Learning about time & a theory of learning

Question 1

associative learning & causality

Answer 1

• All species live in a world in which they need to predict & control the world around them to survive
• Associative learning found across animal kingdom
• All vertebrates & invertebrates, & even in monocellular organisms such as planaria

Question 2

mahoney & ayres (1976)

Answer 2

• rats, tone paired with a 4/8s shock
• forward (CS before shock), backward (after shock) or simultaneous conditioning
• back = learn much if tone after shock
• simultaneous = learned more
• forward = learned most when tone before shock

high latency –> high fear

Question 3

is correlation important for associative learning?

rescorla (1968)

Answer 3

• 3 groups of rats given 5 tones & some shocks
  1. +ve contingency: 2 tone –> shock pairings, no more shocks
  2. zero contingency: 2 tone –> shock pairings, plus extra shocks
  3. -ve contingency: shock never paired with tone
• +ve tone positively corr with shock & vice versa for -ve
• 0 tone uncorr with shock

Question 4

blocking

kamin (1969)

Answer 4

• group 1 light conditioned with a pretrained noise (shock not surprising, predicted by pretrained noise)
• group 2 light conditioned with a novel noise (shock not predicted, surprising)
• less learning in group 1 - pretrained noise blocked learning about the light

Question 5

rescorla & wagner (1972) theory

Answer 5

• describes how much association strength increase on each trial
• equation
• each US can only support so much learning
• learning stops when US no longer surprising

Question 6

acquisition: light –> food

rescorla & wagner

Answer 6

• gradual acquisition of associative strength V
• Learning reaches asymptote determined by lambda
• On each trial sum of V increases & surprise decreases
• Learning stops when lambda = sum V
• The max amount of learning to light will increase with lambda
• As size of food US increases, lambda gets bigger & max amount of learning - asymptote - increases

Question 7

Mackintosh (1976): Overshadowing

Answer 7

• Light (CS1) and noise (CS2) were paired with shock (US)
• Group 1: light presented alone before light-noise pairings
• Group 2: light-noise pairings only
• Group 1 learned to anticipate the shock from the light more than Group 2 (Noise overshadowed the light for Group 2)

Question 8

overshadowing & blocking

competition for strength

Answer 8

• control: light gets all associative strength
• overshadowing: strength divided between light & noise
• blocking: noise all strength in stage 1 so none left for light

response of light: highest to lowest = c –> o –> b

Question 9

extinction

inhibition

Answer 9

• light –> food then light –> nothing
• changes in associative strength -ve
• initiaally light still predicts the food (expected but doesnt happen)
• omission of food is surprising

Question 10

conditioned inhibition

Answer 10

• light –> food
• light + noise –> nothing
• on l–>f trials associative strength to light goes up, learning stops
• on l+n trials light predicts food but nothing happens
• both l & n lose strength, associative strength goes down
• l started +ve, ends lower
• n started neutral, ends -ve

Question 11

why does inhibitor mean US prediction take longer?

Answer 11

it starts -ve

Question 12

why dont warning signals extinguish

Answer 12

even though the warning signal is presented without shock, it is accompanied by the response which is inhibitory

Question 13

periodic timing

learning

Answer 13

learning to respond at a particular time of day

e.g. circadian

Question 14

interval timing

learning

Answer 14

learning to respond after a particular interval of time

Question 15

cockroaches - time

roberts (1965)

Answer 15

• increased activity at dusk
• remove visual cues: cycle drifted until increased activity started 15h before dusk
• restore visualcues: gradual shift back to correct time
• entrainment - light acts as a zeitgeber synchronising the internal clock

Question 16

internal clock environmental or innate?

bolles & stokes (1965)

Answer 16

• subjects born & reared under 19, 24 or 29h cycles
• fed at regular point in own cycle (signalling few hrs after change in lighting)
• animals on 24h cycle leaned to anticipate food
• others didnt

Question 17

suprachiasmatic nucleus (of hypothalamus) & 24h clock

Answer 17

• Metabolic rate appears to vary as a function of the day-night cycle
• Lesions of the SCN abolish the circadian regularity of foraging & sleeping in the rat
• Receives direct & indirect impulses from the visual system, which could keep circadian rhythms entrained with the real day-night cycle

Question 18

physical illness & disruptions to circadian rhythms

Answer 18

• heart disease
• diabetes
• infections
• cancer

Question 19

mental illness & disruptions to circadian

Answer 19

• depression
• SZ
• bipolar illness

Question 20

weber’s law

Answer 20

• JND when you change a stim is proportional to initial intensity/magnitude of changed stim
• –> in absolute terms small amount is judged more accurately than large amounts
• critical point is that % change is more important than absolute change

Question 21

scalar timing theory components

gibbon et al. (1984)

Answer 21

• pacemaker
• working mem
• reference mem
• comparator

Question 22

pacemaker

scalar timing theory

Answer 22

• emits pulses at a roughly constant rate t (random variation)
• when a stim is presented, a switch is operated & pulses allowed to accumulate in WM

Question 23

working mem

scalar timing theory

Answer 23

• when reinforcement occurs, pulses stop accumulating
• another switch allows num of pulses in working mem (N x t) to be stored in refernce mem
• this storage not completely accurate - some mem distortion

Question 24

reference mem

scalar timing theory

Answer 24

• K x N x t
• K represents mem distortion (K less 1 smaller num pulses stored & vice versa for more 1)
• error proportional to duration scalar

Question 25

comparator

scalar timing theory

Answer 25

• on each trial animal compares num of pulses in working mem with a random value drawn from those stored in reference mem (NMx)
• done by comparator
• if values close, animal responds

Question 26

another stim occuring

scalar timing theory

Answer 26

• successive num of pulses stored in WM
• uses 1 of values in reference mem to decide when to respond
• comparator works out how close values are (ratio rule)
• large = dont respond
• small = response

Question 27

problems with scalar timing theory

Answer 27

• no physiological ev for a pacemaker
• conditioning & timing supposedly occur at same time but controlled by diff mechanisms

Question 28

alternatives to pacemaker

problems with scalar timing theory

Answer 28

• timing achieved by a series of oscillators with 2 states (on/off), pattern of activation used to determine exact time
• behavoiural theory of timing

Question 29

behavioural theory of timing

alternatives to pacemaker

Answer 29

• When the animal gets a reward, this simulates beh
• Animal moves across an invariant series of behavioural classes in between reinforcements
• Pulse from an internal pacemaker will change the beh from 1 class to another
• The beh that is occurring when the next reinforcer occurs becomes a signal for that reinforcer

Question 30

conditioning & timing occuring at same time but with diff mechanisms

problems with scalar timing theory

Answer 30

• G&B - calculate rate of reinforcement during stim & background. if CS higher than background –> conditioning
• S&B - sim to regular conditioning but stim is assumed to change over course of its presentation, allowing animal to learn about when a reinforcer occurs