Read the tutorial reading and guidelines before tutorial

1
Read the tutorial reading and guidelines before
tutorial Available on web (http//myweb.dal.ca/agh
orn/biol3062/) TUTORIAL ASSIGNMENTS
STUDENTS DAY ROOM Abdalla -
Beaton Wed. LSC 332 Beauchamp -
Douglas-Farrell Fri. LSC 332 D'Souza -
Griffiths Wed. LSC 4258 Grondin -
Lilly Fri. LSC 4258 Ljungqvist -
Oldford-MacLellan Wed. LSC 5263 Onions -
Shams Fri. LSC 5263 Shears -
Tower Wed. LSC 4263 Tupper-Ring -
Zeggil Fri. LSC 4263
2
OPTIMALITY MODELS
Balance costs and benefits
Lay out logic in a transparent way
Make quantitative (precise) predictions tests
Two general approaches optimality models for
one individual game theory models for gt 1
individual
3
How to do an optimality model
1. Define the behaviour as a simple decision
What size mussel should be preferred?
2. Define constraints
Benefit and cost curves
3. Choose currency
Profitability (energy gained / time)
4
How to do an optimality model
1. Define the behaviour as a simple decision
How close should a tern dive at a fox?
2. Define constraints
Benefit and cost curves
3. Choose currency
Fitness
5
Fitness -gt
Optimum
Benefit
Cost
Distance to predator -gt
6
ESS Evolutionarily Stable Strategy
Found using game theory
Optimum behaviour depends on what other animals
are doing
Explains many situations where theres a balance
of different behaviours
7
Cautious
Relative fitness -gt
ESS
Daring
All daring
All cautious
Proportion of cautious mobbers in population -gt
8
Evolutionarily stable strategy (ESS)
A strategy (behaviour or mix of behaviours) that,
when adopted by most members of the population,
cannot be beaten by any other strategy in the
game.
9
Components of an ESS model
1. Players
2. Strategies
3. Pay-offs
4. ESS -- the strategy that cant be invaded by
any other (i.e. it will persist)
10
Pay-off matrix
Reactor
D
L
W
Actor
W
D
L
L
W
D
11
tough -gt big territory
female colour -gt sneak matings
small territory -gt well defended
12
How to find the ESS
1. Calculate pay-offs of each strategy
2. Set pay-offs equal or walk through logic
of pay-off matrix
3. ESS may be a) proportion of individuals
doing behaviour b) proportion of times
individuals do behaviour
13
PROS CONS OF OPTIMALITY MODELS

• PRO

Explicit, Testable, Precise
General
CON
Hard to tell if model is wrong

• - test might be insensitive
• - model might just need tweaking
• fit might be good enough
• - or approach might be wrong