PSYCH 1X03 - Instrumental Conditioning

Instrumental (Operant) Conditioning

Behaviour is followed by consequence, which changes how likely you do that behaviour again

1. Reward training (Positive Reinforcement)

Adding good to increase behaviour

2. Omission training (Negative Punishment)

Removing good to decrease behaviour 

(ex. taking phone away) 

3. Escape training (Negative Reinforcement)

Removing bad to increase behaviour

4. Punishment training (Positive Punishment) 

Adding bad; decrease behaviour

Positive = add, Negative = remove

Types of Reinforcement Schedules

1. Fixed Ratio: Reward after a set number of responses (ex. FR-5 - every 5 times) 

2. Variable Ratio: Reward after an unpredictable number of responses (ex. gambling) 

3. Fixed Interval: Reward after a fixed amount of time (ex. every 15 minutes

4. Variable Interval: Reward after an unpredictable time intervals (ex. pop quizzes) 

Ratio = number of actions 

Interval = time 

Fixed = predictable 

Variable = unpredictable

Graphs

1. Fixed ratio - Staircase (animal responds and stops to get reward after interval) 

2. Variable ratio - Smooth continous line (animal keeps responding) 

3. Fixed interval - Scallop (more motivated close ot time of payment) 

4. Variable interval - Dotted continous line (gaps occur just after reward) 

Autoshaping

Individuals learn without guidance

Ex. give reward to roommate after they clean their room - not really teaching them HOW to clean room

Shaping by Successive Approximation

Teaching by breaking task into smaller parts (step by step) 

Ex. Teach a cartwheel step by step

Chaining

Learning by sequence (each action triggers the next)

Ex. Teaching a dog to fetch a ball 

1. Go to ball 

2. Pick up ball 

3. Bring it back

4. Drop it - get reward

• Shaping is one behaviour divided into small parts 

• Chaining is multiple behaviours that make up one task

SD (S+)

Signals a behaviour will be reinforced (get reward) 

Ex. A green light means pressing the button will give reward

S Delta (S-)

Signals a behaviour will not be reinforced

Ex. A red light means pressing the button will NOT give reward

Thorndike’s Law of Effect

If a behaviour leads to reward, it is more likely to happen (Positive reinforcement)

Stimulus Generalization

Responding to not only one cue but also similar cues

Contious Reinforcement (CRF) 

Every correct response is followed by a reward

Contrast Effects

Behaviour changes when value of reward changes

1. Negative Contrast:

-Reward drops from high to low; behaviour happens less often

2. Positive Contrast:

-Reward goes from low to high; behaviour happens more often

Mirror Neuron

Brain cell that fires both when you perform an action and see someone else perform that action (learning by observation_ 

Operant Chamber (Skinner box) 

A controlled box where animal performs certain behaviour to recieve a reward

Continuous Reinforcement Schedule (CRF)

Person gets reward after doing a correct response

Overjustification Effect

Giving a reward for something already enjoyable can change the individuals innate motivation, making it feel like work.

Partial Reinforcement (PRF)

Desired behaviour is only reinforced some of the time, not all the time. 

Primary Reinforcer

Reward that satisfies a biological need. 

Ex. Food, water, shelter

Secondary Reinforcer

Something that becomes rewarding through association with a primary reinforcer.

Ex. money - buys food (primary reinforcer) -  money becomes rewarding

Ratio Strain

When behaviour requires too much responses, organism slows down and eventually stops