Animal learning and cognition Flashcards
Darwin
Provided a mechanism of evolutionary continuity between animals and human
Romanes
Pioneered the study of comparative psych, but relied on anecdotes to find egs of intelligence in animals. Intelligence = linear evolution
Lloyd Morgan’s canon 1894
Dismissed Romanes - argued against anthropomorphism.
In no case may we interpret an cation as the outcome of the exercise of a higher physical faculty, if it can be interpreted as the outcome of one which stands lower in psyc state.
Thorndike and his law of effect
First to use experimentation to study animal intelligence - trial by trial learning curves vs learning
Studied trial and error learning in cats using puzzle boxes
If a response leads to a satisfying outcome, it will be strengthened
Watson and little Albert
Emphasised the importance of learning in animal and human behaviour. Advocated behaviourism - studying only observable phenomena. Experiments with little Albert - induced fear when presented with rats due to low voice - fear transferred into rabbit
Skinner and Operant or Instrumental learning
Developed techniques - response shaping, schedules of reinforcement
Developed technology - conditioning chambers
Operant or instrumental learning = studying learning in animals that have to make a response to gain a reward
Pavlov and Pavlovian or Classical conditioning
Studied learning in which a neutral stimulus was paired with a reward.
Also known are type I or respondent conditioning.
Conditioned stimulus
e.g. tone
Unconditioned stimulus
e.g. food
Unconditioned response
e.g. eating
Conditioned response
e.g. salvination
Acquisition
CS and UCS pairing
Extinction
CS aline
Spontaneous recovery
CS alone after 24 hour rest
Generalisation of a stimulus
Gradient and decrement
e.g. lion - all big cats give same response
Banks and Flora 1977 artificial ranks based on
Appearance, Aristotle’s great chain of being, evolution, brain size
Cephalisation index
Ratio of body to brain weight
Anaxagoras and MacPhil
All animals have the same intelligence but different levels of ‘nouse’ bring out intelligence, except language in humans
Learning
A relatively permanent change in behaviour that results from experience
Ranking by speed of learning problems and Shard and Angermeier, Bitterman 1975
Unexpected inter-species difference e.g. rat/human mazes
Difficult to equate perceptual demands of the test
Difficult to equate motivational demands of test e.g. vary rewards and motivation to equal value to an animal
Intra-species differences in speed of learning
Biological relevance - Garcia and Koelling 1966 - Garcia effect
Some stimuli more readily associated than others, depending on CS
Animal memory
When current behaviour is under the influence of past experience
Capacity - Vander Wall 1982
Food storing birds: Clark’s nutcracker stores pinenuts in autumn for winter - can recover 90%, even in snow
Capacity - Vaughan and Greene 1984
Photographs with pigeons - peck at particular slides to get food. All very similar and can remember 160. Everyday scenes, different angles, remember which ones associated with food.
Timing - periodic and internal (Church and Deluty/Gibbon)
Periodic timing - circadian clock - active in day, not night, even when constant light/heat
Interval timing - hear a tone for 4 or 16 s - different levers to press; switch light out at average 4 s - only at 4s reward for pressing lever
Number: Meck and Church 1983, rats
Can distinguish 4 tones (left lever) and 16 tones (right lever) even when same length of time
Number: Brannon and Terrace 2000, monkeys
Gave novel shaped and sized stimuli and press in order of number of dots etc.
Number: Regani, Regotin and Vallortigara 2007, 5-day old chicks
Trained to get food in different wells along a row of food wells, even when trays rotated or begin from new location
Categories: Herrnstein, Loveland and Cable 1976
Pigeons 160 slides with/out trees - reward if pecked at tree
Categories: explanation
Innate categories (certain info) Exemplar learning (learn every example) Feature learning (common component) => exemplar and generalisation
Habituation
A reduction in responsiveness to a stimulus as a result of its repeated presentation
STM: Whitlow 1975
Plethysmograph (change in blood flow)
Tone I, 60s, tone I -> less surprised 2nd time, memory?
STM: Wagner 1976, effector and receptor fatigue
Tone I, 60s, Tone I
Tone I, 150s, Tone I - stronger response - forgot -> memory in active state
Effector fatigue: tone I, 60s, tone II, effector system not fatigued
Receptor fatigue (dishabituation): tone I, 30s, D, 20s, toneI, receptors same as tone 2 - retroactive interference
STM: Radial arm maze: Olton 1978, Beatty and Shavalia
Rats at least 8 times, up to 4hrs - landmarks
Different maze tasks and still remember 1st
STM
Simple procedures involve memory, animals have STM, habituation is found in many species, is a decaying memory trace and varies with tasks
LTM
Consolidation -> retention -> retrieval
Forgetting LTM for various reasons recollection from LTM not conscious
LTM: Miller and Berk
B/W associated with shock African Claw-toed frog remembers from tadpole
LTM consolidation theories: Hebb 1949
Permanent connection of neurons, so consolidation important
LTM consolidation theories: Duncan 1949
Failure to retrieve = no memory - electro-convulsive shock at stage -> unable to perform memory
LTM Retrieval theories: Deweer, Sara and Hars 1980
Rats trained to run through maze, over 300s down to 30s. After 25 days, half allowed 90s reminded (cage next to maze), which meant 40s completion, compared to 150s
Holland and Straub 1979
Noise, food; noise, illness -> noise = food devalued and less response as evokes memory of food and illness
Specific qualities of US
Flavour, intensity
Affective qualities of US
Nice or not - appetitive or aversive
Stimulus substitution - autoshaping
Pigeons, light (CS) 5s, followed by food 5s (US). Begins to respond to CS as if responding to US - 2 responses in conflict. Food vs water pecks - has learned specific qualities of US, changes nature of CR
Conditioned suppression
Decrease in movement -> CS = 60s light ,0.5s footshock - aversive. Food delivered after pressing, light signals to stop lever pressing - prepared for footshock
Compensatory response - Siegel 2005
Reviews role of conditioning in drug tolerance
Injection (CS) -> CR opposes UR -> morphine (US) -> Analgesia (UR) - body prepares for response so compensates e.g. warming body before intake of ethanol in a particular place.
Also, heroin administration in same room = less of an effect
Morphine -> saline -> morphine - less tolerant to drug
morphine -> rest -> morphine
Drug tolerance - Siegel 1977
A decrease in sensitivity to a drug as a result of repeated exposure to a drug - results from compensatory mechanisms: biological, behavioural
Timberlake and Grant 1975
Effect of CS on CR by presenting signal - anaesthetised rate as a substitute for find but instead of eating, engages in social activity. Rat no paired with food delivery = no social interaction
Principle of continuity
Events that occur close together in time or space will become readily associated
Conditioning without continuity - taste aversion conditioning: Smith and Roll 1967
Saccharine leads to illness but could delay gap up to 6 hours and animal still rejects Saccharin taste - can even give other flapper in gap and still avoided
Garcia and Koelling 1966
Biological relevance - some info more easily paired with other
No conditioning, even with contiguity: Rescorla’s truly random control 1967
Contingency is as important as contiguity in learning - regular predictive relationship between contagious events
Contingency
A regular, predictive relationship between contiguous events (CS-US)
Surprise: Blocking experiment 1969 Kamin
Must be ‘mental’ work for animal learning i.e. a surprise, not learnt as much about light in lower shock.
C: Light & noise -> shock means CR when light
E: Noise -> shock; light & noise -> shock means cr when light
S: Noise -> shock; light and noise -> SHOCK means CR when light
Attention: Latent inhibition of light
Light and nothing, followed by light and food, compared to light and food alone
Attention theories: Wagner
Attention is high to novel stimuli and low to familiar stimuli
Attention theories: Mackintosh
Attention high to novel stimuli and important events, low to irrelevant stimuli
Attention theories: Pearce-Hall
Attention high whilst learning about a stimulus and low once learning complete - automatic vs controlled processing - learn that light signals nothing, so attention low later
Attention: Kaye and Pearce 1984
Light followed by nothing, food or food/nothing depending on group. Attention high to low quick for both, nothing and food, but high fluctuations for either (familiar stimuli, but not predictable)
Trial and error learning vs reasoning: Thorndike 1989
Animals solve problems through trial and error, not reading - gradually reducing learning curve, not step like
Deductive reasoning
Conclusion necessarily follows the premises
Inductive reasoning
Conclusion is likely to follow from the premises
Navigation: Wehner and Srinivasan 1981 - Dead reckoning
Desert ants use position of sun in sky to create a vector, pedometer to calculate distance travelled
Navigation: Cartwright and Collett 1982 - Piloting
Gerbils and landmark use - piloting with a single landmark. Found during food, searched in same place. May use relationship between landmarks - map like rep of environment?
Cognitive maps: Tolman
Against trial and error. Block off normal route and went down arm most direct to food. Result of Pavlovian conditioning - light associated wit food?
Cognitive maps: Morris
Water maze - novel route -> original location released but platform moved. Released from new location = able to find platform
Cognitive maps
Allow animals to take a detour and navigate around a problem due to landmarks
Cognitive maps: McGregor
Only use shape of pool - wall length ID and apex location used to orient, metric info about shape but not necessarily used to form complete representation of shape
Insight
Animals capable of working out mechanism of something to solve a problem
Kohler (Gestalt’s psychology) - Chimpanzee Sultan
Novel behaviour. Hung food out of reach and observed solutions to problem -> stacked crates for bananas, not possible for trial and error = multi step process, but had had significant exp with boxes
Epstein et al - pigeons
Prior experience influences behaviour. Response is shapes - peck plastic banana, push box to stand on and peck blue light. Push box under banana = inductive reasoning, behaviours added together = deductive reasoning?
Weir et al - New Caledonian crows
Tool use to extract food from bark
Analogical reasoning: Gillan and Premack - chimpanzee Sarah
Language trained chimpanzee able to solve shape/size relationship and object-use tests - increase intelligence?
When does communication occur?
When one animal transmits a signal to another that is capable of responding appropriately to that signal
Honey bees communication - Von Frisch
Round dance 50-100m, waggle dance >100m from hive. Longer waggle = further. Direction wrt vertical of honeycomb = direction of food wrt sun. Also loner = more food
Veret monkeys communication - Seyfarth and Cheney 1993
Snake, leopard, eagle call all have different sonograms and different behaviours which are learned, not innate
Language: Chomsky and Macphail
Language is unique to humans - all animals have the same intelligence, only differ in language ability
Hockett’s criteria for language
Discrete units
Arbitrary units - no direct relationship to what they signal
Semanticity - each has a specific meaning
Displacement - communicate about things not directly present
Syntax - new position = new meaning
Attempts to train apes to use language: Washoe - Gardner and Gardner 1969
American sign lang and syntax. Baby learned from adult, not trainers
Attempts to train apes to use language: Sarah - Premack 1971
Used cards with symbols - 130 words with syntax
Attempts to train apes to use language: Nim, Terrace et al 1979
19000 utterances and 5235 types - but sentence length did not increase over time - misinterpretation of syntax by researchers - trial and error?
