LEARNING AND MOTIVATION

LEARNING AND MOTIVATION
1. Basic biological processes/Introducon to learning
What is learning?
“An enduring change within an organism brought about by experience that makes a change in behaviour possible”
Enduring – changes are relavely stable
Experience – pracce, previous trials
Learning Performance
Although performance is aected by learning it also depends on;
Opportunity
Movaon
Sensory and motor capabilies
Changes in performance do therefore not always reect a change in learning
Learning is not…
Reexes – these changes in behaviour are not brought about by experience, they are innate
Insncts – these changes in behaviour are also genec, but are more complicated than reexes
Maturaon – these changes in behaviour are brought about by aging
Fague – this change in behaviour is not stable
Reexes
Reexes are automac and usually very fast – learning is not required
Examples;
Airpu Eye blink
Food salivaon
Movement eye turn
Knee tap knee jerk
Touch on baby’s cheek head turn
Pain withdrawal
THE REFLEX ARC
Sensory (aerent) nerves detect smuli
Motor (eerent) nerves smulate muscles
Sensory aerents re at dierent rates depending on the intensity of the smulus
Eg. if you touch something warm – re a lile, touch something hot – re immensely
If this rate exceeds a threshold a response is elicited
Acvang a response pulls the arm away from the smulus
Reducing the input signal reduces smulaon of the eerent’s
insnct
Insncts are a behavioural sequence made up of units which are largely genecally determined and, as such,
are typical of all members of a species
Learning is not required – we are born with them
Note – the dierence between reexes and insnct is in complexity, not type
Eg. Mang rituals
Elicing smuli
Corresponding response
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
Maturaon
Maturaon refers to changes that take place in your body and in your behaviour because you are geng
older
Learning is not required
Fague
Fague is usually transient state of discomfort and loss of eciency as a normal reacon to emoonal strain,
physical exeron, boredom or lack of rest
May lead to physical inability to perform a learned response.. not evidence for lack of learning
Why use animals?
We do not assume animals are like people. We look for similarity between animals and humans in relevant features
for the problem at hand.
Simpler condions
Easily controlled
Less expensive
Wider scope
Types of learning
Habituaon and sensizaon
Classical (Pavlovian) condioning
Instrumental (Operant) condioning
Complex (rule) learning
Social learning
Language mediated learning
Habituaon
Decreased responding produced by repeated smulaon
Eg. rats jump less with each presentaon of loud noise, planes in sydneham
Habituaon is not..
Fague – occurs when muscles become incapacitated so the organism can no longer perform the response
Sensory adaptaon – sense organs become temporarily insensive to smulaon (by bright light or loud
noise)
Sensizaon
Increased responding produced by repeated smulaon
Rats run more in response to the same amount of cocaine when they are pre-exposed to cocaine
Background noise results in more vigorous startle reacons to a tone
Why habituate and sensize?
They help u sort out what smuli to ignore and what to respond to, help us to organise and focus our
behaviour in a world of meaningless smuli
Disorders in habituaon
In people with no mental illness, the neural response in the hippocampus decreases with repeated
presentaons of pictures of emoonal or neural faces
Clinical applicaons;
Diagnosing mental illnesses is oen very dicult – early detecon facilitates treatment and oen leads to
beer outcome
SO, improving diagnoses is very important
Impaired habituaon in people with schizophrenia compared to people with no mental illness – diagnosc
test?
this may enable detecon even before the presence of symptoms
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
2. Classical condioning (Ivan Pavlov 1849-1936)
What is Pavlovian/classical condioning?
A learning process that occurs when two smuli are repeatedly paired: a response which is at rst elicited by
the second smulus is eventually elicited by the rst smulus alone.
Examples – food/pace preference, eye-blink, fear, ancipatory nausea, taste aversions, place avoidance
Pavlov’s dog experiment;
o Animal in restraint, give animal food to make them salivate
o Discovered the dog would start salivang before food was delivered in ancipaon of being food
o Paired the smulus of food with a ringing bell/turning on light
o Eventually just the second smulus was enough to cause salivaon
o Pavlov termed this the ‘psychic reex’ – later referred to as condioned response
o The process has become known as ‘condioning’
What is second order condioning?
A form of learning in which a smulus is rst made meaningful or consequenal for an organism through an
inial step of learning, and then that smulus is used as a basis for learning about some new smulus
Once a condioned smulus has acquired a condioned response, it can also act as if it is a uncondioned
smulus itself

The CR is not always the same as the UR
Eg. condioning of fear and anxiety
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
What is exncon?
Exncon = repeated condioned smulus alone presentaons following acquision resulng in a reducon
of the condioned response
Gradual weakening of the condioned response over me
If you ring the bell and don’t bring food – salivaon will decrease over me – this is called exncon
Acquision and Exncon
How does classical condioning relate to human psychology more generally?
Classical condioning as an experimental model for studying learning processes
Do the same mechanisms extend beyond the transmission of primive reexes
“associave learning” – learning of relaonships between events
What aects classical condioning?
Frequency – more CS-US more learning
Intensity
o More intense CS faster learning
o More intense US greater amount of learning
Conguity (ming) – closer CS and US occur together beer learning
Conngency – higher conngency beer learning
Conguity
Refers to me between onset of condioned smulus and uncondioned smulus
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301

Conngency
Needs two pieces of informaon
What is the probability that the US follows the CS?
What is the probability that the US occurs anyway?
3. Instrumental learning
Edward Lee Thorndike
Interest in animal intelligence – popularised the idea of similarity between humans and other animals
Do animals show insight? – eureka moment, sudden change in behaviour
Dissased with popular descripons and lack of rigour
Thinkbox
Trial and error
Eg. me for cat to escape from the puzzle box
Observed progressive improvement over many trials
No ‘sudden insight’
Law of eect
What a human or animal does is strongly inuences by the immediate consequences of such behaviour in
the past
Thorndike’s (1911) Law of Eect
o Given a parcular situaon, if an acon is met with sasfacon, the organism will be more likely to
make the same acon next me it nds itself in that situaon
Discrete trial procedures
Single trial procedures
Measured objecve dependant on variables such as ‘me’ or ‘errors’
Free operant procedure
Operant condioning = instrumental condioning
Rat placed in skinner box
Free to respond at any me
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
Delivery of food/water conngent upon a response
Looks at connuous changes in behaviour over a long period of me
Radical behaviourism
J.B Watson (1878-1958) and B.F Skinner (1904-1990)
Rejecon of anything unobservable – emoonal thought or cognion
Instrumental learning
Note the important dierence between these procedures and those used by Pavlov
Pavlov – the subject has no control over events, but responds to them
Thorndike/Skinner – the subject has to respond to control the outcome
The behaviour is instrumental in determining what happens
What are reinforcers?
Reinforcers are events that result in an increase in a parcular behaviour
Many reinforcers are intrinsically valued (primary reinforcers) eg. giving dog food
But this is not always the case;;
Secondary reinforcers – acquire their reinforcing properes through experience (eg. clicker with dog)
Social reinforcement – (eg. praise)
Shaping
Principle of successive approximaon of a nal goal
Reinforce behaviours that are closer and closer to target behaviour
Gradually make the condions of reinforcement more stringent, more precise
Can generate enrely novel behaviours
o Bar pressing in rats
o Dog opening door
Response consequence conngencies
Escape learning
A barrier divides the shule box – one half has a grid oor
A warning signal (WS) comes on, followed by a mild foot shock through the grid oor
The subject can escape the shock by leaping over the barrier to the safe area
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
Avoidance
The animal soon learns to jump over the barrier when the warning signal comes on and avoids the shock
altogether
Escape = turning o some currently occurring aversive event
Avoid = prevenng some aversive event from occurring
Negave reinforcement is not punishment
Reinforcement schedules
Instrumental learning
Unlike classical condioning, instrumental learning involved circumstances where behaviour determines the
events that follow
The likelihood that a behaviour will increase or decrease is determined by;
o The nature of the events that follow (appeve/aversive)
o Whether the behaviour produces or terminates these events
Smulus control
Instrumental condioning is another form of associave learning
Smulus Acon Consequence
Shares many properes in common with classical condioning
Instrumental behaviour is also ‘controlled’ by smuli with which acons are associated
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
4. Smulus Control
Thorndike’s Law of eect
According to Thorndike, sasfying outcomes “stamp in” the connecon between the smuli and the
response
The impact sasfying outcome has on behaviour
Smulus control
Instrumental behaviours are ‘controlled’ by smuli with which they are associated
In the future just experiencing/sensing the smulus acvates a response
The “discriminave smulus” – the US that changes to a CS
S-R learning (smulus – response) – SR learning – establishing a mental link between these two things
Skinner’s Triparte Conngency – ABC
Antecedent – the smulus controlling behaviour
Behaviour – what is the response being reinforced?
Consequence – what is the immediate outcome of a behaviour?
Factors important for smulus control
Learning-related
Basic condions that are necessary – principles of associave learning
True of classical and instrumental condioning
Performance-related
Behaviour in novel (but similar) situaons
Factors dictang generalizaon and discriminaon
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
Generalisaon and discriminaon
Every instance in which we encounter a smulus or event is dierent
How is a response previously learned in a given situaon transferred to a new but similar situaon?
Generalisaon = the extent to which behaviour transfers to a new smulus
Discriminaon = the extent to which behaviour DOES NOT transfer to a new smulus
Watson and “lile Albert”
Watson 7 Rayner (1920) set out to test generalizaon of learned fear in an infant, “Albert B”
o US – loud clanging noise
o UR – fear/shock
o CS – white rat
o CR – fear elicited by the rat
Does the CR generalise to other animals and similar objects?
Acquision of a fesh by classical condioning
Rachman (1966): Sexual arousal as a CR
Acquision and exncon
Spontaneous recovery aer 1 week
Generalizaon
The generalisaon gradient
Removal of a sexual fesh by aversive condioning
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
Discriminaon learning
Discriminaon can be learned through training with dierent schedules of reinforcement
Complex discriminaons
Reinforce correct choices made to smuli based on categories; same/dierent, size, shape, number etc
Most animals can learn to discriminate specic smul (exemplars) accurately
Can pigeons and other animals learn ‘high-level’ discriminaons?
Herrnstein et al (1976);
o Showed discriminaon learning occurs with realisc smuli; leaf shapes, water/no water, tree/no
tree, specic person
o Interesngly, this was also the case for real smuli but those that would not normally be
encountered by pigeons eg. sh/no sh
Generalizaon in humans
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
In humans, generalizaon is complicated;
Physical aributes
Semanc similarity
Rules/analogies that link otherwise dissimilar events
Generalizaon on basis of meaning
Razran (1939) study using students;
Words (CS) paired with food (US) salivaon CR
Style, urn, freeze, surf = trained CSs, then tested responses to;
o Fashion, vase, chill, wave (semancally similar
o Sle, earn, frieze, serf (phonologically similar)
Most CRs to items on the rst list
o More generalizaon from the semanc qualies of the words rather than the physical similarity of
smuli
5.
Social learning
Social learning
Instrumental and classical condioning involves learning from direct experience
Humans learn socially by communicang and observing other people
Behaviourists (Skinner/Watson) – all behaviour results from
o Condioning and reinforcement
o Generalizaon of learned responses
“social learning occurs when and organisms responding is inuenced by the observaon of others who are
called models”
acquiring the behaviour of others through observaon
Example – English blue t bird
o Learns to open milk boles and steal cream
o Birds seemed to learn this response by observing others
o Example of observaonal learning? Or complex example of simple learning mechanisms?
o Instrumental condioning – trial and error
R —> pecking the lid
R —> access to milk
Sd —> milkman gone
Social facilitaon vs. social learning?
Goal enhancement
Geng access to some wanted goal might facilitate later trial and error learning – eg. access to cream which
is not usually readily available
Smulus enhancement
Observe others and are oen more likely to approach places that they are – eg. the milk boles
Increased movaon to Act
Try more new things in the company of friends and parents
Contagious behaviour
Mimicking an already established behaviour – eg. yawning
Observaonal condioning
Lab-raised monkeys are not normally afraid of snakes
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
But if a lab-raised monkey sees a wild monkey act afraid of a snake it will acquire a fear of snakes
Two- test acon
Some form of instrumental object that can be manipulated in 2 dierent ways
2 possible outcomes for the animal
Social (instrumental) learning
Mimicry – copying without reference to a goal
Emulaon – understanding here is a goal but not using the same method to gain access to the goal
Imitaon – copying with reference to a goal
Modelling
Children will not only imitate an adult’s specic behaviour but also model general styles of behaviour (eg.
aggressive vs. gentle play)
Suggested cognive aspects of social learning;
o People acvely watch others to gain knowledge about the types of things they do
o Use that knowledge in situaons where its useful
o Informaon is not always used immediately
Bandura (1965)
How does reinforcement inuence modelling?
Three groups;
o Model rewarded
o Model punished
o No consequence
Model observed on TV
Two tests; no incenve and posive incenve
Modelling is reinforcement dependent
Modelling can occur through TV, not just in person
Social cognion theory
1. Aenon to the model
2. Incorporate the models acon into memory
3. Requires having the ability to reproduce the acons of the mdoel
4. The movaon to reproduce the acons of the model
a. Was the model reinforced?
b. Is the reinforcer currently desired?
Applicaons
Adversing campaigns
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
Some real R-R associaons (eg. smoking, drink driving, weight watchers)
Real or contrived R-R associaons
Smacking a child
Who has just bien another child?
6 Biological sources of movaon
Movaon
Why individuals iniate, choose or persist in specic acons in specic circumstances
A necessary condion of behaviour
An energising eect on behaviour
A temporary state that can vary over me dierent from learning
Hebb’s analogy
insnct theories of movaon
From 1st L&M lecture;
Insnct = a behavioural sequence made up of units which are largely genecally determined and as such
typical of all members of a specied
Fixed acon paerns
The same behaviour is displayed by all members of the species in response to the same smulus
A set sequence of behaviours, not a reex
Oen regulated by specic biological state – breeding season, nesng, development
Can be the sign smulus for a reciprocal response in another individual – mang rituals, appeasement signal
Eg. Sckleback mang behaviour is;
o Specic to breeding season (biological state) – hormonal changes
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
o Iniated by sign (or key or releasing) smuli – red belly of other mates, full belly of females,
behavioural triggers
o Filtered by an innate releasing mechanism which acvates acon specic energy or central paern
generator
o Results in a characterisc behaviour called the xed acon paern
Insnct and movaon
Fixed acon paerns (and insncts generally_ are not directly movated by a consideraon of the end goal
Instead they are elicited by a combinaon of environmental and biological circumstances
Studying human insncts
How can we tell if behaviour is insncve?
Biological basis
Cross-species similarity
Cross-cultural similarity
Separated idencal twin studies
Developmental studies
Problems with insncts
Circularity
Proliferaon
What’s the relaonship between learning and movaon?
Drive theories
Drives – exible systems that organise behaviour around a basic need
Specic drive theories
o The drive sensizes the individual to smuli important to sasfy/reduce the drive
o They then movate the individual to behave in a way to sasfy (reduce) the drive
Clark Hull
Organisms suer deprivaons
Deprivaon produces needs
Needs acvate drives
Drives acvate behaviour
Behaviour is determined by learning
Reducon of drive is reinforcing
A “general drive” theory
Advantages of general drive theories
Specic drive theories suer from circularity and the ‘homunculus’ problem
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
General drive theories explain reinforcement as drive reducon
o The organism learns to reduce drive
o Not necessary to infer specic drives for each biological need
Non-homeostac drives
Immediate biological needs homeostac drives eg. hunger or thirst
But not all behaviour is movated to sasfy an immediate biological need
Some (very strong) drives are not relevant to the survival of the individual
Drive for sex?
Most drive theories incorporate sex-drive
Where does sexual drive come from?
o There is no immediate biological need for sex
o Sex as a non-homeostac drive
Sex drive becomes sensible if we think about genec success as a distal movator
Evoluonary tness
The survival of the est
Limited resources / compeon for resources
Individuals who are best suited to their environment will tend to survive
Fitness to reproduce
Darwin recognised ability to produce progeny as crical factor in natural selecon
Physical and behavioural characteriscs passed to ospring (via genes)
Sex drive as an inherited trait
“Psychobiological drives” are determinisc;
o behaviour isn’t pulled towards future possibility
o it is driven by events that occurred in the past
eg. NOT – you have a sex drive in order to pass your genes to the next generaon
BUT RATHER – you have a sex drive because having a sex drive helped your parents to successfully reproduce
and you inherited this trait from them
Problems with Drive Theories
Drive reducon is not necessary for reinforcement (eg. saccharin)
Smulang a drive can be reinforcing
Ignores role for qualitave dierences between reinforcers (eg. liking)
Harlow and maternal deprivaon
Even though biological needs (ie. Hunger) were sased via the wire surrogate, the monkeys preferred the
cloth surrogate
Needs extend beyond the biological
Maternal deprivaon – the pure reducon of a drive is not the only movaon for behaviour
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
7 Long term sources of human movaon
Incenve value
Moment-by-moment incenve value of an outcome is generally determined by an interplay between;
o Liking (hedonic value)
o Biological need
o Current arousal state
o Preferred acvity (not discussed
But what of long term movated behaviour
o Value of rewards decreases with delay
The humanists
Emphasized unique qualies of humans compared with other animals, especially freedom and potenal for
personal growth
Why do people persistently try to test and expand their abilies and capacies?
Why do we long for more?
Carl Rogers
Only one move – innate tendency to try to full one’s potenal for growth
Humans are movated to become fully funconing
o Maintenance
o Enhancement
o Reproducon
Importance of self-concept – how we see ourselves
Needs for posive regard and for posive self-regard
o Uncondional posive regard gives consistency between self-concept and feedback. This leads to
growth.
o Condional posive regard creates anxiety, defensiveness, stasis, distorons and denials of thoughts
Rogers: A fully funconing person
Openness to experience – no need to defend, more emoonal
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
Existenal living – living fully in the present
Trust in one’s own organism – because it feels right
Sense of freedom – to choose what happens
Creavity – and ability to adapt to change
Abraham Maslow
Studied what it meant to be healthy
Proposed that people have a hierarchy of needs
Hierarchy of needs
As the ladder of needs is ascended, the needs become progressively more complex and more human
The needs emerge in order
Frustraon of these needs generates pathology
Self-Actualised people;
– More ecient percepon of reality
– Acceptance of self, others, nature
– Spontaneous
– Problem centering; mission in life, others’ problems
– Detached (need for privacy)
– Independent of culture and environment
– Freshness of apprecia2on; occasional intense experience of basic events of life
– Mysc experience, oceanic feeling; peak experience (transcenders vs nontranscenders)
– Sympathy for humankind
– Close interpersonal relaonships
– Democrac character structure
– Means and ends; interested in goals & process
– Philosophical, unhosle, sense of humour
– Creave
Cricisms of humanisc psychology
– Weak empirically
– The hierarchy itself (Maslow) – can be safe but hungry
– Who is self-actualised (Maslow) – circularity
– Cultural concerns and elism – primarily based on Western ideals, ‘squeaky wheel’ vs ‘protruding nail’
Assessing human goals: qualitave approaches
– Henry Murray (1938) pioneered the use of projecve tests to study people’s needs and goals
– People are asked to describe an ambiguous image, and their descripons analysed for themes
– It is assumed that a person’s preoccupaons, needs, drives and goals will be projected into their
interpretaons
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
Long term human needs
Expectancy-Value Theories
– Why do people with a high nAch only aempt moderately dicult tasks?
– Perhaps we also need to consider a person’s expectancy of actually obtaining the goal;
Expected ulity of an acon
– If the probability of success is high (Ps), then the u2lity of success (Us) is low (easy tasks aren’t sa2sfying):
o Us = 1 – Ps
– But the expected u2lity of an ac2on depends on how likely it is that the ac2on will succeed:
o EU = Us x Ps
o EU = (1 – Ps) x Ps
nAch and Career Choice
According to McClelland, people high in nAch will tend towards tasks that;
– personal responsibility for solving problems
– Sets moderate goals (not too hard or failure too likely, not too easy or success brings no sasfacon)
– Needs concrete rapid performance feedback
The legacy
The humanists
– A reacon to an unappealing predominant paradigm
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301
– Empirically weak but conceptually compelling
Murray, McClelland and Atkinson
– Derived from observaon, theories assessed and modied
– Good at predicng behaviour
Downloaded by Allen L. ([email protected])
lOMoARcPSD|13291301