Transcript Canine Learning - Session 3

CANINE
LEARNING
Canine Behaviour & Training – Session 3
CANINE LEARNING-HOW DOGS KNOW
TO DO WHAT THEY DO!
- Adaptive Learning
 - Effects on Learning
 - History of Learning
 - Ways of Learning


- Classical Conditioning
- Operant Conditioning
 - Latent Learning
 - Social Learning

Habituation
o Insight
o Using understanding of learning with
behaviour problems
o
ADAPTIVE LEARNING
According to Darwin animals are constantly
evolving to become the best adapted to their
environment.
Basically dogs will learn to do what works.
What helps them as individuals and their specie
survive. (find food, protect against threats and
reproduce)
WHAT HELPS DOGS SURVIVE?
Food
 Water
 Being in a group
 Giving clear communication
 Defending self and family from threats
 Keeping clean
 Reproduction

WHAT A DOG NEEDS WILL ACT AS A
MOTIVATOR

They will strive to gain what they are motivated
to, and in the process will learn.
WHAT AFFECTS LEARNING?
The success of learning will be based upon some of
the following:
Innate ability
 Dogs Health (internal environment)
 Environment (external)
 Reinforcement type and schedule

INNATE ABILITY
Having received the genes that allow for vital
instinctive behaviour
Instincts do not require learning, and all animals of
a specie will perform the same instinctive
behaviours, but the success of the behaviour may
be variable.

We can learn about animal instinct (and their
motivations) by
1. Observing,
2. Putting into test situations
3. Factor out learning
DOG’S HEALTH (INTERNAL ENVIRONMENT)
Can affect sensory input
 May affect ability to respond to information
(Illness is a physiological stressor)
 Mental health will also affect learning
(stress inhibits neurotransmission)
 Diet, food allergies and drugs will also affect
behaviour and learning

ENVIRONMENT
This can distract from learning
 Environment causes arousal

REINFORCEMENT TYPES & SCHEDULES
…more about this later, but basically the thing
that encourages (or discourages) a behaviour
(consequence) and to what degree (value of
reinforcer).
COMBINED EFFECT ON LEARNING


Its is most likely that what comes naturally to a
dog (instinct), what a dog has learnt (previous
experience) and influences of internal and
external environment (health and surroundings)
will all have come together to effect what a dog
learns.
This is ultimately the INTERACTIONALIST
approach, a combination of the NATURE and
NURTURE approach.
HISTORY OF LEARNING
LEARNING HISTORY
Greek PhilosophersLearning was about finding ‘truth’
Socrates
Plato
Aristotle
(469-399BC)
(427-347BC)
(384-322BC)
Dialectism
Rationalism
Empiricism
Discovering truth
through
conversation
Discovering
truth through
self-reflection
Discovering
about the
world around
by using senses
LEARNING HISTORY
Roman PhilosophersLearning was about developing skill
Roman Catholic Church (500-1500AD)
• Memorisation
• Recitation
Transmission based Learning
LEARNING HISTORY
Renaissance Philosophers- (15001700AD)
Influential figures at this time and their ideas:
• Descartes (1596-1650): Revival of Plato’s ideas of innate
knowledge, defined reflex action (early nature
approach)
• Locke (1632-1704): Revived ideas of Aristotle that the
mind is a blank canvas to be shaped by environment
(nurture approach)
• Rousseau (1712-1778): Presented ideas of shaping that
occurs through experience of things around (nurture)
• Kant (1724-1804): Extended Plato’s rationalist theory to
include a ‘prior’ knowledge which can be built on with
experience (interactionalist)
LEARNING HISTORY
Psychology based learning- (1800-1900AD)
Scientific approaches
Thorndike
Pavlov
Skinner
(1874-1949)
(1849-1936)
(1904-1990)
Stimulus Response
Association
with trial and error
consequences
(an impulse reaction)
Learning by
Trial and Error
association.
Outcomes and
Natural responses
and learnt responses
LEARNING HISTORY
Skinner and beyond:
• Skinner: “there are certain questions that have to be answered in turning to the
study of any new organism. What behaviour is to be set up? What reinforcers are at
hand? What responses are available in embarking on a program of progressive
approximation that will lead to the final form of behaviour? How can reinforcers be
most effectively scheduled to maintain the behaviour in strength?”
• Piaget (1896-1980): “learning is a developmental cognitive process, that
students create knowledge rather than receive knowledge from the teacher. Students
construct knowledge based on their experiences and how they do so is related to their
biological, physical and mental stage of development”
• Vygotsky (1896-1934): “learning also occurs in a cultural context and
involves social interactions. Culture and language play an important role in developing
new ideas and skills”. Put forth idea of Zone Proximal Development (ZPD) meaning
students learn best just beyond there range of existing experience with assistance
form teacher or peer to bridge of what they know or can do independently to what thye
have yet to learn or need assistance with.
These led to Progressive (Current) Learning Theory
WAYS OF LEARNING
MAIN LEARNING TYPES

Classical Conditioning:
Learning by association; linking non relevant stimulus to
relevant
stimulus to cause the same behavioural response. (I. Pavlov)
 Operant Conditioning:
Learning by association- that behaviours have consequences,
both positive and negative and can increase or decrease the
likelihood of a behaviour occurring again. Trial and error
learning.(B.F Skinner)
 Latent Learning: exploratory learning with consolidation and
without immediate performance
 Habituation: learning not to respond
 Social Learning: Copying the performance of another
 Insight: mental trail and error (humans only)
BIOLOGY OF LEARNING
When an animal makes an association between a
stimulus and a consequence/behaviour neural
connections are made in the brain. When the
association is repeated the connection gets
strengthened allowing for quicker behavioural
responses or habituation.
EXAMPLES OF LEARNING BY ASSOCIATION
CLASSICAL CONDITIONING
UCS= Unconditioned
Stimulus
UCR=
Unconditioned
Response
CS=
Conditioned
Stimulus
CR= Conditioned
Response
*’conditioned’ can be
replaced with ‘learnt’
THEORY APPLICATION IN DOGS

Classical Conditioning
An animal taken to the vet when experiencing pain
can act fearfully of vets in the future.
Pain (UCS)
Fear (UCR)
Pain (UCS)+ Vet
Fear (UCR)
Vet (CS)
Fear (CR)
Likewise a dog that accidently gets a dart from an
electric wire.
Pain from shock (UCS)
Fear (UCR)
Pain (UCS)+ Wire
Fear (UCR)
Wire (CS)
Fear (CR)
Can be positive too...
Food (UCS)
Pleasant Feeling (UCR)
Food (UCS)+ Bowl
Pleasant Feeling(UCR)
Bowl (CS)
Anticipation of food (CR)
CHAINING

The association of component parts into a
sequence that predict the same outcome.
SENSITISATION
Being sensitive to…
 Often a negative association to a stimulus that is
pain or fear inducing

OPERANT CONDITIONING
Skinner carried out many
experiments to show that
animals such as rats and
chicken could learn to make
choices based upon positive
and negative
responses/reinforcers. Rats
learnt to press levers and
chickens learnt to peck keys
to receive food rewards.
THEORY APPLICATION IN DOGS
Operant Conditioning
Positive Reinforcement
Negative Reinforcement
Negative Punishment
Positive Punishment
PRIMARY & SECONDARY REINFORCEMENT
A primary reinforcer is a ‘natural’ reinforcer.
(what need or want)



A secondary reinforcer is paired to a primary and
starts to become as reinforcing as a primary.
A primary reinforcer must always follow a
secondary
REINFORCEMENT SCHEDULES

Continuous- reward every time.

Fixed Ratio- reward at a set interval

Variable Rate (Intermittent)- reward randomly
In positive reinforcement training, in reality a
continuous rate is used, but the value of
reinforcement is variable.
SHAPING
Sequences of successive approximations
(the building up of a behaviour).

Shaping can be forward
Or reverse
POTENTIAL PROBLEMS WITH OPERANT
CONDITIONING


Satiation- as satiation increases the motivation
for the reward decreases
Extinction- when a behaviour stops being
performed as there is not apparent gain of
reward, but can spontaneously recover upon
reward again.
LATENT LEARNING
Gaining knowledge
‘passively’ but
delaying the demonstration of
knowledge or performance
of behaviour
Requires motivation and an evident
Reinforcement, most common example is
maze learning or finding the way home.
Thought to be the association of stimuli within a sequence
THEORY APPLICATION IN DOG

Latent Learning
SOCIAL LEARNING

Observing and copying, carried out through
cultural transmission
THEORY APPLICATION IN DOGS

Social Learning
HABITUATION
A simple form of learning where an animal learns to
make reduce and then stop making behavioural
responses to a specific stimuli.
This is a very functional
behaviour for animals.
When actively performing
habituation it is usually carried out in a
systematic desensitisation procedure and
starts with the stimuli at a mild level
with gradual progression.
Classical conditioning can also be used.
THEORY APPLICATION IN DOGS

Habituation
INSIGHT
A mental form of trial and
error, being able to figure
something out without
trying, just looking.
It is thought by most that
only humans have
insight, but it has been
suggested that larger
apes perform insight
behaviour also.
Sixth sense?...or response to
subtle cues?
Learning Summary
Animals and people learn mostly by
making links or associations and can
often make these associations
through positive or negative
consequences.
UNDERSTANDING OF LEARNING IN
PROBLEM BEHAVIOUR
Systematic desensitisation
Reduce sensitivity to a specific stimulus in
progressive steps aiming for habituation.

Counter-conditioning
Teaching of an alternative (sometimes referred to
in the same way as classical conditioning)

EXAMPLE: SEPARATION ANXIETY
DOG REACTIVITY
ALTERNATIVE TO DESENSITISATION & NOT
ADVISABLE=FLOODING