Fate And Agency: Rod Humble On The Sims 3

Does he have more choice than you?

During EA’s E3 press conference last night there was one moment that was by far the most extraordinary of the evening. Rod Humble, the man in charge of the Sims, came out and began to lecture on free will. The corporate reason for this was the announcement that The Sims 3 is to come out on consoles this Autumn – not something we need to concern ourselves with, unless the changes and improvements of the console version are not also updated into the PC version. But that’s a concern for another time. Because as far as we’re concerned, this was a glorious moment of sunshine amongst the usual pomp and explosions of a gaming press junket. You can watch it below.

Humble’s entrance is magnificent. Rather than jogging out, pumped, and working up the crowd, he shuffles out, stands still, head bowed, and scratches his hair. Finally he speaks:

“Did we have any choice in being here today? Or was it pre-determined since the beginning of the universe that all of us would be here today, in this theatre, thinking these exact same thoughts? In short, do we have any free will at all? In the Western tradition, Greek myth and literature examines this subject to great depth. It’s a subject of fate and agency. And what I love about the Greek gods is they’re not smarter than us, and they’re not wiser, and they’re certainly not more emotionally balanced; they’re just more powerful.”

Silence from the audience.

Humble goes on to argue that in The Sims 3 they’ve created an artificial free will – an unpredictable pattern of behaviours that play out according to myriad circumstances. “Unexpected emergent results.” In fact, in mentioning the experiments of Benjamin Libet (that our unconscious brain appears to prepare to act before our conscious brain chooses to act), Humble postulates that Sims may in fact have more free will than we do. (There’s great stuff about this in RadioLab’s discussions from their Choice and Beyond Time episodes. The former is at the bottom of the post – the latter, directly relevant to Libet’s work, sadly is not embeddable.)

The consequence of all this is my realising I’ve never taken Sims 3 seriously enough, despite protestations from my colleagues. See if it works for you:

What a fantastic moment. I especially love the near complete lack of applause at the end, the audience presumably a little surprised by what happened. Humble is a hero.

Here’s the RadioLab clip about free will. For the Libet material, head here and skip to minute 38.

Big props to BuckSexington for tweeting the Humble clip.


  1. Entriech says:

    I believe when this occurred the best comment I heard was: “Looks like a TED talk accidentally broke out.”

  2. SAeN says:

    It was certainly one of the best moments of the conference. I should probably buy The Sims…

  3. mrmud says:

    Accepting that free will is largely an illusion created by the mind is probably one of the hardest things you can do.

    • AndrewC says:

      If it’s illusory it’s hardly our fault if we can’t accept it is, now is it.

    • Zetetic says:

      Confusing the issue of determinism and agency as being incompatible is apparently one of the easiest things to do, but that’s no excuse for you. Describing free will as ‘illusory’ is really very difficult to make any sense out of.

    • mrmud says:

      Not sure if you were refering to mine or AndrewC’s post but I was refering to how experiments have shown that in many instances we take an action then the mind will make a rationalisation for that action only after the fact.

    • Bret says:

      What Andrew said.

      If free isn’t a thing, then accepting it isn’t an achievement. It’s the end product of years of conditioning ect ect.

      If it does exist, then denying it makes you a dumbass.

      Either way, proclaiming that you know you have no free will is nothing to brag about.

    • Peter Radiator Full Pig says:

      There have been many refutations on the conclusions taken from the experiment.
      There will be a New Scientist article on it, those articles are very good at explaining what happened, and what interpretations are going on.

      The verdict is still out on the meaning of that experiment,

  4. Tei says:

    At the subatomic level, the nature seems random. The only way to manage this world is with statistic. A single particle can do something imposible, like… hum… be created from nothing. Our world, is just the accumulation of this randomness. So yes, theres free will, and “it pre-determined since the beginning of the universe that all of us would be here today, in this theatre?” a big NO.

    Is this randomness soo cool, that can “evaporate” a black hole.

    • Tei says:

      Also, probably Sims 3 is not using random numbers, but pseudorandom numbers so the ‘Sims’ can be predetermined.

      (Fun facts: The Dune 2 map was generated that way, using a random seed, to generate the ground/sand pattern. The level was stored as this single random seed (a integer). )

    • AndrewC says:

      The cat is either dead or not dead, Tei.

    • Zetetic says:

      Accepting that subatomic particles do act randomly (given that there has been considerable effort expended attempting to prove logically that there aren’t hidden variable at work; really the work here is largely beyond my comprehension.), I’m really not sure that has much to say about predetermination in the modal sense (i.e. the idea that things might have happened differently), only in the predictive sense.

    • Tei says:

      In these experiments where a single photon is launch to 2 holes, the photon create a interference image. The photon is crossing both holes and making a interference with himself. Both states are true, the one where the photon use the A hole, and the state where the photon use the B hole. So probably the cat is dead and live (if you don’t open the box).

    • AndrewC says:

      These quantum uncertainties come about because in between the moments when these elementary particles are interacting with another particle, we have no way of knowing what they are doing. They could be anywhere. The universe only knows itself by touch.

      A cat is made up on billions of these particles that are constantly touching each other. So *who* knows if the cat is dead? The *cat* does.

      The point of the experiment was to mock the idea of quantum uncertainty on the human scale.

    • Zetetic says:

      Well, not exactly ‘quantum uncertainty’ but the Copenhagen interpretation as applied to it.

    • Tei says:

      Things like wheater are very sensible to the start conditions. A small change on the start conditions result on a very diferent result. This is the base of the Chaos Theory. These particles doing random things act as seed’s for lots of random numbers on lots of starting conditions that will result on wildly different macroscopic events… where wheater is just one of these systems.

    • Sagan says:

      Regarding pseudorandom numbers:
      The great thing about pseudorandom numbers, and the reason that we can use them at all is, that they can’t be predetermined. The only way to figure out which the thousand’s generated number will be is to generate a thousand numbers and look at the last one. As Rod says: The only way to figure out what the simulation will do is to run the simulation. After you have run the simulation once you can always predict what will happen if you run it again with the same starting conditions. But when given new starting conditions there is no way to predict what random numbers will be generated, and thus there is no way to predict what will happen.

      Actually, often a random number is generated based on the time at which the random number generator is run. And that would make it completely impossible to predict what will happen, because you can’t predict when the player will start playing and when he will stop playing for the evening. And even if you could do that, there would be factors like which background programs are running, and you would have to calculate how much delay they cause. If that delay ever adds up to more than one millisecond all your calculations would be void.

    • Tei says:

      If you have the seed, and the function that use the seed to generate the next “random number”, you have all the serie of numbers that the game will use. The seed, the function that generate the next number, and the code that will use these numbers, are all in memory for everyone to see. I doubt Sims3 use the time for other than seeding the random number generator, since using it more can produce less randomness, since a clock is not a random thing, ….It can fail to a single pattern. If you start asking random number every 1 second, you may start getting the same number. Hence… game devs stay away from abusing time for seeding randoms.

      The Sims model is probably a simple state machine, that is predetermined by this random seed. Two Sims in a closed box withouth details, in a game forced to use the same seed, will act alike. Yes, there can be some framerate interference with gameplay and rules, but that is a bug, the type of interference you don’t want, because may cause big breackage … the game not working with people with low FPS or very high FPS. And anyway, the game can be forced to play below some limits, like forced 30 FPS.

    • Alexander Norris says:

      @Tei: if absolutely everything is random, then we have no choice – our actions are just the result of randomness. We’re just as deprived of free will at the quantum extreme of the scale as we are at the deterministic end.

    • Tei says:

      @Alexander Norris

      I get your point.
      I thinks something like a Strong Questions and a Soft Question. Negating that the universe is deterministic I solve the Soft Question of free will… these that thinks everything is predetermined, …like this Rod Humble guy.
      Theres still the Strong Question, like yours. I can’t solve your problem, and maybe Is a question that is not worth asking, anyway. If the universe is random, and no one is driving my desire to drink beer and drive. I am free as in… no one is commanding my desire, and as in no one know and can know what I will do. Maybe I will drive on my pontiac over a macdonalds, maybe not, on this random universe that is not predetermined.
      I can raise a @Wulf here and say that I have finished my quota of dificult problems with the Soft Question, so I don’t have to solve the Strong one… today.

    • drewski says:

      I’m not going to really make any comments either way because I think both sides of the debate have been adequately covered, but I would like to say that this is more like the kind of discussion I come to RPS for.

      The Sims into determinism into relativity. Excellent.

  5. Mitza says:

    Yes, the lack of applause is almost comical. Nice speech, too bad it got wasted on E3, and not TED.

  6. Clovis says:

    This discussion makes my brain hurt. Let’s complain about the ridiculous DLC and never ending expansions that should have been part of the game itself instead!

  7. zipdrive says:

    I love Radio Lab- it’s one of my two top podcasts (the other being Hardcore History).
    This is typical John Walker stuff, and it’s all great.

  8. Sagan says:

    If it ever turns out that we don’t have a free will, that would be the greatest irony ever. Just imagine if all of those fights for freedom didn’t have a point.

    Also: Rod Humble is awesome, and RadioLab is awesome.

    • Smithee says:

      No irony at all – you’re presumably conflating political/social/economic freedom with free will. It’s more useful here to view them as difference between “willing what we want” and “doing what we want.” There’s nothing incompatible with humans lacking free will but still wanting to live in a world where the range of behavioral options available to them is less limited, especially if having more options allows an individual to be happier or potentially happier.

  9. Kast says:

    That’s… best bit of E3 so far, by a long mile. I love a good surprising think in my games news.

    Also, expanding Sims3 in time and space? Just makes me think of a Space Station/Ship expansion. Which would be fantastic. Extra points if you can import Spore creations for your Sims to interact with.

    • Moonracer says:

      I’m going to hope that expansions related to “time” improve things like how it takes many hours to get out of bed, eat breakfast, take a shower and go to work. expansions related to “space” will probably further removing/improving rabbit holes, adding new places to visit (city/university?) and maybe allowing you to travel between towns (making more use for custom towns).

  10. Dolphan says:

    The universe almost certainly isn’t deterministic, which is interesting, but has nothing whatsoever to do with free will.

  11. Xercies says:

    Very interesing and that experiment with Hot and Cold kind of blew my mind, its so simple really if you think about it. Also really facinated about the Libet experiment and it does change a lot of things i think about how we think about free Will. Facinating.

  12. James G says:

    Actually, Tei touches on one aspect (among many) of quantum physics which confuses me.

    Okay, you you have your two slits set up, and are firing off your particles, be they photons, electrons or whatever you are testing for wave/particle duality. You get your diffraction pattern emerging, despite the fact you’ve only got one thing going through at a time.

    Okay. I get that. It melts my brain which has got used to growing up in the good old world of Newtonian physics, but I get it.

    You then start measuring your photons up by the slits, rather than at the detector, and the pattern collapses. Or at least, so most statements I’ve read of the experiment state. Okay, I understand that, you are collapsing the quantum state. But, and here’s the bit I don’t get, if you are measuring the photons at the slits, how is there still a diffraction pattern to disrupt? Its not like you can count the photons as they pass, once they are detected, that’s it.

    Is it a case of a misleading explanation, and by the pattern collapsing it is just mean that photon will only ever be detected by a single detector? I suppose its not so much the physics that confuses me here, but the experimental method.

    • James G says:

      Wow. Reply fail! Didn’t expect that. I’m usually logged in and using Opera though, whereas this time I’m logged out, and using Chrome. Seen some people suggest that it fails more regularly if you clear your cookies.

    • DrazharLn says:

      From my understanding, it doesn’t matter where you measure the photons. If you turn on a particle detector, then you can’t get a diffraction pattern. There was a really good explanation of this experiment in the NewScientist a couple of years ago. I can’t find that copy, though.

      You don’t disrupt the diffraction pattern, you stop it ever existing.

    • D says:

      I’m no expert on this but I think it would be sufficient to say, that a “particle detector” can detect the existence of a particle without stoppings it’s motion. I think if you look into the uncertainty principle (detecting only the position of a particle with certainty), it should be explained there.
      Anyways I looked stuff up on wikipedia and came across this one
      link to en.wikipedia.org
      which has a really good introduction and history of research into photons, in addition to a stunning result of the experiment. Apparently (and I’m sure I’m butchering it) photons create interference/do not create interference, based on whether the information about the slit taken will be discovered at a later point in time.
      It kinda makes sense to me, only because photons move at the speed of light and thus all time outside of their existence has already happened, from the particles POV.
      Just wanted to share and say thanks for making me read stuff.

    • Peter Radiator Full Pig says:

      “Because photons move at the speed of light and thus all time outside of their existence has already happened, from the particles POV.”

      Er, what? Just because something moves at the speed of light does not put it outside of time, or how ever you would phrase it. There would be some time dilation and stuff going on, relative to the particle, but it doesnt just ignore time.

      Also, photons dont really travel at the speed of light unless in a vaccum, which most of these experiments wouldnt be in, or at least, dont need to be to get the same results.

    • D says:

      By no means an expert opinion, I hope I made that clear. My idea comes from thinking about what happens when something accelerates to the speed of light. The time dilation occurs such that an accelerating object is aging less than objects that are not accelerating with it. As speed comes to 0.9c, the object is aging very little when seen from an outside POV – so the idea is just that when travelling at 1c, the object will not age at all. It will travel its course and collide with a detector without a spec of time passing for it.

      So the two entangled photons moving at the speed of light, if one is later determined to have gone through slit A, then the first will (have) create(d) the interference pattern at an earlier (outside POV) time. But from the entangled photons POV, the time to create the interference pattern is not earlier than the detection. I have no idea what to do with the speed of light in air vs. vacuum – I don’t claim to understand this stuff, I’m just grasping for it to make some sense. Enlighten me if you care :)

    • Peter Radiator Full Pig says:

      You are right that the faster you go, the more pronounced the effects get.
      But time doesnt simply stop once an object reaches c.

      And an object only ages differently from a different frame, not its one (So our POV makes it ages different, in its POV, its ageing regular, and the universe is going nuts)

  13. DrazharLn says:

    98 degrees!? Shit!

  14. KeenanW says:

    Too much science; needs more philosophy.

  15. Bhazor says:

    Will Wright’s bewildering but spiffing presentations has clearly had an effect on him.

    Also the Rod Humble city reminds me that we never did see a follow up for the dimension of Kieron Self Love.
    link to flickr.com

  16. Moonracer says:

    While that speech was quite good, I laughed when Rob started talking about the development team not being able to determine what would happen next in the game. As someone who likes to follow the progress of Awesome Mod and other big modding projects I get the impression that the Sims team has always been a bit sloppy with their coding.

    • Agnes says:

      Agreed. The modding (I call them ‘fixing’) community around the Sims games is fascinating to watch in process, and I respect them greatly – except that there should be no need for such major mods to make a game playable or interesting or what we thought we were getting with TheSims3.

      I truly loved TS1 and 2 and was more depressed than impressed by TS3, which overall felt as if the player was at a further remove from sims created/played. I could be bitter or out of the loop and design issues I had with TS3 could have been addressed by now, but I stopped playing at base game because free will was NOT what I desired, I found it formulaic actually, after a bit, very scripted according to personality traits. There are a host of other reasons as well, technical, design, limits, lack of perceived value or long-term playability, that eventually overrode the very cool things that are in the game and drove me away from playing anymore. I haven’t played it in over 6 months, nor have any of the expansions or stuffpacks or store items spurred me to try again. I’m actually quite sad about this. There’s nothing else out there that looks as absorbing to me as TheSims (I’m that kind of player, yes).

      Call me crazy, but the free will in TS2 surprised me more than in TS3. I suspect it was due to either inadequacies in coding or underlying code never revealed to players that it strikes me that way. I’m not a programmer, or a quantum physicist, or a particularly deep thinker. I’m also not playing TS3 anymore.

      Mayhap they’re overthinking the whole thing?

  17. Bowlby says:

    The faintest idea of a conclusion that The Sims may end up having more free will than actual human beings really just hits home how fucking stupid and half-baked that part of the conference was.

    • AndrewC says:

      I believe the point was that because The Sims are modelled on how we think we make decisions, rather than on how it turns out we may actually make decisions, Sims may actually be closer to our ideal of free-agents than us.

      But I guess that part was just too [very] stupid for you. You must be a [pc gamer].

    • Bowlby says:

      He doesn’t even bother defining what, exactly, he means by free will, and then goes straight into a philosophical discussion about it.

  18. Yellowstone says:

    I see no reason to sink into anomie if the world is deterministic (or random, for that matter). Either too many factors influence my circumstances for me to comprehend, or else everything is to a certain degree random and thus impossible to comprehend.

    Even if either one of these is the case, it’s still better to be free than to be oppressed, to take your example. It’s still worth fighting. After all, your victory might be inevitable.

  19. nine says:

    You have to interview this guy!

  20. Michael says:

    Yeah, yeah, Sims is a meditation on the nature of existence and not at all a shopping simulator.

  21. Zogtee says:

    The Sims 3 already? That was fast. They haven’t beaten The Sims 2 quite to death yet, like they did with the first game.

    Also, no PC version? Fuck that.

    • DollarOfReactivity says:

      Um, Sims 3 has been out exclusively on PC since last summer… Time to come out from under your rock.

  22. Finn says:

    What I want to do is to flip the cause and effect around. If I structure a house a certain way can I change the way their personality develops? I want the Sims to be my lab rats, and I can try to create well adjusted people or crazed neurotics, antisocial OCD sims, whatever.

  23. drewski says:

    Thanks for putting me onto RadioLab, John. Now if only Valve will fix Steam so I can play Torchlight and listen to it…

  24. HairCute says:

    HAHAHA! Wow, zogtee. WOW.

  25. Javier-de-Ass says:

    more like rod humble channeling will wright poorly. bunch of nonsense

  26. Stromko says:

    What does it matter if there is ‘free will’ or not? What is certain is that our past experiences, our habits, and the information available to us are all factors, among many others, in what we do. If each of us produces, in some small way, a ‘unique’ result, then what’s to be worried about?

    There really is no answer. Even if we proved or disproved the presence of the soul, a quality that exists beyond physical laws, we could continue to drill down and question just WHAT that truly means. Personally I don’t think we can ever prove free will beyond a shadow of a doubt, as certainly elements that are beyond our control are powerful factors, but by the same token we can’t disprove that there can be some component that is unpredictable.

    It’s enough for me that we’re asking these questions. It doesn’t prove free will but it seems closer to free will than is even necessary.

  27. social media says:

    Yeah, yeah, good video ! i used to take nitro-tech i’m not going to lie this shit works if u workout. thanks a lot..

    no shotgun results

  28. Josh W says:

    There’s another way to coming up with the idea of free will, that is different from what’s been mentioned before, and it if someone can make a pattern to predict what you will do, then you can hear about it and change what your doing.

    That doesn’t sound like much of a solution, but it actually is; it means the moment anyone feels predestined to do something, the very fact that they have had that thought changes the game, because they were predetermined the moment before they realised, and now they can look ahead at that predetermined path and decide if they like the result.

    If they don’t, then the very understanding of themselves that makes it seem predetermined can be acted upon. If you understand why you want stuff, where different feelings and desires or dislikes come from then you might realise that your being self-defeating or whatever and develop a new desire to stop doing it.

    In this veiw, human will is free because it keeps feeding back it’s own descriptions into itself, we’re free not to be a twat because we can notice when we are and work against it etc. Now maybe there are limits to the extent people can change themselves, and in fact there probably are, which is why I don’t subscribe to Satre’s “Human nature does not exist, because you can change it” idea, but I think it’s a pretty good starting point.

    It means that not only is the human mind bloody complicated, and possibly too chaotic to humanly predict, even if you could humanly predict it, that prediction itself would have to be included in the model and accepted by the person your trying to predict, or you run the risk of them breaking the prediction on purpose. (You can try to keep it secret, but stratergies often reveal the thinking behind them)

    On another note, (and it might be a bit harsh to do this) but look at how his form of expression changes when he gets from “me playing the sims” to “our game the sims” he goes all bored and robotic, ironic in a talk about freedom. I wonder if he actually wrote the last few paragraphs, or if a marketing guy reworked it. There’s a weird thing that happens now where marketing people have to temporerally turn developers into their avatars, because no one wants to listen to them. I suppose it’s sort of like what happens to politicians and their speechwriters.