Talk:Ultimatum game

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Game theory (Rated B-class, High-importance)
WikiProject icon This article is part of WikiProject Game theory, an attempt to improve, grow, and standardize Wikipedia's articles related to Game theory. We need your help!
Join in | Fix a red link | Add content | Weigh in
B-Class article B  This article has been rated as B-Class on the quality scale.
 High  This article has been rated as High-importance on the importance scale.
 
WikiProject Economics (Rated B-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Economics, a collaborative effort to improve the coverage of Economics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
 

An anonymous user changed a single 'her' to 'his'. Since there are two players I used both sexes in order to maintain clarity across the example. Since the anonymous user only changed a single instance, the sex of the first player changes throughout the article (see the example in the equilibrium analysis section). I have reverted it, not solely for a political reason, but in order to maintain consistency. I don't think the sex should be uniformly changed for two reasons: 1. it helps to distinguish between the first and second player while also using pronouns and 2. it helps to counter a sexist linguistic habit without introducing awkward phrases like "them". Thanks. --Kzollman 02:22, Mar 24, 2005 (UTC)

I agree with the above, but feel compelled to ask you to rethink you're "factual inaccuracy" claim:

  1. Has there every been an experiment in the Ultimatum game where the endowment was greater than 20 years salary, even in local currency?
  2. Do you believe that experimental economics has a comparable budget to experimental physics?
  3. Was there some other point I made in the sentences you deleted which could be described as factually inaccurate, or even questionable?

I believe the answer to all of those question is no, and that you therefore have no "factual" basis for removing my remarks, although you could very well do so on the grounds of "being too colourfull".

I'm very sorry to be so precise on this point, but you can image how Wikipedians don't appreciate being called "factually inaccurate" when that is clearly not the case. Wragge 00:50, 2005 May 17 (UTC)

This is probably not the best place to raise this but I just cannot understand your position on gender pronouns. How is using a singular they 'awkward' yet randomly hopping between masculine and feminine somehow not so? Why use 'her' or 'him' at all when you've got a perfectly good gender-neutral pronoun to play with? I can kind of appreciate your attempt to redress the balance of people using 'him' too much but I think it's misguided, I mean if using 'her' caught on you'd only get a counterreaction that would switch back to using 'him'. The only way to settle the problem is to move towards a middle ground and get used to using 'they'. I'm not going to edit the article though because you obviously believe in this and though I think it looks really weird I believe in diversity! Trent 900 00:57, 21 May 2006 (UTC)
It's worth noting that designating one player as one gender and the other player as another, then referring to them by their respective gender-specific pronouns, makes the page smaller while carrying acrost the same amount of information in a simpler format. I don't particularly see any reason to use a gender-neutral pronoun and then call out first and second players as first and second players over and over again when you can just use "him" and "her". Just seems awkward to me. 216.82.142.13 (talk) 23:39, 6 October 2009 (UTC)

High stakes ultimatum game[edit]

The above makes my objection sound more like personnel offense rather than an attempt to write something important and interesting (I was offended, but that's not important now).

What happened was that I wrote that no very high stakes Ultimatum game had ever been played (I mentioned the figures of 1 million and 9 million dollars, which has been left in place). Another editor took the view that 1 - 3 days salary is comparable to a million dollars, and that this experiment actually has taken place (he therefore deleted my claim that it had not). I must admit this made me slightly angry, but it also made me think:

  1. Possibly, there has been proof that beyond a certain threshold people's responses don't change with absolute amounts (I don't think there has, but I'd be interested to see it.)
  2. Has there (at any time) been an ultimatum game played for stakes beyond, say one thousand dollars? I don't think there has in the laboratory, but I remember watching a television program on British television (Channel 4) in which the climax was very similar to an Ultimatum or trust game, and this was played for several thousand pounds.
  3. In terms of specific one-off situations which were similar to the ultimatum game I can think of extortion cases revolving around the the bad-beat jackpot rule in poker cardrooms which are similar: The bad-beat jackpot specifies that the loser of an exceptionally unlucky hand receives a very large prize, accumulated over several months of play (this can be over 100,000 USD). The winner (with an even stronger hand) gets a much smaller prize, say 10,000 USD. There have been cases where the winning player has threatened to fold his hand on the end (forfeiting the prize for both) unless the "loser" (who would win the lions share of the money) agrees to split the prize with him. So far, the loser has given in rather than sacrific $45,000+. Can anyone top that?
  4. Perhaps the other editor is the CEO of a multinational company, in this case I can see why he could compare three days salary to a million dollars (that might even require a cut in pay).

Am I completely wrong about this? Was I wrong to bring up the issue of high stakes Ultimatum game play?

Wragge 03:15, 2005 May 17 (UTC)

Wragge - I was the user who changed your entry. First let me thank you for adding to the entry. Your addition made me realize that I had mistakenly not mentioned the fact that behavior changes when the stakes are raised. As a result of your comment this omission has been fixed. I am deeply sorry if my comment that the addition was "factually inaccurate" offended you. It was not my intention, and if I could change that comment I would. As an explanation (not a defense) I thought that your original sentence suggested that no one had studied the effect of increasing the size of the amount, I wanted to be sure that no one left with that impression.
With respect to your assertion above. No experimental economist that I am aware of has suggested that there would be a cut off amount above which people's behavior would change. The general belief is that as stakes increase behavior approaches the subgame perfect equilibrium. But no experiment, that I am aware of, has shown that all players play the subgame perfect equilibrium after a certain amount. If I am in error, please reference the article. Otherwise I think your suggestion constitutes original research.
No I don't think that the worldly budget for experimental economics compares to that of experimental physics, although I'm unsure of its relevancy to the ultimatum game. Perhaps the entry on experimental economics?
Your comment about poker confuses me. Do you have any reference for that event? In all the casinos I have played (only in the US) any discussion of the bad beat jackpot would immeadiately void the jackpot and so no one could make that threat. But, I'd be interested in hearing about it. In fact, it does sound like something that would be worth discussion on this page as an actual example of a real world ultimatum game. If you have a reference for it, please add it to this page!
Again sorry for the conflict. --Kzollman 04:02, May 17, 2005 (UTC)
P.S. For others who are interested, this sentence was the one removed: Unfortunately, experimental economics doesn't have the same resources as experimental physics, and this theory has not yet been tested. --Kzollman 04:13, May 17, 2005 (UTC)


No problem, Kzollman, I was just a bit irritated at the time (as you could tell). Anyway, experimental physics has nothing to do with experimental economics, I just brought the comparison in as the biggest contrast I could think of for a difference in research 'endowments'.
It's not important to include the irrelevant physics-allusion in the article, but I still think it would be encyclopedic and interesting to mention the theory of high absolute-amounts producing different results, and stating the largest cash amounts ever tested with.
My case-history of a bad-beat jackpot was something I heard on the rec.gambling.poker channel about two years ago. Although deja-news would have this it's hardly worth looking up, given the low signal/noise ratio there. I always thought it was an unlikely story at the time (though I have no evidence it was untrue, and the discussion was fairly lengthy about the supposed incident). Anyway, it wasn't exactly the same as the ultimatum game. On the other hand, it was for pretty high stakes.
One comparison to (tournament) poker might be relevant - the tendency of players to agree on a split in advance when the stakes get too high - controlling for this effect could be difficult as higher stakes increase the temptation to corrupt the test.
It seems most likely that the largest amount ever publically put on the line will have been on a gameshow somewhere, and we should have a look into this because it probably gives the highest upper bounds on the endowment used in an experiment. Sorry I can't do this now - have to work sometime.
There must be comparable equivalents in the business world, which would have stakes over a million dollars. I can't think of any but some types of one-off contract negotiation might present the case history here.
Would researching the above commit the sin of originality? If so, I suggest we state the upper limit (in days of labour/PPP-terms) ever invested in an experimental setting. I don't think it's all that much (the highest figure I've heard is three day's pay - and for me, that really isn't a lot.)
Cheers, Wragge 07:30, 2005 May 17 (UTC)
I actually think that the game theory articles are woefully lacking for examples of the games in actual real world situations. This is something a friend and I are planning to start working on tomorrow, in fact. We are planing to start tracking down all the examples we can find of signaling games and the stag hunt in animal populations. Similar stuff for this game would be good. Also for the nash bargaining game (this is probably more analogous to the tournament splitting example). Anyway, if you can think of some, I think it would be a great addition. Glad there are no hard feelings. best, --Kzollman 07:09, May 18, 2005 (UTC)

Real-world examples[edit]

If I come across any good analogs in the real world I'll be sure to add them, but I'll also be looking out for more experiments in economics not listed on Wikipedia yet - there don't seem to be nearly enough interesting experiments from such a rich field. Anyway, thanks for your additions; they were in line with what I was planning to do.

I think one thing you have to be careful of in Wikipedia is that a claim is totally unsalvage-able before deleting it (some editors NEVER delete any claim from another user, even then). One reason is that many Wikipedians love fact s more than their own children, and take an accusation of factual error more seriously than aspersions against their parentage. (Hopefully I don't have as much of a problem as that but its all an interesting experiment in group dynamics.)

Good luck, .Wragge 09:16, 2005 May 18 (UTC)

Sarah's Results: Me 75 Lara 35 Yes Me 75 Greg 35 Yes Me 80 Jackie 20 Yes Me 80 Rish 20 No Me 80 Jess 20 No

Variants of the game[edit]

The first sentence of the article is mistaken; neither anonymity nor single-play are requirements of the game or part of the definition of the game. Many variants of the game have been staged, including face-to-face games and repeated-play games.

I guess that the further article contradicts this assumption (anonymity and single-play). In a single play the best strategy for first person of sharing x is x-1:1 and for second to accept(whatever share he gains bigger then 0). Uzytkownik (talk) 00:54, 28 September 2008 (UTC)

Moved addition[edit]

Ultimatum Game and the Power of Precommitment

If the second player can pre-commit (see entries on pre-commitment and on Thomas Schelling) to a minimum acceptable offer (and this is common knowledge), then the first player is in the weaker position. To some extent, strong emotions such as anger at unfairness or envy may serve as a pre-commitment device. This follows Robert Frank's arguments in "Passions within Reason". (Note : rejection of unfair offers has even been observed in certain monkey species, so it might be evolutionarily advantageous in social animals).

Ironically, if you were playing a *computer* (not a person) that was programmed to reject any offer of less than say 70 %, you might be more inclined to accept a low percentage than if you were playing a person.

Willingness to reject an offer is often related to the ability of the rejection to HARM the opponent. (There have been many game-variants that have examined this issue). For this reason, it may be beneficial for an organization to set up a bureaucracy that does not gain or lose directly from the outcomes of ultimatum games, but that is pre-committed by policy (or script) to only offering (or accepting) proposals that disproportionally favor its employer. When the behavior of rejecting an offer does not cause the proposer pain and may not even be noticed, it is less likely to persist.

The Ultimatum Game has been widely studied in the psychology of judgement and decision making. Many important contributions have been made in this area of research - including work by Kahneman, Thaler and Camerer. (Hopefully the author of the page can add more of this work to the article).

Sociological applications[edit]

I don't think the quote in this section is relevant as the source is very much coloured and certainly not impartial? The linked Foundation for Teaching Economics is a free market vehicle/think tank which received funds from the Scaife Family Foundation, the Charles G. Koch Charitable Foundation amongst others. It appears to push a very contraindicitive message about free market economics and international trade.

http://www.mediatransparency.org/recipientgrants.php?recipientID=1362

The quote from Prof. P.J. Hill of the private Evangelical Protestant Wheaton College and the attending link to the above foundation's 'explanatory' and rather one-sided 'educational' document brings it's validity into question. Perhaps it should be deleted or a proviso added?

I'm not sure what the concern over bias is here. The quote begins with "I see...", which can't be biased. He knows how he sees the world, and is free to comment on it. Also, your concern about bias is not allowed at wikipedia, because it constitutes original research. If, however, you have some third party sources which call the particular quote into question, please do add them to the article. --best, kevin [kzollman][talk] 23:12, 5 January 2007 (UTC)



I agree, and would add that the "explanations" section seems to suffer from the same problem. Should the "explanation" of an experiment be an attempt to reconcile the results with economic theory? That could go under the heading apologetics, maybe. 71.208.70.12 13:13, 12 April 2007 (UTC)matt

Intention to delist as GA[edit]

This article was listed as a GA, as far as I can tell, without any discussion or comment. I do not believe that it currently meets the criteria, so I intend to delist it unless we can improve it in the next couple of days. The lead is just a definition, and is inadequate as a summary of the article. The article is not as accessible as it could be, the references are difficult to extract, and the standard of citation is poor. I am concerned that the article may be viewed as original research, since it appears to include a synthesis of primary sources. I will try to address some of these concerns myself over the next couple of days, but I won't be able to fix it entirely myself. Geometry guy 21:45, 12 July 2007 (UTC)

GA review (see here for criteria)
  1. It is reasonably well written.
    a (prose): b (MoS):
  2. It is factually accurate and verifiable.
    a (references): b (citations to reliable sources): c (OR):
  3. It is broad in its coverage.
    a (major aspects): b (focused):
  4. It follows the neutral point of view policy.
    a (fair representation): b (all significant views):
  5. It is stable.
  6. It contains images, where possible, to illustrate the topic.
    a (tagged and captioned): b (lack of images does not in itself exclude GA): c (non-free images have fair use rationales):
  7. Overall:
    a Pass/Fail:


I was able to draw out the references, but wasn't able to improve any other aspects. Geometry guy 20:10, 14 July 2007 (UTC)

Er... huh?[edit]

Who puts up the money in the first place, and why? That person's motivation strikes me as potentially the most interesting aspect of the whole game, yet there's no mention of it. -88.111.83.69 08:45, 9 August 2007 (UTC)

The money is taken to be an externality. Some money that is stumbled upon by both parties. Situations exist in trades, for instance. If you have a widget that is worth $5 to you, and your widget is worth $10 to me, we have a $5 surplus. If you sell me your widget for $5, you have made no money (you traded an object worth $5 to you for $5), but I "make" $5 because I give you $5 and gain something worth $10. Alternatively if you sell it for $10, then you make $5 and I make nothing. You and I have stumbled on a $5 surplus and we must negotiate how to split it. The ultimatum game presents a circumstance where you state a price for your widget and I either buy it or leave. --best, kevin [kzollman][talk] 17:10, 9 August 2007 (UTC)

Possible source[edit]

Just stumbled across this, not sure if it will be useful to you guys or not:

Politizer talk/contribs 14:09, 28 January 2009 (UTC)

Probabilities[edit]

What is the probability of rejection as a function of x in a split x:(1-x)? Are these probabilities consistent when comparing the results of different experimenters, at least if the ratio of the amount of money at stake and the income of the sample population is the same? Icek (talk) 14:05, 12 February 2009 (UTC)

Testosterone and the ultimatum game[edit]

http://www.nature.com/news/2009/091208/full/news.2009.1131.html —Preceding unsigned comment added by ParasiteNetwork (talkcontribs) 20:54, 11 December 2009 (UTC)

I do not like this game[edit]

This game considers money to have an absolute value. It is not the case.

Two persons. Person A gets 10M, person B gets 1M. Then person B may have made an absolute gain of 1M, but they made a lost when compared to person A. You may call that status, buying power or whatever, it is still a lost that is not taken into account by the model. Plus, revenge by refusing to give money to an unfair a**hole is a huge gain worth at least 9M of fake money.

If I had a piece of bread and were in the desert with a starving person, I will give that person half the bread (or more if I'm alright and the person obviously needs more). Let's take a look at the possible motives:

Altruism: could I be altruistic because I feel it's the way to go. If I do not act altruistically, I feel wrong. Therefore, I get the priviledge to feel good because I was altruistic.

Status: Maybe I did it because of how people will perceive me once they know about my "altruism".

Future gain: Maybe I will get paid back somehow by that person (by Karma?).

Feeling good, a good status and future gain are all "gains". They may have no monetary value but they also have relative values.

All in all, I believe the ultimatum game to be mostly idiotic (at least in the described uses). —Preceding unsigned comment added by 64.229.248.232 (talk) 04:07, 28 February 2011 (UTC)

Looking for Superrationality & Ultimatum journal article.[edit]

Somebody should update this with a journal article on linking [superrationality] and this game; if one can be found. Playing with utility function can't possibly explain the way human mind works.

75.142.101.74 (talk) 06:57, 26 July 2011 (UTC)

Self esteem[edit]

It seems to me that willingness to accept a smaller share is very much tied up with the experimental pair's individual self-esteems. I posit that a person with high self-esteem is far less likely to take a small share offered by a person with low self-esteem than vice-versa. It would be interesting to set up an experiment with people of varying degrees of self esteem and testing high-high, high-low, low-high and low-low pairings to see what happened. -- Derek Ross | Talk 21:27, 17 January 2012 (UTC)

Bias in the article[edit]

Parts of the article are biased towards showing an innate desire for fairness among players in the ultimatum game. This is clear in the section on Experimental Results. First, it is generally untrue that in experiments "people offer "fair" (i.e., 50:50) splits". I have never seen this experimental result in players developped economies. The argument of fairness needs to be balanced by the fact that fair splits almost never occur. Second, an explanation for the deviations from the Nash result due to the small size of the sums involved cannot be discarded with obsolete experiments. The much better known Slonim and Roth (1998) result --that appeared at the same time as the not very well known articles cited in Wikipedia (¡from slides!)-- is not mentioned (Roth is the foremost researcher in the area). Finally, there is a recent paper by S. Andersen, S. Ertaç, U. Gneezy, M. Hoffman, and J. A. List in American Economic Review 2011 (3427-3439) that confirms the results of Slonim and Roth, with experiments over a much larger range of sums in play in the game by villagers in India. The range is from 20 to 20.000 rupees (a year's salary), and the results confirm that as the amounts increase, the fraction offered falls. — Preceding unsigned comment added by Fyg50 (talkcontribs) 01:20, 2 July 2013 (UTC)

References to previous note on Bias[edit]

Slonim and Roth (1998)[1]

S. Andersen, S. Ertaç, U. Gneezy, M. Hoffman, and J. A. List (2011)[2] — Preceding unsigned comment added by Fyg50 (talkcontribs) 01:34, 2 July 2013 (UTC)

Invalid reference on "industrialized countries..."[edit]

In industrialized[2] cultures,[which?] people offer "fair" (i.e., 50:50) splits, and offers of less than 20% are often rejected.[3]

The "industrialized cultures" part needs clarification. Reference 2 (Sanfey, Alan; Rilling, Aronson, Nystrom, Cohen (13). "The Neural Basis of Economic Decision-Making in the Ultimatum Game". Science 300 (5626): 1755–1758. check: [1]) does not even mention which countries are those. However, the add the following references to support the paragraph:

  • W. Guth, R. Schmittberger, B. Schwarze, J. Econ. Behav. Organ. 3, 376 (1982).
  • R. H. Thaler,J. Econ. Perspect. 2, 195 (1988).

GameEcon.Behav.10, 95 (1995).

  • G. E.Bolton,R.Zwicic,
  • A. E. Roth, in Handbookof ExperimentalEconomics, J. H. Kagel, A. E. Roth, Eds. (Princeton Univ. Press, Princeton,NJ, 1995).
  • See J. Henrichet al. [Am. Econ. Rev. 91, 73 (2001)] for Ultimatum Game research in simple societies.

If someone wants to clarify the section, digging on these sources and specifying the actual countries would be a good start. And the current reference should probably be removed.

-85.55.62.39 (talk) 05:51, 9 July 2013 (UTC)

  1. ^ http://www.jstor.org/discover/10.2307/2998575?uid=3737784&uid=2&uid=4&sid=21102421832591
  2. ^ http://www.aeaweb.org/articles.php?doi=10.1257/aer.101.7.3427