This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
Author's personal copy
Canine Aggression Toward Unfamiliar
People and Dogs
Lore I. Haug, DVM
South Texas Veterinary Behavior Services, 2627 Cordes Drive, Sugar Land, TX 77479, USA
Dog aggression is a serious public health issue in the United States. More
than 4 million dog bites to humans are estimated to occur each year ,
and up to 42% of dogs presented to behavior clinics do so for aggression
toward other dogs . Aggression places a serious strain on the human–animal
bond. Dogs frequently are surrendered to shelters for behavioral reasons,
including aggression . Additionally, injuries to victims can result in owners’
incurring signiﬁcant ﬁnancial and legal burdens. Although aggression is a nor-
mal behavior in all animal species, it becomes problematic when it develops in
abnormal intensities or contexts, manifests toward aberrant targets (eg, is self-
directed), becomes dangerous to other people and animals, and/or interferes
with the human–animal bond.
DIAGNOSIS AND CLASSIFICATION
Different authors have classiﬁed aggressive behavior in various ways using
either functional or categorical divisions. Common categorical terminology
can facilitate professional communication; however, such a scheme does not ac-
curately describe all patients—even humans (as evidenced by the number of
‘‘disorder unspeciﬁed’’ labels found in the American Psychiatric Association’s
Diagnostic and Statistical Manual of Mental Disorders). In a reductionist sense, canine
aggression towards unfamiliar people and dogs generally occurs because of
fear, resource guarding (protection of territory, owners, or other animals), or
predation. In many cases, dogs present with multiple forms of aggression.
Fear-motivated aggression is the most common diagnosis in dogs aggressive
toward unfamiliar stimuli, even when elements of territoriality are present.
Offensive posturing by the dog does not rule out anxiety or fear as an under-
lying cause . The distance to the stimulus and previous learning affect the
dog’s behavioral presentation. Many dogs show highly offensive posturing
when behind a barrier or when the trigger stimulus is far away. As the stimulus
approaches or the barrier is removed, the dog’s behavior may become more
ambiguous and ﬁnally reﬂect outright fear. It is common for dogs to be highly
E-mail address: firstname.lastname@example.org
0195-5616/08/$ – see front matter ª2008 Elsevier Inc. All rights reserved.
Vet Clin Small Anim 38 (2008) 1023–1041
SMALL ANIMAL PRACTICE
Author's personal copy
reactive or aggressive toward other dogs while on leash but then to interact
appropriately while off leash. Several theories are postulated to explain this
behavior. First, the dog may feel trapped by the conﬁnes of the leash, which
limits the dog’s movements, including its ability to retreat. Second, a tight leash
(especially if the owner also is pulling actively) while the dog is approaching or
greeting another dog may alter the dog’s posture sufﬁciently to send misleading
signals. These signals may trigger the recipient dog to react agonistically, with
a scufﬂe ensuing. Over time, the dog learns that on-leash greetings are unpre-
dictable and potentially dangerous, and the dog becomes preemptively defen-
sive. Third, excitable, but friendly, dogs often are punished with leash
corrections for overly exuberant behavior around other dogs. Again, over
time the dog learns that the approach of other dogs predicts unpleasant and
potentially painful circumstances, generating defensive behavior.
Territorial behavior manifests primarily in the dog’s home and yard but also
may occur in the car or in areas where the dog is walked habitually. Territorial
behavior tends to be most intense directly along the boundary line, and dogs
may protect small territories more intensely than large ones . Unlike fear
aggression, which often manifests at an early age, territorial and protective
behavior are not expected to occur until 6 months of age or older, when the
dog approaches social maturity ; however, these latter types of aggression
frequently have elements of fear as well.
Dogs showing apparent protective behavior more commonly are fear ag-
gressive but become more offensive in the presence of their owner. It is spec-
ulated that this change occurs because the owner may have reinforced the
dog inadvertently or, alternatively, has punished the dog in the presence of
strangers or other dogs, intensifying the dog’s emotional reaction to the stim-
ulus. Dominance-related aggression typically is directed toward dogs with
which the dog has frequent close, social contact. On occasion, however,
dogs do seem to engage in status conﬂicts with strange people and, more
commonly, with unfamiliar dogs. This behavior occurs in relatively close
proximity to the stimulus, where postural signaling is most effective; thus
dominance probably is not the diagnosis if the dog shows aggressive behavior
toward the stimulus from a distance. Predatory reactions are more likely to
be directed toward small dogs and fast-moving objects such as joggers and
ETIOLOGY AND DEVELOPMENT
The development of aggressive behavior frequently is complicated and multi-
factorial. Problems associated with aggression in dogs fall into two broad
categories: (1) normal dogs expressing normal but unacceptable behavior or
(2) abnormal dogs reacting out of context to the environment . The bound-
aries of ‘‘normal’’ behavior are not ﬁxed rigidly: perinatal factors (intrauterine
environment, maternal and sibling interactions), experience (socialization and
learning), and biologic correlates (genetics, hormones, and neurophysiologic
factors) all affect the expression of the behavior.
Author's personal copy
Genetics and Breed Inﬂuences
Selection of phenotypic and behavioral characteristics in dog breeds has re-
sulted in various changes in social competency. Some breeds do show tenden-
cies toward certain forms of aggression [4,9]. Behavioral traits, including
aggression, have been identiﬁed as clustering in lines or families within a breed
 or even to be related to coat color patterns . The heritability of owner
impressions of aggressive behavior toward dogs and humans in Golden
Retrievers has been estimated as high as 81% . A group of studies by Svart-
berg [13,14] identiﬁed consistent heritability of a boldness/shyness personality
factor in dogs. Genetics also inﬂuences behavior through effects on neurotrans-
mitter systems and other biologic correlates.
Biological Correlates of Aggression
Hormones and gonadectomy
A vast body of literature has examined the effects of sex steroids, particularly
testosterone, on aggression in various species. Although testosterone does inﬂu-
ence the expression of aggressive behavior, there is a complex interplay
between testosterone, social status, neurotransmitters systems, gender, and en-
vironmental context . Castration of male dogs affects sexually dimorphic
behaviors and will reduce mounting, urine marking, and roaming. Reductions
in territoriality and aggression toward other dogs (particularly other males)
occurs, but to a lesser degree [9,16,17]. Ovariohysterectomy in females does
not inﬂuence aggressive behavior signiﬁcantly or consistently . Kim and
colleagues  evaluated seven intact and seven ovariohysterectomized Ger-
man Shepherd bitches for reactivity and aggression and found that 5 months
after spaying the spayed bitches showed signiﬁcantly more reactivity than
intact bitches. Gonadectomy should not be expected to play a major role in
controlling aggression in dogs.
Neurotransmitters and neural correlates
The biologic basis of aggression is complex. Studies of violence and aggression in
humans have focused heavily on the neurotransmitter serotonin (5-HT). The
5-HT system is associated with behavioral inhibition . Evidence links
5-HT deﬁciency to aggression, but this effect is difﬁcult to isolate from its effects
on impulsivity and social behavior, because serotonin also tends to improve both
these traits . Reisner and colleagues  found lower levels of the serotonin
metabolite 5-HIAA in the cerebrospinal ﬂuid of dominant aggressive dogs than
in nonaggressive dogs. Other studies of aggressive dogs also have found differ-
ences in serotonin receptor densities and function in various brain regions .
Biologic correlates may be particularly relevant for the classically ‘‘reactive’’ dog.
These dogs respond to even mild or apparently nonthreatening stimuli in a vol-
atile manner, and this reaction may be intensiﬁed if the stimulus appears sud-
denly. Intermittent explosive disorder (IED) in humans may serve as a model
for such dogs. Human patients who have IED are deﬁned by impulsive
aggressive behavior and are highly reactive to even low-level provocation. These
patients rate higher on general anger and hostility than do groups with other
1025CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS
Author's personal copy
psychiatric diagnoses . Patients who have IED also are impaired in their rec-
ognition of some facial signals , which can affect their social proﬁciency. Dogs
have been shown to have reduced competency in social signaling compared with
wolves . Perhaps dogs have deﬁcits in signal interpretation contributing to
the comparatively higher level of aggression in dogs than in wolves.
The limbic system, chieﬂy the amygdala, processes threat and emotional
responses. As part of the temporal lobe, the amygdala has a low seizure thresh-
old. Partial seizures in the temporal lobe can trigger feelings of fear, anxiety,
irritability, and anger. If a hypersensitivity develops in the amygdala so that
a subseizure threshold of neuronal excitability exists, emotional disturbances
could arise. This possibility is supported by the fact that human patients who
have this behavioral and emotional proﬁle improve when taking anticonvulsant
medication . A hyperresponsive amygdala easily could describe the ‘‘reac-
tive’’ dog mentioned previously. Essentially, the amygdala sends a high rate
of false alarms that activate the ﬁght–ﬂight system and the regions of the brain
responsible for vigilance, attention, anxiety, and fear. Some support for this
conjecture comes from studies by Jacobs and colleagues  showing that
aggressive dogs have higher basolateral nucleus group volumes and neuronal
densities in the amygdala than do nonaggressive dogs. Basolateral nucleus
groups of aggressive dogs also were shown to have more neurons containing
neurokinin 1, which is involved in regulation of aggressive behavior . Fur-
thermore, there are dense concentrations of 5-HT receptors in the amygdala.
Serotonin has a net inhibitory effect in the amygdala, so 5-HT–deﬁcient states
would result in compromised braking of amygdalar reactions .
Perinatal environment and early experience
A puppy’s perinatal environment can have a lasting impact on its adult behav-
ior. Maternal stress or early postnatal stress can permanently alter an animal’s
reactivity to future stress [29,30]. Studies indicate that low-level postnatal stress
(brief maternal separation and neonate handling) is protective. It reduces hypo-
thalamic-pituitary-adrenal (HPA) reactivity and increases hippocampal 5-HT.
In contrast, more severe stress (prenatal stress, prolonged maternal separation,
perinatal illness) can increase the HPA axis responsiveness to physiologic and
psychologic insults in the future [29,31]. Therefore, breeders should be coun-
seled carefully on the perinatal environment of their litters.
Socialization deﬁcits are arguably the most prominent factor in the develop-
ment of aggression in physiologically normal dogs. Unfortunately, the amount
of socialization required for optimal development of any individual is un-
known. Roll and Unshelm  noted that 44% of a population of dog-
aggressive dogs had few or no interactions with conspeciﬁcs from 5 weeks to
5 months of age. Deﬁcits in social interaction may become more problematic
as the animal matures and neophobia and competitive interactions become
more salient. Mere exposure to other people and dogs is not sufﬁcient to guar-
antee adequate social skills. Interactions must be monitored to ensure that the
puppy has a positive and enriching experience.
Author's personal copy
Inﬂuence of learning
All forms of aggression are modiﬁed by learning. Aggression is about local
control of the environment. If an animal learns that aggression will alter the
environment in a desirable way, reinforcement occurs, and the animal will
show that behavior pattern in a similar circumstance in the future. The power
of reinforcement emphasizes the importance of avoiding trigger situations dur-
ing management and treatment. See the article by Horwitz in this issue for fur-
In dogs, the origin and progression of aggression to unfamiliar stimuli can vary,
as can the associated behavioral presentation. Selecting the most appropriate
treatment course depends on the animal’s behavioral phenotype and the
owner’s resources and capabilities. Clinicians should explain each step of the
treatment process carefully. Techniques should be demonstrated when appro-
priate and feasible. Owners must understand that altering the dog’s behavior
will take time, and improvement may not occur in a linear fashion. Although
most owners do not want a lesson in neurophysiology, a brief and simple ex-
planation of the persistence of neural circuits, particularly those associated with
fear-related behaviors, can help owners understand their dog’s behavioral re-
sponses. Setbacks are a typical part of most therapy programs, although the
program should be designed and modiﬁed periodically to minimize them.
For some owners, the number of environmental changes and interventions
can be overwhelming. Breaking the interventions down into progressions
will help owners accomplish goals successfully and see more rapid response.
This early positive reinforcement for the owner can improve compliance
greatly. Treatment programs can be divided into three phases: management,
foundation exercises, and stimulus-speciﬁc behavior modiﬁcation exercises.
Environmental management involves addressing the animal’s biologic needs
and preventing further rehearsal of inappropriate behavior patterns. Safety pre-
cautions also must be implemented.
Exercise and enrichment
Many dogs live in environments either grossly deﬁcient in stimulation or re-
plete with inappropriate stimulation. Additionally, as a dog’s behavior becomes
more problematic, the dog tends to be even more isolated from the environ-
ment. Many owners cease walking their dogs altogether, and dogs with territo-
rial behavior often are relegated to spending large amounts of time crated or
penned outside. The profound lack of mental and physical exercise compounds
the dog’s frustration and agitation and decreases the latency to arousal around
triggering stimuli. Owners must ﬁnd ways to exercise their dogs safely. They
must walk the dog at times and in places where they are unlikely to encounter
other people or dogs, even if the owner must drive the dog to an acceptable
area. As well as burning off excess energy, exercise may help by elevating levels
1027CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS
Author's personal copy
of norepinephrine and 5-HT in the brain and releasing endogenous endorphins
, the latter two of which have calming and anxiolytic effects [34,35].Dogs
ideally should receive at least 30 continuous minutes of aerobic exercise per
day, because research indicates that prolonged aerobic exercise is more effec-
tive in triggering opioid-mediated effects on mood and sympathetic activity
Mental stimulation through environmental enrichment helps occupy dogs
that have limited physical exercise routines and that are left alone for long
periods. Enrichment increases behavioral adaptation , in part by improving
the animal’s problem-solving skills. Rotating toys, feeding from food-dispensing
devices, and engaging the dog in activities requiring problem solving (eg, train-
ing and discrimination tasks) all should be part of the dog’s normal routine.
Training even simple tricks is excellent mental stimulation and helps strengthen
the dog–owner bond as well as increasing the dog’s skill set.
Preventing inappropriate behavior
Dogs that have a long-standing history of aggressive behavior have developed
a learned, conditioned reaction to trigger stimuli. Accordingly, owners also
have become conditioned to anticipate unpleasant encounters. Most aggressive
outbursts occur repeatedly in a handful of contexts such that these environ-
ments alone can predict the appearance of unfamiliar dogs and people.
When the dog and the owner are exposed to these environments, both undergo
anticipatory changes in autonomic arousal that push the dog closer to the reac-
tive threshold even in the absence of triggering stimuli . Temporarily re-
moving the dog from these contexts (and from exposure to triggering
stimuli) will facilitate the conditioning of more desirable behavioral responses.
Avoidance also reduces the risk of injury to other people and dogs. The dog
should not be exposed to any such stimuli until later in the rehabilitation pro-
cess and only during controlled training sessions. For dogs that are aggressive
when away from home, exercise modalities and locations must be altered, or
the dog must be kept beyond its threshold distance for the stimulus. If the
dog is aggressive inside the car, car rides should be minimized or stopped alto-
gether. Some dogs are less reactive if crated while in the car, and the crate can
be covered to prevent the dog from seeing stimuli outside. Similarly, inside the
house, the dog should be prevented from patrolling windows and doors for
passing people or dogs by blocking windows (eg, closing blinds) or gating
the dog away from the front of house, especially in the owner’s absence. If nec-
essary, the dog can be crated or closed into a room with no or few windows
and protected from outside noises. While the owner is home, the dog can be
handled more safely and will respond more reliably if ﬁtted with a head collar
and dragline, which can be used to interrupt inappropriate behavior immedi-
ately but calmly. When visitors arrive, the dog should be conﬁned before
the visitor actually enters the house, ideally in an area where the dog cannot
see the doorway through which the visitor arrives. If the dog is aggressive
only as the visitor enters, but not afterward, the dog can be allowed out of
Author's personal copy
conﬁnement, under supervision, and on leash with a head collar, once the dog
is quiet and the visitor is settled.
Helping the owner gain some sense of control over the dog is a valuable step
early in a behavior program. Muzzle-loop head collars such as the Gentle
Leader (Premier Pet Products, Inc., Richmond, Virginia) (Fig. 1) are especially
advantageous for large and/or aggressive dogs. These collars provide excellent
control over the dog’s head, thereby allowing the owner to manipulate the
direction of the dog’s focus. Additionally, the owner can close the dog’s mouth
gently but ﬁrmly, which will prevent a bite in an emergency situation and allow
humane correction of inappropriate behavior.
Dogs with a previous bite history, with severe or escalating aggression, and/
or with owners that have difﬁculty controlling them should be trained to wear
a muzzle. The muzzle must allow the dog to pant and accept food treats. Pro-
vided the dog cannot separate its canine teeth enough to grip another person or
dog, a nylon sleeve muzzle can be used as effectively as a basket muzzle. If the
muzzle its to be left on for long periods of time, the basket muzzle may be the
preferable choice. Both types of muzzles limit panting, and care must be taken
when they are used in hot weather. The dog must be adapted to both head col-
lars and muzzles gradually in a manner that associates the devices with pleasant
experiences. Neither piece of equipment should ever be placed on the dog as
a form of punishment.
Not all dogs can wear a muzzle or head collar because of behavioral, medi-
cal, or conformational limitations. Other collar types and harnesses (eg, Easy
Walk, Premier Pet Products, Inc.; Zuba Dream Walker, Zuba Pets, Menlo
Park, California) are available that may improve the owner’s control. Punitive
collars such as a prong, slip chain, or electronic stimulation should be avoided.
Punitive actions that elevate fear and/or cause the dog pain may be associated
Fig. 1. The Gentle Leader head collar (Premier Pet Products, Inc., Richmond, Virginia).
1029CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS
Author's personal copy
with the trigger stimulus rather than with the dog’s own behavior . This
misdirected association is particularly likely if the owner has poor timing and
mechanical skills, because the dog will be unable to associate the correction
consistently with a speciﬁc behavior. This unpredictable punishment actually
will increase the dog’s anxiety level.
Dogs should be handled on a 4- or 6-foot nylon or leather leash. Retractable
leashes are inappropriate and dangerous, because they provide poor control and
can cause injury to the owner or the dog if the cord becomes wrapped around
part of the body. Cotton long lines can be used to control the dog for exercise
purposes, because aggressive dogs should never be off leash in public.
Dogs that are visually reactive may beneﬁt by reducing the clarity of their
visual ﬁeld. The Calming Cap (Premier Pet Products, Inc.) is an elastic, semi-
transparent cloth ‘‘hood’’ that covers the dog’s eyes. This device can be
extremely useful during car rides and also can be used in the home or on walks.
TTouch body wraps (Linda Tellington-Jones, Santa Fe, New Mexico) and the
Anxiety Wrap (Animals Plus, Huntington, Indiana) have proven effective in
calming some excitable or anxious dogs, although no studies have evaluated
them in a controlled manner. These products provide tactile pressure over
the dog’s body for a swaddling or acupressure effect.
Dealing with unplanned exposures
One goal of good management is to reduce uncontrolled stimulus exposures;
however, unexpected contacts do occur even with highly dedicated and atten-
tive owners. Owners may carry a pop-open umbrella or Direct Stop citronella
spray (Premier Pet Products, Inc.) for dealing with free-ranging dogs. Some
dogs can be discouraged with a ﬁrm, ‘‘No! Go home!’’ and others may be dis-
tracted by throwing a large handful of treats directly at the dog. Well-meaning
people should be directed gently but ﬁrmly to avoid approaching the dog.
Training the dog in advance to perform an emergency U-turn allows calm
but rapid escape from a potentially volatile situation. The muzzle-loop head
collars permit the owner to control the dog’s head and mouth to prevent
a bite (to the target or the owner if the dog is prone to redirect) without the
need for punitive measures should another person or dog approach too
In a number of cases, alterations in diet and exercise (mental and physical)
and reduced exposure to provocative situations improve a dog’s behavior
sufﬁciently that the owner is content with management alone. This strategy
is a viable one, particularly for time-restricted owners of dog-aggressive dogs,
when avoiding contact with other dogs is relatively easy.
The second level of intervention focuses on training foundation exercises, which
increase the dog’s skill set and give the dog alternative ways to respond to stim-
uli. The exercises also are designed to amplify the owner’s general control over
the dog and to improve the dog’s focus on and responsiveness to the owner.
Author's personal copy
Basic cue response
Although many dogs previously were enrolled in a puppy or basic obedience
class, an amazingly large number of owners have never sought any type of
training for their dogs, even for dogs that have serious behavior issues. Few
dogs with aggressive behavior are sufﬁciently proﬁcient at even basic obedience
behaviors. Although obedience itself will not resolve an aggression problem,
these cues are important as a way for an owner to request alternative responses
from the dog . In situations where the dog is uncertain as to the most ap-
propriate behavioral response, basic behaviors can provide the dog with clarity
and safety if the behaviors have been trained previously and practiced in a clear
and consistent manner. The goal of training is twofold: (1) to obtain reliable
response to the cues, and (2) to condition the dog to become calm and relaxed
when performing the behaviors. The latter is crucial and is done by rewarding
the dog only for relaxed responses once the dog has a basic understanding of
the behavior itself. At a minimum the dog should be able to respond to cues for
‘‘sit,’’ ‘‘down,’’ ‘‘stay,’’ and ‘‘come.’’ The dog should be able to walk calmly on
leash by the owner’s side and also respond to its name by orienting to the
owner. Targeting exercises (eg, the dog touching its nose to a target stick or
the owner’s hand) also are valuable. These behaviors are easy to teach and
are easy for the dog to learn, typically resulting in highly reliable behavior.
Among other things, targeting can be used to reorient a distracted dog and
to lead or lure the dog away from a problematic situation.
All behaviors should be trained using positive reinforcement. Positive rein-
forcement training establishes a classically conditioned positive emotional re-
sponse (a ‘‘pleasure’’ feeling) to both the cue and the performance of the
behavior. Training based on punishment may be associated with higher levels
of behavior problems . The addition of a bridge signal or conditioned rein-
forcer (eg, clicker, whistle) improves reinforcement clarity and can be used in
future exercises as discussed later.
Establishing owner-focused interactions
Leadership programs frequently are recommended to establish command–
response interactions between the dog and the owner and stress the importance
of interacting with the dog only when the dog is calm. The owner begins to estab-
lish consistent behavioral criteria for any interaction with the dog (ie, petting, feed-
ing, starting a training session, putting on the dog’s collar, opening doorways).
Although it is most important that family members participate in these rules, vis-
itors and other unfamiliar people are encouraged to abide by the protocol as well.
Four basic criteria are required of the dog. These criteria can be introduced singly
or together, depending on the dog’s baseline behavior and the owner’s skill:
1. Respond to any requested cuebehavior (eg, sit) within an established time frame.
2. Remain calm during the entire interaction.
3. Remain focused on the owner during the interaction. The dog is encouraged
to make eye contact and look to the owner rather than focusing on another
resource or target.
1031CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS
Author's personal copy
4. Remain outside a previously designated ‘‘personal space’’ around the owner.
This behavior keeps the dog from crowding the owner (eg, to get through a door-
way) and also reduces nuisance behaviors such as jumping and mouthing.
The criteria for focus and calmness are by far the most important of the four.
In all situations, if the dog fails to maintain an established criterion throughout
the interaction, the owner aborts the interaction and directs the dog again. No
verbal or physical punishment is applied. Once the dog has attained criterion
again, the interaction can resume or start over.
Relaxation Tasks and Safety Cues
Owners of aggressive dogs frequently state that the dog becomes so aroused
that the dog is unresponsive in the presence of the triggering stimulus. Owners
frequently try to calm or reprimand the dog to halt the aggressive reaction. The
ﬂaw in this approach is that such dogs lack emotional control and generally do
not know how to relax and self-regulate their arousal, even on a daily basis.
Therefore the owner’s attempts to calm the dog will be futile. In fact, the
owner’s mounting tension and frustration typically raises the dog’s arousal
even further. Relaxation must be taught to the dog in a methodical manner
in an environment initially free of distraction.
Structured down-stay (or sit-stay) exercises should be practiced as a baseline
relaxation task . The dog is trained to maintain a short, relaxed down-stay
and then gradually is exposed to increasing levels of generic environmental dis-
tractions and human activities. To further increase the dog’s baseline relaxa-
tion, behaviors that a dog exhibits voluntarily when normally relaxed can
be reinforced and placed on cue. Because the dog already is inclined to per-
form these behaviors, they are relatively easy to put under stimulus control.
Canine massage and TTouch also are excellent exercises to establish changes
in relaxation in association with a safety signal. Voluntary lateral recumbency
is associated with relaxation in dogs. This ‘‘play dead’’ behavior (Fig. 2) can be
placed on cue to allow the owner another tool for lowering arousal in the face
of a provocative stimulus. Because this position is highly vulnerable for the
dog, it is imperative that the dog never be physically forced into this position,
either during the training phase or during a real situation. If the dog will not
perform the behavior when cued, the behavior either is not sufﬁciently re-
hearsed or the dog has been placed in a situation that is too stressful for its
stage of training. Forcing the dog into this position will seriously erode the
dog’s trust in the handler (Relaxation tasks should serve as another form of
safety signal.) The effectiveness of these behaviors can be enhanced by aug-
menting them with other safety signals such as conditioned odors or having
the dog perform them on a ‘‘relaxation rug,’’ which can be transported to var-
All exercises are trained ﬁrst within the owner’s home in a quiet environ-
ment. Once the dog is proﬁcient, the tasks are repeated in other areas both
on and off the owner’s property. The dog never should be asked to perform
Author's personal copy
in an environment that it is not yet ready to handle. Asking the dog to hold
a relaxation position when it is in a stressful environment will erode the value
of the behavior as a safety cue.
Safety cues and signals
Animals readily make associations between contextual (environmental) stimuli
and emotional experiences that occur when those stimuli are present. For ex-
ample, a dog in a veterinary examination room receiving a painful injection
while resting on a blue rug may become afraid of blue rugs. Even though
the rug was neutral and did not harm the dog, the rug became associated
with the context in which the dog was hurt or frightened. Through a similar
learning process, a dog can associate environmental stimuli with pleasant,
Safety signals are environmental stimuli that become paired with relaxed
physiologic states in safe environments. Safety signals can be tactile, olfactory,
visual, or auditory. They also can be previously trained behaviors (cues). The
stimuli themselves eventually generate a relaxed state in the dog when the
animal is exposed to them. Exercises addressing stimulus-speciﬁc responses re-
volve primarily around classical conditioning paradigms such as countercondi-
tioning. Classical conditioning is a powerful tool for establishing baseline
changes in physiologic and psychologic relaxation and in establishing the safety
signals used during the last phase of training.
Conditioning safety cues involves choosing a speciﬁc stimulus (eg, a speciﬁc
dog bed, small rug, or odor) and pairing its presence with pleasant activities
and the relaxation tasks. For instance, the dog can be cued to lie on the dog
Fig. 2. The lateral recumbency (‘‘play dead’’) position is used as a relaxation task. This dog
has learned to perform this behavior reliably even in the presence of some fear-inducing dis-
tractions, and obeying this command helps control her arousal. Relaxation is shaped during
the training of the behavior; however, note the tucking of the dog’s tail and the slight ﬂexion
of the right hind leg up toward the dog’s body. These signs indicate that shaping for further
relaxation is needed.
1033CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS
Author's personal copy
bed and then rewarded when it does so in a relaxed manner. During the con-
ditioning process, the dog is never asked to lie on the bed when it is agitated, as
a punishment, or while anything unpleasant to the dog (eg, nail trimming) is
being performed. With repetitions the dog becomes conditioned to relax
when asked to lie on the bed or when other safety cues are present.
Safety cues should be portable and easy to reproduce but also fairly unique
to the environment in which they eventually will be used (eg, out on walks, at
the veterinary clinic, when visitors come to the house). This speciﬁcity pre-
vents the dog from habituating to their presence in the environment. The
dog should be exposed to the safety signal only during conditioning sessions
to ensure that the pairing of cue and relaxation remains as consistent as
Stimulus-Speciﬁc Behavior Modiﬁcation Exercises
Stimulus-speciﬁc exercises center on desensitization-counterconditioning
(DCC) drills. Typical methodology has both classical and operant conditioning
components, although variations may focus heavily on one element over the
other. In traditional DCC, the animal is exposed to a low-level stimulus, and
the presence of the stimulus is paired with something the dog ﬁnds rewarding,
such as food or play. The previously described down-stay relaxation task
serves as the foundation for stimulus-speciﬁc DCC. The trigger stimulus
becomes a new distraction added to the protocol. The dog is asked to sit or
down-stay, preferably in the presence of a previously established safety signal,
and then the dog is exposed to a low-level stimulus (eg, a dog or person) at a dis-
tance such that the dog brieﬂy alerts but then returns focus to the owner. If the
dog reacts to the stimulus, the stimulus is too close or too intense. The dog then
is rewarded for remaining calm in the cued position.
For each stimulus category (eg, dogs or people), the owner should develop
a hierarchical list with the stimulus composition least likely to arouse the dog
at the top and the stimulus composition most likely to trigger arousal at the bot-
tom. The more intermediary stimuli listed, the better. The owner also should
determine the thresholds at which the dog (1) alerts/orients to the stimulus, (2)
barks/growls, and (3) lunges or tries to bite. This list becomes the dog’s general
DCC sessions are divided into four base criteria: distance (between the dog
and the trigger stimulus), duration (that the dog is exposed to the stimulus dur-
ing any one trial), intensity (of the behavior or physical characteristics of the
stimulus), and number (of stimuli present at one time during the trial). During
any one trial, only one criterion should be manipulated. For example, if an
owner ﬁnishes a trial with a child 30 feet away from the dog, on the next trial
the child should not move closer to the dog and change his/her behavior.
Rather, the child either should be asked to move closer or to alter his/her
behavior. Once the dog can master each criterion individually, sessions can
begin to incorporate multiple criteria at one time.
Author's personal copy
Highly aroused dogs may beneﬁt from beginning DCC with audiotapes of
sounds associated with the trigger stimulus (eg, dog tags, barking, footsteps
on the sidewalk, human voices). This technique allows the owner to begin
the process in the safety of the dog’s home. For dogs with territorial aggression,
sessions should include sounds of doorbells and knocking. The dog also should
be trained to sit or lie calmly away from the door when the door is opened and
someone enters. This behavior is accomplished ﬁrst with family members, then
with familiar visitors, and ﬁnally progresses to unfamiliar visitors.
Some dogs are so reactive that any visual exposure results in a dramatic
aggressive display even if the stimulus is hundreds of yards away. For these
dogs, a purely classical conditioning paradigm using a previously conditioned
bridge stimulus (ie, clicker or whistle) may be more appropriate initially. The
dog is placed in a sit-stay position and is controlled by a head collar. The stim-
ulus (eg, a person) steps into view from behind a solid barrier at a great dis-
tance for only 1 or 2 seconds before stepping back behind the barrier. (This
brief appearance reduces the likelihood that the dog’s arousal will continue
to escalate.) As the person comes into view, the owner immediately applies
the bridge stimulus, irrespective of the dog’s behavior, and then offers the
dog food or a toy. The dog may be so aroused by the sight of the person
that it refuses the food. In the absence of the bridge signal, this level of arousal
means conditioning may not occur, because the dog may refuse the food or
toy. The bridge signal allows the beginning of conditioning even if the dog re-
fuses to eat the food or play with the toy. The dog is allowed to return to base-
line arousal before the process is repeated. Over time, this method can establish
an ‘‘auto-look’’ to the owner after the person comes into view. This looking to
the owner is the beginning of a threshold, in that there is a brief period of non-
reaction. At this point traditional DCC can begin.
Diet and Nutrition
There is considerable controversy and conﬂicting data on the inﬂuence of di-
etary factors on aggressive behavior. Few controlled studies have evaluated
nutritional effects in dogs. Dodman and colleagues  evaluated the inﬂuence
of dietary protein level on aggressive behavior and found that reductions in
protein may help reduce territorial behavior associated with fear, but the effect
was not robust, and the diet did not affect other types of aggression studied.
Studies in humans have shown changes in aggression and violence with dietary
tryptophan supplementation  and one study indicated a possible effect in
dogs . Anecdotal reports indicate possible beneﬁts of raw food diets,
grain-free diets, and low-protein diets; however, no controlled studies have
been done with the former two diets. How any individual animal responds
to dietary change is unknown and seems to be a matter of trial and error.
Gesch and colleagues  noted improvements in violence and antisocial be-
havior in prisoners receiving a supplemental vitamin-mineral and fatty acid
preparation. Similar results might be obtainable in dogs.
1035CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS
Author's personal copy
Pheromone and Aromatherapy
The canine olfactory system is well developed and represents a signiﬁcant por-
tion of the dog’s brain mass. The olfactory system is highly connected to the
limbic system. Dogs have a functional vomeronasal organ that transmits infor-
mation to the accessory olfactory bulb and then on to the amygdala . Ol-
factory stimuli can play a substantial role in the development and resolution
of behavior issues. Dog-appeasing pheromone (DAP; Ceva Sante
bourne, Gironde, France) is a synthetic analogue of the pheromone secreted by
lactating bitches. Recent studies have shown merit in its use for increasing
adaptability in newly adopted puppies , for improving performance in
puppies attending puppy classes , for reducing signs of fear or anxiety in
veterinary settings , and for treating fear of ﬁreworks . Wells  has
demonstrated that lavender scent can reduce excitability during car rides,
and it also increases relaxation in shelter settings . Lavender can be used
spontaneously or conditioned as a safety cue to be used in the home or applied
to a bandana the dog wears while away from home.
Currently there are no medications labeled for treating aggression disorders in
dogs. There are few controlled clinical studies evaluating drug therapy in
aggressive dogs. Virga and colleagues  found no beneﬁt with amitriptyline
use in aggressive dogs as compared with behavior modiﬁcation alone. White
and colleagues  also found no effect beyond placebo in the use of clomipr-
amine for dominance-related aggression. One study evaluating the use of ﬂuox-
etine in dogs with dominance-related aggression did ﬁnd a small effect, but the
improvement also could be attributed to placebo effects .
Despite the lack of data supporting clinical efﬁcacy, anecdotal reports indi-
cate that pharmacologic intervention can facilitate or expedite behavior therapy
in some cases. Beneﬁt may be most likely if (1) the aggression is related to high-
anxiety states or fearful behavior, (2) the animal appears to have a concurrent
impulse-control disorder, or (3) the dog is truly ‘‘reactive,’’ that is, the behav-
ioral proﬁle supports the possibility of amygdalar hyperreactivity.
Selective serotonin reuptake inhibitors (SSRIs) manipulate serotonin concen-
tration in the synaptic cleft, and their effect is relatively speciﬁc for serotonin.
They have antidepressant, anxiolytic, and anticompulsive effects . SSRIs
currently are the primary class prescribed for aggression problems in dogs
(Table 1). Fluoxetine, recently approved for use in dogs for separation anxiety
under the name Reconcile (Eli Lilly, Indianapolis, Indiana), is the SSRI with the
longest history of use for behavior problems in dogs. Its use for aggression is
extra-label. All SSRIs require continuous prolonged administration to produce
therapeutic changes. Fluoxetine typically is well tolerated; however, reported
side effects include sedation, gastrointestinal upset, anorexia, irritability, agita-
tion, and seizures . Fluoxetine and paroxetine  inhibit various cyto-
chrome P-450 enzymes; therefore, potential drug interactions should be
monitored carefully. Other commonly used SSRIs include paroxetine,
Author's personal copy
sertraline, ﬂuvoxamine, and citalopram. Sertraline and ﬂuoxetine often are use-
ful choices for older dogs because they do not have the anticholinergic effects of
paroxetine that may interfere with cognitive function .
Tricyclic antidepressants (TCAs) also have a long history of use for behavior
problems in dogs. Amitriptyline and clomipramine (labeled for separation
anxiety in dogs under the name Clomicalm [Novartis Animal Health, Greens-
boro, North Carolina]) are the two most frequently prescribed. These drugs
have both serotonin and norepinephrine reuptake properties, with clomipr-
amine being more speciﬁc for serotonin . TCAs also have anticholinergic,
antihistaminic, and alpha-adrenergic blockage effects, which are responsible for
most of the observed side effects and can include sedation, constipation, uri-
nary retention, vomiting, diarrhea, agitation, hypotension, and lowered seizure
TCAs seem to be more effective for anxiety- and fear-related disorders.
Given the lack of therapeutic effect in the few studies evaluating TCAs for
aggression, it might be wiser to reserve these drugs for use in dogs that have
concurrent severe anxiety disorders or in cases in which SSRIs have failed to
produce any therapeutic response.
Although not commonly used, anticonvulsants such as carbamazepine or
gabapentin may have some utility in dogs that seem to have amygdalar
Dosages for common psychotherapeutic agents in dogs
Drug Dosage Reference
Buspirone 1.0–2 mg/kg every 8–12 hours
Carbamazepine 4–8 mg/kg every 12 hours
Gabapentin 10–30 mg/kg every 8–12 hours Plumb 
Pindolol 0.125–0. 25 mg/kg every 12–24 hours Plumb 
Propranolol 5–40 mg/dog every 8 hours Plumb 
Alprazolam 0.02–0.1 mg/kg every 8–12 hours
Clorazepate 2 mg/kg every 12 hours
Diazepam 0.55–2.2 mg/kg every 8–12 hours
Selective serotonin reuptake inhibitors
Citalopram 0.5–1.0 mg/kg every 24 hrs
Fluoxetine 1.0–2.0 mg/kg every 24 hours
Paroxetine 0.5–1.5 mg/kg every 24 hours
Sertraline 0.5–4.0 mg/kg every 24 hours
Amitriptyline 1.0–4.0 mg/kg every 12 hours
Clomipramine 1.0–3.0 mg/kg every 12 hours
Data from Crowell-Davis SL, Murray T, Seibert LM. Veterinary psychopharmacology. Ames (IA): Blackwell
Publishing; 2006; and Simpson BS, Papich MG. Pharmacologic management in veterinary behavioral
medicine. Vet Clin North Am Small Anim Pract 2003;33(2):365–404, unless otherwise noted.
1037CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS
Author's personal copy
hyperreactivity . These drugs sometimes are used in conjunction with
SSRIs to control explosive aggression. Buspirone, an azaspirone, is a presynap-
tic 5-HT1A agonist. It also has partial agonist properties at postsynaptic
5-HT1A receptors. Buspirone has been used to control mild anxiety disorders
and generalized anxiety . Its effect on aggressive behavior has not been
evaluated, but clinical experience indicates that as a sole therapy it has little
role in controlling aggression in dogs.
Serotonin modulators, discussed previously, all require continuous adminis-
tration for therapeutic effects. Some drugs can be used on a situational basis to
control anxiety and frustration that may contribute to aggressive responses.
These medications can be given on an as-needed basis, for example, before out-
ings during which avoidance of problematic stimuli is impossible or to enhance
success during a controlled training situation. Situational drugs include opioids,
beta-blockers, and benzodiazepines. As mentioned previously, opioids can
modulate serotonergic transmission and sympathetic activation, thereby reduc-
ing heart rate, blood pressure, and anxiety . Similarly beta-blockers have
been postulated to reduce anxiety by controlling heart rate changes associated
with anxiety, although some beta-blockers (eg, pindolol) have direct serotoner-
gic actions as well. Benzodiazepines are very effective for reducing anxiety, but
they also may produce disinhibition of aggression [62,63], particularly in an an-
imal that is highly offensive. Therefore, their use probably should be restricted
to animals whose aggression is purely defensive.
There also is little evidence that most natural therapeutics are useful in the
treatment of aggression, except for tryptophan, as mentioned earlier, and
a milk hydrolyzate, alpha-casozepine, which seems to be beneﬁcial in reducing
some forms of anxiety . Crowell-Davis and colleagues  provide a more de-
tailed review of medications and their usage in veterinary psychopharmacology.
Canine aggression toward unfamiliar people and dogs is a common behavior
problem. Although a variety of factors are involved in the development of
this problem, genetics and socialization deﬁcits play a major role. This problem
typically can be well controlled with targeted changes in the animal’s environ-
ment, implementation of appropriate behavior modiﬁcation exercises, and
adjunctive pharmacologic support where indicated. Owners should be encour-
aged to seek professional help early in the problem before the dog actually in-
jures another animal or human.
 Overall K, Love M. Dog bites to humans—demography, epidemiology, injury, and risk.
J Am Vet Med Assoc 2001;218(12):1923–34.
 Sherman CK, Reisner I, Taliaferro L, et al. Characteristics, treatment, and outcome of
99 cases of aggression between dogs. Appl Anim Behav Sci 1996;47:91–108.
 Miller DD, Staats SR, Partlo C, et al. Factors associated with the decision to surrender a pet to
an animal shelter. J Am Vet Med Assoc 1996;209(4):738–42.
Author's personal copy
 Lindsay S. Aggressive behavior: basic concepts and principles. In: Handbook of applied
dog behavior and training. Etiology and assessment of behavior problems, vol. 2. Ames
(IA): Iowa State University Press; 2001. p. 161–201.
 Overall K. Canine aggression. In: Clinical behavioral medicine for small animals. St. Louis
(MO): Mosby; 1997. p. 88–137.
 Reisner I. Differential diagnosis and management of human-directed aggression in dogs.
Vet Clin North Am Small Anim Pract 2003;33(2):303–20.
 Bowen J. Miscellaneous behaviour problems. In: Horwitz DF, Mills DS, Heath S, editors.
BSAVA manual of canine and feline behavioural medicine. Quedgeley, Gloucester (UK):
British Small Animal Veterinary Association; 2002. p. 119–27.
 Hart B, Hart LA, Bain MJ. General approaches to behavioral pharmacology. In: Canine and
feline behavior therapy. Ames (IA): Blackwell Publishing; 2006. p. 63–73.
 Hart B, Hart LA, Bain MJ. Aggression toward people. In: Canine and feline behavior
therapy. Ames (IA): Blackwell Publishing; 2006. p. 103–28.
 Reisner I, Houpt K, Shofer FS. National survey of owner-directed aggression in English
Springer Spaniels. J Am Vet Med Assoc 2005;227(10):1594–603.
 Podberscek AL, Serpell J. The English Cocker Spaniel: preliminary ﬁndings on aggressive
behaviour. Appl Anim Behav Sci 1996;47:75–89.
 Liinamo A-E, van den Berg L, Leegwater PAJ, et al. Genetic variation in aggression-related
traits in Golden Retriever dogs. Appl Anim Behav Sci 2007;104:95–106.
 Svartberg K. A comparison of behaviour in test and in everyday life: evidence of three
consistent boldness-related personality traits in dogs. Appl Anim Behav Sci 2005;91:103–28.
 Svartberg K, Tapper I, Temrin H,et al. Consistency of personality traits in dogs. Anim Behav
 Haug LI. Androgens and 5-HIAA in dogs with intraspeciﬁc aggression [master’s thesis]. Col-
lege Station (TX): Texas A&M University; 2003.
 Hart B, Hart LA, Bain MJ. Canine and feline behavior therapy. 2nd edition. Ames (IA): Black-
well Publishing; 2006.
 Neilson J, Eckstein R, Hart B. Effects of castration on problem behaviors in male dogs with
reference to age and duration of behavior. J Am Vet Med Assoc 1997;211(2):180–2.
 Mertens P. Canine aggression. In: Horwitz DF, Mills DS, Heath S, editors. BSAVA manual of
canine and feline behavioural medicine. Quedgeley, Gloucester (UK): British Small Animal
Veterinary Association; 2002. p. 195–215.
 Kim HH, Yeon SC, Houpt K, et al. Effects of overiohysterectomy on reactivity in German
Shepherd dogs. Vet J 2006;172:154–9.
 Hashimoto S, Inoue T, Koyama T. Effects of conditioned fear stress on serotonin neurotrans-
mission and freezing behavior in rats. Eur J Pharmacol 1999;378:23–30.
 Higley JD, King ST, Hasert MF, et al. Stability of interindividual differences in serotonin
function and its relationship to severe aggression and competent social behavior in Rhesus
Macaque females. Neuropsychopharmacology 1996;14:67–76.
 Reisner I, Mann JJ, Stanley M, et al. Comparison of cerebrospinal ﬂuid monoamine
metabolite levels in dominant-aggressive and non-aggressive dogs. Brain Res 1996;714:
 Badino P, Odore R, Osella MC, et al. Modiﬁcations of serotonergic and adrenergic receptor
concentrations in the brain of aggressive Canis familiaris. Comp Biochem Physiol A Mol
Integr Physiol 2004;139:343–50.
 McCloskey MS, Berman ME, Noblett KL, et al. Intermittent explosive disorder-integrated
research diagnostic criteria: convergent and discriminant validity. J Psychiatr Res
 Best M, Williams JM, Coccaro EF. Evidence for dysfunctional prefrontal circuit in patients
with an impulsive aggressive disorder. Proc Natl Acad Sci U S A 2002;99(12):8448–53.
 Keele NB. The role of serotonin in impulsive and aggressive behaviors associated with
epilepsy-like neuronal hyperexcitability in the amygdala. Epilepsy Behav 2005;7:325–35.
1039CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS
Author's personal copy
 Jacobs C, Van Den Broeck W, Simoens P. Increased volume and neuronal number of the
basolateral nuclear group of the amygdaloid body in aggressive dogs. Brain Res
 Jacobs C, Van Den Broeck W, Simoens P. Neurokinin-1 receptor in the basolateral nuclear
group of the canine amygdala—comparative study in normal and aggressive dogs. Brain
 Sanchez MM, Ladd CO, Plotsky PM. Early adverse experience as a developmental risk
factor for later psychopathology: evidence from rodent and primate models. Dev Psychopa-
 Chapillon P, Patin V, Roy V, et al. Effects of pre- and postnatal stimulation on development,
emotional, and cognitive aspects in rodents: a review. Dev Psychobiol 2002;41:373–87.
 Anisman H, Zaharia MD, Meaney MJ, et al. Do early-life events permanently alter
behavioral and hormonal responses to stressors? Int J Dev Neurosci 1998;16(3/4):
 Roll A, Unshelm J. Aggressive conﬂicts amongst dogs and factors affecting them. Appl Anim
Behav Sci 1997;52:229–42.
 Lindsay S. Neurobiology of behavior and learning. In: Handbook of applied dog behavior
and trainingAdaptation and learning, vol. 1. Ames (IA): Iowa State University Press; 2000.
 Guszkowska M. Effects of exercise an anxiety, depression and mood. Psychiatr Pol
 Hebb AL, Poulin JF, Roach SP, et al. Cholecystokinin and endogenous opioid peptides:
interactive inﬂuence on pain, cognition, and emotion. Prog Neuropsychopharmacol Biol
 Thoren P, Floras JS, Hoffman P, et al. Endorphins and exercise: physiological mechanisms
and clinical implications. Med Sci Sports Exerc 1990;22(4):417–28.
 Hoffman MD, Shepanski MA, Ruble SB, et al. Intensity and duration threshold for aerobic
exercise-induced analgesia to pressure pain. Arch Phys Med Rehabil 2004;85:1183–7.
 Lindsay S. Neurobiology and development of aggression. In: Handbook of applied dog
behavior and training. Procedures and protocols, vol. 3. Ames (IA): Blackwell Publishing;
2005. p. 279–345.
 Lindsay S. Impulsive, extrafamilial, and intraspeciﬁc aggression. In: Handbook of applied
dog behavior and trainingProcedures and protocols, vol. 3. Ames (IA): Blackwell Publish-
ing; 2005. p. 433–555.
 Clark GI, Boyer WN. The effects of dog obedience training and behavioral counseling upon
the human-canine relationship. Appl Anim Behav Sci 1993;37:147–59.
 Kabaila A. The effects of current training techniques and environmental factors on dog
behavior. Anim Welf Sci Ess 2004.
 Overall K. Treatment of behavioral problems. In: Clinical behavioral medicine for small
animals. St. Louis (MO): Mosby; 1997. p. 274–92.
 Dodman N, Reisner I, Shuster L, et al. Effect of dietary protein content on behavior in dogs.
J Am Vet Med Assoc 1996;208(3):376–9.
 Bjork JM, Doughtery DM, Moeller G, et al. Differential behavioral effects of plasma
tryptophan depletion and loading in aggressive and nonaggressive men. Neuropsycho-
 DeNapoli JS, Dodman N, Shuster L, et al. Effect of dietary protein content and tryptophan
supplementation on dominance aggression, territorial aggression, and hyperactivity in
dogs. J Am Vet Med Assoc 2000;217(4):504–8.
 Gesch CB, Hammond SM, Hampson SE, et al. Inﬂuence of supplementary vitamins,
minerals and essential fatty acids on the antisocial behavior of young adult prisoners.
Br J Psychiatry 2002;181:22–8.
 Pageat P, Gaultier E. Current research in canine and feline pheromones. Vet Clin North Am
Small Anim Pract 2003;33(2):187–211.
Author's personal copy
 Gaultier E, Bonnafous L, Vienet-Legue D, et al. Efﬁcacy of dog appeasing pheromone in
reducing stress related behaviors of newly adopted puppies coming from a pet shop. Pre-
sented at the American College of Veterinary Behaviorists/American Veterinary Society
of Animal Behavior Scientiﬁc Paper Session, Washington, DC, July 16, 2007.
 Denenberg S, Landsberg G. Evaluation of the effect of dog appeasing pheromones on the
reduction of anxiety and fear in puppies during training. Presented at the American College
of Veterinary Behaviorists/American Veterinary Society of Animal Behavior Scientiﬁc Paper
Session, 3–4. Washington, DC, July 16, 2007.
 Mills DS, Ramos D, Estelles MG, et al. A triple blind placebo-controlled investigation into the
assessment of the effect of dog appeasing pheromone (DAP) on anxiety related behaviour of
problem dogs in the veterinary clinic. Appl Anim Behav Sci 2006;98:114–26.
 Sheppared G, Mills DS. Evaluation of dog-appeasing pheromone as a potential treatment
for dogs fearful of ﬁreworks. Vet Rec 2003;152(14):432–6.
 Wells DL. Aromatherapy for travel-induced excitement in dogs. J Am Vet Med Assoc
 Graham L, Wells DL, Hepper PG. The inﬂuence of olfactory stimulation on the behaviour of
dogs housed in a rescue shelter. Appl Anim Behav Sci 2005;91:143–53.
 Virga V, Houpt K, Scarlett JM. Efﬁcacy of amitriptyline as a pharmacological adjunct to
behavior modiﬁcation in the management of aggressive behaviors in dogs. J Am Anim
Hosp Assoc 2001;37:325–30.
 White MM, Neilson J, Hart B. Dominance-related aggression in dogs: effects of treatment
with placebo or clomipramine. Presented at American Veterinary Society of Animal Behav-
ior, New Orleans, LA, July 11, 1999.
 Dodman N, Donnelly R, Shuster L, et al. Use of ﬂuoxetine to treat dominance aggression in
dogs. J Am Vet Med Assoc 1996;209(9):1585–7.
 Crowell-Davis SL, Murray T. Selective serotonin reuptake inhibitors. In: Veterinary psycho-
pharmacology. Ames (IA): Blackwell Publishing; 2006. p. 80–110.
 Simpson BS, Papich MG. Pharmacologic management in veterinary behavioral medicine.
Vet Clin North Am Small Anim Pract 2003;33(2):365–404.
 Bourin M, Chue P, Guillon Y. Paroxetine: a review. CNS Drug Rev 2001;7(1):25–47.
 Stahl SM. Classical antidepressants, serotonin selective reuptake inhibitors, and noradren-
ergic reuptake inhibitors. In: Essential psychopharmacology: neuroscientiﬁc basis and
practical applications. Cambridge (MA): Cambridge University Press; 2000. p. 199–244.
 Crowell-Davis SL, Murray T. Tricyclic antidepressants. In: Veterinary psychopharmacology.
Ames (IA): Blackwell Publishing; 2006. p. 179–206.
 Ben-Porath DD, Taylor SP. The effects of diazepam (Valium) and aggressive disposition on
human aggression: an experimental investigation. Addict Behav 2002;27:167–77.
 Bond A, Curran HV, Bruce MS, et al. Behavioural aggression in panic disorder after
8 weeks’ treatment with alprazolam. J Affect Disord 1995;35:117–23.
 Beata C, Beaumon-Graff E, Diaz C, et al. Effects of alpha-casozepine (Zylkene) versus
selegiline hydrochloride (Selgian, Anipryl) on anxiety disorders in dogs. J Vet Behav
 Crowell-Davis SL, Murray T, Seibert LM. Veterinary psychopharmacology. Ames (IA): Black-
well Publishing; 2006.
 Plumb DC. Veterinary drug handbook. 5th edition. Ames (IA): Blackwell Publishing; 2005.
1041CANINE AGGRESSION TOWARD UNFAMILIAR PEOPLE AND DOGS