Mechanisms and Function in Dog

4
Behaviour
Per Jensen

Introduction

Ethology is the science in which we study animal behaviour, its causation and its bio-
logical function, and this entire book is devoted to the behavioural biology of dogs.
But it is not entirely clear what we understand by behaviour in the first place, and
some examples may help to reflect on this problem of definition of the subject. In its
simplest form, behaviour may be series of muscle contractions, perhaps performed
in clear response to a specific stimulus, such as in the case of a reflex – for example,
a dog scratching itself with the hindleg. However, in the other extreme, we find very
complex activities, such as a pack of wolves seizing a prey, continuously assessing
their directions and positions with the help of various cues from the environment,
from the behaviour of the prey and from the rest of the pack.
We would use the word behaviour for both of these extremes, and for many
other activities in between in complexity. It will include all types of activities that
animals engage in, such as locomotion, grooming, reproduction, caring for
young, communication, etc. Behaviour may involve one individual reacting to a
stimulus or a physiological change, but may also involve two individuals, each
responding to the activities of the other. This makes the science of ethology a
complex one, and there is a need to structure the way in which to approach the
huge task of analysing the biology of behaviour.
Since behaviour is such a complex biological phenomenon, it follows that it
can be studied from a number of different perspectives. Any science dealing with
behaviour must therefore limit its purpose, or formalize the framework of the
science. The most influential formulation of a research programme for ethology
was put forward by the Nobel Prize winner Niko Tinbergen (Tinbergen, 1963).

© CAB International 2007. The Behavioural Biology of Dogs 61

(ed. P. Jensen)
62 P. Jensen

This programme is frequently referred to as ‘Tinbergen’s four questions’, and it

outlines the specific aspects that he felt were ethology’s central task to study.
Assume that we are interested in studying the behaviour of barking in dogs (see
also Chapter 12 in this volume for a more thorough discussion of barking). What
sorts of questions should we pose? According to Tinbergen, ethology can ask four
different types of questions, which are not mutually exclusive. A complete etho-
logical understanding of barking requires that we provide answers on all of them,
albeit not necessarily in one and the same study:
1. What causes barking? The answer to this question refers to the immediate
causes, such as which stimuli elicit or stimulate a particular behaviour. In this
question, we may also ask which physiological variables, such as hormones, cause
or mediate the performance of barking.
2. What is the function of barking? In this case, the answer describes how the
performance of a behaviour affects the reproductive success, the fitness, of the
dog. It therefore has to do with evolutionary aspects and consequences.
3. How does barking develop during ontogeny? Studies of this question aim at
describing the way a behaviour is modified by individual experiences. It will also
cover questions such as how barking matures, and how it changes with the age of
the dog, and how it is affected by learning.
4. How has barking developed during phylogeny? This is a clearly evolutionary
question, and usually calls for comparative studies of related species. It is of course
not possible to examine the vocal behaviour of fossils, but comparing vocalizations
of different now living, related canids, may help to understand when the behaviour
developed and how it has been preserved in different lineages of species.
The interested dog biologist – who is likely to have a close personal relation
with dogs – might object that Tinbergen’s questions are incomplete, since they do
not consider what animals perceive, feel and know in relation to their own behav-
iour, i.e. the emotional and cognitive aspects of behaviour. This aspect of animal
behaviour was long considered to be inaccessible to science. However, over the
last decades, scientists have developed methods and concepts to allow investiga-
tion into this area. This has led to a new branch of the science, emerging in the
1970s, known as cognitive ethology (Bekoff, 2000). In this book, Chapter 12 is
largely devoted to these aspects of dog behaviour. However, proper understand-
ing of dog behavioural biology requires that the Tinbergen questions are exam-
ined systematically and, based on the answers to them, we can proceed and pose
new questions regarding cognitive aspects of behaviour.

Nature and Nurture

An important question concerns the relative contribution of inherited traits

(genetic factors) and acquired traits (for example, learned) in the expression of a
particular behaviour. This is sometimes referred to as the nature versus nurture
debate. Chapter 5 in this book is particularly concerned with the genetic aspects
 Mechanisms and Function in Behaviour 63

of dog behaviour. Here, I will just emphasize that all behaviour, whether we refer
to it as innate or learned, is necessarily always a product of both genetic and envi-
ronmental factors, i.e. it is always a question of nature AND nurture. This double
necessity is particularly evident in some developmental processes, such as the mat-
uration and socialization of dog pups, which is treated later in this chapter.
We can easily understand the wider aspects of the ‘nature-and-nurture view’
by considering a typical learned behaviour, such as retrieving game. In order for
the dog to learn to retrieve on command, it needs some basic behavioural dispo-
sition to chase and carry objects. This disposition can be selected for, as we know
from specialized retrieving breeds such as labradors, so obviously there is a
genetic basis for this behaviour. However, we can teach the dog to connect a
signal (for example, a spoken command) with the performance of the behaviour,
which is an acquired process – no dog is born with the knowledge that ‘fetch’
means ‘run and fetch that dead bird and bring it to me’. In order to be able to
form this connection, the dog needs sensory organs that can perceive the signal
(ears in this case), a nervous system that can process it and form the connection
between the correct behavioural response and a reward, and a locomotory system
(muscles and joints) that allows it to retrieve. All these features are constrained by
genetic factors. Therefore, any behaviour will always be the result of an interac-
tion between genetic predispositions and environmental variables.

Behaviour is a Brain-product
Imagine a dog performing a typical canine behaviour: it carries a bone away from
the family, digs a hole in the soil, drops the bone in the hole and covers it with soil,
using the nose to push the soil in place. It is self-evident that the behaviour of the
dog is controlled by the brain, but how? Information reaches the brain from the
sense organs (smell and shape of the object, visual information from the surround-
ings) and is interpreted there in some meaningful way to the dog (‘the stimuli rep-
resent a bone’; ‘this is a suitable place to dig’). Following this interpretation, the
brain sends nerve impulses to the appropriate muscles, which make the dog move
and start digging. From watching many dogs performing this type of behaviour,
one can easily conclude that the pattern is similar for different members of the
species: holes are normally dug with specific movements with the forelegs, the bone
is normally always covered with the help of the nose. The reason is that the
complex of movements is guided by specific programmes by which the brain con-
trols motor activity. If we want to understand the mechanisms of dog behaviour, it
therefore seems reasonable to start exploring the brain and the nervous system.

The Brain and the Nervous System

In all vertebrates, dogs included, the brain develops in the embryo from the so-
called neural tube (Alcock, 2001). Early, it differentiates into three different parts,
64 P. Jensen

the forebrain, midbrain and hindbrain. Somewhat later, the forebrain differenti-
ates further into the telencephalon and the diencephalon, the hindbrain differen-
tiates into metencephalon and myelencephalon, and the midbrain stays as a
functionally intact unit in the adult (the mesencephalon). Of course, the relative
specialization of the different brain parts varies between species. In mammals,
telencephalon develops into cerebrum, which includes the cerebral cortex, and
this structure is extremely variable between species. It is the part of the brain con-
taining centres for integration of sensory stimuli, and for conscious reasoning and
reflection – in particular a part of the cortex called the neocortex, which is unique
to mammals. Humans have the most developed cerebral cortex, and the neocor-
tex makes up about 80% of our total brain mass. Dogs, like other mammals, have
the same structures, but they constitute a relatively much smaller part of the
brain.
One part of the cerebral cortex is common to all mammals (and reptiles), and
is referred to as the limbic system (Alcock, 2001). This consists of the hippocam-
pus and the olfactory cortex (where olfactory input is integrated, see Chapter 6),
and parts of the thalamus and hypothalamus of the diencephalon. This region of
the brain contains centres involved in the immediate control of basic behavioural
programmes, related to feeding, aggression and sexual behaviour. The limbic
system also connects to sensory areas in the neocortex and is responsible for
attaching emotions, feelings, to behaviours. Human patients receiving electro-
stimulation in different nuclei of the limbic system report sudden feelings of fear
or anger, and so on, depending on the exact localization of the stimulation
(Carlson, 1981).
Whereas dogs differ considerably from humans in the structure and size of
the cerebral cortex and neocortex, the limbic system is very similar both in struc-
ture and relative size. This is, of course, a strong neurobiological argument for the
assumption that dogs are able to perceive more or less the same range of feelings
as we do, but have a limited ability to reflect consciously on these feelings.

Neurons and their Organization

The information which the brain receives is carried via neurons, specialized cells
which are able to propagate an electric current, called an action potential
(Carlson, 1981). Neurons connect to each other via a so-called synapse. A synapse
may either be excitatory, meaning that the action potential in the first neuron
elicits a similar action potential in the second, or inhibitory, in which case the
action potential of the first neuron prevents the second from releasing any elec-
trical activity. The central nervous system (CNS) consists of the brain and the
spinal cord, both of which are made up of billions of interconnected neurons,
which in turn are organized into different nuclei and areas specialized in handling
certain types of information processing.
The part of the nervous system not included in the brain and spinal cord is
called the peripheral nervous system (PNS). This is the system which receives
 Mechanisms and Function in Behaviour 65

information from sensory organs of different kinds. Some of them monitor events
outside the body, and these are treated in more detail in Chapter 6. Others
monitor internal events in the body, such as blood glucose levels, gut extension
and muscle tension, sends this information to the brain and allows the CNS to
consistently maintain a scheme of the state of the body and its surroundings.
The PNS also has outgoing, efferent, neurons, which innervate muscles and
other organs. This division of PNS is divided into two separate systems, the
somatic and the autonomic systems. The somatic system controls the aspects of
motor activity which humans can consciously affect, for example the skeletal
muscles. The autonomic system is responsible for the control of smooth and
cardiac muscles, and the gastrointestinal and endocrine systems, and can usually
not be controlled consciously. The autonomic system is further divided into the
sympathetic and the parasympathetic branch, which together determine the
degree of arousal and tension of the body. Neurons from both branches innervate
most organs of the body, and the relative strength of activation of each of the
branches determines the effect on the organ. For example, the sympathetic
neurons increase heart rate, inhibit digestion and increase breathing; parasympa-
thetic neurons do the opposite (Fig. 4.1).

Fig. 4.1. An overview of the innervation of different organs by sympathetic (solid

lines) and parasympathetic (dotted lines) parts of the autonomic nervous system. A
particular organ is normally innervated by both sympathetic and parasympathetic
neurons, which exert opposite effects – if one stimulates the activity of the organ,
the other inhibits. In this picture only a few examples of innervated organs are
shown: salivary glands, bronchs, heart, stomach, pancreas, adrenal medulla,
intestines, sexual organs and bladder.
66 P. Jensen

The protrusions of neurons (axons) from brain and spinal cord are collected
in bundles, nerves, which are often embedded in myelin, a protein which protects
the neurons and increases transmission speed of action potentials. Some neuron
bundles are not myelinated, and have a slower transmission rate.

Stimuli

Stimuli from the surroundings constantly flow over every individual. A small part
of all available information is recorded by specialized sense organs, and is trans-
mitted to the brain in the form of visual, olfactory, auditory and mechanic stimuli.
The fraction of the available information that is detected and perceived by an
animal is first limited by the structure and function of the sensory organs – this is
dealt with extensively in Chapter 6.
When the stimuli reach the brain, they are interpreted in a meaningful way
to the animal. The male dog brain may identify a collection of olfactory stimuli
as, for example, a female in heat, a potential prey, or the territorial mark of
an intruding male. Interestingly, the interpretative patterns are to a large
extent innate and do not have to be learned. This is the basis of so-called key
stimuli – a collection of stimuli is automatically interpreted in a specific, functional
way. A male three-spined stickleback (a small fish) reacts to anything roundish with
a red ventral part as if it was an intruder – the key stimulus for a rival is simply
‘roundish and red at the bottom’. Similarly, dogs do not need specific training to
identify the meaning of social signals such as a wagging tail. On the other hand, it
takes extensive experience before a dog can learn that a cat wagging its tail is not
in a friendly mood, since this violates the innate interpretation pattern of this
signal.
Key stimuli are usually linked to specific behavioural responses. A small,
furry object which moves away from the dog very quickly will release chasing
behaviour. Racing dogs, such as whippets, are fooled into running fast by the key
stimuli of a prey attached to an arm on an engine. A dog can learn not to react
to a key stimulus, but if not specifically trained, they will behave according to the
innate interpretation of the sensory input.

The Motor Apparatus

Motor activity is the central aspect of all behaviour. It occurs as a result of syn-
chronized pathways of muscle contractions, which cause the animal to move in a
functional manner in relation to a particular stimulus. The simplest forms of coor-
dinated muscle activity are commonly referred to as reflexes, and these were dis-
covered and studied extensively in dogs by the Russian physiologist Pavlov (see
Chapter 8 for further discussion of Pavlov’s findings). A reflex is a simple, non-
variable, motor response to a specific stimulus (Toates, 2002). Usually, the
reflex involves neurons that do not necessarily connect to the CNS, so
 Mechanisms and Function in Behaviour 67

Mylohyoideus

Geniohyoideus

Posterior tongue

Palatopharyngeus

Superior constrictor

Thyrohyoideus

Thyroarytenoideus

Middle constrictor

Cricothyroideus

Inferior constrictor

Diaphragm

Fig. 4.2. The sequence of contractions of a number of muscles, highly coordinated

in order to form the swallowing reflex of the dog. Reprinted with permission from
McGraw Hill.

the reflex motion can be carried out without any conscious awareness or
feelings. In humans, the knee-jerk reflex is a common example, and in dogs, the
swallowing reflex has been extensively studied (Fig. 4.2). Although simple, the
reflex may involve a large number of muscles which contract and relax in a highly
coordinated and patterned fashion, as can be seen in the swallowing reflex of
dogs.
A somewhat more complicated pattern of muscular activity is the so-called
modal action pattern (MAP), often referred to as the fixed action pattern. MAPs
are again highly structured, but usually more complex than reflexes. They nor-
mally involve the CNS, but once elicited, the full motor programme proceeds
with minimal interference from the CNS. This means that the MAP usually pro-
ceeds to the end once it has been released. For example, a dog will often attempt
to roll and rub in substances with a strong smell. Once the behaviour has been
released, it will proceed largely in the same manner every time; first the dog bends
68 P. Jensen

one foreleg, then it places the shoulder part against the odorous substance and
finally rolls over on its back while wriggling the body. The neural programme
controlling the behaviour is tightly wired, which makes it virtually impossible for
the dog to perform it in a different way (for example, a dog normally can not
decide to rub its belly in the smell instead). However, there is some flexibility built
into the control system, for example allowing the dog to vary duration and inten-
sity, or to interrupt the behaviour if strongly stimulated – that is why the pattern
is preferably called ‘modal’ rather than ‘fixed’ (Toates, 2002).
The total motor output of a dog, or any animal, is to a large extent a complex
of more or less flexible combinations of series of reflexes and MAPs. Taken
together, this can be referred to as ‘motor programmes’, a term which implies that
there is certain level of neural organization of the order and structure of the
muscle contractions, but also that there is some flexibility to allow the animal to
adjust the motor output to the prevailing circumstances.
In summary, this means that what an animal can do is highly constrained by
the organization of the neural system and the muscles. This is one important
reason for why animals have clearly identifiable species-specific behaviour. The
dog which we met earlier will have great problems in deciding to cover its bone
with soil in any other manner than by using its nose – a cat would rather use its
forelegs to scratch soil over something it attempts to cover.

Endocrine Moderation of Behaviour

As we have seen, the brain interprets incoming stimuli and controls motor pro-
grammes. A particular set of stimuli (key stimuli) send, to give rise to a particular
motor programme. However, this system is controlled in all parts by the chemi-
cal environment of the body. Hormones moderate and affect behavioural output
in a number of ways, and they are all part of the body’s endocrine system (Toates,
2002).
Hormones of various types are secreted from glands in different parts of the
body, and they have a variety of biological functions. Some control metabolic
pathways and gastro-intestinal activity, others affect sex organs and yet others are
linked to the function of the immune system. Most hormones serve many differ-
ent purposes, but in this chapter we are mostly concerned about those hormones
that affect behaviour in various ways.
Some of the most important hormones in this respect are those which are
controlled by, or secreted directly by, the hypophysis (pituitary). This gland is
located just under the brain, with nerve projections directly from the hypothala-
mus. It secretes hormones which have a stimulatory effect on other glands in the
body, and thereby has a central regulatory role in the endocrine system. For
example, the hypophyseal hormone FSH (follicle-stimulating hormone) stimu-
lates the testes of male dogs and causes them to increase testosterone secretion.
Testosterone has a broad behavioural effect on the dog; for example, it increases
its sexual responsiveness and lowers its threshold for aggression (Toates, 2002).
 Mechanisms and Function in Behaviour 69

Another hypophyseal hormone, oxytocin, exerts a direct behavioural effect by

calming the dog (Uvnas Moberg, 1994).
Hormones do not cause or inhibit behaviour by themselves. Rather, they
affect the sensitivity of the neural pathways involved in different behaviours. For
example, testosterone sensitizes the responsiveness for stimuli which are related
both to aggression and sexual behaviour. In the male dog, less olfactory stimuli
from females in heat, and less visual stimuli from a rival are needed to start
the sexual and aggressive motor programmes when testosterone levels are
high. Furthermore, testosterone strengthens the motor output, so sexual and
aggressive behaviour is performed at a higher intensity when the hormone levels
are higher.
Although the endocrine system is dynamic and can change dramatically over
a short time, some effects are more lasting than others. For example, testosterone
and oestrogen have several important functions in modulating the nervous system
during development. A male embryo which is exposed to high levels
of oestrogen in the uterus will become feminized in a number of ways (vom
Saal et al., 1983). For example, it will generally be less aggressive when adult.
This modulation of the nervous system goes on at least up to puberty, and this
is the reason why aggressive male dogs castrated late in life may not respond as
the owner wishes. When the dog approaches puberty, the nervous system is to
some extent already ‘calibrated’ on a certain aggression threshold, and the
daily fluctuations in testosterone levels become less important for the level of
aggressiveness.

Development of Behaviour

A newborn dog is a rather helpless creature – the eyes are closed, acoustic abili-
ties utterly limited, and the olfactory system very immature. The motor pro-
grammes are limited to a few essential movement patterns necessary to obtain
food and get rid of metabolic waste products. However, within a few weeks, the
dog has developed a wide array of perception abilities and motor skills, and a few
weeks later it has finished some of the most important developmental pathways in
its life. This developmental process is a beautiful example of how genetics and
experience are both necessary parts of normal behavioural development. The
process is often divided into four periods: the neonatal period, the transition
period, the socialization period and the juvenile period (Braastad and Bakken,
2002).
During the neonatal period, the pups are completely dependent on their
mother. They react to tactile stimuli and possibly smell. Movement is undevel-
oped and consists of slow crawling and head oscillations. They also emit whining
or yelping vocalizations.
During the transition period, the adult behaviour patterns start to develop.
The eye-opening at approximately 13 days of age marks the start of the period,
which ends when the ear channels open at 18–20 days. During the transition
70 P. Jensen

period, the motor abilities develop, and eventually the pup will start walking.
Eliminative behaviour no longer requires tactile stimulation from the mother.
Play-fighting and tail-wagging also start to develop during this period, and the
vocalizations become more variable.
During the period of socialization between 3 and 8 weeks, the young dog
starts to display most adult behaviour patterns. During this period, the first signs
of fear can also be observed. Social behaviour develops, as can be seen by the ten-
dency to coordinate activities, and the first signs of aggression show up. During
this phase, social bonds with other pups and with the mother and humans
develop. The most sensitive period in this respect is 4–8 weeks of age. If the pups
are deprived of human contact during this period, they may be extremely difficult
to tame later on. During this limited time, the nervous system is predisposed for
learning what a familiar flock and family member looks and smells like, and this
capacity is largely lost after the socialization period.
By 8 months of age, the dog is more or less fully grown. During the juvenile
phase, which follows after the socialization period, basic behaviour is relatively
invariant, although motor abilities are improved.
Between 6 and 14 months of age, dogs of most breeds enter sexual matura-
tion. Sexual as well as aggressive behaviour develops rapidly. The male dog starts
performing raised leg urinations (RLU) and regardless of sex, social rank disputes
are common.

Development and Calibration of the Behavioural System

The processes going on during pup development can be formalized with the help
of a developmental model suggested by the Canadian scientist Jerry Hogan
(Hogan, 1988). Following the reasoning from the previous parts of this chapter,
the behaviour system can be said to be made up of three different parts: the per-
ceptional system, the motor system and the central system (Fig. 4.3).
The reaction norms of some parts of each of these systems are genetically
predetermined, whereas others require environmental input, learning. For
example, the newborn dog probably has no specific knowledge of the concept of
‘mother’, this must be learnt. But the pathways for which stimuli that should be
combined into a ‘mother’-concept are predetermined, so the pup will specifically
be sensitive for certain combinations of sensations. As soon as it opens it eyes it
will start looking for stimuli complexes that make sense and eventually this
complex will be assigned the concept of ‘mother’. Note that the genetic predispo-
sitions steer in considerable detail what and when to learn – it is therefore impos-
sible to call the developmental process either genetic or acquired – it is both at the
same time.
In the same way, some motor output is pre-wired at birth, whereas other
output depends on training and acquisition of skills. This ensures that the pattern
which develops is both species-specific and adapted to the precise circumstances
under which the individual grows up.
 Mechanisms and Function in Behaviour 71

Central
mechanisms

Stimuli Behaviour

Perceptual Motor
mechanisms mechanisms

Fig. 4.3. A conceptual model for a behaviour system. Stimuli from the external
world are analysed by perceptual mechanisms, which are integrated by central
mechanisms or relayed directly to motor mechanisms, the output of which equals
behaviour. Central mechanisms may affect several behaviour outputs and be
related to several types of sensory input. Redrawn after Hogan (1988).

Motivation and Conflict

Now that we have examined how a particular behaviour develops and is con-
trolled, what remains to be investigated is the central process whereby a dog
determines whether or not a particular stimulus is salient enough to warrant
action. This will, of course, also depend on which other stimuli impinge on the
dog at the moment, and which other activities it is occupied by. Simply spoken,
the dog needs to prioritize its activities.
The processes involved in making these kinds of decision are usually referred
to as motivation (Toates, 2002). The concept can be loosely defined as the likeli-
hood that an animal will perform a particular behaviour at a certain time. For
example, the probability that a dog will start drinking in a certain time period
could be taken as a measure of its drinking motivation, and the probability that it
will ingest food, of its feeding motivation. Often, motivation is used with a wider
72 P. Jensen

meaning to include all processes and rules that determine an animal’s decision of
whether to engage in a particular behaviour or not.
If we could watch a dog in a situation where only stimuli relevant for one par-
ticular behaviour system is present, we might be able to measure its motivation
for that behaviour rather precisely – it would be an inverse function of the neces-
sary strength of stimuli needed for releasing a response. For example, imagine
that a dog is kept in a closed, evenly lit room without windows. A bowl of food is
placed in front of the dog, and we want to measure how long it takes before it
starts feeding as a measure of its feeding motivation. If we had been able to keep
all other stimuli than food constant, we might then achieve a good measurement.
However, we are not likely to succeed. Even in this highly controlled situa-
tion, the dog is exposed to a series of conflicting motivations. First of all, it is
alone, so the motivation to gain social contact with other dogs or humans might
be high. Second, we cannot stop physiological processes, so the dog may be dehy-
drated and have a high motivation to drink. Similarly it may be motivated to
explore the room, to escape the situation or to test its social status in relation to
the person providing the food bowl. So, even in this highly artificial situation, the
dog’s behaviour is a result of the solution of a conflict between many different
competing motivational systems.
Motivations are thought to be neurally arranged in a hierarchical fashion
(Fig. 4.4). On the top level, the central motivational decisions are made, such as
whether to be active or to sleep. Once a high-level decision has been made, the
lower levels follow in a more or less automatic fashion. When the dog has decided
to follow its motivation to feed, other motivations are closed down, and the
feeding system has the exclusive control over the motor system until some other
motivation takes over.
Whereas the strength of some motivational systems are mainly a function of
the prevailing sensory input, others fluctuate independent of stimuli. These moti-
vations form the basis of rhythmical behaviour. For example, sleep motivation
follows a diurnal rhythm – even in a room with light all day, a dog will settle into
a regular diurnal sleep–wake rhythm (Toates, 2002). Some motivational systems
are to a certain extent self-regulated – the motor system will be activated at
regular intervals, or at a certain time since it was last active (Hogan, 1988). This
has the consequence that animals will have an urge to perform some behaviour
patterns regularly even in the absence of relevant stimuli. In dogs, chasing and
biting may be partly regulated in this way. Behaviours with this type of motiva-
tional control are sometimes referred to as ethological or behavioural needs, and
the failure to allow for these needs may cause development of abnormal behav-
iours of different kinds (Jensen and Toates, 1993) (see also Chapter 13).

Stress and Stereotypies

Biological stress is one of the most well investigated physiological and behavioural
phenomena, and it has a high significance for dog welfare. Stress can be defined
 Mechanisms and Function in Behaviour 73

Highest level
(reproduction)

Intermediate level
(copulation)

Muscle level
(pelvic muscles)

Fig. 4.4. Tinbergen’s view of the hierarchical organization of behaviour, in this

example applied to the reproductive behaviour of dogs. Behaviour is thought to
be controlled on different hierarchical levels, and within each level, the systems
are mutually inhibitive. If, for example, reproductive behaviour is in control at the
top level, other systems, such as sleep, feeding etc., will be inhibited. The same
applies on all levels, so the reproductive system controls the use of the individual
muscles, for example to allow the dog to copulate.

as the state of an animal when it is not able to act in accordance with its motiva-
tional state (Jensen and Toates, 1997). In this situation, there will normally be ele-
vated levels of cortisol, a hormone secreted by the adrenal cortex following
stimulation by the hypophysis hormone ACTH. There will also be an increased
activity in the sympathetic nervous system, which is associated with increased
heart rate and blood pressure, plus a number of other sympathetically regulated
states.
Dogs suffering prolonged periods of high stress will run the risk of develop-
ing various diseases. For example, the immune system may be down-regulated,
increasing the susceptibility for infections, and the cardiovascular system may be
damaged. When a stress situation is chronic or persistent, dogs, like other animals,
will try alternative ways to cope with the stress situations, and they may develop
various behaviour disorders. A more comprehensive description of behaviour
disorders in dogs, and their causes, can be found in Chapter 13.
A typical class of stress-induced behaviour disorders is stereotypies, defined
as movements which are repeated over and over again, without any obvious func-
tion (Lawrence and Rushen, 1993). These abnormal behaviours are common
in zoos, on farms and in laboratory units, where animals often have limited
possibilities to express some highly motivated behaviours. They are also not
74 P. Jensen

uncommon in dogs living under conditions which do not allow a full expression
of normal behaviour. Typically, an animal with a highly motivated, but thwarted,
behaviour, will attempt to perform parts of the behavioural programme, and then
repeat this part for an extended period of time. Horses with limited locomotory
possibilities may develop weaving (swaying from side to side while putting the
weight on alternative front legs), and horses with limited possibilities for foraging
behaviour may develop crib biting (biting edges of the crib, stall or fence).
Common stereotypies in dogs are, for example, tail chasing and wall jumping.
Stereotypies may be difficult to cure once they have developed, so to avoid them,
preventive measures are most efficient.
Stress can also lead to increased aggression, which the dog may direct either
towards other dogs, humans, including the owner, or both (see Chapter 13). The
mechanism here may be that the increased sympathetic nerve activity puts the
dog in a state of ‘fight or flight’, which lowers the threshold for aggression-induc-
ing stimuli. However, increased aggression as such is not diagnostic for stress,
since many other conditions may give the same result (see Chapters 13
and 14).

Final Remarks

The behaviour of dogs is – just as for any animal – a result of how the nervous
system interprets sensory information and transforms this into muscle activity.
Sensory organs, the structure of the nervous system, and the ways in which
muscles and bones build up the body are all processes that are under the control
of genetic factors. However, as we have seen, both the development of the dog,
and the processing of information, depend on environmental input. Behaviour is
therefore truly an effect of both genes and environment. Understanding the way
in which behaviour is controlled and executed is essential for understanding how
dogs experience their lives and why some individuals sometimes develop abnor-
mal and unwanted behaviour.

References
Alcock, J. (2001) Animal Behavior – An Evolutionary Approach, 7th edn. Sinauer
Associates Inc., Sunderland, Massachusetts.
Bekoff, M. (2000) Animal emotions: exploring passionate natures. Bioscience 50,
861–870.
Braastad, B. and Bakken, M. (2002) Behaviour of dogs and cats. In: Jensen, P. (ed.)
The Ethology of Domestic Animals – An Introductory Text. CABI, Wallingford,
UK, pp. 173–193.
Carlson, N.R. (1981) Physiology of Behavior. Allyn and Bacon, Toronto, Ontario.
Hogan, J.A. (1988) Cause and function in the development of behavior systems. In:
Blass, E.M. (ed.) Handbook of Behavioral Neurobiology. Plenum Press, New
York, pp. 3–15.
 Mechanisms and Function in Behaviour 75

Jensen, P. and Toates, F.M. (1993) Who needs ‘behavioural needs’? Motivational
aspects of the needs of animals. Applied Animal Behaviour Science 37,
161–181.
Jensen, P. and Toates, F.M. (1997) Stress as a state of motivational systems. Applied
Animal Behaviour Science 54, 235–243.
Lawrence, A.B. and Rushen, J. (eds) (1993) Stereotypic Animal Behaviour –
Fundamentals and Applications to Welfare. CAB International, Wallingford,
UK.
Tinbergen, N. (1963) On aims and methods of ethology. Zeitschrift für
Tierpsychologie 20, 410–433.
Toates, F.M. (2002) Physiology, motivation and the organization of behaviour. In:
Jensen, P. (ed.) The Ethology of Domestic Animals – An Introductory Text. CAB
International, Wallingford, UK, pp. 31–50.
Uvnas Moberg, K. (1994) Role of efferent and afferent vagal nerve activity during
reproduction: integrating function of oxytocin on metabolism and behaviour.
Psychoneuroendocrinology 19, 687–695.
vom Saal, F.S., Grant, W.M., McMullen, C.W. and Laves, K.S. (1983) High fetal
estrogen concentrations: correlation with increased adult sexual activity and
decreased aggression in male mice. Science 220, 1306–1309.