L6 Flashcards

1
Q

What is optimality theory?

A
  • In order to maximise individual fitness animals should forage as efficiently as possible
    -Animals should balance benefits of caloric intake and costs eg time and energy expense
    • Usually graphical models
    • Can predict how and animal should go about optimising its or its offspring’s food intake
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Example: North Western Crows

A
  • Feed on whelks
    • Drops welk at height and eats soft body of whelk
    • Which welk should it choose, height of dropping it, how many times should it drop it before it gets another?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Observations of Whelk feeding

A
  • Crows select large whelks
    • Drop from 5m height
      -Don’t move onto another welk after it doesn’t break
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Predictions

A
  • Large whelks should be more likely to break at 5m than smaller whelks
    • <5m results in fewer breakages
    • > 5m breakage plateaus
    • Each time a whelk is dropped it has the same chance of cracking
    • Large whelks take fewer drops to crack than smaller sizes
    • This amounts to a large reduction in cost
    • Larger also provide more calories
    • 5m was the optimum drop height, going higher doesn’t increase breakage chance that much more
    • Also takes longer to retrieve whelks at higher height so more chance another animal may steal it
    • Dropping it x amount of times is independent
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What causes a mismatch between cost benefit logic - why might an experiment prove an animal’s foraging not to be optimal? (4 reasons)

A
  1. The animal may not have been well ‘designed’ by selection
    • Environment may have changed an evolution hasn’t caught up
    1. Observations weren’t appropriate
      - Experimental design needs to be reassessed
    2. An important factor may have been omitted from the model
    3. The assumptions may not have been valid
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Example: Oystercatchers feeding on mussels

A
  • Smaller mussels were chosen than it was predicted
    -Birds behaviour did not match model as large mussels were impossible to open
    • Largest mussels should theoretically be taken by oystercatchers
    • Size chosen is far less than theoretically optimal as they cannot open them
    • When mussels they can open is considered then their foraging is close to optimal
    • Food availability has placed a constraint on an individuals ability to forage optimally
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Nutrient quality is more important to herbivores than carnivores: Moose

A
  • Moose feed in two habitats
    • Deciduous forest leaves which are high in energy but low in sodium
    • Aquatic vegetation which is high in sodium but low in energy
    • Herbivore digestive system can only process so much as it is hard to digest
    • Stomach space is an additional constraint
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What combination of aquatic and terrestrial vegetation should moose eat for optimality?

A
  • Large amount for aquatic and small terrestrial or vice versa
    • Have to take in enough of both
    • Minimum sodium intake means a certain amount of aquatic vegetation
    • Digestive bottleneck means they cannot take in too much food
    • Moose cant exist just on aquatic plants as its rumen is too small
    • Maximising sodium intake requires more aquatic material
    • Maximising energy intake requires more terrestrial vegetation
    • Moose maximise energy intake whilst meeting sodium requirements

-There is an optimal diet to be above sodium constraint, below rumen constraint and above the energy constraint

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

Charnov’s marginal value theorem

A

-Do animals balance cost benefits?
-Foraging environments tend to be patchy with food distributed in clumps
- Eg a swarm of krill
- Patch feeding requires knowledge of how long to spend in patch before moving on
- As time goes on in patch rate of food intake decreases
- Loading curve
- Curve of diminishing returns
- Returns diminish as eg carrying weight, less food resource

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

When to give up on a feeding patch?

A
  • If it gives up too early on a patch there is too much travel time travelling and there is little reward
    -Too late wastes time foraging ineffectively
    • Optimal patch time
      -Travel time to patch needs to be least as possible for greatest resource size
    • Maximum benefit means travel time has to hit loading curve at greatest intake of energy
    • A longer travel time means longer in patch to maximise energy gain
    • Bigger shop for greater travel time
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the loading curve in Charnov’s marginal value theorem?

A

Loading curve

- Curve of diminishing returns
- Returns diminish as eg carrying weight, less food resource - as longer time is spent in patch the fitness diminishes to a plateau
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Do animals behave according to charnov’s theorem? : Starlings

A
  • Starlings with chicks have to fly long distances
    • How many prey items should be collected on each trip to maximise food delivery?
    • It is harder to find food whilst carrying prey so they get diminishing returns
    • Shorter travel time means patch time should be shorter and vice versa so less food collected
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Experiment

A
  • Trained starlings to collect mealworms from a tray at which they could be dropped at a certain rate giving birds diminishing returns
    • Load size increases with increasing distance from the nest
    • Close correspondence between observed load size and predicted loads
    • Starlings have to be concerned about predation, self feeding etc
    • Charnov is accurate and a robust model
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

The four assumptions of the MVT

A

1.Travel time between patches is known

2.Travel costs = patch time costs

3.Patch profitability is known

4.No predation

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

Travel time between patches is known

A
  • Assumption is true if going backwards and forwards to feed, but if searching for eg a patch of krill distance is often unknown
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Travel costs = patch time costs

A

may be true for hummingbirds flying when travelling and hovering while feeding, but is not likely to be true of seedeaters that fly to sight and sit passively feeding

17
Q

Travel costs = patch time costs experiment

A
  • 30 patches arranged on 5 trees with different spacing to vary travel times
    • Blue tits feeding
    • Found that time as a measure was not a good fit
    • Travel costs greater than patch costs?
    • By identifying a mismatch differences in travel costs and patch costs affect foraging behaviour
18
Q

Patch profitability is known

A
  • Assumes animal knows how much food is in the patch
19
Q

Patch profitability is known experiment

A
  • Animal is unlikely to know how profitable a patch is in advance
    • This is done by sampling different patches
    • Trained downy woodpeckers to forage from logs each with 24 holes - empty or with seeds
    • 2 logs presented in each trial , one with no food and one with varying food
      -In trials with 0 and 24 logs, it was easy to determine seed presence
    • Had to observe very little holes to judge whether there were seeds
    • Logs containing more seeds meant more observations
    • Had to observe more holes as there was more food
    • Woodpeckers used information to maximise net energy intake and minimise search cost
20
Q

No predation

A

No stochastic environment things going on

21
Q

If assumptions are violated …?

A

The model may need to be modified

22
Q

Optimality models and behaviour benefits

A
  • Testable quantitative predictions
    • Explicit assumptions that may not always work
    • Illustrate the generality of decision making
23
Q

What to do when the model fails to predict observations? 3 rules

A
  • Ignore it (count as acceptable error)
    • Accept animal is sub-optimal
      -Re-build model