week 11 - intro to learning

motivation and emotion

motivation

translates knowledge into action

3 components of emotion

1. affect: the subjective feeling

   • fear, pain, hunger, etc.

   • NOT directly observable

2. physiological response

   • autonomic and hormonal changes

   • heart rate, stress hormones, salivation, etc.

3. behavioural responses

   • observable actions

   • fleeing, freezing, avoiding, etc.

the automaton problem

• animals might behave as if they feel

• but could theoretically function without subjective experience

• like very sophisticated machines

• we must be careful not to equate behaviour with affect

Behaviour ≠ proof of feeling

judgment bias

a method to infer an animals emotional state by examining how it responds to ambiguous stimuli

drive theory

drive = internal state arising from a physiological need

• the drive energizes behaviour toward need reduction

hull’s formula

behaviour = learning x drive

• both learning and drive are necessary for behaviour to occur

incentive learning

occurs when an animal learns that a particular action or reinforcer actually reduces a drive

• its not enough to simply perform an action; the animal must experience the reinforcer while in the relevant motivational state to understand its value

textbook rats example:

• rats that were given food while hungry were more likely to press the lever for food then those who were only given food when already satiated

• had to associate the food as being the solution to hunger

negative contrast

if reward decreases in size or value, the animal is less likely to work for it

• ex: previously received 5 pellets → now only 2 → performance drops

positive contrast

if reward increases, the animal is more likely to work for it

• ex: previously received 2 pellets → now 5 → performance increases

combined model of behaviour

behaviour = drive + learning + incentive

• captures both immediate needs and historical motivation

rG-sG mechanism

RG = response to goal

rG = residual response to goal

sG = stimulus generated from rG

example:

• jimmy got a new job he likes (the liking of the job = RG)

• next time hes on his way to work he gets that good feeling of liking again, though less strong than the first time (smaller feeling of liking = rG)

• that feeling of liking his job makes him rush to get to work faster (that feeling turns into a stimulus = sG)

partial reinforcement extinction effect (PREE)

behaviour that was partially reinforced persists longer during extinction than behaviour that was always reinforced

• uncertainty pushed you to keep trying

opponent process theory

every strong emotion automatically triggers an opposing emotion

• initial emotion = a-process

• opposing emotion = b-process

ex:

• people dont feel pleasure during the nightmare, they feel it when it ends

• fear (a-process) → relief/pleasure (b-process)

  • the end of something bad produces a good feeling

**after repeated exposure, opponent emotion begins sooner and gets stronger, altering motivation and behahviour