Slashdot Mirror
People Become More Utilitarian When They Face Moral Dilemmas In Virtual Reality
First time accepted submitter vrml writes "Critical situations in which participant's actions lead to the death of (virtual) humans have been employed in a study of moral dilemmas which just appeared in the Social Neuroscience journal. The experiment shows that participants' behavior becomes more utilitarian (that is, they tend to minimize the number of persons killed) when they have to take a decision in Virtual Reality rather than the more traditional settings used in Moral Psychology which ask participants to read text descriptions of the critical situations. A video with some of the VR moral dilemmas is available, as is the paper."
So, we're assuming that all participants considered the death of (virtual) humans to be a bad thing?
In games like Counter-Strike: Global Offense, I take hostages, set up bombs are willing to give up my virtual live to protect it from being defused, kill people.
While in reality I'm not a suicide bombing terrorist. Who would have guess?
... (they tend to minimize the number of persons killed)
Anyone who's played Black & White knows that's not true. They don't even minimize the number of persons killed by poop.
In the various versions of the train dilemma, you have two options 1) don't act and five people will die; or 2) act and only one person will die. While I see the logic of your argument, and tend to agree that it is the best overall or numerical result. It does seem to be a rather chilling choice. It avoids the premise that by taking action the actor becomes a murderer; having taken action that directly resulted in the death of another. In the other case the actor is only a witness to a tragic event.
Until you're faced with the choice of saving your sister versus five anonymous others.
Utilitarianism is false, because no human being can know how to globally maximize the good. They just believe they do, and then use "the end justifies the means" to commit atrocities.
Our quirky affective behavior is arguably an optimal heuristic in a world where you only have a peep-hole view of the global state of things. For example, in those trolley dilemma you're _told_ that the trolley is random. But we're hard wired to believe that nothing is random, which means you have to fight a belief that the trolley was purposefully sent to kill those five individuals. Maybe the lone individual would save the world. In any event, maintenance of the status quo (letting the five get killed) is, again, arguably an optimal behavior when there is insufficient information to justify doing something else.
He is not only a witness if he KNOWS that he had the power to prevent the five deaths at the cost of one other. Inaction is also an action by itself.
-=This sig has nothing to do with my comment. Move along now=-
Harm and benefit according to whose definition? Utilitarianism is incredibly subjective.
"Become more utilitarian", i.e. they choose to save more lives, which is already at 88% in a non-VR, simple textual scenario like the trolly switch issue.
This is odd, because in most scenarios of VR, people seem to want to throw a switch to deliberately divert a trolly from one person to kill 5 instead, as long as they have a chat line where they can type "lolf49z!"
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
He is not only a witness if he KNOWS that he had the power to prevent the five deaths at the cost of one other. Inaction is also an action by itself.
Yes, and that ultimately leads to the "next level" of utilitarian dilemmas. What if you're a doctor, with five terminal patients who all need different organs. In walks a healthy person who is (miraculously) compatible with all of them.
Should you kill the healthy person, harvest the organs, and save the five terminal patients? (For the sake of argument, we assume that the procedures involved have a high chance of success, so you'll definitely save a number of people by killing one.)
Many people who say we should flip the switch in the trolley problem think it's wrong to murder someone to harvest their organ and ensure the same outcome. Why is "inaction" appropriate for the doctor, but not in the case of the trolley?
(I'm not saying I have the right answers -- but once you start down the philosophical path of utilitarian hypotheticals, there's a whole world of wacko and bizarre situations waiting to challenge just about anyone's moral principles. I can't wait until the "I was kidnapped and forced to keep a famous violinist alive" scenarios come up!)
It's a chilling choice, but the train dilemma is flawed when you consider that it would never happen in real life anyway. I'm not saying that the 5 vs. 1 scenario wouldn't happen, but I highly doubt someone is even going to consider the second option at all if presented with the scenario. If the thought doesn't even cross the person's mind, there's not a choice being made between the options. If no choice is being made in reality, the thought experiment is worthless as a way explain human behavior. The whole concept of the thought experiment is undermined when you realize that it's not something any person would ever end up doing because of another variable that the thought experiment does not consider.
Maybe it's best to try and switch the lines but fail. Then you only have one count of attempted murder.
One issue that studies never seem to take into account is responsibility.
If a group of people will be killed but you could decide to kill a single person, there is a third option: you could choose not to decide.
When you switch the tracks you are taking responsibility for making the decision, and all consequences thereto. There will be an inquest, you will be brought up under charges for manslaughter, your actions will be made public in the newspaper... all sorts of bad things will happen, and your life will be forever changed.
For a recent example, consider the recent Asiana Airlines Flight 214, where a woman was run over by a fire truck. The battalion chief responsible for directing operations was put through the wringer by over-zealous bureaucrats looking for someone to blame. His helmet cam footage was all that saved him. Blameless, he only narrowly escaped taking the blame.
If you simply walk away, then it's not your problem. The responsibility lies somewhere else, no one can blame you for not making the decision. You weren't expected to handle it, it's not your fault.
This makes perfect sense in the current study: there's no consequences for killing virtual people, so it's easy to make the moral choice.
Real morality takes courage, and the willingness to sacrifice.
I can't believe that people still think that these trolley car "thought experiments" are telling them anything novel about human moral instincts.
All they are are less-visceral variations on Milgram's famous work. An authority figure tells you you must kill either the hot chick on the left or the ugly fatty on the right and that you mustn't sound the alarm or call 9-1-1 or anything else. And, just as Milgram found out, virtually everybody goes ahead and does horrific things in such circumstances.
Just look at the videos in question. The number of laws and safety regulations and bad designs of the evil-mad-scientist variety in each scenario are innumerable. They take it beyond Milgram's use of a white lab coat to establish authority and into psychotic Nazi commander territory. In the real world, the victims wouldn't be anywhere near where they are. If they were, there wouldn't be any operations in progress at the site. If there were, there would be competent operators at the controls, not the amateur being manipulated by the experimenter; and those operators would be well drilled in both standard and emergency procedures that would prevent the disaster or mitigate it if unavoidable -- for example, airline pilots trained to the point of instinct to avoid crashing a doomed plane into a crowded area.
The proper role of the experimenter's victims ("subjects") is to yell for help, to not fucking touch critical safety infrastructure in the event of a crisis unless instructed to by a competent professional, to render first aid to the best of their abilities once help is on the way, and to assist investigators however possible once the dust has settled.
Yet, of course, the experimenter is too wrapped up in the evil genius role to permit their victims to even consider anything like that, and instead convinces the victims that they're bad people who'll kill innocents when ordered to. Just as we already knew from Milgram.
How any of this bullshit makes it past ethics review boards is utterly beyond me.
Cheers,
b&
All but God can prove this sentence true.
The healthy person isn't part of a potentially doomed set unless you harvest his organs.
You cannot ethically *start* the process of saving lives by unnecessarily killing someone.
In the train scenario, either 5 people die, or 1 person dies. There is no other option, because there's no way to stop the train in time. Your choice is simply whether to:
a) minimize the deaths by action, or
b) maximize them by inaction.
In the organ harvest scenario, you have a potentially doomed set, and a non-doomed set. You also have numerous options beyond:
a) kill the healthy guy for his organs, or
b) don't kill healthy guy for his organs.
For example, you also have:
c) convince the healthy guy to donate a subset of his organs which can be spared in order to save some of the terminal patients.
d) continue looking for compatible harvested organs.
e) harvest organs from the first terminal patient to pass on in order to save some of the other terminal patients.
There's more, but I think you can see the difference between the two scenarios.
Harm and benefit according to whose definition? Utilitarianism is incredibly subjective.
Exactly. I recognize full well that killing 1 will save 5, and in general I do not have a moral problem with choosing to alter fate to change the outcome to favor the 5, but I do not view any of the participants in the video cases as being faultless.
You and others are walking down the train tracks, a train is coming, and none of you move. Why arent you moving? Maybe that lone guy on the side track knows that the train isnt going to run down his track, which full well makes me a murderer if I divert the train to his track. The larger group has to take responsibility for their own damn actions.
That, my friend, is utilitarian in my eyes.
"His name was James Damore."
The first time I heard this dilemma it was posed by a bible basher attempting to recruit me into his church. In that version the individual is your own child. The point is that God would flip the switch and sacrifice his son to save everyone else whereas a mere human would normally save their child. Why an omnipotent God could not break the rules and save both the individual and the group was left unexplained.
Disclaimer: Grandad to three. The instinct to protect your child can overcome the instinct to defend yourself, sacrificing a bunch of strangers to save your own child is a no-brainer for most parents.
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
The one person wasn't "potentially doomed" at the start either. He was just crossing a set of empty railroad tracks.
Is 1563649 a prime number?
Right. So you're a billionaire and I work for you. My embezzling a hundred thousand dollars from you to send my kid through school is moral because you won't really miss it and it does a great deal of good for my kid and no discernible harm to you. That's the *pure* utilitarian way of looking at it, although such purity in outlook is something at least very rare and very probably non-existent.
There's another way of looking at this problem that seems built into human beings which philosophers call deontological ethics -- the ethics of rights and responsibilities. I have no right to your money, furthermore in agreeing to work for you I have a responsibility to discharge my duties faithfully. Yet while most people would agree that embezzling money from you would be wrong, that doesn't mean they're pure rights-based thinkers. If we simply change what is stolen their thinking is apt to shift.
Suppose instead of money, I steal a loaf of stale bread from you to feed my starving child. Normally that bread would go to feed a pig. Framed this way, I think a lot of people would consider it immoral for me NOT to steal from you, to let bread that could save a starving child go to a pig.
I should mention there's another important and overlooked style of ethical reasoning: aretaic, or "virtue" ethics. Somethings we do because we want to be the kind of person who does such things, and some things we don't do because of what that would do to our character.
In practice everyone seems to mix these different styles. Those who claim to use only one system of ethics to guide their behavior inevitably do little tricks to import other outlooks into their "pure and simple" philosophical framework. Utilitarians for example may conceptualize a rights violation as a harm, thus allowing them to argue in deontological mode when it suits them. Self-described deontological thinkers are apt to invent responsibilities that allow them to argue in utilitarian mode or aretaic mode when it suits them.
I can't say I've ever met anyone who has managed to put all of of their morality a set of geometry-like postulates and who reasons morally exclusively from those postulates. I've met plenty who've deluded themselves into thinking they do exactly that, but somehow the claimed mathematical proof is never forthcoming. If you press people, they inevitably argue from rules of thumb, by paradigms, by analogies, and other convenient but not necessarily consistent means of getting to a workable answer. They never resort to pure reasoning from a simple set of moral postulates such as utilitarianism.
And I suspect that's the best we'll ever manage at justifying all the things we feel in our gut. If nature declines to provide us with any system of arithmetic that is both complete and consistent, why should we expect her provide us with morality that is so? There may be some dilemmas that can't ever be solved, either because our postulates lead to contradictory answers, or they don't lead to any answer at all.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
Utilitarianism is false, because no human being can know how to globally maximize the good.
This is like saying "mathematics is false, because no human being can know if a statement should be an axiom or not". In both cases the subordinate "because" clause is trivially true, but not logically related to the independent clause it pretends to justify. Mathematics is a tool for generating models, some of which are useful for approximating how the real world behaves; utilitarianism is a subtool within mathematics that's appropriate for generating models of the part of reality we call "human morality".
They just believe they do, and then use "the end justifies the means" to commit atrocities.
Every proposed moral system has been used to justify at least an atrocity or two at some point: utilitarianism, deontology, moral relativism, moral absolutism, every goddamn religion you care to name — even Buddhism! (What the hell, right?) The truth is that people choose an action, then they justify their action by creating a post hoc story that rationalizes why the chosen action was Right, and it makes no sense to blame the justification instead of the choice.
Morality itself is a pattern in the brain that shapes what one chooses — how one resolves the balance between conflicting goals — and it's not actually an object-level belief that one can directly observe with conscious thought. If you give people books to read about object-level moral beliefs, the readers don't become more moral or less moral, they just get better at crafting post hoc justifications.
(Also, as it turns out utilitarianism was not a great model for human behavior by itself, but it actually does pretty well if you extend it with uncertainty in the Bayesian sense. Moreso if you go the extra step and add causality to the model (fixing the edge cases that crop up in more nai:ve decision theories that treat actions as evidence). If the space of possible futures is small enough, you can even wrestle the conditional probabilities into submission, e.g. using Judea Pearl's causal networks, and get concrete answers that take that uncertainty into account — still a high bar, but more tractable than "noooo, it's not worth doing unless it's perfect". Many human behaviors that seem irrational in a Homo economicus utilitarian calculus suddenly look perfectly rational if you model the study participant as, say, a Pearlian estimator with a low computed probability for P(stranger will actually give $100|stranger says they'll give $100 if I were to do X AND I counterfactual-do(X)).)
Range Voting: preference intensity matters
There's more, but I think you can see the difference between the two scenarios.
Most of your argument makes the assumption that the patients are not close to death. What if they are? (Disaster scenario or something.) And what if the healthy guy says, "No!" even to the idea of donating some organs?
If you find it repugnant to kill him, do you still favor forced "donation" of his organs if it won't kill him, but will save the lives of other people in imminent danger of dying? After all, you seem in favor of killing a guy in one scenario to save five people, what's wrong with stealing a kidney?
(To be clear, I'm not arguing either side of anything in these debates. I'm just bringing in the kind of questions that moral philosophers do....)
No, it is not. To quote Thoreau: "It is not a man's duty, as a matter of course, to devote himself to the eradication of any, even the most enormous wrong; he may still properly have other concerns to engage him; but it is his duty, at least, to wash his hands of it, and, if he gives it no thought longer, not to give it practically his support. If I devote myself to other pursuits and contemplations, I must first see, at least, that I do not pursue them sitting upon another man's shoulders."
The first imperative is that you don't do anything evil yourself. If you fail to distinguishing between doing something and "letting something happen", you can be scared into supporting any injustice or atrocity by the intangible ghosts of ever greater ones.
We are not masters of the world of outcomes. You can pull the lever any way you choose, and in the real world, as opposed to trolly problems, you don't know what will happen. This is especially important if it relies on another person's decision, but even if it's not, it's still true.
We are, however, masters of our own decisions (or it is at least morally imperative to assume we are). You can choose to divert the trolley from the one, or the five; as long as your intent is to divert the trolley away from one (and not towards the other) then you cannot be morally condemned for either. Nor can you be condemned for doing nothing and letting the universe explode (or whatever the trolly problem writer dreamed up for that eventuality).
xkcd is not in the sudoers file. This incident will be reported.
As a utilitarian, I find that lots of people fail to consider long-term consequences of actions, concentrating on the short-term. Therefore, deontological ethics are useful because they establish expectations, which are useful, and because such things as embezzlement tend to tick people off, reducing happiness. Aretaic ethics are useful because they cover long-term consequences. There are things I simply do not do, and I believe both I and others are happier for that.
It is my opinion that all ethical systems, including divinely ordained, fit nicely under utilitarianism*. I haven't a clue how to prove that, and I don't have any set of postulates, and very strongly doubt any could be made to be useful. Only careful observation of the situation and careful consideration will determine if an action is actually good - except, of course, that having to do all that investigation and thinking before acting would be a pain, and hence it's usually better to use shortcuts. Morality ain't simple when you try to apply it.
*Yes, this does mean I reject numerous concepts of God.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes