PSYC2050

Final exam

habituation

getting used to it response

what is Watson's Methodological Behaviourism

• Measure behaviour to infer learning

  • Limited to observable effects

what is Skinner's Radical Behaviourism

• Measure behaviour to infer learning

• Limited to observable effects

  • Complex processes can be broken into small behavioural stimulus and response units
    

what is US: unconditioned stimulus 

A stimulus that naturally and automatically triggers a response without prior learning.

UR: unconditioned response

The natural reaction to an unconditioned stimulus that occurs without prior conditioning.

CS: conditioned stimulus

• Stimulus to learn to respond

CR: conditioned response

• Response to learnt CS

INNATE vs learned (conditioned or unconditioned)

unconditioned vs conditioned

. Phases of a typical conditioning experiment

1. habituation

2. aquisition

3. extinction

what happens during Habituation

CS presented alone

what happens during Acquisition

• Present CS + US

what happens during Extinction

• CS presented alone

 Learning processes during extinction

Spontaneous recovery + Renewal effect + Reinstatement

Spontaneous recovery VS Renewal effect VS Reinstatement

reintroduce CS after break. CR reappears VS when extinction is context specific VS Present US alone after extinction

Factors influencing acquisition curve

1. Intensity of US (more intense, more rapid learning

   1. Order and timing (CS coming before US is better)

describe delay conditoning short

describe delay conditioning long

describe trace conditioning

describe simultaneous conditioning

describe backward conditioning

describe temporal conditioning

Excitatory

CS predicts occurrence of US

exam excitatory 

 if 'A' is bell… A-US, A-US, A-US -> A lead to CR

Inhibitory

• CS predicts absence of US or different US

example of Inhibitory

E.g. if 'B' is light, so no US… A-US, A-US, AB, A-US, AB

• Animal learns that B predicts absence of US, so no CR

How do we know if animal has learned smth about inhibitory stimulus?

summation and retardation tests

Retardation test (what takes place, how to test

Inhibitory conditioning takes place

A-US, A-US, AB, A-US, AB

B becomes inhibitor 'I'

To test: Train 'I' and neutral stimulus 'N' -> excitatory

I-US, I-US, I-US

N-US, N-US, N-US

• Slower learning to inhibitor: I< N

  • You don’t learn as well to it because it was originally an inhibitory stimulus
    

summation test (what takes place, how to test

Inhibitory conditioning takes place

A-US, A-US, AB, A-US, AB

 

To test:

• New excitatory CS alone: N

• New excitatory CS + inhibitor: N + I

• CR time is faster for N compared to N + I

retardation vs summation

• What it is: A test to see if a stimulus takes longer to become excitatory.

• How it works: After a stimulus has been trained as a conditioned inhibitor (meaning it signals the absence of an outcome), it is then paired with an outcome to become a conditioned excitor. If it functions as a true inhibitor, it will take more trials for the organism to learn the excitatory association compared to a novel stimulus.

• Purpose: To determine if the prior inhibitory training has retarded the acquisition of an excitatory response. VS Inhibitory presented WITH new stim

Equipotentiality

any two stimuli can be paired together

Contiguity

 the more two stimuli are paired, the stronger the association

Contingency

 conditioning changes trial to trial in a regular way

Blocking

when a neutral stimulus and excitatory stim r paired with the US

no learning of neutral stimulus with the US

superconditioning

when a neutral stimulus and inhibitory stim r paired with the US

strong association of neutral stimulus with the US

example of blocking

E.g. rats in either 'blocking' and 'control group'

Control: light + noise -> shock , repeat till CR

Just light-> avoid

 

Blocking: noise -> shock, repeat till CR THEN light + noise -> shock

Just light-> no avoidance

 

Blocking effect: history of learnt blocks learning of new stimulus

example of superconditioning

Noise-> no shock , so safe. (inhibitor)

Tone + light -> shock

Rats showed stronger conditioning to light then control

CS pre-exposure (latent inhibition) ( retardation, summation?)

Longer you are presented with CS before learning protocol -> less likely to learn

• Context specific

• Passes retardation test

• Does not pass summation test

testing Generalisation

Test:

• CS₁ (e.g. tone)- US

• Test different groups with CS₁(e.g. tone w diff frequency) CS₂(e.g. tone w diff frequency) _, CS₃(e.g. tone w diff frequency) _…

• -> generalisation

 

E.G. Rabbit eye blink response

• US: mild electric shock

• UR: eye blink

• CS+: 1200 Hz tone

-> Tone-shock; Tone-shock; Tone-shock

-> response

-> playing other frequency (e.g. 400 Hz, 800Hz) -> learning occurred, but learning is less compared to 1200Hz

 

*other similar stimuli may produce CR

More similar to CS, more likely to elicit CR

* as learning continues, organism learns the best CS associated with US (discriminate)

Garcia effect (preparedness)

Some association are learnt faster

US sickness (taste- sickness > noise + light- sickness)

US shock (taste- shock> noise + light- shock)

Because noise + light are commonly associated with lightning, which is like common sense

Rescorla-Wagner model

• Explains how organism learns prediction of US

• Expected VS actual strength of US

  • Big difference-> surprise

• Expectation is based on prior experience with US

• Strength of US is fixed

Describe a new fear conditioning experiment and use it to illustrate the renewal effect

CS: whistle

US: angry dog

UR: flinching

CR: flinching

 

Renewal effect: get extinction in one condition, one context, but in another context it comes back

 

E.g. administer shock in context A till extinction-> context B, CR comes back

Classical conditioning vs Operant conditioning

Classical conditioning	Operant conditioning
Ivan Pavlov	B.F. Skinner
Learning via association	Learning via reinforcement
Relies on reflective associations between stimuli	Relies on consequences of past actions influencing behaviour
Involuntary responses	Voluntary behaviours

Chaining (teaching new behaviour)

Best to shape behaviour, best to start with the last behaviour on the chain (learning backwards)

Positive reinforcement

adds smth to increase behaviour (e.g. star for good behaviour)

Negative reinforcement

taking smth away to increase behaviour (remove discomfort)

 

Positive punishment:

adds smth to decrease behaviour (e.g. getting told off)

Negative punishment:

takes smth away to decrease behaviour (losing licence) 

Strongest rate of learning: VI, FR, VR, FI

Strongest rate of learning: VR, FR, VI, FI

Schedules of reinforcements

• Continuous (CRF)

  • More effective

• Partial, Intermittent (ratio- instances of behaviour & interval- time of behaviour; fixed & variable)

  • Fixed ratio: every nth action is associated with an action (e.g. every 10th visit get free coffee)

  • Variable ratio: on average, every nth action  (e.g. gambling 'wins' after placing bets)

    • Most resistant to extinction

  • Fixed interval: first behaviour after n seconds (e.g. pocket money every week for cleaning room)

    • Variable interval: on average, first behaviour after n seconds (e.g. checking messages)

Partial, Intermittent (Schedules of reinforcements)

ratio- instances of behaviour & interval- time of behaviour; fixed & variable

Fixed ratio

every nth action is associated with an action (e.g. every 10th visit get free coffee)

Variable ratio

• on average, every nth action  (e.g. gambling 'wins' after placing bets)

  • Most resistant to extinction

Fixed interval

first behaviour after n seconds (e.g. pocket money every week for cleaning room)

Variable interval

on average, first behaviour after n seconds (e.g. checking messages)

Successful punishment:

No escape or avoidance As intense as possible (within limits) Continuous schedule No delay Short period of time No subsequent reinforcement Reinforce incompatible, appropriate behaviour concurrently Watch for side effects

• understand the various reward variables that affect reinforcement

Drive: Reinforcement depends how much the subject wants it (motivation)

 

Magnitude/size: size matters (e.g. amount of food given)

 

Delay: delayed reinforcement reduces learning

 

Reinforces work better when:

– Drive/desire is higher

 – Reinforcer is larger (but this tapers off)

– Reinforcer is given right away

Three term contingency

1. The discriminative stimulus - Sets the occasion serves to signal the occasion when a particular behaviour will be reinforced/punished 2. The operant response - The behaviour 3. The outcome (reinforcer/punisher) that follows - The consequence

• understand the mechanism of stimulus control

occurs when your behaviour comes to be under the control of the stimulus Behaviour happens when the stimulus is present and doesn’t happen when the stimulus is absent

Shaping (successive approximations)

Gradually develop new behaviour Reinforce behaviour that leads to desired behaviour

Operant conditioning:

• You don't need to experience responses or consequences for learning to occur

Behaviour regulation theory (premack

what if stimuli aren't reinforcers but behaviours are

E.g. water X -> drinking water

TV X -> watching TV

Escape learning

emit a response that terminates an aversive consequence (negative reinforcement)

Avoidance learning:

after escape learning has been learnt enough, emit a response that prevents an aversive consequence altogether

Avoidance learning- anxiety example

E.g. fear of snakes

• Avoid going on bushwalks where snake could appear

  • Avoid the aversive feeling of snakes being around

Flooding method

Flooding (e.g. exposure to snakes with no negative consequence)

• escape can't take place

Learned helplessness

Effect of unavoidable aversive experience

Functional analysis

What is the problem of behaviour? When does it occur? Identify reinforcers Stimuli, behaviour, consequence

Positive parenting program

Using operant conditioning

Using chaining and reinforcement

 

E.g. sending kids to school

1. Teach kids behaviour to perform

2. Set a routine

3. Reward for desired behaviour

4. Avoid nagging or hassling

   1. Turn it into a game

Selective attention

Ignoring irrelevant stimulus

MORE difficult the more similar the two things are

Endogenous/goal-directed (top-down) control

E.g. tuning out of dull convo and tuning into another

Exogenous/stimulus-driven (bottom-up) control

E.g. attention captured by shattered glass

Endogenous VS Exogenous

Endogenous/goal-directed (top-down) control VS Exogenous/stimulus-driven (bottom-up) control

Inattentional blindness

When you focus your attention, you miss other changes in plain sight

Change blindness

Changes are missed because they occur alongside brief visual disruption (e.g. blinks, windshield wipers)

Balint's syndrome simultanagnosia

Bilateral occipital/parietal lobe damage prevents patients from perceiving more than 1 stimulus at once, but you can link them together.

Describe evidence that attention spreads across objects

Balint's syndrome simultanagnosia Red dots- participants identify red Green dots- participants identifies green Red and green dots- participant only identifies red Linking red n green dots- participant identifies red n green

Overt

Divert eyes to location of attention

Covert

Eye focuses on else where but attention is else where

Late selection

Distractor influences target processing

• Something I didn't pay attention to but still extract meaning from it

Early selection

Distractor doesn't influence target processing

• I know its colour, its blah

  • You target with something in mind before already

Perceptual load

Low load- numbers are higher, slower

High load- numbers are lower

 

Describe attention and the early vs. late selection issue

What impacts early or late selection is  due to perceptual load.

 

Under what condition do you get late selection under the load theory

Conditions of Low perceptual load

High working memory load

Late selection Much easier to be distracted

Low perceptual load =
VS
High working memory load =

Low perceptual load = late selection

 

VS

 

High working memory load = late selection

Low working memory load task

1. Number (e.g. 7

2. Reaction time task

   1. Was the number a 7

High working memory load task

1. Number (E.g. 197365

2. Reaction time task

   1. Was 7 a number

Broadbent's filter theory

• Structural

  • Predicts early selection

Attentional limitations

Kahneman

Attention -> resources

Number of current task that can be performed depends on difficulty (resource demands)

Available resources increases under arousal/motivation

 

Automaticity -> training improves attention

Feature vs conjunction search

Feature  (disjunctive)	Conjunction
Parallel	serial
Pop-out (doesn't matter how many distractors)	Increase distractors- longer search time
E.g. red X amongst black X	E.g. red X amongst black X and red O
Unaffected by search set size	Affected by search set size
Pre-attentive	Attentive
Efficient	Inefficient

 

Target absent, conjunction search - takes longest time

Feature  (disjunctive)

Feature  (disjunctive)

Parallel

Pop-out (doesn't matter how many distractors)

E.g. red X amongst black X

Unaffected by search set size

Pre-attentive

Efficient

Conjunction search

Conjunction

serial

Increase distractors- longer search time

E.g. red X amongst black X and red O

Affected by search set size

Attentive

Inefficient

Feature integration theory (FIT)

E.g.  red X amongst black X and red O E.g. feature is colours (red or black) or (O) Automatic processing of all features on stimulus Individual feature 'maps' (based of features you know where every stimulus is Attention required to bind features together   OVERALL: when required to identify two + features (e.g. red and X), you integrate features together from your feature map with attention

Rapid serial visual presentation (RSVP)

Only one location but time-pressured, stimulus after one another almost immediately   Limitations: Bind information to wrong thing (e.g.XY-> report as blue X

Two target RSVP- attentional blink (AB)

Task: report the 2 letters amongst numbers

 

What varies is the number of distractors (e.g. 3 distractors in between -> lag 3)

People are good at lag 1

 

Easier identified T1-> reduce AB as less time to process

 

AB: rejecting distractions (if you remove distractions, there is no AB)

Task switching

E.g. T1: letter? T2: colour? Task set: rules to do a task Cost: disabling one task set/disengaging from prior task for another   Moving assembly line: focus on one task/pp The more similar two tasks are, the more difficult to switch (vise versa Practise reduce cost    Switching to EASIER task is MORE DIFFICULT -> hard to disengage from hard task

Theories of switch cost

Disengagement theory (impairs adoption of alternative task set Endogenous + exogenous factors Endogenous (preparation, requires time, switch cost reduced) Exogenous (cost cannot be removed, triggered by suitable stimulus) Endogenous only (residuals cost due to inadequate preparation   The theories differ in: Role of active preparation (endogenous Role of interference effects from prior tasks and task sets Whether exogenous factors play a role

Automaticity

The more you learn a task, consumes fewer resources

Baddeley's Working memory (short term memory) model

CENTRAL EXECUTIVE:

CENTRAL EXECUTIVE:

• Attentional controller

• Links working memory and long term memory

• Switching attention

• Mental manipulation of attention

• No storage capacity