7

Have studies been done that demonstrate the limits of human memory on reputation in the Iterated Prisoner's Dilemma?

i.e. at what point (on average) does a player start to "forget" the actions of a previous opponent?

Jason Nichols
  • 3,555
  • 1
  • 18
  • 38
  • Have you ever taken a course on human memory? I'm asking because "limits" is a term that someone who doesn't know anything would use. But that's not the major problem with your question. The question is actually about psychology not economic and so this is the wrong forum for it. "Cognative Sciences" is probably the best place for it. – JenSCDC Nov 20 '14 at 03:21
  • 1
    You mean like the Availability Heuristic? – Steve S Nov 23 '14 at 06:53
  • 2
    @AndyBlankertz: K-A-H-N-E-M-A-N. He won the Nobel Prize in Economics in 2002. And yet... get this... the guy's actually a Psychologist. In other words, "Cognative" [sic] Scientists actually do have a place on this forum. – Steve S Nov 23 '14 at 06:59
  • I agree with Steve, cognitive scientist do have a place here. As for "limits" (agree, not best word), the key word is "chunking". If one chunks (see A. Miller and his 1950's paper on the number seven) properly, there are no restriction on working memory. (Working is the key word, there are other kinds of memory, and the other kinds are in most senses more important, but some working memory is always required.) –  Nov 24 '14 at 09:54

1 Answers1

1

I'm not sure what you mean by risk pricing, but because you mention a repeated game, you might find the following papers useful. They present some theory about repeated games in which players have limited memory.

jmbejara
  • 9,345
  • 5
  • 30
  • 80
  • I meant risk pricing in terms of discounting the value or utility of a given outcome based on its likelihood. – Jason Nichols Nov 19 '14 at 18:00
  • Do you have a link or suggestion of where I might find a formal definition of "risk pricing?" – jmbejara Nov 19 '14 at 21:36
  • http://en.wikipedia.org/wiki/Expected_utility_hypothesis – Jason Nichols Nov 19 '14 at 21:39
  • Right, it's just that the term "risk pricing" is probably too broad. I'm not sure what it's referring to because it sounds like you could be talking about a stochastic discount factor or state price vector that discounts payoffs in different states by both the likelihood that that state occurs as well as other conditions. http://en.wikipedia.org/wiki/State_prices – jmbejara Nov 19 '14 at 21:44
  • Perhaps you want to know about how risk aversion might change (so that the risk premium might change)? – jmbejara Nov 19 '14 at 21:45
  • roughly. If you play the game repeatedly (against different opponents), how long until memory fades, and prices return to "normal" for a given opponent, and how would that curve look? – Jason Nichols Nov 19 '14 at 21:48
  • 1
    Let us continue this discussion in chat. – jmbejara Nov 19 '14 at 21:53