theory of acquired motivation (LoLordo & Seligman, 1997),
all using results obtained from dogs.
Solomon published voraciously on avoidance
conditioning in dogs. With Lyman Wynne, Solomon
reported on avoidance conditioning in normal dogs and dogs
whose autonomic nervous system had been surgically or
pharmacologically disrupted (Solomon & Wynne, 1955). In
1953, Solomon and colleagues published their rst papers
on avoidance conditioning on dogs using a shuttlebox
preparation (Solomon et al., 1953; Solomon & Wynne,
1953). A theoretical paper soon followed (Solomon &
Wynne, 1954) in which they expanded on the two-process
theory of conditioning. In this theory, emotional fear
responses result from Pavlovian conditioning and skeletal
responses from operant conditioning, which are maintained
through fear reduction.
One of the most well-known research paradigms to emerge
from Solomon’s lab was that of Bruce Overmier and Martin
Seligman who coined the term “learned helplessness,”
(1967). In this work, Overmier and Seligman investigated
the effects of prior shock history on a dog’s behavior in
later avoidance training. Dogs that had previously been
administered unavoidable shocks showed three changes in
behavior during the avoidance learning phase: “(1) many
failures to initiate any escape response—that was interpreted
as a ‘motivational’ decit; (2) when an escape response did
occur, the animal did not show any benecial learning from
the experience—that was interpreted as an ‘associative’
decit; and (3) a marked quiet passivity in the presence of
the continuing shocks that were not being escaped—that
was interpreted as an ‘emotional’ decit” (Overmier, 1996,
p. 332). Steven Maier joined Seligman in identifying that
the uncontrollability of the initial shocks was a critical
feature for producing the later decit in avoidance learning
(Seligman & Maier, 1967). Overmier (1996) indicated that
this led the researchers to make an appeal for contingency
analysis and a contingency-based learning theory in their
theoretical paper on these phenomena (Maier, Seligman, &
Solomon, 1969). The learned helplessness research showed
History of Dogs 62
direct applicability to human behavior, serving as a potential
model for clinical depression and post-traumatic stress
disorder (e.g., Maier, 2001). This human connection was
always an emphasis for Solomon (Overmier, 1996).
Abraham Black expanded the laboratory’s interest in
avoidance conditioning by investigating the extinction
of avoidance responses in curarized dogs (Black, 1958).
Avoidance-trained dogs that had experienced extinction
while curarized required many fewer trials for the avoidance
response to extinguish later when they were not curarized,
as compared to dogs that had not been administered curare
and could continue to make the avoidance response, even
though the contingency had been removed. Black later
pursued an impressive research program involving dogs
in which he studied the interaction of conditioning, overt
behavior, and brain activity. In this line, he experimentally
demonstrated reinforcement of specic brain patterns using
either avoidance learning (negative reinforcement) or
brain stimulation (positive reinforcement) (Black, 1971).
In 1970, Black, Young and Batenchuk reported that they
successfully trained pharmacologically paralyzed dogs to
emit an operant avoidance response of either more or fewer
theta waves in their hippocampus. Furthermore, when the
dogs were presented with the discriminative stimulus when
they were not paralyzed, the effects of the prior learning
manifested themselves in skeletal responses. Reversing
the direction of the question, Black and Young (1972)
additionally demonstrated that reinforcing overt behaviors
could simultaneously select different brain wave patterns.
Robert Rescorla also earned his Ph.D. under Solomon
investigating the effects of overlaying a Pavlovian fear
conditioning procedure on a dog that was already responding
on an unsignaled, or Sidman, avoidance schedule (Rescorla,
1967a). This work again investigated the interplay between
operant and Pavlovian conditioned responses. He followed
this with research aimed at dissecting the interaction of
an operant Sidman avoidance schedule and a Pavlovian-
conditioned fear response (Rescorla, 1968), as well as the
establishment of a positive reinforcer through contrast with
shock (Rescorla, 1969). Rescorla’s work on dogs also led
to a seminal methodological paper in which he took issue
with the then current control procedures used in Pavlovian
conditioning (Rescorla 1967b). The papers Rescorla
produced with dogs and the theorizing that came out of that
work were of inestimable value.
Without doubt, the work on dogs in Richard Solomon’s
lab moved the eld of psychology forward. Not only did
Solomon’s lab tackle important theoretical issues of the
day, they produced performances in dogs that resulted in
meaningful and powerful advances in our understanding
of both dog and human behavior. Moreover, their thorough
and methodologically precise work serves as a beacon for
today’s psychologists who use dogs as subjects.
Martin M. Shapiro
Whereas Solomon and colleagues’ work used more of
Bekhterev’s preparation (shock) to produce conditioned
responses, Martin M. Shapiro used Pavlov’s salivary
preparation when investigating the interaction of operant and
Pavlovian conditioning (Shapiro, 1960a, 1960b). Shapiro
used dogs’ salivary response in an operant preparation
in order to investigate the potential articiality of the
distinction between operant and Pavlovian conditioning
(Shapiro, personal communication, August 12, 2010).
Shapiro, Miller, and Bresnahan (1966) investigated the
relationship between the discriminative and Pavlovian
conditioned functions of a stimulus. Similarly, Shapiro,
Mugg, and Ewald (1971) investigated the effect on operant
behavior when a CS correlated more or less highly with
the presentation of the US, which simultaneously served
as the reinforcer for the operant response. Finally, Shapiro
and Herendeen demonstrated the possibility of inhibiting a
conditioned response by adding a response contingency to
the CS-US delivery (Herendeen & Shapiro, 1975; Shapiro &
Herendeen, 1975). In this experiment, dogs were only given
food after presentation of a conditioned stimulus if they did
not salivate during the 10-second CS. This demonstration
raises questions of the inuence of operant conditioning
procedures on responses typically thought of as mainly
subject to Pavlovian conditioning control.
The nal tradition in psychology that we will follow here
is the cognitive tradition into which much of the current
work on dogs falls. Two research lines within this tradition
are apparent: that of Harry Frank, and that emanating from
a group of francophone Canadians. We will rst consider
Harry Frank’s research, which he credits as having been
an extension of Scott and Fuller’s work (Frank, personal
communication, July 29, 2010). Frank’s research on dogs is
intimately tied to an interest in dogs themselves and started
on a personal note: After acquiring a malamute/elk hound
cross in graduate school, and being erroneously informed
that malamutes are at least 1/8 wolf, he started noticing
similarities to wolves in his dog’s behavior. This turned
itself into a research line when he later discovered that the
Inuit tribe known for breeding the malamute was diligent
about keeping wolf blood from entering their breeding lines.
He was eventually persuaded to take a wolf pup to rear,
and from those experiences was born a research program
investigating the comparative development of wolves and
History of Dogs 63
dogs from a cognitive perspective.
Frank (1980) hypothesized that natural selection favored
group hunting in wolves, so that they evolved a cognitive
subsystem, including the capacity for foresight, mental
representation, and the beginnings of an understanding of
means-ends relations. Dogs, on the other hand, having been
relieved of many of the selective pressures to which wolves
are exposed, lost this cognitive complexity, and instead were
selected for tractability in the domestication process (Frank
& Frank, 1987). From this hypothesis, Frank articulated
a number of predictions testable by comparisons between
wolves and dogs. He predicted “that dogs should perform
better than wolves on learning tasks in which a) cues are
arbitrarily selected by the experimenter, b) reinforcement is
administered by the experimenter, and c) the to-be-learned
behavior has no perceptible, functional connection with the
outcome” (Frank & Frank, 1987, p. 144). Wolves on the other
hand should perform better on “problem solving tasks” that
required “cognitive processes, such as foresight, planning,
mental representation (imagery), and serial organization of
behavior” (Frank & Frank, 1987, p. 144).
The Franks conducted a longitudinal study comparing
the behavior of socialized wolves and malamute dogs on a
battery of performance tests starting when the pups were 6
weeks old and continuing until they were 25 weeks old. The
training tasks represented a subset of those used by Scott
and Fuller (1965) and were designated as either training or
problem solving tasks.
While Frank and Frank generally found support for their
initial hypothesis that dogs performed better on what the
authors designated as training experiments and the wolves
would perform better on those tasks designated as problem
solving tasks, they also astutely noted several caveats in their
1987 summary of their work (Frank & Frank, 1987), some
of which apply as much to current work as to the studies
at which they were originally directed. The rst caveat was
that attempts to equalize the early experience of the dogs
and wolves were not wholly successful. For example, wolf
pups were much more active during their rst six weeks of
life, which might have affected their behavioral repertoires
in unknown ways and brought them into contact with
contingencies not encountered by the dog pups. Thus, it is
possible that the higher level differences (such as problem
solving ability) found in older puppies are the products of
earlier and possibly lower-level psychological differences
and that the cognitive differences noted might be explained
without necessitating appeals to evolutionary selection of
higher psychical levels to explain those differences.
Second, this research involved only one breed of dog that
has been articially selected for specic traits and other
breeds might perform differently. Indeed, Frank and Frank
(1987) noted that Scott and Fuller’s basenjis performed nearly
as well and sometimes better than the wolves on problem-
solving tasks, suggesting that there might not be wolf−dog
differences on such tasks at all. Third, comparisons were only
made until the pups were 25 weeks old, which might or might
not be equivalent in dogs and wolves and the differences seen
might be a result of differential rates of development and
not indicative of differences that would be found in mature
animals. Fourth, the differences seen could be explained by
differences that do not impinge on information processing,
such as sensitivity to the stress of connement, or sensitivity
to contingencies of reinforcement or punishment (Frank
& Frank, 1983), especially those administered by humans,
rather than a difference in some cognitive capacity. We
will add a nal caveat, which is that the distinction between
training and problem solving tasks is not well-dened and
thus amenable to arbitrary manipulation: the authors noted
that some tasks used could be interpreted as belonging to
either the training task designation or the problem-solving
designation (Frank, Frank, Hasselbach, & Littleton, 1989),
meaning that results would either conrm or disconrm the
hypothesis depending on how the task was viewed.
The authors were shrewd observers of their subjects’
behaviors and noted small behavioral differences on tasks
that might have contributed to observed differences in the
overall task performance, such as the observation that wolf
pups watched the food bowl and dog pups watched the
manipulandum on the manipulation task (Frank & Frank,
1987). For this reason alone, reading Frank and Frank’s
work is useful for guidance in identifying more discrete and
well-dened behavioral differences between wolves and
Figure 10. Harry Frank with wolf puppies from his research
project in Connecticut. Frank investigated cognitive differ-
ences between wolves and Malamute dogs. Photo taken by
Erich Klinghammer. Copyright Harry Frank. Used with per-
History of Dogs 64
dogs that might be fruitfully investigated.
Frank and Frank placed their work in both information
processing (Frank & Frank, 1987) and Piagetian stage
frameworks (Frank & Frank, 1985). This tracked the
departure of psychology in general from a behavioral to a
cognitive approach. Although conditioning was addressed in
this formulation, the emphasis was placed on the prepotent
inuence of genetic factors acquired through natural
selection. The program eventually ended, but the inuence
of this research line can certainly be seen in much of the
current research on canid cognition.
François Y. Doré and Students
The second strand of the early North American foray into
canid cognition comes from a French lineage that found
its way to Canadian universities. The major progenitor
of the research and the students who currently continue
this research is François Y. Doré. Doré published several
papers outlining a neo-Piagetian approach to comparative
cognition (Doré, 1991; Dumas & Doré, 1989, 1991). It
would be this Piagetian approach that Doré and his students,
including Sylvain Gagnon and Sylvain Fiset, would take in
investigating dog behavior. Gagnon demonstrated that dogs
were successful on invisible displacement tasks, but not
as successful as on visible displacement (Gagnon & Doré,
1992). The dogs’ success rate decreased as a function of the
complexity of displacement and the delay between when
the displacement occurred and when the dog was allowed to
begin its search (Gagnon & Doré, 1994). Finally, Gagnon and
Doré (1994) reported a cross-sectional study to identify the
age at which most dogs begin to solve invisible displacement
problems, allowing for a preliminary understanding of
the development of that complex behavior. Doré and his
students have produced a cohesive oeuvre using Piaget’s
framework applied to animals and have demonstrated the
potential fruitfulness of investigating complex behavioral
repertoires in animals, and in dogs in particular.
The history of dogs as psychological subjects tells the tale
of psychology as a science. We see the rise and fall of different
theories and different areas of inquiry; we see the issues
that were of greatest import or controversy in psychology
at a given time; and we see the immensity of work on dogs
that has often gone unnoticed. The research reviewed here
was restricted to that emanating from North American
laboratories, and there is surely an equivalent wealth of
experimental research on dogs from other countries. For
example, a search by the authors on the term “dog” using the
Psycinfo database returned hundreds of papers in Russian. It
is our hope that this paper will encourage others to delve into
the great research resources not covered herein.
We have traced work using dogs as experimental
psychological subjects in research originating from a
Pavlovian tradition, and an operant tradition using dogs, to
a developmental approach, through a cognitive approach, as
well as research programs that have taken a clearly integrative
approach combining several of these lines. Along the way,
we have delved into research that clearly bears on current
research interests, as well as introducing old discoveries
worthy of more thorough examination, and suggesting new
phenomena to explore.
Some of the research ndings reviewed here concerned
general issues, such as the investigation of avoidance
conditioning by Richard Solomon and his students, or the
effects of early deprivation from Donald Hebb and his
students. Others investigated questions more particular to
dogs, such as the relationship between humans and dogs in
W. Horsley Gantt’s effect of person, or the social relations
between dogs examined by William T. James. Additionally,
much of this research addressed behavioral and cognitive
differences between individuals, strains, and breeds of dogs,
and the development of these differences.
Despite the fact that much of the research reviewed here
is between 40 and 70 years old, it has much to offer current
researchers. The results of many studies still stand today,
and others present well-documented phenomena, even
as their theoretical explanations might have fallen away.
Such studies would benet from reinterpretation in today’s
understanding of behavioral and cognitive processes. For
example, the work of Gantt and the effect of person might
be usefully situated in our advancing understanding of
Pavlovian conditioning effects, potentially even from the
work of other researchers mentioned in this paper, such as
Furthermore, given the current and historical interest
in the provenance of behavioral and cognitive differences
between individuals, it is necessary to reassess some of the
conclusions from the research we reviewed in light of our
greater understanding of the interaction between genetics
and the environment. For example, the Arkansas researchers
concluded that the nervous behaviors in the pointer line
were likely a result of simple Mendelian inheritance of
a dominant trait. Given the benet of further advances
in our understanding of genetics, this conclusion seems
overly simplistic. The nature-nurture dichotomy has been
discarded, and the inuence of both the environment and
genetics on producing morphological, physiological, and
behavioral traits is undeniable. In particular, the Arkansas
researchers were working before two modern views of
inheritance were available: (a) epigenetics, which has
History of Dogs 65
recently shown nongenetic inheritance of behavior passed on
transgenerationally from both the maternal (e.g., Champagne
& Meaney, 2007) and even the paternal side (e.g., Alter,
Gilani, Champagne, Curley, Turner, & Hen, 2009) allowing
for a more Lamarckian mode of inheritance; and (b) the
concept of the “ontogenetic niche” (West & King, 1987) in
which inheritance is viewed as including not only genes, but
environmental variables such as resources, and peers. In fact,
their own study (McBryde & Murphree, 1974) investigating
therapeutic interventions for nervous dogs, pointed to the
inuence of the environment.
Finally, some of the phenomena we examined transcend
in importance any particular tradition and bear on the study
of dogs in general. For example, Gantt’s research on the
Pavlovian phenomenon of effect of person is relevant to
all researchers working with dogs. While such an effect
relates to research themes that seek to elucidate the dog-
human relationship, it also serves as a cautionary note for
experimenters when designing studies in which the dog is in
contact with a human. Regardless of whether the research is
addressing social issues between human and dogs, or asking
entirely different questions, the presence of a human, the
interaction with the human, and the dog’s history with the
human might differentially affect results. James’ observation
of similar effects of a human in an operant situation points to
the generality of this concern.
In reviewing the considerable work from North America
that has already been conducted in dogs, we hope that
forgotten works have been dusted off that can inform current
research, that forgotten questions might be reinvigorated and
might nd new life with new researchers, and that current
researchers nd themselves part of a larger, longer-lived,
and vibrant tradition of using dogs as psychological subjects.
Many of the reasons given by these earlier researchers for
their choice of subject species are the same reasons current
researchers cite in describing their choice of using dogs:
their complex behavior, their sociality, and the breadth of
morphological and behavioral differences.
This history has demonstrated the huge breadth of research
in dogs, pulling from many traditions in psychology, all
of which contribute to our understanding of dog behavior
and cognition, regardless of the tradition in which we work
individually. To this end, we turn to Paul Feyerabend and his
circumspect view on the interaction of science and history:
“It is clear that the increasing separation of the history, the
philosophy of science and of science itself is a disadvantage
and should be terminated in the interest of all these three
disciplines. Otherwise we shall get tons of minute, precise,
but utterly barren results” (Feyerabend, 2010, p. 28). None of
the areas of psychological research focusing on dogs seems
immune to the critique that, as a eld, we are largely unaware
of the substantial work that has gone before that of today.
We have endeavored to bring to light the multifaceted, and
often forgotten history of dogs as psychological subjects. In
so doing, we hope that our eld can understand its already
well-established roots and synthesize the historical and
modern work, and avoid the pitfalls foretold by Feyerabend
of an ahistorical science.
Agrawal, H. C., Fox, M. W., & Himwich, W. A. (1967).
Neurochemical and behavioral effects of isolation-rearing
in the dog. Life Sciences, 6(1), 71-78. doi:10.1016/0024-
Alter, M. D., Gilani, A. I., Champagne, F. A., Curley, J. P.,
Turner, J. B., & Hen, R. (2009). Paternal transmission of
complex phenotypes in inbred mice. Biological Psychiatry,
66, 1061-1066. doi:10.1016/j.biopsych.2009.05.026
Anderson, S. L., & Gantt, W. H. (1966). The effect of person
on cardiac and motor responsivity to shock in dogs.
Conditional Reex, 1(3), 181-189.
Angel, C., Murphree, O. D., & DeLuca, D. C. (1974). The
effects of chlordiazepoxide, amphetamine, and cocaine
on bar-press behavior in normal and genetically nervous
dogs. Diseases of the Nervous System, 35(5), 220-223.
Angermeier, W. F., & James, W. T. (1961). The inuence
of early sensory-social deprivation on the social operant
in dogs. Journal of Genetic Psychology: Research and
Theory on Human Development, 99, 153−158.
Bacon, W. E., & Stanley, W. C. (1963). Effect of deprivation
level in puppies on performance maintained by a
passive person reinforcer. Journal of Comparative and
Physiological Science, 56(4), 783-785.
Bacon, W. E., &Stanley, W. C. (1966). Resistance to
extinction after continuous and partial-reinforcement in
basenji puppies. Psychonomic Science, 4(9), 327-328.
Bailey, R. E., & Bailey, M. B. (2006). Patient Like the
Chipmunks: The Story of Animal Behavior Enterprises
(Long Version). Hot Springs, Arkansas: Animal Behavior
Black, A. H. (1958). The extinction of avoidance responses
under curare. Journal of Comparative and Physiological
Science, 51, 519-524.
Black, A. H. (1971). The direct control of neural processes
by reward and punishment. American Scientist, 59, 236-
Black, A. H., & Carlson, N. J. (1959). Traumatic avoidance
learning: A note on intertrial-interval responding. Journal
of Comparative and Physiological Science, 52, 759-760.
Black, A. H., & Young, G. A. (1972). Electrical activity
of the hippocampus and cortex in dogs operantly trained
to move and to hold still. Journal of Comparative and
Physiological Science, 79, 128-141.
History of Dogs 66
Black, A. H. Young, G. A., & Batenchuk, C. (1970).
Avoidance training of hippocampal theta waves in
axedilized dogs and its relation to skeletal movement.
Journal of Comparative and Physiological Science, 70,
Breland, K., & Breland, M. (1966). Animal Behavior. New
York, NY: Macmillan Company.
Browne, J. (2003). Charles Darwin: The power of place.
Princeton, NJ: Princeton University Press.
Champagne, F. A., & Meaney, M. J. (2007). Transgenerational
effects of social environment on variations in maternal
care and behavioral response to novelty. Behavioral
Neuroscience, 121, 1353-1363. doi:10.1037/0735-
Clarke, R. S., Heron, W., Fetherstonhaugh, M. L., Forgays,
D. G., & Hebb, D. O. (1951). Individual differences in
dogs: Preliminary report on the effects of early experience.
Canadian Journal of Psychology, 5(4), (150-156).
Compton, J. M., & Scott, J. P. (1971). Allelomimetic behavior
system: Distress vocalization and social facilitation
of feeding in telomian dogs. Journal of Psychology:
Interdisciplinary and Applied, 78(2), 165-179. doi:10.10
Darwin, C. (1859). On the Origin of Species by Means of
Natural Selection. London: John Murray.
Darwin, C. (1871). The Descent of Man; and Selection in
Relation to Sex. London: John Murray.
Darwin, C. (1872). The expression of emotions in man and
animals. London: John Murray. doi:10.1037/10001-000
Dewsbury, D. A. (2009). Origins of behavior genetics: The
role of the Jackson Laboratory. Behavioral Genetics,
35(1), 1-5. doi:10.1007/s10519-008-9240-1
Dewsbury, D. A. (in press-a). John Paul Scott (1909−2000)
and the study of genetics, development, and social
behavior. In W. E. Pickren, D. A. Dewsbury, & M.
Wertheimer (Eds.), Portraits of Pioneers in Developmental
Psychology. New York, NY: Taylor & Francis.
Dewsbury, D. A. (in press-b). A history of the behavior
program at the Jackson Laboratory: An overview. Journal
of Comparative Psychology. doi:10.1037/a0021376
Doré, F. Y. (1991). A neo-Piagetian theory can contribute to
comparative studies of cognitive development. Behavioral
and Brain Sciences, 14(2), 368-370.
Dumas, C., & Doré, F. (1989). Cognitive development in
kittens (Felis catus): A cross-sectional study of object
permanence. Journal of Comparative Psychology, 103(2),
Dumas, C., & Doré, F. (1991). Cognitive development
in kittens (Felis catus): An observational study of
object permanence and sensorimotor intelligence.
Journal of Comparative Psychology, 105(4), 357-365.
Dykman, R. A., Murphree, O. D., & Peters, J. E. (1969).
Like begets like: Behavioral tests, classical autonomic
and motor conditioning, and operant conditioning in two
strains of pointer dogs. Annals of the New York Academy of
Sciences, 159, 976-1000. doi:10.1111/j.1749-6632.1969.
Feyerabend, P. (2010). Against Method (4th ed.). London,
Fox, M. W. (1964). The ontogeny of behaviour and
neurologic responses in the dog. Animal Behaviour, 12(2-
3), 301-310. doi:10.1016/0003-3472(64)90016-8
Fox, M. W. (1968). Neuronal development and ontogeny of
evoked potentials in auditory and visual cortex of the dog.
Electroencephalography & Clinical Neurophysiology,
24(3), 213-226. doi:10.1016/0013-4694(68)90002-3
Fox, M. W. (1969). Behavioral effects of rearing dogs with
cats during the ‘critical period of socialization.’ Behaviour,
35(3-4), 273-280. doi:10.1163/156853969X00242
Fox, M. W. (1971). Integrative Development in Brain and
Behavior in the Dog. Chicago, IL: University of Chicago
Fox, M. W. (2009). Wild canids: Their systematics,
behavioral ecology, and evolution. Wenatchee, WA:
Dogwise Publishing. (Original work published 1975)
Fox, M. W., Beck, A. M., & Blackman, E. (1975). Behavior
and ecology of a small group of urban dogs (Canis
familiaris). Applied Animal Ethology, 1(2), 119-137.
Fox, M. W., & Stelzner, D. (1966a). Approach/withdrawal
variables in the development of social behaviour in the
dog. Animal Behaviour, 14(2-3), 362-266. doi:10.1016/
Fox, M. W., & Stelzner, D. (1966b). Behavioural effects of
differential early experience in the dog. Animal Behaviour,
14(2-3), 273-281. doi:10.1016/S0003-3472(66)80083-0
Frank, H. (1970). The interrelations among three personality
systems. Dissertation Abstracts International, 30(10-A),
Frank, H. (1980). Evolution of canine information processing
under conditions of natural and articial selection.
Zeitschrift für Tierpsychologie, 53, 389-399.
Frank, H., & Frank, M. (1983). Inhibition training in
wolves and dogs. Behavioural Processes, 8, 363-377.
Frank, H., & Frank, M. (1985). Comparative manipulation-
test performance in ten-week-old wolves (Canis lupus)
and Alaskan malamutes (Canis familiaris): A Piagetian
interpretation. Journal of Comparative Psychology, 99,
Frank, H., & Frank, M. (1987). The University of Michigan
canine information-processing project (1979-1981). In
H. Frank (Ed.), Man and Wolf: Advances, Issues, and
Problems in Captive Wolf Research (pp. 143 – 167),
History of Dogs 67
Dordrecht, Netherlands: Dr W. Junk Publishers.
Frank, H., Frank, M., Hasselbach, M., & Littleton, D. M.
(1989). Motivation and insight in wolf (Canis lupus)
and Alaskan malamute (Canis familiaris): Visual
discrimination learning. Bulletin of the Psychonomic
Society, 27, 455-458.
Frank, H., &Tubbs, R. (1973). Relationship of rod-and-frame
test performance to two dimensions of the personality
assessment system. Perceptual and Motor Skills, 37, 747-
Frank, Harvey, & Verdun (2000). American responses to
ve categories of shame in Chinese culture: A preliminary
cross-cultural construct validation. Personality and
Individual Differences, 28, 887-896. doi:10.1016/S0191-
Fuller, J. L. (1948). Individual differences in the reactivity
of dogs. Journal of Comparative and Physiological
Psychology, 41(5), 339-347. doi:10.1037/h0057680
Gagnon, S., & Doré, F. (1992). Search behavior in various
breeds of adult dogs (Canis familiaris): Object permanence
and olfactory cues. Journal of Comparative Psychology,
106(1), 58-68. doi:10.1037/0735-7036.106.1.58
Gagnon, S., & Doré, F. (1994). Cross-sectional study of
object permanence in domestic puppies (Canis familiaris).
Journal of Comparative Psychology, 108(3), 220-232.
Gantt, W. H. (1944). Experimental Basis for Neurotic
Behavior: Origin and Development of Articially
Produced Disturbances of Behavior in Dogs. New York,
NY: Paul B. Hoeber. doi:10.1037/11517-000
Gantt, W. H. (1973). Reminiscences of Pavlov. Journal
of Experimental Analysis of Behavior, 20, 131-136.
Gantt, W. H., Newton, J. E. O., Royer, F. L. & Stephens,
J. H. (1966). Effect of person. Conditional Reex, 1(1),
Gillaspy, J. A., Jr., & Bihm, E. M. (2002). Marian Breland
Bailey (1920-2001). American Psychologist, 57(4), 292-
Goodwin, C. J. (2004). A History of Modern Psychology
(2nd ed.). Hoboken, NJ: Wiley and Sons, Inc.
Gurguis, G. N., Klein, E., Mefford, I. N., & Uhde, T. W.
(1990). Biogenic amines distribution in the brain of
nervous and normal pointer dogs: A genetic animal model
of anxiety. Neuropsychopharmacology, 3(4), 297-303.
Gurski, J. C., Davis, K. &, Scott, J. P. (1980). Interaction of
separation discomfort with contact comfort and discomfort
in the dog. Developmental Psychobiology, 13(5), 463-467.
Hare, B., & Tomasello, M. (1998). Communication of food
location between humans and dogs (Canis familiaris).
Evolution of Communication, 2, 137-159. doi:10.1075/
Hare, B., & Tomasello, M. (2005). Human-like social
skills in dogs? Trends in Cognitive Sciences, 9, 439-444.
Harlow, H. & Zimmerman, R. S. (1959). Affectional
responses in the infant monkey. Science, 130, 421-432.
Harvey, A. M. (1995). W. Horsley Gantt--A legend in his
time. Integrative Physiological and Behavioral Science,
30, 237-243. doi:10.1007/BF02698578
Herendeen, D. L., & Shapiro, M. M. (1975). Extinction
and food-reinforced inhibition of conditioned salivation
in dogs. Animal Learning and Behavior, 3(2), 103-106.
Herrnstein, R. J. (1969). Method and theory in the study
of avoidance. Psychological Review, 76(1), 49-69.
Horvath, M. (Ed.) (1987). Publications of W. Horsley Gantt.
Integrative Physiological and Behavioral Science, 22, 52-
74. DOI: 10.1007/BF02700417
Hull, C. L., Livingston, J. R., Rouse, R. O., & Barker,
A. N. (1951). True, sham, and esophageal feeding as
reinforcements. Journal of Comparative and Physiological
Psychology, 44(3), 236-245. doi:10.1037/h0063258
Igel, G. J., & Calvin, A. D. (1960). The development
of affectional responses in infant dogs. Journal of
Comparative and Physiological Psychology, 53(3), 302-
James, W. T. (1936). The effect of the presence of a second
individual on the conditioned salivary response in dogs
of different constitutional types. Pedagological Seminary
and Journal of Genetic Psychology, 49, 437-449.
James, W. T. (1941a). Experimental observations indicating
the signicance of work on conditioned motor reactions.
Journal of Comparative Psychology, 32(2), 353-366.
James, W. T. (1941b). Morphological form and its relation
to behavior. In The Genetic and Endocrinic Basis for
Differences in Form and Behavior (525-643). C. R.
Stockard (ed.). Philadelphia, PA: Wistar Institute of
Anatomy and Biology.
James, W. T. (1943). The formation of neurosis in dogs
by increasing the energy requirement of a conditioned
avoiding response. Journal of Comparative Psychology,
36(2), 109-124. doi:10.1037/h0054870
James, W. T. (1950). Operant conditioning apparatus for
dogs. Journal of General Psychology, 43, 143-145. doi:10
James, W. T. (1951). Social organization among dogs
of different temperaments, terriers and beagles, reared
together. Journal of Comparative and Physiological
Psychology, 44(1), 71-77. doi:10.1037/h0061218
James, W. T. (1952a). Observations on behavior of new-born
puppies: Methods of measurement and types of behavior
History of Dogs 68
involved. Pedagological Seminary and Journal of Genetic
Psychology, 80, 65-73.
James, W. T. (1952b). Observations on behavior of new-born
puppies: II. Summary of movements involved in group
orientation. Journal of Comparative and Physiological
Psychology, 45(4), 329-335. doi:10.1037/h0062288
James, W. T. (1953a). An improved operant conditioning
apparatus for dogs. Journal of General Psychology, 48,
James, W. T. (1953b). Morphological and constitutional
factors in conditioning. Annals of New York Academy of
Sciences, 56, 171-183. doi:10.1111/j.1749-6632.1953.
James, W. T. (1956). The geotropic reaction of newborn
puppies. Journal of Genetic Psychology: Research and
Theory on Human Development, 89, 127-130.
James, W. T. (1963). An analysis of esophageal feeding as a
form of operant reinforcement in the dog. Psychological
Reports, 12, 31−39.
James, W. T. (1971). A methodological study of conditioning
and extinction of an avoidance response via social
facilitation in dogs. Journal of General Psychology, 84(2),
James, W. T. (1988). Autobiography of William T. James.
William T. James Papers (1927-1998) (Box 17), Hargrett
Library (UGA 04-024/ UGA 05-036), Athens, GA.
Courtesy of Hargrett Rare Book and Manuscript Library/
University of Georgia Libraries.
James, W. T., & Gilbert, T. F. (1955). The effect of social
facilitation on food intake of puppies fed separately and
together for the rst 90 days of live. British Journal of
Animal Behaviour, 111, 131-133. doi:10.1016/S0950-
James, W. T., & Ginsburg, B. E. (1949). The effect of
prostigmine on the conditioned response of ‘inhibited
dogs.’ Journal of Comparative and Physiological
Psychology, 42(1), 6-11. doi:10.1037/h0061847
Klein, E., Steinberg, S. A., Weiss, S. R. B., Matthews, D.
M., & Uhde, T. W. (1987). The relationship between
genetic deafness and fear-related behaviors in nervous
pointer dogs. Physiology & Behavior, 43(3), 307-312.
Klein, E. H., Tomai, T., & Uhde, T. W. (1990). Hypothalamo-
pituitary-adrenal axis activity in nervous and normal
pointer dogs. Biological Psychiatry, 27(7), 791-794.
Klein, E., & Uhde, T. W. (1988). 24-hour motor activity
in genetically nervous and normal pointer dogs.
Biological Psychiatry, 23(5), 523-527. doi:10.1016/0006-
Köhler, W. (1999). The Mentality of Apes. E. Winter
(Trans.). Retrieved from http://books.google.com/books
(Original work published 1925)
Kubinyi, E., Virányi, Z., & Miklósi, Á. (2007). Comparative
social cognition: From wolf and dog to humans.
Comparative Cognition & Behavior Reviews, 2(1), 26-46.
Liddell, H. S. (1938, March 24). [Letter to W. T. James].
Howard S. Liddell Papers, #14-23-1466 (Box 2, Folder
8), Division of Rare and Manuscript Collections, Cornell
Liddell, H. S. (1958). History and prospects of the behavior
farm laboratory at Cornell University. In H. Gantt (Ed.)
The Physiological Basis of Psychiatry. Springeld, IL:
Chas C Thomas.
Lindsley, O. R. (1956). Conditioned suppression of behavior
in the dog and some sodium pentobarbital effects.
(Unpublished doctoral dissertation). Harvard University,
Lindsley, O. R. (2002). Our Harvard pigeon, rat, dog, and
human lab. Journal of Experimental Analysis of Behavior,
77(3), 385-387. doi:10.1901/jeab.2002.77-385
Lindsley, O. R. (2004, May 28). Interview with Ogden
Lindsley. Interviewed by Alexandra Rutherford. Copy in
possession of rst author.
LoLordo, V. M., & Seligman, M. E. P. (1997). Richard
Lester Solomon (1918-1995). American Psychologist,
52(5), 567-568. doi:10.1037/0003-066X.52.5.567
Lubbock, J. (1884a). Teaching Animals to Converse. Nature,
29(740), 216. doi:10.1038/029216a0
Lubbock, J. (1884b). Teaching Animals to Converse. Nature,
29(754), 547-548. doi:10.1038/029547c0
Lubbock, S. J. (1889). On the senses, instincts, and
intelligence of animals: with special reference to insects.
K. Paul, Trench & co.
Lynch, J. J. (n.d.). An Interview with James J. Lynch.
Part I: Speaking Heart to Heart. Interview by Paul J.
Rosch. Retrieved from http://www.stress.org/interview-
Lynch, J. J. (1987). W. Horsley Gantt’s effect of person.
In F. J. McGuigan & T. A. Ban (Eds.), Critical Issues in
Psychology, Psychiatry, and Physiology: A Memorial to
W. Horsley Gantt (pp. 93-106). Amsterdam, Netherlands:
Gordon and Beach Publishers.
Lynch, J. J., & Gantt, H. W. (1968a). Comparison of the CR
to the UR in classical heart rate conditioning: The effect of
person and its integration with shock. Conditional Reex,
Lynch, J. J., & Gantt, H. W. (1968b). The heart rate
component of the social reex in dogs: The conditional
effects of petting and person. Conditional Reex, 3, 69-
Lynch, J. J., & McCarthy, J. F. (1967). The effect of petting on
a classically conditioned emotional response. Behaviour
Research and Therapy, 5(1), 55-62. doi:10.1016/0005-
History of Dogs 69
Lynch, J. J., & McCarthy, J. F. (1969). Social responding in
dogs: Heart rate changes to a person. Psychophysiology,
5(4), 389-393. doi:10.1111/j.1469-8986.1969.tb02838.x
Maier, S. F. (2001). Exposure to the stressor environment
prevents the temporal dissipation of behavioral depression/
learned helplessness. Biological Psychiatry, 49, 763-773.
Maier, S. F., Seligman, M. E. P, & Solomon, R. L. (1969).
Pavlovian fear conditioning an learned helplessness:
Effects on escape and avoidance behavior of (a) the CS-
US contingency and (b) the independence of the US and
voluntary responding. In B. A. Campbell & R. M. Church
(Eds.) Punishment and Aversive Behavior (pp. 299-342).
New York, NY: Appleton-Century-Crofts.
McBryde, W. C., & Murphree, O. D. (1974). The
rehabilitation of genetically nervous dogs. Pavlovian
Journal of Biological Science, 9(2), 76-84.
McIntire, R. W. (1965). The training of dogs for eld
reconnaissance. (Report No. 0724674). Retrieved from
McIntire, R. W. (1967). The relative effectiveness of
stressful and rewarding characteristics of primary and
social reinforcers in the control of canine behavior.
(Accession number AD0832920). Retrieved from http://
McIntire, R. W. (1968). Dog training, reinforcement,
and behavior in unrestricted environments. American
Psychologist, 23(11), 830-831. doi:10.1037/h0037742
McIntire, R., & Colley, T. A. (1967). Social reinforcement
in the dog. Psychological Reports, 20(3 pt. 1), 843-846.
Melzack, R. (1954). The genesis of emotional behavior: An
experimental study of the dog. Journal of Comparative and
Physiological Psychology, 47(2), 166-168. doi:10.1037/
Melzack, R. (1972). Effects of early perceptual restriction
on simple visual discrimination. Science, 137, 978−979.
Melzack, R., & Burns, S. K. (1965). Neurophysiological
effects of early sensory restriction. Experimental Neurology,
13, 163-175. doi:10.1016/0014-4886(65)90106-8
Melzack, R., & Scott, T. H. (1957). The effects of
early experience on the response to pain. Journal of
Comparative and Physiological Psychology, 50(2), 155-
Melzack, R., & Thompson, W. R. (1956). Effects of early
experience on social behaviour. Canadian Journal of
Psychology/Revue canadienne de psychologie, 10, 82−90.
Miklósi, Á., Polgárdi, R., Topál, J., & Csányi, V. (1998). Use
of experimenter-given cues in dogs. Animal Cognition, 1
(2), 113-121. doi:10.1007/s100710050016
Morgan, C. (1894). Introduction to Comparative Psychology.
London, UK: Walter Scott Ltd. doi:10.1037/11344-000
Murphree, O. D. (1974). Procedure for operant conditioning
of the dog. Pavlovian Journal of Biological Science, 9(1),
Murphree, O. D., DeLuca, D. C., & Angel, C. (1974).
Psychopharmacological facilitation of operant
conditioning of genetically nervous catahoula and pointer
dogs. Pavlovian Journal of Biological Science, 9(1), 17-
Murphree, O. D., Dykman, R. A., & Peters, J. E. (1967a).
Genetically-determined abnormal behavior in dogs:
Results of behavioral tests. Conditional Reex, 2(3), 199-
Murphree, O. D., Dykman, R. A., & Peters, J. E. (1967b).
Operant condition of two strains of the pointer dog.
Psychophysiology, 3(4), 414-417.
Murphree, O. D. & Newton, J. E. O. (1971). Crossbreeding
and special handling of genetically nervous dogs.
Conditional Reex, 6(3), 129-136.
Murphree, O. D., Peters, J. E., & Dykman, R. A. (1967).
Effect of person on nervous, stable, and crossbred pointer
dogs. Conditional Reex, 2(4), 273-276.
Murphree, O. D., Peters, J. E., & Dykman, R. A. (1969).
Behavioral and autonomic results from crossing nervous
and stable dogs. Conditional Reex, 3(3), 140.
Newton, J. E. O., & Chapin, J. L. (1978). Operant performance
patterns of normal and nervous water-deprived dogs.
Pavlovian Journal of Biological Science, 13(1), 22-28.
Newton, J. E. O., Teitelbaum, H. A., & Gantt, W. H. (1968).
An attempt to condition the cardiac response to person.
Conditional Reex, 3(2), 140-141.
Nonidez, J. F. (1941). Foreword. In C. R. Stockard (Ed.),
The Genetic and Endocrinic Basis for Differences in Form
and Behavior (viii-x). Philadelphia, PA: Wistar Institute
of Anatomy and Biology.
Overmier, B. J. (1996). Richard L. Solomon and learned
helplessness. Integrative Physiological and Behavioral
Science, 31, 331-337. doi:10.1007/BF02691436
Overmier, J. B., & Seligman, M. E. P. (1967). Effects of
inescapable shock upon subsequent escape and avoidance
responding. Journal of Comparative and Physiological
Psychology, 63, 28-33. doi:10.1037/h0024166
Pavlov, I. P. (1927). Conditional Reexes: An Investigation
of the Physiological Activity of the Cerebral Cortex.
Oxford, England: Oxford University Press.
Pfaffenberger, C. J. (2002). The New Knowledge of Dog
Behavior. Wenatchee, WA: Dogwise Publishing. (Original
work published 1963)
Pfaffenberger, C. J., & Scott, J. P. (1959). The relationship
between delayed socialization and trainability in guide
dogs. Journal of Genetic Psychology, 95, 145-155.
Pfaffenberger, C. J., Scott, J. P., Fuller, J. L., Ginsburg, B.
E., & Bielfelt, S. W. (1976). Guide dogs for the blind:
Their selection, development, and training. Amsterdam,
History of Dogs 70
Netherlands: Elsevier Science Ltd.
Pryor, K. (1999). Don’t Shoot the Dog: The New Art of
Teaching and Training (Rev. ed.). New York, NY: Bantam
Razran, H. S., & Warden, C. J. (1929). The sensory
capacities of the dog as studied by the conditioned reex
method (Russian schools). Psychological Bulletin, 26(4),
Reese, W. G. (1985). Pavlovian Society Awards. Pavlovian
Journal of Biological Science, 20, 155-162.
Reese, W. G., Peters, J. E., & Dykman, R. A. (1987). Gantt
on Gantt. In F. J. McGuigan & T. A. Ban (Eds.), Critical
Issues in Psychology, Psychiatry, and Physiology: A
Memorial to W. Horsley Gantt (pp. 17-56). Amsterdam,
Netherlands: Gordon and Beach Publishers.
Rescorla, R. A. (1966). Predictability and number of pairings
in Pavlovian fear conditioning. Psychonomic Science,
Rescorla, R. A. (1967a). Inhibition of delay in Pavlovian fear
conditioning. Journal of Comparative and Physiological
Psychology, 64(1), 114-120. doi:10.1037/h0024810
Rescorla, R. A. (1967b). Pavlovian conditioning and its
proper control procedures. Psychological Review, 74, 71-
Rescorla, R. A. (1968). Pavlovian conditioned fear in
Sidman avoidance learning. Journal of Comparative and
Physiological Psychology, 65(1), 55-60. doi:10.1037/
Rescorla, R. A. (1969). Establishment of a positive reinforcer
through contrast with shock. Journal of Comparative
and Physiological Psychology, 88(2, pt. 1), 260-263.
Rescorla, R. A. (1995). Richard L. Solomon (1918-1995).
Biographical Memoirs. Washington, D. C.: National
Academies Press. Retrieved from http://www.nap.edu/
Rescorla, R. A., & Solomon, R. L. (1967). Two-process
learning theory: Relationships between Pavlovian
conditioning and instrumental learning. Psychological
Review, 74(3), 151-182. doi:10.1037/h0024475
Romanes, G. (1883). Animal Intelligence. New York: D.
Appleton & Co.
Romanes, G. (1887). Experiments on the Sense of Smell in
Dogs. Nature, 36, 273-274. doi:10.1038/036273a0
Ruiz, G., Sánchez, N., & De la Casa, L. G. (2003). Pavlov
in America: A heterodox approach to the study of his
inuence. Spanish Journal of Psychology, 6, 99-111.
Schlosberg, H. (1934). Conditioned responses in the white
rat. The Pedagogical Seminary and Journal of Genetic
Psychology, 45, 303-335.
Scott, J. P. (1958). Critical periods in the development of
social behavior in puppies. Psychosomatic Medicine, 20,
Scott, J. P. (1978, September 29). [Letter to J. Holstein].
John Paul Scott Collection (Box 36, Folder 4), Joan Staats
Library, Jackson Laboratory, Bar Harbor, ME.
Scott, J. P., & Fuller, J. L. (1965). Genetics and the Social
Behavior of the Dog: The Classic Study. Chicago, IL: The
University of Chicago Press.
Scott, J. P., & Marston, M.-V. (1950). Critical periods
affecting the development of normal and mal-adjustive
social behavior of puppies. The Pedagogical Seminary
and Journal of Genetic Psychology, 77, 25-60.
Seligman, M. E., & Maier, S. F. (1967). Failure to escape
traumatic shock. Journal of Experimental Psychology,
74(1), 1-9. doi:10.1037/h0024514
Shapiro, M. M. (1960a). Classical salivary condition in dogs.
Dissertation Abstracts, 20, 4188-4189.
Shapiro, M. M. (1960b). Respondent salivary conditioning
in dogs during operant lever pressing in dogs. Science,
132, 361-364. doi:10.1126/science.132.3427.619
Shapiro, M. M., & Herendeen, D. L. (1975). Food-reinforced
inhibition of conditioned salivation in dogs. Journal of
Comparative and Physiological Psychology, 88(2), 628-
Shapiro, M. M., Miller, T. M., & Bresnahan, J. L. (1966).
Dummy trials, novel stimuli, and Pavlovian-trained
stimuli: Their effect upon instrumental and consummatory
response relationships. Journal of Comparative and
Physiological Psychology, 61(3), 480-483. doi:10.1037/
Shapiro, M. M., Mugg, G. J., & Ewald, J. (1971). Instrumental
preferences and conditioned preparatory responses in dogs.
Journal of Comparative and Physiological Psychology,
74(2), 227-232. doi:10.1037/h0030343
Shepard, E. (2001). Finding aid to The Charles Rupert
Stockard, PhD (1879-1939) Papers. Retrieved from http://
Shepherd, W. T. (1915). Adaptive intelligence in dogs and
cats, as compared with adaptive intelligence in rhesus
monkeys. American Journal of Psychology, 26, 211-216.
Shepherd, W. T. (1919) On sound discrimination in dogs.
The American Journal of Psychology, 30, 291-294.
Sidman, M. (1954). The temporal distribution of avoidance
responses. Journal of Comparative Physiological
Psychology, 47(5), 399-402. doi:10.1037/h0059155
Sinn, D. L., Gosling, S. D., & Hilliard, S. (2010). Personality
and performance in military working dogs: Reliability and
validity of behavioral tests. Applied Animal Behaviour
Sciences, 127, 51-65. doi:10.1016/j.applanim.2010.08.007
Solomon, R. L., Kamin, L. J., & Wynne. L. C. (1953).
Traumatic avoidance learning: the outcomes of several
extinction procedures with dogs. Journal of Abnormal and
History of Dogs 71
Social Psychology, 49, 291−302. doi:10.1037/h0058943
Solomon, R. L. & Wynne. L. C. (1953). Traumatic avoidance
learning: acquisition in normal dogs. Psychological
Monographs, 67(4), 19.
Solomon, R. L., & Wynne. L. C. (1954). Traumatic
avoidance learning: the principles of anxiety conservation
and partial irreversibility. Psychological Review, 61(6),
Solomon, R. L., & Wynne. L. C. (1955). Traumatic
avoidance learning: acquisition and extinction in dogs
deprived of normal peripheral autonomic function.
Genetic Psychology Monographs, 52, 241-284.
Spady, T. C., & Ostrander, E. A. (2008). Canine behavioral
genetics: Pointing out the phenotypes and herding up the
genes. American Journal of Human Genetics, 82(1), 10-
Staff. (1950). Manual of dog testing techniques. Bar harbor,
ME: Roscoe B. Jackson Memorial Laboratory.
Stanley, W. C. (1966). The passive person as reinforcer in
isolated beagle puppies. Psychonomic Science, 2, 21-22.
Stanley, W. C., Morris, D. D, & Trattner, A. (1965).
Conditioning with a passive person reinforcer and
extinction in Shetland sheep dog puppies. Psychonomic
Science, 2, 19-20.
Stockard, C. R. (Ed.) (1941). The Genetic and Endocrinic
Basis for Differences in Form and Behavior. Philadelphia,
PA: Wistar Institute of Anatomy and Biology.
Svartberg, K., & Forkman, B. (2002). Personality traits
in the domestic dog (Canis familiaris). Applied Animal
Behaviour Science, 79(2), 133-156. doi:10.1016/S0168-
Thomas, K. J., Murphree, O. D., & Newton, J. E. (1972).
Effect of person and environment on heart rates in two
strains of pointer dogs. Conditional Reex, 7, 74−81.
Thompson, W. R., & Heron, W. (1954a). The effects of early
restriction on activity in dogs. Journal of Comparative
Physiological Psychology, 47(1), 77-82. doi:10.1037/
Thompson, W. R., & Heron, W. (1954b). The effects of
restricting early experience on the problem-solving
capacity of dogs. Canadian Journal of Psychology, 8(1),
Thompson, W. R., Melzack, R., & Scott, T. H. (1956).
‘Whirling behavior’ in dogs as related to early experience.
Science, 123, 939. doi:10.1126/science.123.3204.939
Thorndike, E. (1898). Animal Intelligence: An Experimental
Study of the Associative Processes in Animals. New York,
NY: Columbia University Press. doi:10.1037/10780-000
Thorndike, E. (1899). Do animals reason? Popular Science
Monthly, LV, 480-490.
Todes, D. P. (2000). Ivan Pavlov: Exploring the Animal
Machine. New York, NY: Oxford University Press.
Todes, D. P. (2001). Pavlov’s Physiology Factory:
Experiment, Interpretation, Laboratory Enterprise.
Baltimore, MD: The Johns Hopkins University Press.
Topál, J., Miklósi, Á., & Csányi, V. (1997). Dog-
human relationship affects problem-solving
behavior in the dog. Anthrozoös, 10(4), 214-224.
Townshend, E. (2009). Darwin’s Dogs: How Darwin’s Pets
Helped Form a World-Changing Theory of Evolution.
Tryon, R. C. (1940). Genetic differences in maze learning
ability in rats. Yearbook of the National Society for the
Study of Education, 39(I), 111-119.
Watson, J. B. (1916). The place of the conditioned-reex
in psychology. The Psychological Review, 23, 89-116.
Weiss, E., & Greenberg, G. (1997). Service dog selection
tests: Effectiveness for dogs from animal shelters. Applied
Animal Behaviour Science, 53(4), 297-308. doi:10.1016/
West, M. J., & King, A. P. (1987). Settling nature and nurture
into an ontogenetic niche. Developmental Psychobiology,
20, 549-562. doi:10.1002/dev.420200508
Wynne, C. D. L. (2007). What are animals? Why is
anthropomorphism still not a scientic approach to
behavior? Comparative Cognition & Behavior Reviews,
2, 125-135. doi:10.3819/ccbr.2008.20008