Animal learning and cognition Flashcards

Question

What is variable ratio reinforcement?

Answer 1

On average every 30 presses but could vary Gambling - don't know when you'll get rewarded so quickly keep pressing

Answer 2

CS = conditioned stimulus (light) US = unconditioned stimulus (food) UR = unconditioned response (salivation) CR = conditioned response (salivation) Didn't actually ever really use a bell - used more controlled stimuli like metronomes Skinner = doing Pavlov = predicting

Answer 3

Even when hammer didn't hit knee, reflex still happened because person associated noise with hammer

Answer 4

Conditioned stimulus alone - response is extinguished Extinction not due to forgetting, due to new association with the conditioned stimulus and no response

Answer 5

If you wait 24 hours after bell not being associated with food there is a bounce back - a spontaneous recovery of conditioned responding

Answer 6

The more similar a stimulus is to trained stimulus, the bigger the response

Answer 7

UCS = pornography UCR = arousal CS = leather boot CR = arousal Measured with thismograph

Answer 8

Pavlov also showed that the conditioned response was sensitive to theproperties of the conditioned stimulus. For example, if a tone of 1200 Hz was used as the conditioned stimulus then conditioned responding was maximal when a 1200 Hz tone was presented. However, there would also be a high CR when 1000 and 1400 Hz tones were presented, although responding would be lower than the maximum produced with a 1200Hz CS. There was lower responding as the frequency of the tone deviated more from the CS. Plotted on a graph this is known as a Generalisation gradient and is the result of Stimulus generalisation. When conditioned responding is lower due to a slight difference between the trained CR and the presented CR (such as a change from 1200 to 1000Hz) this is known as Generalisation decrement.

Answer 9

- Increasing complexity of animals led them to be more intelligent - In terms of whether animals had blood, how many legs they had etc

Answer 10

- Simpler forms of life gave way to more complex forms - these were more intelligent - As if intelligence has a purpose to improve on previous forms - Continuity between species - linear form with humans the most intelligent and invertebrates the least However this is a gross oversimplification - animals don't involve in a sequence and evolution does not have a purpose Hinde and Stevenson-Hinde (1973) - because of the time scale of evolution, there must be multiple chances for intelligence to evolve Clayton and Emery (2005) - ‘feathered apes’ crow problem-solving - species of crow are as intelligent as some apes and young children

Answer 11

Jerison (1973). The principle of proper mass - the more important a function is, the more of the brain is devoted to it Animals with different brain sizes have different body sizes as well

Answer 12

Brain weight (E)/ body weight (P) K = E/ P to the power of 2/3

Answer 13

Human Dolphin Chimpanzee

Answer 14

Increase Innovation e.g. different ways of foraging or building nests Brain size may relate more to boldness (low neophobia) than intelligence

Answer 15

The larger the group size, the more interactions: larger brains and more intelligent

Answer 16

Keeping track of how ripe the fruit is - more intelligent

Answer 17

Birds that store food have larger hippocampus in relation to their body size compared to non-storing Adult birds have larger hippocampus due to experience of storing food

Answer 18

Always have a null hypothesis Inspired by the Greek philosopher Anaxogoras who reasoned that all animals were equally intelligent, but that some were better able to express that intelligence

Answer 19

Unexpected between-species differences - Skard (1950) - rats and humans in complex maze - rats were much quicker and finding their way through the maze than humans and made fewer mistakes - Warren (1965) - goldfish, chickens, cats, horses, monkeys in a stimulus discrimination - Angermeier (1984) - operant response in various mammals, birds, fish Papousek (1977) added 5 month-old children Menzel & Erber (1978) added bees Results showed exact opposite of predictions from cephalisation index

Answer 20

Difficult to equate perceptual demands of test Difficult to equate motivational demands of test - make sure both species will value the reward equally Bitterman (1975) systematic variation - vary perceptual and motivational demands of the test to limit effect of contextual variables - however this is very time consuming

Answer 21

Effects of radiation on the brains of animals - Garcia used x rays and lithium chloride to study effects Saline = salty water: Illness - small volume drunk Shock - large volume drunk LTW = water in presence of lights and tones, only when animal's tongue makes contact with water Illness - large volume drunk Shock - small volume drunk Reason - certain types of associations are made more readily than others Internal experience of flavour associated with internal experience of illness External experience of lights and noises associated with external experience of shock

Answer 22

Because ‘intelligence’ is conflated with the concept of a general IQ (in humans) we tend to avoid the term in relation to animals We tend to talk about ‘cognition’ instead - processing information Psychologists tend to think of cognition as the ability to process information in an adaptive way Cognitive processes can be modelled in terms of computations Animal cognition as information processing: - Memory - Learning - Reasoning - Navigation - Communication

Answer 23

Capacity Time Number Categories

Answer 24

Habituation

Answer 25

Consolidation theories Retrieval theories

Answer 26

When current behaviour is under the influence of past experience

Answer 27

Location of Hidden Food Vander Wall (1982) – Clark’s nutcracker Nutcracker = foraging bird - Take pine seeds from pine cones in autumn and store them in locations - difficult for others to find all their food, also difficult for them to find food - They are much more successful at finding own caches of food than other's caches - memory is involved

Answer 28

Bird pecks at screen - some squiggles are associated with food rewards 80 images associated with food, 80 not - birds could remember and discriminate Pigeons could also discriminate between very similar photographs when associated with food or no food - 320 images were remembered - animals can have very large capacity for memory

Answer 29

Animals show particular cycle of activity over 24 hours - time of day and amount of light - when to be active, when to seek food, when to rest

Answer 30

- Roberts glued material over eyes of cockroaches and kept everything else the same - Their circadian rhythms were 23 1/2 hours, continued to get shorter to 15 hours - When he uncovered eyes, they recovered activity by about an hour a day Is this innate or learned?

Answer 31

Bred mice and let them develop in either 19 or 29 hour cycle of daylight - even though they got used to this they found it very difficult to anticipate when food would be

Answer 32

- Experiments with rats - trained them to press on a left hand lever for 2 sec tone, right hand lever for tone 16 secs long - Switched lights off then switched them on, rats would press lever and would only get a reward if lights were off for 4 secs, they on average learned to only press lever if lights had been off for 4 secs

Answer 33

Clever Hans trained by Von Osten tested by Pfungst (1908) - Horse could count - knew how many people would be wearing a hat and clopped his hooves to say the number - Could even do fractions - If trainer was behind a screen, clever Hans could not answer the question - He was reacting to the people asking the question and mimicking their behaviour - their anticipation and body language was a cue for the horse Supports Lloyd Morgan's canon

Answer 34

Left or right lever press depending on number of tones presented Equated length of time to present tones Changed this to use same length of time and different number of tones and animals could identify number of tones

Answer 35

- Trained monkeys to touch screen in correct order (ascending number order) - Are they learning a number or pattern of dots? - Changed stimuli so it was not possible to learn a pattern, and monkeys still pressed pictures in correct order - the monkeys could learn the numbers

Answer 36

- Food wells in order - trained that a certain food well was always rewarded - Spatial or number task? To test this, they rotated the stimuli by 90 degrees - chicks learnt four from the left or four from the right

Answer 37

- Alex the parrot could count number of stimuli and use language to say number - Could also say categories like colour Subitising - not counting just identifying

Answer 38

Presented pictures to pigeons - some associated with food reward - Some trained to discriminate trees from non-trees, water and not water, Margaret and not Margaret Responded more to new pictures containing the thing they had been trained to recognise

Answer 39

Animals could discriminate between oak leaves and non-oak leaves

Answer 40

Innate categories Exemplar learning Feature learning - learning specific parts of pictures e.g. particular shade of green Exemplar learning and stimulus generalisation

Answer 41

A reduction in responsiveness to a stimulus as a result of its repeated presentation

Answer 42

Fodor proposed the possibility of Innate categories that involve no learning. Face recognition in humans is one example of a proposed innate category – we naturally recognise faces and we do not have to learn to do so. However, the idea of innate categories is not plausible with some of the examples we have just examined.

Answer 43

The animal remembers every instance or exemplar of the category. Given Herrnstein et al.’s and Cerella’s results with novel items Exemplar learning is not very plausible either.

Answer 44

If a pigeon learns about features that the rewarded slides have in common with each other then it will form a positive association with features that slides share, but neutral associations with features that common to both rewarded and unrewarded slides. If this is how pigeons achieve categorisation then we would not only expect them to transfer performance to novel pictures, but we would expect performance to be as good as during training. In fact we find that while the pigeons are good with new photographs, their performance is not as good as with the training photographs.

Answer 45

Stimulus generalisation is when we show a strong conditioned response to items that are similar to the trained stimulus, but as similarity decreases responding drops off. Exemplar learning with stimulus generalisation does a good job of explaining categorisation, but there is still controversy about which of these accounts is correct.

Answer 46

Effector system - the physiological system responsible for producing a response This gets tired - habituation

Answer 47

Receptor system - the physiological system responsible for perceiving the stimulus This gets tired

Answer 48

STM lasts for relatively short periods and concerns information about the immediate past. While in this state the information can take part in associative learning, and it may influence the current performance of the animal. LTM is believed to endure for much longer periods of time

Answer 49

In order to determine whether or not an animal has retained information about a previously presented stimulus, the experimenter must provide some sort of recognition test. If the subject’s behaviour toward a stimulus should change, after being exposed to it, then this may be due to the existence of a memory of the initial exposure to the stimulus. The exposures could be to the stimulus itself (habituation), or the stimulus could serve as a signal for a biologically important event (conditioning).

Answer 50

Whitlow (1975) presented a loud tone to rabbits in a chamber and measured the change of blood flow in their ears in response to each presentation of the tone. He found that on the first presentation (S1) there was a strong response, but on the second presentation (S2), after a 60 second delay, there was a weaker response. If the interval was increased from 60 to 150 seconds the response to S2 was just as high as the response after S1 S1 - Tone 1 = startle 60 secs S2 - Tone 1 = weak startle (habituation) S1 - Tone 1 = startle 150 secs S2 - Tone 1 = startle

Answer 51

Following Whitlow's experiment, Wagner (1976) explained habituation in terms of memory, saying that the offset of S1 will leave a decaying representation of itself

Answer 52

According to Wagner (1976, 1981) the representation of a stimulus can be in one of three states: two active states, A1 and A2, and one inactive state, in which the memory is not modifiable and is unable to influence the animal’s behaviour. A1 = short term store Representation can be thought of as being at the centre of the animal’s attention As it decays, it moves to the A2 state A2 = rehearsal Can be thought to be at the periphery of the animal’s attention. It then decays to the inactive state I = long term store Because of the way Wagner’s model works, a representation in the A2 state cannot move to an A1 state before it has completed its decay to the inactive state Once in the inactive state the representation can move to the A1 state

Answer 53

If S2 is presented before the decay from A1 to A2 is complete – that is, within 150 seconds – and it matches the representation of S1, then habituation will be observed. However, if the delay between S1 and S2 is longer than 150 seconds, then no habituation is observed, which is what Whitlow found.

Answer 54

If the physiological system that is responsible for overt behaviour is fatigued then it will not be able to respond to any stimulus within the period of fatigue. Whitlow showed that Effector fatigue was not a good explanation for his results by changing the properties of the tone presented at S1 and S2: While habituation was normal when the same tone was presented at S1 and S2, with a 60 second delay, there was no habituation if different tones were used. Since overt behaviour was not impaired, Effector fatigue could be ruled out.

Answer 55

Cells responsible for the reception of S1 might be fatigued after the offset of S1 and would therefore be less sensitive at the presentation of S2. If the two tones presented at S1 and S2 are different then different receptors would be involved, but if the two tones at S1 and S2 are the same then the same receptors would be involved to process the two tones. If these receptors are fatigued then they would be unable to respond appropriately at S2. To determine whether Receptor fatigue could explain his results, Whitlow presented the same tone at S1 and S2, with a 60 second delay between the two. Mid-way through the delay, however, there was a 2-second distractor – a flashing light and a buzz near the eye. The distractor disrupted habituation, producing dishabituation, consequently showing that S2 is as well perceived as S1, and therefore ruling out Receptor fatigue as an explanation for the original habituation results. If it was due to receptor fatigue and not habituation, we would expected a lesser response on this experiment.

Answer 56

Retroactive interference The retroactive interference made the animals forget S1, and therefore there was no habituation at S2

Answer 57

In this an animal is allowed to visit four out of eight arms on the maze. At the end of each visited arm is a food reward. In between each visit to an arm the animal returns to a central hub. Rats are readily able to complete this task with 8- or 12-arm mazes, but find the task difficult with 17 arms. The rats could solve the problem of which arms to enter by following a stereotyped route, but observation shows that they don’t do this. To test how long the memory for previously visited arms lasts, some tests involve removing the animal after it has visited four arms for a short retention interval. When it is replaced into the central hub all eight arms are open and the animal has to avoid the previously visited arms to find food in the remaining arms.

Answer 58

Rats remember these visits for a duration of up to about four hours and a capacity of at least 8 items (Olton, 1978). Memory during the radial maze task seems to be more resilient to retroactive interference than in habituation.

Answer 59

Sea slugs release substance when threatened to hide themselves in response to vibrating surface - once habituated to surface, they stop releasing substance

Answer 60

Trained rats to complete a normal radial maze task, with a one-hour delay after the fourth arm had been entered. In the interval the rats were required to complete a second radial maze task in a different room. This second task appeared to have no effect on the rats’ memories for the first task, and Roberts (1981) was only able to produce retroactive interference in the radial maze by making the rats complete three distractor radial maze tasks in the interval of the first task.

Answer 61

Yes The two examples of short-term memory given here (habituation and radial maze) show that the properties of short-term memory can vary according to task. STM occurs in variety of tasks STM is decaying memory trace (Wagner)

Answer 62

Habituation = Short-lasting and susceptible to interference Radial arm maze = Lasts hours and less susceptible to interference

Answer 63

This animal undergoes a metamorphosis, over a period of 35 days, from a tadpole to an adult frog. Miller and Berk trained tadpoles to move from a black to a white compartment in order to avoid an electric shock. When tested for their memory of the task after their metamorphosis into frogs, the animals were still able to remember to avoid the black compartment.

Answer 64

Consolidation Retention Retrieval

Answer 65

Forgetting is due to an inadequate trace being laid down or to the passing of time Consolidation theorists such as Hebb (1949) and Duncan (1949) claimed that long-term changes in the nervous system must occur for long-term retention of information. According to Hebb memory depends on the virtually permanent formation of circuits of interconnected neurons. Consolidation and rehearsal is needed to complete these connections.

Answer 66

Forgetting is due to the animal’s inability to retrieve the information

Answer 67

Support for this idea comes from Duncan’s (1949) studies with Electroconvulsive shock (ECS). He trained rats to move from one end of a box to the other when a light came on in order to avoid a shock. He found that ECS immediately after a trial severely impaired the rats’ ability to remember the significance of the light. One explanation for these findings is that the ECS prevented the consolidation of learning.

Answer 68

Deweer et al. trained rats to run through a maze with six choice points in it. At the beginning of training the rats took 300 seconds to find their way to the end, but by the end they took only 30 seconds. One group of rats were then given a 25-day delay before running the maze again. They took, on average, 150 seconds to find their way to the end. A second group were treated in the same way as the first, with the exception that, just before they were required to run the maze after the 25-day delay, they were placed into a wire mesh cage next to the apparatus for 90 seconds. This treatment apparently had the effect of reactivating the memory because rats in this group took just 40 seconds to run to the end of the maze. This experiment suggests that the reactivation treatment facilitated the retrieval process that consolidation theorists thought unimportant. We can conclude that forgetting from long-term memory occurs for a variety of reasons.

Answer 69

- Light placed above cage where receptive female was - light was signal for a receptive female - Males would go near the female As a result, quails approached the light alone as they associated this with a receptive female, OR because the associated response is to approach the light

Answer 70

Stage 1: Tone came on = wintergreen Noise came on = peppermint Stage 2: Rat drinks wintergreen without tone, this was mixed with chemical to make them sick and rat became sick Stage 3: Tone and sucrose = gaping response aversive reactions (like that with bitter foods) Noise plus sucrose = lip smacking hedonic reactions (like that with palatable foods) - He gave rats the tone with sucrose but no flavour, and animals still produced this gaping response - Rats did not do this with noise and sucrose - Disproves that conditioning is only associating one thing with another - Rat is remembering something about perceptual qualities of the stimulus

Answer 71

USs have Specific and Affective qualities Affective qualities can be Appetitive or Aversive - Specific perceptual qualities - make USs different to each other - Affective qualities - how you feel about something, good = appetitive, bad = aversive Specific qualities of the US are things that make the US unique – flavour, if the US is food or liquid, where the US is presented, its duration, intensity, etc. Affective qualities are those characteristics that the US has in common with other USs. Food, water, and the opportunity to mate are all things that might be said to lead to a state of satisfaction in the animal, and are collectively called Appetitive USs. Mild electric shocks, illness and loud noises are all things that might be said to be unpleasant stimuli, that the animal might try to avoid if possible, and are therefore known as Aversive USs.

Answer 72

Evidence of stimulus substitution - look at autoshaping Inside conditioning chamber, pigeon sees light, then food is delivered Light = pavlovian signal for food being delivered Animal does not have to do anything to receive the food - pavlovian, not instrumental However, pigeon pecks at light as if it is eating grain, and different beak movement if stimulus is water In pigeons if a CS is presented for a few seconds before the delivery of a US then over time the bird will start to make CRs (pecks) to the CS. The bird doesn’t have to respond to the CS to gain a reward, but does anyway. This is autoshaping. If the CS signals a US of food then the CR that the pigeon makes looks a lot like it’s pecking at grain. If the CS signals a US of water then the CR looks like the sort of pecks pigeons make to drink. Konorski also said that when the CS activates a representation of the Affective properties of the US then we can expect a Preparatory Response

Answer 73

Salivation during CS that signals US general increase in activity Conditioned Suppression decrease in movement e.g. rat presses lever for reward, CS light comes on and rat stops pressing in preparation for footshock to follow, then starts pressing again

Answer 74

Konorski said that when the CS activates a representation of the Specific qualities of the US then we can expect the CR to mimic the UR. This is known as Stimulus Substitution.

Answer 75

Tolerance is a decrease in sensitivity to a drug as a result of repeated exposure to it. This tolerance results from compensatory mechanisms: Biological: Metabolic changes Synaptic changes Psychological: Conditioning affects homeostasis mechanisms - Compensatory mechanism - homeostasis - Body works to reduce the effects of drug by some metabolic and synaptic changes to get it out of the system and return the body to homeostasis - If the body can predict changes, conditioning might affect metabolic and synaptic changes

Answer 76

Injected animals with morphine Injection = CS Morphine = UCS Analgesia = UCR If animal predicts effect of morphine, CR opposes UCR When drug was injected in a familiar environment to be under effects of that drug, animals were tolerant, but they were not tolerant in a different environment Becoming used to environment in which drug is administered, become tolerant to drug

Answer 77

- Rats injected with saline or ethanol - Ethanol - reduces body temp - Injections of each in room a or b - Tests - injected again in same or different context - If expecting ethanol in context a (group 1), increase in body temp should happen - preparatory response - If expecting saline in context a (group 2), no increase in body temp should happen - no preparatory response

Answer 78

Expecting alcohol but did not get it, their body temperature and blood pressure decreased - Measured pulse transmission time and skin temperature (Conditioning = vodka tonic, test = just tonic)

Answer 79

- Put rats on hot surface - raised paws up and licked paws If they feel effects of morphine they won't feel pain from heat so it takes longer for them to lift their paws up from the hot surface - Morphine, saline, Morphine - CS is injecting and is no longer associated with pain relief - M rest M - Injection always associated with morphine - build tolerance - M rest M - stay tolerant to morphine - MPM - gone from being tolerant to being non tolerant as injection no longer predicts effects of morphine Extinction - breaking link between CS and US relationship by only presenting CS

Answer 80

Breaking link between CS and US relationship by only presenting CS

Answer 81

That compensatory CRs will be most evident in contexts associated with drug-taking. Therefore when drugs are taken in unusual contexts the compensatory CR will be less evident and likelihood of overdose is greater Extinction of drug-taking is context-dependent. Drug users often self-report relapse after returning to normal environment When drug is taken in novel environment, there is no conditioned compensatory response, so this can much more easily lead to overdose than when in a normal environment - Build up conditioned response that opposes UCR - environental tolerance opposes effects of drugs - Take more drugs to get hit that they need because they have become tolerant - either biologically or through learning - Take large amount of drugs in different environment - higher effect of drugs and could lead to overdose

Answer 82

Much fewer rats died in heroin same group because they built environmental tolerance

Answer 83

Events that occur close together in time or space will become readily associated Conditioning needs to happen close together in time or no association will be made Shown in og pavlovian conditioning

Answer 84

- Saccharin (sweet) then given lithium chloride to make them feel ill - When paired straight away or when they feel ill 8 hours later, rats still learn an aversion to saccharin - sometimes conditioning can happen without continguity (bit of an exception) Also, Saline associated with illness but not associated with shock Light tone water associated with shock but not illness Even though principle of contiguity was in effect - depends on modalities

Answer 85

- Tone and footshock - Animals only learn to make association if tone preceds the footshock - standard group - Contingency is as important as contiguity - predictive relationship

Answer 86

A regular predictive relationship between contiguous events (CS-US)

Answer 87

Leon Kamin showed that for conditioning to be successful the US has to be surprising, and not just contiguous with, or contingent on, the CS. The discovery of blocking was the evidence for Kamin’s (1969) belief that the US should be unexpected for learning to occur. Kamin’s experiment: Stage 1 - a group of rats (blocking group) learned that a noise (CS) signalled a mild footshock (US). Stage 2 - a light and the noise were presented in compound, and training continued. When the rats were tested for the presence of a CR (freezing behaviour) when the light was presented alone, they showed a much lower response than a group that were always trained with the compound CS of a light and a noise (control group). Learning about the significance of the light was said to have been blocked by the previously learned relationship between the noise and the footshock. Blocking did not occur in a third group of rats (group surprise) that had the same training as the blocking group but in which the footshock was more intense in stage 2 of training. The results show that for the blocking group, blocking occurred because the US was not unexpected by the rats in stage 2. This was because of their prior training in stage 1. For group surprise the more intense footshock in stage 2 was unexpected so new learning about the significance of the light occurred. For the control group, whose learning began in stage 2, both the light and the noise signalled an unexpected US, so both were learned about.

Answer 88

- Also seen in humans - Mines signal tanks to explode - Then missiles and mines together signal tanks will explode, but people who have already learned the mines and tanks association ignore missiles

Answer 89

Blocking effect less prevalent in people with schizophrenia

Answer 90

If some cues are less attended to then learning about them won’t progress as quickly as those that are receiving the animal’s full attention. This seems to be the case in a procedure known as Latent Inhibition (LI).

Answer 91

For one group, the experimental group, a light is presented in stage 1 that does not signal an outcome for the animal. In stage 2 the light suddenly becomes a signal for the imminent delivery of food. When compared with a control group that only begin training in stage 2, and always learn that light signals food, the experimental group are slower to learn about the relationship between the light and the food in stage 2. The explanation for this result is that during stage 1 the experimental group learn that the light doesn’t signal anything and come to ignore it. In stage 2 they fail to notice its significance as quickly as the control group, because they’re not paying it any attention, and therefore learn more slowly.

Answer 92

Attention is high to novel stimuli and attention is low to familiar stimuli In the LI procedure the light is familiar to the experimental group in stage 2, but is novel to the control group. This accounts for the different rates of learning in the two groups.

Answer 93

Attention is high to novel stimuli Attention is high to signals for important events Attention is low to irrelevant stimuli In LI, for the experimental group the light is novel at first so attention is high, but it is irrelevant in stage 1, so attention to it drops off. In stage 2 the light signals an important event so attention is gradually paid to it, accounting for the fact that the rats do eventually learn the relationship between light and food. However, for the control group the light is both novel and important from the beginning of training, so learning is faster than in the experimental group.

Answer 94

Attention is high while learning about a stimulus Attention is low once learning is complete Pearce-Hall made use of some principles of attention in human cognitive psychology and predicted that attention would be high whilst learning about a stimulus is ongoing, but low once learning is complete. This was because, they said, there is a distinction between controlled and automatic processing. In controlled processing deliberate attention must be paid to the task at hand, but once the task is learned the processing becomes automatic, and attention paid to the task is low. In terms of LI, Pearce-Hall say that attention is high to the light in stage 1 for the experimental group, but that this attention drops off as the rats learn that it signals no outcome. In stage 2 attention is low to the light at first, but when the rats realise that food is delivered attention becomes high again. For the control group attention is high to the light, but once learning about the relationship between light and food is complete attention to the light is low.

Answer 95

Kaye & Pearce (1984) conducted an experiment with three groups. For Group None a light signalled no outcome. For Group Continuous a light always signalled the delivery of food. For Group Partial the light signalled food on half of the trials, and nothing on the other trials. Kaye & Pearce measured the Orienting Response (OR) of the rats – how often they looked at the light when it was on. The vigour of the OR is thought to provide an index of how much attention is being paid to the light. The results showed that the OR was high for all three groups at first, but that it dropped off for groups None and Continuous, but always remained high for Group Partial. Since Mackintosh says that attention should remain high for important events, the finding that attention decreased for Group Continuous is a bad result for his theory. Wagner says that attention should be low to familiar stimuli, but attention to the light remained high for Group Partial, which is a bad result for him. The only theory that is consistent with the results is Pearce-Hall: once learning is complete for Groups None and Continuous attention is low, and since animals can’t learn what the light signals for Group Partial attention remains high.

Answer 96

Wagner - Attention is high to novel stimuli and attention is low to familiar stimuli Mackintosh - Attention is high to novel stimuli Attention is high to signals for important events Attention is low to irrelevant stimuli Pearce-Hall - Attention is high while learning about a stimulus Attention is low once learning is complete This one is most likely to be correct

Answer 97

There is evidence for both conscious and unconscious processes in human conditioning, so some theorists have suggested a 2-system theory

Answer 98

Got shown a picture of either snake or spider for 30ms (too quickly to consciously process). Received shock with snake, no shock with spider. There was shock anticipation with snake even though the picture was presented too quickly to consciously process.

Answer 99

Anecdotal evidence - smokers often report reduced desire to smoke where smoking is forbidden anyway (e.g., hospitals) Droungas et al. (1995) Two groups of smokers shown a video of smoking-related cues. One group was allowed to smoke and the other was forbidden Group smoke reported a greater urge to smoke, greater feelings of withdrawal, and were quicker to smoke after the session

Answer 100

Taste of cigarette happens before the absorption of nicotine Just the act of picking up or tasting cigarette makes smokers happier even though nicotine has not taken effect

Answer 101

Not abstinence - as association still sits there in LTM (abstaining rates are worse for smoking than heroin and alcohol) Extinction - Removing UCS - association is lost Removing nicotine from cigarette - taste of cigarette no longer associated with pleasurable nicotine Or counterconditioning (aversion therapy) Taste of cigarette becomes associated with nausea due to emetic drugs CR becomes nausea instead of pleasure

Answer 102

The extinction of a(n undesirable) response through the establishment of an association between the CS and another (more desirable) response A new CR is established by pairing the CS with a more salient US

Answer 103

- Successful attempts but usually short term - In real world, CS can still be associated with pleasure Very difficult to remove all originally associated cues

Answer 104

“Irrational” fear of an “objectively harmless” stimulus DSM-IV categories: Agoraphobia - Public places outside home, e.g. shops trains Social Phobia - Being watched/appraised by other people Specific Phobia - Grouped into: - Animals and insects - Blood/injury/medical (e.g. dentist) - Situational (e.g. driving, crowds, enclosure, air travel) - Natural Environment (e.g. heights, water)

Answer 105

CS = sight of drill US = pain CR = anxiety However, phobias are often to stimuli other than just immediate CS e.g. just thinking about going to the dentist, not just sight of actual drill

Answer 106

Some reinforcers aren’t effective from birth. Their value has to be learned, and for this reason they are called secondary or conditioned reinforcers. The obvious example of this in humans is money. Babies would not value money, but as they learn that it can buy them other items, they begins to want it. It ends up having a value of its own to people, even though it’s worthless if not exchanged for something else of value.

Answer 107

One example of the use of social reinforcement to change someone’s behaviour is that of Ann, studied by Allen et al. (1964). She was a nursery school child who engaged in a number of odd behaviours, such as talking very quietly, poking and pinching her skin, and talking only to adults. One may conclude Ann was an unhappy and insecure child, but Allen et al. realised such behaviour led to social contact with adults. Instead of reinforcing her strange behaviours with attention Allen et al. instructed teachers only to give Ann attention when she was playing with other children. Within a day her contact with children went from 10 to 60%.

Answer 108

No Even without direct experience of a traumatic US or pain-causing CS, humans could form an association via vicarious conditioning, even if the UR is evoked via story-telling

Answer 109

Systematic desensitisation (Joseph Wolpe) Cues associated with anxiety listed and ranked Relaxation paired with anxiety cues New association with relaxation reduces effectiveness of cues Flooding VR therapies - no real danger - Baby Peter frightened of furry animals - Exposed him to animals - He would observe other children playing happily with animals - Interrupted by scarlet fever outbreak - Paired animals with sweets instead of fear - Once he was in state of relaxation she moved rabbit closer to him - process of desensitisation 'Mother of behaviour therapy' - Mary Cover Jones

Answer 110

- Wild rhesus monkeys are fearful of snakes Lab reared rhesus monkey observes wild monkey response and produces same fear response

Answer 111

Reinforcers - increase tendency to repeat target behaviour Positive reinforcers Negative reinforcers Punishment - decrease tendency to repeat target behaviour

Answer 112

- Butler - monkeys responded by pressing door to see the room for 9 hours straight - Jones - humans more likely to press unpredictable lights People will work just to be interested in something - interesting sensory input works as a reinforcer

Answer 113

Preferred activities can act as reinforcers for less preferred activities Preference measured by likelihood to engage in particular behaviour Homme et al - Children would sit quietly on carpet if they get opportunity to kick the bin after

Answer 114

Acquire reinforcing properties through experience Paired with primary reinforcers Need to avoid extinction

Answer 115

- Delayed reinforcement - less association and less repetition of behaviour Rat does not know what behaviour led to food - cannot associate lever pressing with food because they showed many different behaviours in 70s

Answer 116

e.g. assignment grades - Humans have language which bridges time - can rationalise what behaviours led to reinforcer

Answer 117

R1 : 2s access to food immediately R2 : 4s access to food after 4s Time between trials constant so responses to R2 would lead to twice as much food Despite this the pigeons pecked key 1 on 95% of trials It’s still important to ensure reinforcement is delivered quickly, because delay reduces incentive

Answer 118

Smokers offered $1 for foregoing a puff every 5 minutes Half told money given at end of session, others in 3 weeks Delay caused 20x more smoking Rewards of not smoking are too far in the future for people to conceive of the value

Answer 119

If children behave in class the token (conditioned reinforcer) can be delivered immediately Can be exchanged for a wide range of primary reinforcers, so different children can be reinforced differently

Answer 120

Fox, Hopkins, and Anger (1987) Reduction of injuries in mine workers using token economy 12-year period, injuries fell by 68% and 85% Cost of injuries per year $260k Cost of tokens $12k Unions insisted on writing tokens into contracts

Answer 121

The ability to combine two or more pieces of information to draw a novel conclusion

Answer 122

Deductive reasoning is said to occur when the conclusion is necessitated by the premises. For instance if the statements ‘All ice is cold’ and ‘This is ice’ are true, then we can conclude that ‘This is cold’.

Answer 123

Inductive reasoning is when the conclusion is likely from the premises. The conclusion can be drawn from the presented facts, but is not necessarily true – the conclusion follows with some degree of probability. So if ‘this ice is cold’ then we can reason by induction that ‘all ice is cold’

Answer 124

We can look at the ability of animals to solve problems

Answer 125

Navigating by taking account of one’s own body movements The desert ant, Cataglyphis, has the problem of returning home as quickly as possible, after finding food, to avoid the heat of the Sahara. Wehner & Srinivasan (1981) showed that the ant uses dead reckoning to achieve efficient navigation, rather than use landmarks or follow a scent trail along its outward path. They allowed ants to find food a fixed distance and direction from their nests. When the ants were at the food location the experimenters moved them 600m away. The ants searched for their nests as if they had not been displaced at all. To do this the animals would have to have recorded their own body movements with respect to some external directional information, and also to register the distance they had travelled, so they could compute how far it was back to their nests. We now know that the ants gain their directional sense from the position of the sun in the sky. Wehner and colleagues later showed, by manipulating the ants’ leg length, that they have a kind of internal pedometer for recording the distance they have travelled: - Doubled size of ants legs - doubled size of legs and they ran double the distance - Navigation due to own body movements Longer travel by path reckoning = more errors

Answer 126

Collett and his colleagues trained gerbils to find food a fixed distance and direction from a landmark. The landmark and goal moved from trial to trial, and the gerbils were released from different points of the arena, so they were unable to locate the food by path integration (dead reckoning). Even so, when the food was removed for a test trial, the gerbils still searched in the correct location with respect to the landmark. This ability to plot a course to a hidden goal using landmarks is known as piloting. Piloting can also take place with multiple landmarks.

Answer 127

If animals possess a cognitive map then they should firstly be able to select a novel route, or shortcut, to a goal, and secondly make a detour around an obstacle that blocks a previously taken path. Tolman et al could not find conclusive evidence of this - only took a novel route when light signalled goal Morris (1981) trained rats to locate a hidden platform in a circular pool of water. They were always trained to find the platform in the same location, and always released from the same point on the edge of the pool. The experimental group were released from a new point on the edge of the pool. To navigate efficiently the rats would have to select a novel route to the goal. Compared with a control group which was released from the same place as during training, they did appear to find the platform quickly. The results of another group, for which the platform was put in a new location during the test, showed that the rats weren’t able to detect or sense the platform in the water without learning its location. However, nearly all the rats in the experimental group began their path to the goal by heading in the wrong direction. It is possible that the rats only altered their course once they had explored the pool for a time, and crossed a path they had previously taken during training. Taken together, Morris’s and Tolman’s experiments do not provide good evidence for the existence of cognitive maps in rats.

Answer 128

Reward always in top right corner of rectangle Now reward was in novel shape Animals still searched in the appropriate corner with the long and short edges equivalent to what they were in the rectangle

Answer 129

Wolfgang Köhler was an early objector to Thorndike’s assertion that animals learned through trial and error. He showed that chimpanzees were able to use poles and boxes to get closer to fruit hung out of the chimps’ reach. He said this behaviour was evidence of insight, that the novel behaviour came as a result of a flash of inspiration after they had reasoned out the problem internally. One problem with Köhler’s conclusion is that the chimps were not naïve to interacting with boxes and poles. Perhaps they had had experience of using these items to solve other problems at some earlier time. Later experiments in the 1950s showed that chimps without prior experience of boxes and poles were unable to reproduce Köhler’s chimps’ behaviour

Answer 130

Conducted an experiment with pigeons. Some were trained to push a box towards a spot on the edge of an arena to gain a food reward. They were then trained to stand on a box that was placed underneath a plastic banana and peck at the banana to get food. In the test trial, the box was moved so it was not underneath the banana. Birds that had prior experience of pushing the box around were quicker to move the box underneath the banana, so they could stand on it, than birds that had no experience of moving the box. The fact that the birds produced a novel behaviour without appearing to learn through trial and error could be seen as evidence of insight. However, without prior experience the birds were not able to complete the task efficiently, so it seems experience might be key to explaining apparently insightful behaviour.

Answer 131

In the laboratory, one bird, Betty, was given a straight wire to use to retrieve a bucket containing food from the bottom of a cylinder. After some experience of the wire she spontaneously bent it to make a hook that she used to lift the bucket. The fact that she had experience of the wire, and was wild-caught, might be important to her production of a novel behaviour, but it is impressive, nevertheless.

Answer 132

A subset of inductive reasoning is reasoning by analogy. For example, in the question ‘Cat is to kitten as cow is to….?’ the answer, by analogy with the cat-kitten exemplar, might be calf. This is on condition that the cat-kitten relationship is thought to be an example of a mother-baby relationship. It could equally be an example of a big-small relationship, so the answer could plausibly be rabbit.

Answer 133

If animals can be shown to understand relationships of this sort then it’s possible they have the capability for abstract or symbolic thought, which is a key trait in human intelligence. Gillan and Premack and their colleagues (1981) trained a chimpanzee, Sarah, to use language by pointing to symbols on a board. She learned that one symbol meant ‘same as’, and was given a symbolic analogical reasoning task. The example was two symbols that were identical except for their size. She was then presented with a large triangle with a spot in it, and asked what the correct choice was between a small triangle with a spot in it, and a large triangle with no spot. She understood, by analogy with the example relationship, that the correct answer (‘same as’) was the smaller of the two triangles. It is possible that the experimenters had inadvertently cued Sarah to give the correct answer, since she had never been presented with an analogical reasoning task before, but if the results were to be replicated then this experiment perhaps provides evidence of the most remarkable problem solving skills we have yet seen in an animal. Also chose other equivalents like tin and tin opener

Answer 134

Communication occurs when one organism transmits a signal to another organism that is capable of responding appropriately to that signal

Answer 135

The foraging bee returns to the hive and recruits other workers to search for food in the same location using one of two dances. Round dance - food source is less that 100m from the hive - encourages random search by the recruited bees Waggle dance - food is more than 100m from the hive: The bee runs in a straight line waggling its abdomen rapidly, then turns 360o to the left so it ends where it started. It then makes the straight run again but this time turns to the right. Altogether this dance forms a figure of eight pattern. The dance tells the workers both the distance and direction of the food from the hive: Distance is revealed by the length of the straight run. Since the rate of waggling is constant, the distance of the food from the hive is also proportional to the number of waggles, and how long it takes to complete the whole figure or eight pattern. Researchers watch the dance then ran after recruited bees to see how far they went

Answer 136

The dance is performed on the vertical surface of the honeycomb and the direction of the straight run has a fixed angle with respect to vertical. The angle subtended between vertical and the direction of the straight run is the same as the angle between the sun’s position in the sky and the food source.

Answer 137

Seyfarth and Cheney (1993) 3 different alarm calls in response to different predators: When a monkey sees a snake its call causes the others in its troop to look around on the ground; when a leopard is spotted all the members of the troop run to the trees in response to the alarm call; and when an eagle is detected, the alarm call causes the others to look up to the sky. Knowledge of what the calls mean is not innate – the calls are learned in childhood.

Answer 138

Chomsky - asserted that because all language conforms to rules of grammar, humans must have a language acquisition device in their minds, that will not be found in animals Macphail - argued that there is no evidence for differences in the intelligence of different species, other than between humans and all other animals. This difference is due entirely to our ability to use language. It follows from this argument that if animals can acquire language it blurs the distinction between human and animal intelligence.

Answer 139

Discrete units Arbitrary units Semanticity Displacement Productivity through syntax

Answer 140

Composed of discrete units Are arbitrary in that the calls have no direct (unlearned) relationship with what they signal They have semanticity - each call has a specific meaning Does not have displacement or productivity through syntax

Answer 141

Language allows us to refer to events that are displaced in space or time; the honey-bee dance allows the signaller to communicate information about food that is not present in the hive

Answer 142

Language is structured through rules of grammar, or syntax. Words can have many different meanings when put together in different orders. It is syntax that makes language flexible and varied, but no natural animal communication is structured by rules of grammar as far as we know.

Answer 143

Furness and Hayes & Hayes attempted to train an orangutan and a chimp (Vicki), respectively, to speak, but these studies had very limited success, partly due to the physical limitations of apes’ ability to produce speech.

Answer 144

By the time she was five years old, after four years of training, she was able to produce 132 words, including verbs, nouns and pronouns. She could string these words into sentences like ‘you tickle me’ and ‘open key food’, an instruction to open the fridge. One day, while out with her trainer, Washoe saw a swan on a lake, a bird she had never seen before. The trainer asked, ‘What’s that?’, and Washoe replied, ‘Waterbird’. It is possible that this was a demonstration of Washoe using syntax to create a new word. Later in her life, Wahsoe adopted a chimp called Loulis. Loulis learned 22 words from interactions with Washoe, and there is some evidence that Washoe actively trained Loulis to form signs with his hands.

Answer 145

David Premack and his colleagues trained Sarah, another chimpanzee, to use symbols on cards that each meant a different word. Sarah learned 130 words including nouns, adjectives and verbs, and had to place the cards in the correct order (syntax) to be rewarded.

Answer 146

Sue Savage-Rumbaugh and her colleagues trained a number of apes including Lana, a common chimp, and Kanzi, a bonobo to press symbols on a board, called a lexigram, to create sentences.

Answer 147

In one interaction Premack, using card symbols, asked Sarah what colour chocolate was. In reply, Sarah chose the symbol for brown. This demonstrated Sarah’s use, not only of discrete and arbitrary units (the symbols), but of semanticity and displacement – she knew the meaning of the symbols and the chocolate was not present. Many of the researchers who have trained apes to use symbols believe they fulfil all of Hockett’s criteria for language, including productivity through syntax.

Answer 148

Concluded that simpler mechanisms could account for many instances of apparent productivity through syntax. Washoe - simple sentences like ‘you tickle me’ that require syntax for the correct meaning, could have been learned through trial and error. The instance where she signed ‘Waterbird’ could simply be a demonstration of her signing the two things she could see – the lake and the bird.

Answer 149

Terrace trained a chimp, Nim, to sign ASL, and recorded over 19000 multiword utterances, of 5235 types over an 18-month period. Despite this number and variety, Terrace concluded that there was no grammatical structure to what Nim said. He found that position habits could make certain sentences look grammatically correct: Nim was in the habit of signing the word ‘more’ before the noun in a sentence Nim also copied sentences from his trainer, and then altered one or two words, so the new sentence had the correct syntax Most importantly, Nim did not elaborate in his sentences - in children who are learning to speak the mean length of utterances increases with age. This was not the case with Nim. However, despite Terrace’s conclusions there’s no doubt that, although difficult to show, some apes, like Sarah, are able to produce grammatically correct sentences.

Answer 150

Alex, the parrot (Pepperberg), Kanzi, the bonobo, and various dolphins in Louis Herman’s lab have been shown to comprehend the grammatical rules of instructions.

Answer 151

Yes Overall, evidence that animals can acquire and reproduce grammatically correct language is limited

Animal learning and cognition Flashcards

(175 cards)