4: Optimal foraging

Question 1

why can we make precise predictions on behaviour?

Answer 1

behavioural ecology has strong theory

Question 2

what is the optimality logic?

Answer 2

selection will favour animals that forage most effectively

Question 3

what is optimality modeling used for? (3)

Answer 3

• used to determine the best course of action for an animal
• insights into optimal foraging (often graphically)
• predict how and animal should optimise its own food intake or food provisioning to young per unit time

Question 4

what ideas did John Maynard Smith introduce?

Answer 4

ideas from economic theory and applied these models to animal behaviour
- developed optimality modelling

Question 5

in a Zach study 1979 what was the study and what were the observations ? (3)

Answer 5

whelk choice by North western crows in USA

• always select large whelks (3.5-4.4cm)
• drop them from 5m onto rocks to break them open and eat body inside the shell
• keep dropping the same whelk until it breaks

Question 6

in a Zach study 1979 what were the predictions? (3)

Answer 6

• large whelks should break easier at 5m than small whelks
• whelks dropped at <5m should be less likely to break but if >5m no more likely to break
• chances of whelk breaking should be independent of number of times its dropped - each time its dropped it has the same chance of breaking

Question 7

in a Zach study 1979 after dropping whelks from a tower on the beach from different heights what were the results found? (6)

Answer 7

• large whelks take fewer drops to crack (5 compared to 18 for small)
• larger whelks have more calories
• 5m is the optimal height with 5.6 being the min energy expenditure
• if drop from too high then whelk may be stolen before reaching it
• chance of whelk breaking not affected by number of previous drops
• optimal behaviour by dropping from 5m and using the same whelk until it breaks

Question 8

what can a hypothesis based on cost benefit logic that is incorrect lead to?

Answer 8

further insights

Question 9

why may a hypothesis based on cost benefit logic be incorrect? (4)

Answer 9

• animal may not have been well designed by selection (not yet evolved to changing environment)
• observations may have been inappropriate so need to re-assess experimental design
• important factor may have been omitted from the model
• assumptions may not have been valid for the species studying

Question 10

describe the predictions of the oystercatcher study on constraints on foraging

Answer 10

predicted to select large mussels that yield the most food

Question 11

what were the results of the oystercatcher study on constraints on foraging

Answer 11

• birds selected mussels smaller than predicted as cant open the largest ones
• if only mussels that can be opened are considered then the optimum size is 50mm which matches observations

Question 12

what do moose need to eat and why must they carefully select what they eat?

Answer 12

energy rich low sodium deciduous forest leaves or sodium rich low energy aquatic vegetation surrounding lakes
- plants may be deficient in nutrients

Question 13

in the moose what can set a limit on the total food it can process daily?

Answer 13

stomach space- rumen constraint

Question 14

In Belovsky 1978 optimal foraging approach on moose what was found? (5)

Answer 14

• energy gained from either lots of aquatic vegetation or a little terrestrial
• cant just exist on aquatic plants
• only a small area satisfies all constraints
• constraints include energy, sodium and rumen
• diet lies trying to maximise energy intake

Question 15

what is the general economic model also known as?

- what does it predict?

Answer 15

Charnovs marginal value theorem

- how animals should forage in a patchy environment

Question 16

what do foraging environments tend to be?

Answer 16

patchy with food disturbed in clumps

Question 17

what are 2 examples of a patchy foraging environment?

Answer 17

• hummingbirds move from bush to bush to collect food

- swarm of krill is a patch for humpback whales

Question 18

what is the loading curve in MVT?

Answer 18

• diminishing returns of continuing to feed in a patch

- rate of food intake decreases over time as the patch may become depleted or the animal overloaded with food

Question 19

what is the optimal patch time in MVT and how does it vary?

Answer 19

• if didn’t take long to reach the patch there will be a steeper tangent to the curve
• longer travel time means animal should spend longer in the patch to maximise energy intake

Question 20

what happens if an animal leaves a patch before or after its optimal time?

Answer 20

the rate of energy intake is lower

Question 21

if you plot travel time against time spent in a patch what is the curve?

Answer 21

curve of diminishing return

Question 22

what are the 4 main assumptions of the MVT?

Answer 22

• travel time between patches is known (unless searching fo next flower in a sparse patch)
• travel costs = patch costs (true for hovering hummingbird but not seed eater )
• patch profitability is known
• no predation

Question 23

what did Kacelnik discover in 1984 about what starlings get as they forage and why?

Answer 23

diminishing returns
as harder to find food when carrying prey
- more it carries the less efficient it is at probing for more

Question 24

what did Kacelnik discover in 1984 about what starlings feed on and how does their optimum prey number to collect differ between taking a long or shorter travel time?

Answer 24

leatherjackets- daddy longlegs larvae

• longer = 7 prey
• shorter = 4/5

Question 25

what was Kacelnik 1984 set up for starlings?

Answer 25

trained to feed from mealworms from artificial patches with diminishing returns placed at 8-600m from nest - dropped mealworms successively slower to create diminishing returns

Question 26

what did Kacelnik discover in 1984 for how the starling load size changed in the experiment?

Answer 26

increased with increasing distance from the nest

Question 27

Cowie 1977 tested the MVT assumptions using great tits- what was done/found? (5)

Answer 27

- expend more energy during travel time 30 patches with 5 experimental trees - different spacing of trees varied travel time between patches - travel and foraging times closer than predicted by the model - differences in travel/patch costs influence foraging behaviour

Question 28

what was the Lima 1984 study on Downy woodpeckers to test MVT assumptions? (8)

Answer 28

- trained them to forage for seeds from logs each with 24 holes - logs were either empty or with seeds - birds sample different patches to work out profitability - 0 or 24 seeds 1.7 looks - 0 or 6 seeds 6.3 looks - 0 or 12 seeds 3.5 looks - observed means close to predicted - used information during foraging to maximise energy intake and minimise search costs

Question 29

what do optimality models used to study behaviour provide?

Answer 29

testable quantitative predictions | - so can test theories of an animal's behaviour

Question 30

what do optimality models and behaviour involve?

Answer 30

explicit assumptions - can alter model to learn more about subtle behavioural differences

Question 31

what do optimality models and behaviour illustrate?

Answer 31

the generality of decision making - can be applied to different species and not just based on foraging behaviour (e.g. how many eggs should be laid)

Question 32

what are the 3 responses when a model fails to predict observations?

Answer 32

- ignore it and count as error - accept animal is sub optimal - rebuild model

Question 33

how is the rate at which an animal gains energy quantified?

Answer 33

dividing caloric value by time taken to find food

Question 34

how may leafcutter ants appear to be acting non-optimally?

Answer 34

reduced daytime foraging by larger ants due to attacks from parasitic flies