2010 Scraps

relevant to gig/todd heuristic ecological rationality: are conditionals frame shifts? I'm identifying counterfeits of different currencies from different countries: with fake $ I should look at feature A first, feature B second and C third. But with ¥ I should look at C first than A and B.

So, when classifying a mix of bills as fake or real, do I shift between the $ and ¥ frames and then do TTB (gig and Todd) or is denomination a Feature D that I am evaluating first? This is a case where the order of the cues varies with their nature. What is the damn prediction here?

style issue: think about relative strengths of climactic and anti-climatic forms of presentation. theory theory vs. simulation theory. I am aruging for theory theory and will present problems withsimulation thery. Do I lay out sim theory first t o introduces problems that I solve with theory theory? Or do I lay out a whole comprehensove account of theory theory and then show how sim theory is a inadequate? Well, whichever I addres first will get the most space, because that is where I'll be developing the concepts that damn sim theory. so if sim theory is dominant and needing to get knocked down but everyone buys it, I might take the climacic form. If theory theory needs a cohesive account, then i should do it first, and go back to sim theory to show that it is less powerful. APA recommends anticlimactic form (for clarity, not for reasons I lay out)

still thinking about modeling the sense of agency that people play in games. if i feel like i have agency over outcomes, than I am more likely to act in a way s consistent with pareto. other wise, i will play defensiely. factors such a model would take into account: number of bits of communication (allowing IPD to be more cooperative than PD) ratio of individual ampact to range of totaly outcomes. first is the difference in the net outcome for individual playing most pareto and most myopic move (that is really poorly defined). second is the worst and best outcome for all players playing pareto consistent or maximally shortsighted. i guess. need to pin some things down: when I say outcome, do I mean the individuals payoffs or group payoffs, or opponenet's payfofs? do i model group size explcitly or is it enough to do this ratio thing?

  fehr's intention bsaed based models are an important competitor to these ideas, and they account for  things like revenge.  i don't think this agency approach does.  but i don't like the idea of a kindness function and other things like that.

Lin says something relevant to this in the poteete/janssen/ostrom