Newcomb's paradox

Discussion in 'General Philosophy' started by James R, Jan 25, 2017.

?

Would you choose to take Box B only, or to take both boxes A and B?

This poll will close on Jan 25, 2018 at 7:08 AM.
  1. Take Box B only.

  2. Take both boxes (A and B).

  3. I would try to randomise the decision process and let someone/something else decide for me.

  4. I can't decide how to choose, so I would refuse to play and go home with no money at all.

  5. I refuse on principle to play silly philsophical games, so I would go home with no money at all.

Results are only viewable after voting.
  1. Sarkus Hippomonstrosesquippedalo phobe Valued Senior Member

    Messages:
    7,410
    There isn't, there is correlation: both your choice now and the SIP's prediction were based on exactly the same things.
    The only possibility of you outsmarting the predictor would be if something affected your decision making in a manner unforeseen by the predictor - such as brain injury - between the the time the SIP made its choice and the time you make yours, such that it is basically a different person than the one the SIP made its prediction on. Or if there is some change in the set-up to the game itself.
    Otherwise the predictor is assumed to be always right, and as such choosing only box B is the optimal solution.
    They will want me to take both, but in this instance you are no longer assuming that the SIP is always right. I.e. it seems to be a material deviation from the original set-up, and introduces additional information to my decision making that was not available to the SIP.

    This basically seems to hinge on how strong one thinks the predictor's accuracy is.
    If it is assumed to always be correct then it will have foreseen everything that could possibly happen and thus choosing both boxes yields just USD 1,000 and choosing just one box yields USD 1,000,000.
    If it is assumed to be fallible then you introduce expected results.
    If you change the set-up of the game between when the SIP makes its prediction and when you make yours, then you change the argument, and possibly the outcome, entirely.

    But I will always choose the one box and be USD 1,000,000 better off.


    I also think the disparity of the amounts makes this less of a dilemma than it could be. Say it was 10k in box A and another 10k in box B... your only way to get more than 10k is to outsmart the predictor.
    But assuming the predictor is always right, if taking box B yields any higher amount than is in box A then it would seem that taking just box B is the way to go.
     
  2. Google AdSense Guest Advertisement



    to hide all adverts.
  3. C C Consular Corps - "the backbone of diplomacy" Valued Senior Member

    Messages:
    1,798
    I'll assume #7 is in conjunction with the suggestion of #6 that the SIP is not perfect in its predictions. I'd probably still choose box-B alone due to the SIP yet being correct 80% of the time. Which is to say, I'd still be carried by the momentum of what should be done when formerly confronted with SIP having a long background history of never failing. However, I'd be more vulnerable to last moment impulses or concerns ("Take the sure thing of a grand by selecting both A & B!").

    As noted above by the reference to still riding on the impetus of the past, the "me" of this here and now that is outside of these imaginary situations has been influenced / tampered-with beforehand by the intimidating scenario of SIP being perfect in its predictions (or appearing so). That scenario then being withdrawn and replaced by a slightly fallible version of SIP. As a result, I can't fully or confidently assert what a "pristine me" that had not been formerly influenced by relationship to a scenario with a perfect SIP would actually do if / when encountering only the one with a fallible SIP.
     
  4. Google AdSense Guest Advertisement



    to hide all adverts.
  5. rpenner Fully Wired Staff Member

    Messages:
    4,833
    The payoff matrix of this description of the game seems to be:
    \(\begin{array}{c|c|c} \backslash & \textrm{SIP fills box B} & \textrm{SIP doesn't fill box B} \\ \hline \\ \textrm{take box B} & 10^6 & 0 \\ \hline \\ \textrm{take both} & 1.001 \times 10^6 & 10^3 \end{array}\)

    If this is the game, then the strategy of taking both boxes dominates the strategy of taking just box B. But, crucially, it ignores that SIP is claimed to both predict your choice and hide the million dollars in box B only if you would choose to take just box B. Since ignoring that it tantanmount to believing the illusion of free will is stronger than machine prediction, we can assign a dollar value to reinforcing your belief in free will, representing a diminishing in the SIP reputation and a gain in your belief in human uniqueness.
    \(\begin{array}{c|c|c} \backslash & \textrm{SIP fills box B} & \textrm{SIP doesn't fill box B} \\ \hline \\ \textrm{take box B} & 10^6 & f \\ \hline \\ \textrm{take both} & 1.001 \times 10^6 + f & 10^3 \end{array}\)
    So if \(10^3 < f < 10^6\), then the worst outcome of the strategy of taking just box B is better than the worst outcome of taking both boxes.
    For \( 10^6 < f \) this analysis breaks down as the SIP no longer can force our worst outcome by always picking to not fill box B.
     
  6. Google AdSense Guest Advertisement



    to hide all adverts.
  7. iceaura Valued Senior Member

    Messages:
    24,066
    That accounting or reasoning holds only under the assumption that true free will decisions are unpredictable - that the SIP is therefore making its predictions based on a reductionist or Newtonian deterministic calculation, in which the will involved is an inconsequential epiphenomenon ("illusion").

    Human beings can predict each others decisions, some of them with startling accuracy, without making such calculations, and without dropping human uniqueness (individual or collective), or the role and freedom of will, from the description of the situation.
     
    Last edited: Jan 27, 2017
  8. iceaura Valued Senior Member

    Messages:
    24,066
    The assumption of fallibility in the SIP may (depending on how it's formulated, which isn't specified exactly above) include the possibility that it has mistakenly left B empty, as well as mistakenly filling it. That would have a very significant effect on the decision, for most people.

    The possibility of coming away with nothing is a disproportionately powerful influence on most people, and a source of much "irrationality" in human decisionmaking. Whether that is actually a mistake, something a human "should" handle differently, is another wide field of discussion entirely.

    You seem to have considered that, but I'm not sure - just emphasizing, in case.
     
    Last edited: Jan 27, 2017
    C C likes this.
  9. wellwisher Banned Banned

    Messages:
    5,160
    The underlying motivation for playing any game is winning the game. In the example used, the path toward victory is set in advance, and is very obvious. There appears to be not choice, since there is only one path to victory, which can be inferred.

    Say we modify the game setting, and make the game, more complicated and less obvious, like being a defenseman defending in hockey. The offensive player, with the puck, is not making the path to victory easy for the defenseman. He is skating at full speed, faking left and right, with his body, eyes and puck. He also has two fellow skaters, one on each side of him. He can shoot or pass.The goal of the defenseman is to win this challenge, by breaking up the play. However, is not clear or obvious, the proper path to victory.

    He will not always choose wisely. However each choice will help him learn the best path to victory. If gets beaten to the left, one time, the next time he may go right. The offensive player knows this, and will not just repeat the same thing. The defensive player, also knows this. He has to not commit himself, to either side or the middle, but stay very loose so he is never off balance, when is reacts to any side in real time. Free choice appears.
     
  10. Sarkus Hippomonstrosesquippedalo phobe Valued Senior Member

    Messages:
    7,410
    Wellwisher, any chance, just once, you actually answer the question posed without venturing off onto a barely relevant (if at all) tangent? To be honest, I was somewhat surprised you managed an entire response without mentioning consciousness or the two centres of the brain, or even the logical v artistic hemispheres of the brain.
    Small steps, eh.
     
  11. river Valued Senior Member

    Messages:
    9,182
    Change your personality analysis midstream . You were told .
     
  12. wellwisher Banned Banned

    Messages:
    5,160
    If I played the SIP game, I would choose box B. The reason is, although there are two ways to win, choosing B looks more impressive. That choice, to the untrained yes, shows me willing to take a risk. This is better for the audience, who will marvel at my winning using a care to the wind attitude.

    The reason I can extrapolate the game, beyond the choice, all the way to the needs of the audience, is the SIP is telling me in advance the two solutions to the problem. The result is people go into the game with insider information, telling them what to pick. It would be like me saying, when you go into the boss's office, make sure you smile, or else he will be grumpy. Even if the boss is always in a good mood, and does not care if you smile or not, I could make people smile, since this is considered the optimized path, based on my insider information. The term super intelligent predictor, SIP, gives the source of the insider information, prestige; subjective enhancement. If you called it the dim witted predictor, lowering the prestige, you would get more people willing to deviate. If you tried the same game without the SIP, the results will also not be the same, since without the credible insider information, the choices will randomize more among players.

    In my last post, I did not detour, but showed another scenario, where you don't know the outcome in advance. I did a scenario where the SIP is sleeping. In that scenario, one is dealing with other humans who will not play in a predefined way. The defenseman will go into the game, without infallible scouting reports, implicit of a SIP. The offense will randomize their attack, to make it harder on the defenseman. The defenseman has no insider information, leading to more open approach.

    Political propaganda appears to work the same way, where the SIP come from the leaders or media personalities, who are assume do have superior intelligence and insider information. This can cause the base to behave in predictable way, making their choices, based what their SIP predefines for them.

    If you know how the human mind and brain works then this is not hard to figure out. Many people don't want to know how the mind works because this results in more responsibility, since your choices are not pre-made for you based on insider information and prestige. I randomize more than most, because I don't aways accept the SIP, if there is data and/or logic that shows it is fallible. The result is I will pick others options, until I figure out my own version of SIP. Then with that set, options coming to me quickly, since these are defined by my own version of insider information I trust, due to lots of thinking and trying.
     
    Last edited: Jan 29, 2017
  13. Sarkus Hippomonstrosesquippedalo phobe Valued Senior Member

    Messages:
    7,410
    You would choose box B because it "looks more impressive" to do so? Wow. Just... wow.
    There is also just one way to win (unless you call USD 1,000 "winning" when you could have had USD 1,000,000. And it has nothing to do with what "looks more impressive" per se, but simply with whether you would choose both boxes or just box B, and the ability of the SIP to predict accurately.
    The needs of the audience are irrelevant, other than being what might, among other influ noes, drive you to picking one or both boxes. The audience is no more relevant than any other personal justification. The only relevant thing is that the SIP will know/accurately predict what your personal justification will lead to.
    If you provide another scenario then you are, by definition, going on a detour. Did you go into exams and answer questions with "well, I don't know the answer to that question, but if you asked me about a different scenario then the answer would have been...."?
    Eh? The SIP is simply something that accurately predicts what you are going to choose. Propaganda would be what informs your choice that the SIP will then predict. So it works nothing like you say.
    ??? So basically you can't stick to what is being discussed, but instead throw out random nonsensical noise in an attempt to make sense of it? And yes, if one knows how then mind and brain works then it wouldn't be hard to figure out. But you don't know those things. You might think you do, you might have read some out of date notions of how the brain is split into hemispheres, or how there is an animal, lizard and other parts of the brain etc. But you're just a mess when it comes to making sense of it or applying it. Although the frequency with which you seemingly try to apply it to any discussion means that some of it might even be relevant.
    Just a shame about the the other 90% of the time.
     
  14. Tralay Registered Member

    Messages:
    86
    This is an absolutely incredible insight into the thought process of the people on this site. Some get it right away, some over think it, some miss the whole subject matter, some just don't get it. Very interesting, and gives me an insight into whether or not these individuals would be able to truly handle a truly complicated subject or not.
     
  15. Tralay Registered Member

    Messages:
    86
    by the way, i chose box b only, if he is always right then he guessed that i would pick box b only and therefore would have put the million dollars in there. The question wasn't about whether it was possible or not, the op gave you all of the factors to the particular equation that was being formulated, it wasn't based on reality, only on the given factors.
     
  16. Michael 345 Valued Senior Member

    Messages:
    3,217
    B

    It's $1000 or $1001000 in picking A and B

    B only is the better option

    If B = nothing is break even

    Came with nothing go home with nothing

    If B = million go home million

    Of course if you KNOW SIP has a perfect record you roll with sip

    Pick B

    Please Register or Log in to view the hidden image!

     
    Last edited: Jul 25, 2017
  17. Yazata Valued Senior Member

    Messages:
    4,619
    I worry that this is like the Monty Hall problem, and that the correct solution is counter-intuitive and not at all what one would initially suspect.

    But overcoming that, I think that I'd choose both A and B, for much the same reason that JamesR gave.

    That's assuming that we don't know that SIP's prediction before we choose. If we did, then our choice would be trivial, since we would already know how the machine loaded the boxes.

    If we don't know what the machine predicted, then our choice would depend on how well we trust that the predictor does know what we are going to do.

    I don't trust it, so I would go with the choice that would guarantee me a payout of at least 1,000. Choosing A&B eliminates the possibility of a 0 payout. If the machine is wrong and predicts that I choose B, then I win 1,001,000.

    If one does trust it, then it would probably be smartest to go with B, since the predictor would know that you would do that and would have loaded a 1 million in the opaque box. So as the opposing argument goes, we shouldn't blind ourselves by trying to avoid 0 so much that we settle for 1000, a very suboptimal payout. But again, that depends on our trusting the predictor. If we are wrong about that, we might end up with nothing.

    Prediction............... Choice.........Win
    A&B........................ A&B.............1,000
    A&B........................B...................0
    B.............................A&B..............1,001,000
    B.............................B...................1,000,000

    I guess that the apparent 'paradox' here is the conflict between an abstract optimal winning strategy based on the SIP premise on one hand, and an optimal winning strategy in real life.
     
    Last edited: Jul 28, 2017
  18. Tralay Registered Member

    Messages:
    86
    the sip is never wrong Yazata, that was part of the deal. Therefore, if you pick B, there will be a million dollars waiting for you, it's not a hard one to figure unless you try to overthink it.
     
  19. Yazata Valued Senior Member

    Messages:
    4,619
    Neither is the Bible. Its inerrancy is part of the fundamentalist deal.

    If we remain strictly within the rules of the game, no matter how much they differ from real life, I agree (and said above) that choosing B every time would be the best strategy. That's because the infallible predictor would know beforehand that I would choose B and would put a million dollars in box B every time we played.

    But if we embed the game within the broader context of what's possible in real life, then choosing A&B would seem to me to be the best strategy, since it would guarantee that we come away with at least a thousand dollars even if the predictor put nothing in box B. So the best real life choice would seem to be A&B, provided that we don't believe that the predictor can accurately predict the future and that our subsequent choices exert no control over its earlier actions.

    I think that's why this is called a "paradox", despite it not literally being a logical paradox.

    The question in the initial poll asks which option I would personally choose. I live in real life and that fact shapes my decisions.
     
    Last edited: Jul 28, 2017
  20. Tralay Registered Member

    Messages:
    86
    But that's not what this exercise was about. It was just a simple follow the rules of the exercise, that's it.
     
  21. Yazata Valued Senior Member

    Messages:
    4,619
    Then why is it so often referred to as a "paradox"? What accounts for the paradoxical aspect that many very smart people perceive?
     
  22. Tralay Registered Member

    Messages:
    86
    because it's not that advanced of a paradox. It is a quite simple one, one not worthy of this site actually because everyone here is trying to stretch it into something that it just isn't.
    A real paradox is one that actually calls for more advanced thinking....this was more like a little jig saw puzzle.
     
  23. Tralay Registered Member

    Messages:
    86
    wikpedia: In philosophy and mathematics, Newcomb's paradox, also referred to as Newcomb's problem, is a thought experiment involving a game between two players, one of whom purports to be able to predict the future. Whether the problem actually is a paradox is disputed.
     

Share This Page