2009 Week 3 Social
Using examples from your supplements and the chapters (Krebs & Davies 1993) about living in groups, fighting/assessing and sexual competition,….
Using examples from your supplements and the chapters (Krebs & Davies 1993) about living in groups, fighting/assessing and sexual competition,….
June 14th, 2009 at 1:40 pm
….how have the models (hypotheses) tested by behavioral ecologists progressed from relatively simple assumptions about a static environment (e.g. optimality models) to address variations in costs and benefits depending on the social/physical environment (e.g. game theory)?
How is the logic of natural selection an underlying thread that is consistent throughout all these models, even as the assumptions about a static environment are challenged?
June 15th, 2009 at 11:36 am
Hi, we were chatting today about how the changing conditions addressed by optimality theory models in Chapt. 6 about Groups, serve as foreshadowing for the game theory models in Chapt 7 about Fighting. If you would like to read more about game theory….
http://en.wikipedia.org/wiki/Game_theory
June 18th, 2009 at 7:12 am
The Game Theory Models approach adds another layer of complexity to the Optimality Models. Cost and benefits are dependent on the relative frequency of others in the population and the fitness of the genotypes is the sum of the pairwise encounters within a population. The similarities between Optimality and Game Theory Models are that both models assume several constants: stable environments, fixed payoffs, static strategies, and complete heritability of variation.
The optimality model discussed by Krebs and Davis on page 48 refers to the optimal load size of foraging starlings. The model is a simple theoretical prediction of optimum foraging decisions of starlings. The models utilize the time and energy as currencies to predict the cost and benefits of starling feeding (Krebs & Davis 48).
The Game Theory Models of evolution are discussed in Krebs and Davis on page 151 using Hawk and Dove strategies as examples. Hawks always fight to kill and injure opponents; however, doves always display but never engage in serious fights (Krebs & Davis 152). These two extremes help to discuss evolutionary stability of displays. Table 7.2 of Krebs and Davis displays the matrix of average pay-offs between the two species. When Hawks meet Hawk, it is assumed on half the occasions it wins and on half the occasions it is injured. Hawks always beat Doves, and Doves always retreat against Hawks. Therefore when a Dove meets a Dove, there is always a display and it wins of half the occasions. Neither the Dove nor Hawk strategy would be an ESS; however, a mixture of Hawks and Doves would be stable (Krebs & Davis 152).
The confining assumption of Optimality models are that the models do not consider the conditions of the environment. The more complex Game Theory Model allows for changes in the environment. Conditions will change as a result of how many Hawk and Dove genotypes are in the gene pool. However, both models still assume genotype heritability in variation of behavior traits even as the assumptions about a static environment are challenged.
June 18th, 2009 at 8:09 am
Kate, thanks for such a clear explanation of two very difficult concepts! One point I thought we might want to discuss more is the assumption that the environment does not change. That might seem to contradict the statement that the Game Theory models “allow for change in the environment”.
To clarify, I like to point out the distinction between the SOCIAL environment (i.e. other strategies and tactics in a population) and the PHYSICAL environment (i.e. food, shelter, predators, cover, temperature, etc.). The Game Theory models explicitly model changes in the SOCIAL environment, assuming that the fitness payoffs remain the same. The fitness payoffs theoretically reflect the influence of the PHYSICAL environment on how many copies of a genotype are made within the lifetimes of individuals.
Has anyone else been puzzling over this?
June 18th, 2009 at 1:26 pm
We began this course with relatively simple assumptions based on a static environment. By evaluating the function of the behavior we considered the optimum balance between the cost and the benefits of behaviors. This week we progressed to using hypotheses based on a scale of social costs and benefits. An example from chapter seven is the hypothesis of the conditional strategy of assessing the strength of competitors involving sequences to determine the relative strength of the competitor. This hypothesis looks at more than what the optimality model could measure. The issue at this point is more than to fight or not fight, but also to consider the actions of the competitor. The example provided from Krebs and Davies (1996) was the red deer stag. The cost of fighting for the red deer is normally very high. However, the sequence that the red deer go through to assess the strength of their competitor determines how high that cost truly is. If the competitors are equally matched throughout the process of roaring at each other, then they continue to assess each other’s strength by conducting a “parallel walk”. If the competitors are unevenly matched, one competitor will likely retreat as the cost appears to be too high. There are the rare circumstances that the contest reaches the stage of a fight. It becomes necessary in this circumstance to examine the social interactions between individuals to determine the behavior.
Natural selection remains an underlying thread that is consistent throughout all the models that we examined. This week for example, in the particular case of the red deer stags, if there was a genotype that did not assess the strength of their competitors and chose the extreme “Hawk” strategies, the fitness would be reduced. The amount of fatal wounds a stag would receive as a young stag would reduce the amount of recruitment of that genotype. The conditional strategies performed by the alternative genotype would increase the differential fitness.
June 19th, 2009 at 1:12 pm
The website referenced earlier by Dr. Packard did help clarify things a bit for me because game theory is a concept that I find very interesting but have a little bit of trouble understanding. One explanation used on the Wikipedia page refers to game theory as an attempt “to mathematically capture behavior in strategic situations”. To highlight the difference between game theory and optimality modeling, I believe the “strategic situations” could be further categorized as strategic social situations. That, to me, is the key between optimality modeling and game theory: while one predicts trade offs between costs and benefits to maximize net benefit for an individual (Krebs & Davies, 1993), the other is a more dynamic theory that is influenced not only by the physical environment but more importantly by the social environment. Game theory takes into account what strategies are available for use as well as the frequency of those strategies occurring throughout the population.
That’s not to say that optimality models are obsolete and should be discontinued. These types of models still provide valuable insight into behavior and are a quantitative way of testing hypotheses. The point is that each technique, whether game theory or optimality modeling, offers unique perspective into animal behavior, adaptation, and genotypic frequency within a population. By using both methods, a more complete understanding of animal behavior can be obtained. For example, one optimality model could predict the age and size at which it is most beneficial for a subordinate male meerkat to leave his natal group and join another. The game theory models could be used to take into consideration the strategy of other male subadults in the same litter as well as in other litters. The two approaches together would give a better representation of all the various forces at work in the group: intrasexual competition, predation pressure, altruism, and group size (both subadults and pups) would be a few. From these models then, a more accurate depiction of genotypic frequencies within the population could be reached, thereby furthering our understanding of the process of natural selection in meerkats.
June 19th, 2009 at 1:18 pm
Two ideas stood out for me this week. 1: The concept of continuously drilling down into the basis of a behavior until the function and mechanism can be addressed. 2: The covariance concept that the genotype for sexual selection behavior has two genes, one for the female’s preference of the trait and one for the trait (Krebs and Davies 1993,p193). The first concept was demonstrated in the barn swallow, where the researchers first answered the question of do females prefer to mate with males with longer tail feathers (Krebs and Davies 1993,p198). After they concluded yes, they proceeded to answer the question of why. A correlation between tail feather link and parasite resistance may not be an intuitive relationship, but it is a very important in order to better understand the complexity of animal behavior. An example of the covariance concept of genotypes may be the white wing patch size in female ducks which is positively correlated to increased levels of male parental care (Hegyi et al. 2008).
Hegyi, G., Garamszegi, L., and Eens, M. 2008. The roles of ecological factors and sexual selection in the evolution of white wing patches in ducks. Behavioral Ecology 19: 1208-1216
June 19th, 2009 at 3:38 pm
In the third week of this course the focus was on the social aspects incorporating categories like living in groups, fighting, and sexual competition. I found the fighting and assessment aspect to be very interesting. The Rohwer (1978) experiment of Harris sparrows demonstrated the uses of displays and there effects on fighting. In the sparrows the dominate male has darker plumage which reduce the fighting allowing for an increased share of the food supplies. The study concluded that the badge of dark plumage does not result in dominance it must be associated with an aggressive behavior. I also enjoyed the discussion on the Hawk-Dove game. The Hawk-Dove game (H&G) is an evolutionary game theory, which addresses the situation where there is a competition for resources. The H&G places numerical point values on measures of fitness, in order to simplify the payoffs.
June 19th, 2009 at 3:44 pm
Sorry forgot citation
Rohwer, S. & Rohwer, F.C. 1978. Status signaling in Harris sparrows: experimental deceptions achieved. Anim. Behav. 26 1012-22
June 21st, 2009 at 2:14 pm
The behavior models such as Optimality Theory and Game Theory models have progressed from simple foraging decisions (such as Starlings and leatherjackets) to the Hawk:Dove Game Theory of fighting and displays. The Optimality models studied in Chapter 3, Economic Decisions can best be summarized as relatively simple decisions that a species makes to optimize decisions and associated currencies in a stable environment. In these models (as defined by Krebs & Davies on p. 75) benefits are maximized, constraints of the animal are defined, and a currency for measuring success are established. These Optimality models are easy quantified and predictions can be made to determine possible natural selections decisions.
In contrast, Game Theory models, a species’ success and/or failure can be impacted by the choices of others (changes in the social environment). And they impact fitness, which makes choices crucial (Physical environment). Out of my supplemental materials (Childress & Lung 2003), I discovered that female elk with calves (as part of a herd) spend a much higher percentage of their time scanning for predators than do females w/o calves. Their primary concern becomes protecting the calves (Parental Effort), even at the expense of replenishing their own energy reserves; fitness in ensuring the growth and survival of the calves being their primary goal. In stark contrast, male elk continued to feed even while predators where just meters away. Their primary goal is to replenish depleted energy stores in preparation for the fall rutting season (Mating Effort) vice scanning for the herd; fitness in mating being of top concern. Thus Game Theory Model can analyze animal communication, breeding strategies, and territorial defenses.
Childress, M.J. & Lung, M.A. 2003. Predation risk, gender and the group size effect: does elk vigilance depend upon the behaviour of conspecifics? Animal Behaviour. 66, 389-398.
June 22nd, 2009 at 5:00 am
Through out the third week, we discussed fighting, sexual competition and living in groups. Through these discussions i found the assessing aspect and fighting within species very interesting. In the red deer stag example each male competitor begins assessing each other by roaring. If the two competitors are matched in roaring strenght then the two males progress to parallel walking. this allows the two males to get a closer look at each other, their size, possible fighting ability and antlers. the assessment process continues unless one male retreats if they feel out matched. The cost of fighting is very high. if a male is injured they will be less likely to be able to over take a harem of females in order to breed. In most cases, one male will retreat during one part of the assessment phase. If one does not retreat then the assessment phase will progress in to a fight of pushing and shoving with head butts and wrestling. The winner then has control over a group/harem of females to breed with multiple females and pass his genes through the gene pool.
The game theory is a more complex model than the optimality model, where the resources influence what the species potental. both models fail to represent changes within an environment but the game theory shows the cost and benefits in the social enviornment and the chance of heritable traits being passed on.