Lecture 3 Flashcards

1
Q

What does the prey model explain?

A

Why predators select certain types of prey when they are foraging

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Who created the prey model?

A

MacArthur and Pianka (1966)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the aim of the prey model?

A

To maximise fitness

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What does the prey model try to calculate?

A

If a predator encounters one less rewarding prey regularly but another more rare prey which provides greater gains to fitness, which would be the best strategy.
So whether collecting everything it finds or search only for more rewarding prey is the better strategy

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the 3 possibilities considered in the prey model?

A

1) only take the most profitable prey (maximises fitness gain for each prey but more search time. So, every time you eat the prey you will get large energy intake but have a longer time between)
2) take all prey encountered (reduces search time but decreases fitness gain)
3) only take the least profitable prey (obvs not a rational strategy

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is the constraints of the prey model?

A

. The model presumes that all things will evolve to be optimal.
. The model assumes that search behaviour and handling behaviour are mutually exclusive (so might mean the time it takes to load up a beak for example or the amount of time it takes to crack open a nut). However, probably in many cases this will be violated (e.g. if while a bird is breaking a nut it may be looking around for the next good resource)
. Prey are encounter sequentially (so they find one prey then they find another, so don’t ever find two prey at the same time)
. The animal has all necessary info regarding the available prey, encounter rates with each prey grow and how profitable each is (so is presuming a perfect cognitive system where they know everything)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Give the crab example of the prey model

A

. Crabs were offered different sized mussels in controlled ratios (low, medium and high density)
. At low and medium densities of mussels the intermediate mussels were preferred
. However, the model would predict that at high densities the crab should take only large mussels but this did not happen they took mostly large with some intermediate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Because the prey model example of the experiment done on crabs and mussels did not accurately predict behaviour can we discard the model?

A

. No because some of the constraint assumptions were not met by the experiment
. Multiple prey instead of sequential prey
. Measure of encounter could be incorrect as crabs may not actually encounter all mussels in the experimental set up

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Who carried out the crab and mussels experiment on the prey model?

A

Elner and Hughes (1967)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What were the predictions of the ratio they should take to maximise their fitness/ energy intake from the optimal foraging theory of the crab and mussels prey model experiment?

A

At low density they should take a few high-quality individuals, a bit more medium quality ones and a lot of low-quality ones.
At medium density they should take many medium quality ones
At high density they should take only high-quality ones

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What did they find when they used crabs and mussels to experiment the prey model?

A

Found that at low density the model predictions were fairly right, same at medium density, but the model was not fully right at high density because they took medium quality ones as well.
So, this experiment did ok qualitatively but not quantitively, the numbers don’t match perfectly but in general terms the patterns are kind of the same way the model predicted

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Who experimented the prey model using great tits?

A

Krebs et al (1977)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How did Krebs et al. Experiment the prey model using great tits?

A

. Sections of mealworm were placed on a conveyor to create an encounter rate and they could take which ones they wanted and the worms were if different quality
. Mealworms were either 4 or 8 segments long

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What did Krebs et al find when they used great tits to experiment the prey model? And what does this mean?

A

. As expected the birds took the larger prey
. Demonstrates that abundance of most profitable prey is the most important factor
. So roughly agreed with the model
. The birds still took some smaller prey when larger prey were in high abundance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What were the issues with Krebs et al. experiment of the prey model using great tits?

A

. Model does not accurately predict behaviour
. Could be errors by the bird
. Could be sampling behaviour
. The birds (nor the crabs in the other experiment) followed the all or nothing duke that you should only take high quality resources
. Worked qualitatively but not quantitatively, never gave an exact match in numbers but generally the pattern is in the correct direction

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Who first proposed the patch model?

A

Eric Charnov (1976)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is the patch model also known as?

A

Marginal value theorem

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is the decision that the patch model makes for us?

A

That when you have a bunch of patches of food with areas in between them that don’t have any food, this tells you how long you should stay in a patch before you move on. So, when you forage in a patch the prey is depleted/ the density of the prey is depleted, so the profitability of being in that patch goes down. So, the cumulative number you can catch at any one time because the insolence** of the prey goes down. So, there reaches a point where you are putting in a lot of effort for not capturing many more prey and this model tells you when you should leave that patch and move on

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What are the three parts to the patch model?

A

1) decision to stay in patch or move on
2) currency- long term net energy intake
3) constraint assumptions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What happens to the energy intake as the patch is depleted?

A

. Energy intake starts to decrease

. Sooner or later this will reach a critical point or marginal value when that patch is no longer profitable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

When a patch reaches a critical point or marginal value when the patch is no longer profitable what is it better to do?

A

It is at this point that it is better to find another patch than to expend energy in the depleted area

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What does the patch model show?

A

That both travel time between patches and patch richness will have a large impact on when the marginal value is reached, so if you have a patch that has a lot of resources in it to start with then you would spend a lot of time in there compared to an area with little resources. If you have a long time to travel that will expend a lot of energy so you may want to spend a bit longer in that patch

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What is the issue with the patch model?

A

It works qualitatively but not quantitatively

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is central place foraging?

A

Is a model where organisms collected food in patches and then return to a central place to utilise them.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

Give an example of central place foraging

A

Blackbirds in the spring where they will load their beaks with mealworms from a patch and then return to the nest to give them to the young

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What does the central place foraging model suggest?

A

That as travel time increases the animal should spend a longer time loading up at any given patch because you will expend more energy the further you go. Short trips require less energy so can be done more regularly and with smaller loads. Patch richness will also affect loading time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

Who came up with the. Entrap place foraging model?

A

Orians and Pearson (1979h)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

Give two main features of the central place foraging model

A

1) loading time- time taken filling beak, or pouches etc.

2) travel time- time spent moving from patch to burrow (includes unloading time)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

In the central place foraging model what does patch richness affect?

A

Loading time as animals spend less time searching the patch for food

30
Q

Give the eastern chipmunk study example of marginal value theorem so essentially when they should leave that patch

A

. Chipmunk fills up cheek pouches with seeds
. Transports seeds to burrow where they are cached
. These caches are used to get the individual through the winter
. Enters torpor for 4-5 days at a time then emerging to eat a bit and then go back to sleep over winter
. Body temp can drop to almost 0 degrees

31
Q

Who carried out the eastern chipmunk study on marginal value theorem?

A

Giraldeau and Kramer (1982)

32
Q

What were the predictions made when eastern chipmunks were used to model marginal value theorem?

A

. That as the distance from the burrow increases the amount of time you spend in the patch should increase
. That the weight of seeds should increase as the distance increases
. That the rate in which take on food grams should decrease depending on your travel time

(3 different model parameters states, they just tested the model at three different states)

33
Q

What was the issue with the eastern chipmunk study of marginal value theorem?

A

Loads were much smaller than those predicted by the model. So, the time they spend in the patch and the amount of seeds taken and the rate of seed is all lower than was predicted by the model
Energy expenditure, predation and competition were not taken into account

Qualitatively the results are ok but there was a big difference between the models predictions and what was seen, so there is quantitatively discrepancy between what was seen and what the model predicts

34
Q

How did they carry out the study on eastern chipmunks looking at marginal value theorem?

A

Used sunflower seed trays as patches to measure intake. Done in the lab where they moved different seed trays different distances

35
Q

What is the issue with optimality models?

A

Very general but not usually very accurate quantitatively

36
Q

What is the issue with increasing the accuracy (quantitatively) of optimal models?

A

Accuracy can be improved but this reduces the generality of the model, so can add other things to the model but it will make them very specific and then you won’t be able to use them on other animals.
As the model becomes more specific it is less applicable to a wide range of scenarios- different organisms.
Some additions can be useful however such as sampling or stochastic resources

37
Q

Explain the experimental set up known as the two-armed bandit using great tits (Krebs et al) (how they did it, predictions and results)

A

. Great tits given 2 patches to forage in
. Birds had not experienced either patch before and didn’t know which was more profitable
. Model predicted that movement between patches (sampling) should decrease as information is gathered
. Results showed this was the case

38
Q

How can foraging accuracy be improved?

A

By sampling the environment to increase information about resources

39
Q

What is meant be stochastic resources?

A

This means random/ unpredictability of the resources. So, if there is a certain amount of unpredictability to the quality of resources when you meet them

40
Q

If animals aren’t getting enough food then what behaviour are they likely to show?

A

Risk prone behaviour

41
Q

Describe optimality models (include positives and negatives)

A

. Had some qualitative success
. In most ways they fail to capture the behaviour acutely in a quantitative way
. This inaccuracy is a large part of why science is moving away from this type of model into more complex areas such as game theory and the use of neural networks to model behaviour
. Optimality models can provide a general overview of some behaviours but should be used only as a guide if at all

42
Q

Give an overview of social foraging

A

. Group foraging comes with costs- when foraging on your own there is no intraspecific competition
. There are two types of competition:
- scramble competition
-interference competition

43
Q

Describe scramble competition

A

Competitors foraging on the same resources reducing possible gains/ depleting the resources. Long- lives effect- when competitor leaves resources remain depleted

44
Q

Explain interference competition

A

Could be dominant individual causing increased vigilance or simply extra individuals scaring off prey. However, if another individual comes into your patch and scares the prey off then when that competitor leaves the prey comes back and you can start foraging normally. Short-lives effect- when the competitor moves on the resource returns to the higher value

45
Q

Why do animals join a group in terms of predation?

A

. Predation reduction (dilution effect)- The more targets available to a predator the smaller the chance any given individual will be attacked
. Conspecifics can also provide a shield to hide behind when predators attack- selfish herd
. Selecting a target from many moving individuals can cause the confusion effect reducing predator efficiency
. Predator detection is easier when many pairs of eyes are watching rather than just one individual- vigilance behaviour reduces for each individual in a group vs the single individual
. In some species mobbing behaviour or group defensive behaviour is seen

46
Q

Why join a group as a predator?

A

. More efficient predation strategies (African wild dogs)
. Many animals hunt in groups
. Allows them to take larger prey

47
Q

Why join a group in terms of encounter rate?

A

. Increases the encounter rate through the sharing of information which provides two benefits:

1) increases every individuals chance of exploiting a patch
2) it can, but does not always increase energy intake (for this to be true the rate of patch location must be high and patches must contain enough resources for all members)

48
Q

Give an example of animals living in a group having a positive effect

A

African wild dogs live in very large packs and have a hunting success rate/ efficiency of between 65-90% (most place it at around 80-85%) but is they were in a smaller group then that would go down significantly

49
Q

What does ideal free distribution explain/ predict?

A

Explains how animals should disperse themselves in a group/ predicts how animals should distribute themselves in the environment to maximise their gains.

50
Q

How has ideal free distribution been used?

A

Has been used for social animals to determine what the optimal group size is and how they should organise themselves

51
Q

What does ideal free distribution ‘forget’ about and assume?

A

Competition and just assumes that to maximise gains individuals in a group should spread out as much as possible

52
Q

Ideal free distribution assumes that to maximise gains individuals in a group should spread out as much as possible. What is this called?

A

Dispersion economy

53
Q

According to ideal free distribution when does aggregation occur?

A

When aggregation occurs it is due to lack of suitable patches

54
Q

Who came up with ideal free distribution?

A

Fretwell and Lucas (1969)

55
Q

Why was ideal free distribution developed?

A

To look at how animals are dispersed throughout habitats

56
Q

What can you use ideal free distribution to determine?

A

If you have a bunch of resources in patches you can use this model to see how the animals in that environment will go into these patches/ how they will group within these patches

57
Q

What does the ideal free distribution model assume?

A

. Patches have an intrinsic quality in the absence of competitors (so doesn’t take competition into account)
. Individuals will move to the patch of highest quality (they are ‘free’)
. Individuals will know the value of each patch (‘ideal’)
. Competition will reduce the value of a patch
. Each individual is equal in ability to exploit patches

58
Q

What are the predictions of the ideal free distribution model when looking at the patches?

A

. If a group has a choice between 2 equally profitable patches they will distribute equally between them
. Any difference to this distribution is transient
. Individuals in an overexploited patch will move to maximise gains
. When equal numbers of individuals are in each patch no migration will occur as it would not be advantageous (would just increase competition)
. This makes the ideal free distribution an evolutionarily stable strategy
. Patches don’t have to be equal value. Equilibrium is reached when the number of individuals in each patch is such that no benefit will be gained by joining the other group

59
Q

When is the problem with IFD?

A

It is a qualitative match but not quantitative

60
Q

What are the assumptions by the ideal free distribution model looking at individuals?

A

. No difference between individuals- obviously not true (different personalities). Some individuals will be better competitors than others in the group
. Individuals have perfect knowledge of patch quality- again hugely unlikely which is why we see sampling behaviour
. Individuals can accurately assess patch quality- in reality some limitations will exist. Sensory limitations could reduce the ability of an individual to measure the quality of a patch- limitations as to how they can do that

61
Q

What can alter the results of an ideal free distribution model?

A

Variable interference can alter the results- when resources are a set value and not a continuous input individuals can maximise their gains by being aggressive to conspecifics

62
Q

Explain stable group size

A

Optimal group size will provide the maximum gain for each individual in the group. This is not seen in nature. Groups are often bigger than this which is known as stable group size

63
Q

What is the key concept of classic game theory? What is that?

A

Nash equilibrium, it is when you’ve found the optimal strategy

64
Q

What is the key concept of evolutionary game theory?

A

Evolutionary stable strategy

65
Q

How is the optimal strategy determined in classic game theory?

A

Analytical mathematics or computer algorithms applied directly to the payoff matrix to determine optimal strategy

66
Q

What is the issue with game theory when tests on real animals?

A

There are constraints in the ability of individuals to perceive and process information from the environment then memorise and recall it perfectly is another issue that is frequently ignored in the game theory models.

67
Q

Give an example of where game theory would be incorrect on real individuals

A

You may have a game where there are 10 iterations and players are expected to remember every single behaviour that their opponent did in the past moves. This is not realistic but many game theories assume this. Basically, assumes perfect cognition in their ability to take in and process information and produce the best behaviour. So, there are criticisms for game theory that both apply to economics and behavioural ecology

68
Q

How do you train a neutral network?

A

Generally through natural selection

69
Q

What happens to the rate eastern chipmunks load seeds into their pouches the longer stay in a patch and why?

A

Will load seeds into their pouches at a high rate when they first go into a patch but the longer the time they spend in that patch the rate at which they load into their cheeks goes down because the latch is being depleted

70
Q

When they tested the marginal value theorem on eastern chipmunks and they found that loads were smaller than predicted so the time in the patches and the amount of seeds taken and the rate of seeds is all lower than predicted by the model, what could have been the reasons for these results?

A

. Could be the measure I the currency (g/s)- energy intake isn’t everything you may need nutrients as well
. Energy expenditure was not taken into account at any point
. Predation- if loading in an exposed area than shorter foraging trips would be safer (predation isn’t taken into account in this theory, in the lab they are in a very exposed area, mag expect smaller foraging trips off they feel threatened)
. Competition- both on the patches and cache need to be defended if rivals are present (all competition wasn’t present in the experiment the chipmunks may be expecting competition so that may affect their behaviour)

71
Q

Why do animals sample? What is the issue with models and sampling?

A

Often animals have to sample before they can decide what resources are high or low quality but in models it is presumed that the animals already know this and have all the information