Changing belief in free will can cause students to cheat

I see only one possible problem, if I understand the experiment correctly; a few "super-cheaters" from the determinism groups could skew the entire test, since they might choose to take more than $15. With each division having around 24 students, if a super-cheater in a group of five took all of the remaining money (15 for himself, and around 8 for four other supposedly honest students), one person would skew the entire average by a little more than two dollars above the baseline of around seven dollars. If one individual could have that much significance, I think it's a bit questionable.

One might also be tempted to use these results to argue that belief in free will is important from a moral perspective (whether or not it actually exists).

No...no I wouldn't. That would be making an argument that depended on its own conclusion: IOW, 'free will is important because thing that depends on free will existing would be important if does exist.'

Let's be really clear here: There is NO evidence that the concept of 'Free Will' has any objective existance what-so-ever. Before you build your castle of argument from consequences, work on the foundation first.

I'm curious about how the participants' beliefs on free will prior to the experiment could affect how much they cheat. Participants who walk in with a well developed belief in determinism might not be as impacted by the determinist readings as those who have never really considered the idea before. You can see that the participants' beliefs after taking the survey are influenced by the reading before it--I'd speculate that having a pre-existing belief in determinism would not have much of an effect, but a sudden adoptation of the idea would create the results seen here.

I'd like to see a variation where there is discussion of what might be called deterministic will: A notion that though the brain is a mechanism, it is a mechanism that can pre-calculate effects from causes if that process is "kicked in" and acted upon.

IMV, the notion of "free will" applies a pressure to force this computation (because ignoring the computation leads to less thoughtful behaviour), but unfortunately, I'm determined to have this view.

GZZZZT! Gödellean feedback! *snap* *poof*

I find it difficult to believe that simply reading a passage from Crick would be e nough to change what they believe about the nature of free will, even if they are students.

It seems possible to me that the passage might prompt a wide range of more plausible beliefs. For example, the belief that "the examiner doesn't believe that free will exists."

David,

That's an interesting possibility. However, that would require one group's result to be a dramatic outlier compared to other groups. I suspect if that was the case the researchers would have reported it. I'm going to email the authors and ask, though, because it is an interesting question.

Benjamin,

I agree, it's probably not a very sound argument -- but indeed it's a tempting one, just not one I'd make.

Re: 2
I think a distinction needs to be drawn between whether free will exists and whether we ought to believe it exists. The positive and normative arguments are separate in questions of ethics. We can surely formulate arguments that it is better to believe in free will than to believe in determinism, and that argument is not the same as the one you criticize, that free will exists.
The two are related - ie, the nonexistence of free will may be a compelling reason to believe determinism, but it doesn't force us to believe determinism.

The two are related - ie, the nonexistence of free will may be a compelling reason to believe determinism, but it doesn't force us to believe determinism.

If you're looking, it does. Willful ignorance aside, if the evidence points strongly to one position, it forces you to take that position. I'm thinking of the arguments made against Pascal's wager, that beliefs cannot be chosen, but are just what we've been convinced must be true. If you choose to believe something, you are acknowlegding that it's not a truthful belief.

I have considered the matter personally and decided I have free will. However, it is constrained by my genetics, developmental history, raising, education, culture, physical condition, situation, etc. So I have come to the conclusion that I have bounded free will, and I am aware of only a few of the boundaries.

I think in time we'll come to have a more sophisticated understanding of free will versus the lack thereof, but that's a long ways off. Currently, pretty much everyone implicitly believes in free will, and that belief is deeply tied to many different aspects of our culture.

Let me point to an example of that lack of sophistication: 'a case can be made that free will is simply an illusion, and that every "decision" we make is completely controlled by factors outside of an individual's control.'

I don't believe in free will. But I also know that it's wrong to say that my decisions are completely controlled by factors outside myself. I am by far the biggest determinant of my future actions (decisions and otherwise) -- it's simply a matter of direct versus indirect causality.

Here's an analogy. For the sake of argument, let's consider weather to be something that is unaffected by any "free will". The quoted statement would be analogous to saying that San Francisco's weather system is completely controlled by factors outside of San Francisco -- so the weather conditions within San Francisco have no effect on its future state. This is of course false. Although San Francisco's weather is not subject to free will, and is affected by all sorts of external factors, the largest factor of all is in fact San Francisco's current weather state.

So, too, with us. An individual is not outside the laws of causality. But the most significant and direct causes for an individual's behavior lie in that individual.

I believe in free will.

However, I'm a compatibilist, so that belief looks a lot like the position laid out in comment # 10.

Being a compatibilist is fun that way.

Given that our culture inculcates the idea that we are "responsible" for what we do *because* we have "free will," these results are what you'd mostly expect.

#2:

To me, it seems your arguments do not counter the statement that you quote. The statement is that one could argue that the belief in free will is important, not that free will itself is important. I don't see how this is 'free will is important because thing that depends on free will existing would be important if does exist' in other words. It seems to me to be a completely different statement. In particular, "belief in free will is important" doesn't in any way imply "the existence of free will is important" or "free will is important". To me the statement "One might also be tempted to use these results to argue that belief in free will is important from a moral perspective (whether or not it actually exists)" merely says "an erroneous belief could nonetheless have desirable consequences", which is a reasonably arguable statement.

Let's be really clear here: There is NO evidence that the concept of 'Free Will' has any objective existance what-so-ever.

I agree with this absolutely. I'm no defender of free will (well, that would depend on how it's defined, but under most reasonable definitions of the hard determinism flavour, no. But if, say, your definition of "free will exists" is "Wednesday is the next day after Tuesday", then I'd say, yes, you are right...). But I don't get this:

Before you build your castle of argument from consequences, work on the foundation first.

I don't see how the existence or non-existence of free will is a foundation for an argument that "belief in free will might be necessary (or "important") for morality". They seem to me to be entirely unrelated questions.

El Christador: I don't see how the existence or non-existence of free will is a foundation for an argument that "belief in free will might be necessary (or "important") for morality". They seem to me to be entirely unrelated questions.

'Morality' is a dependant concept of 'Free Will'. If there is no objective existance of 'Free Will', then 'Morality' is a null concept (meaning that it in turn describes something that also has no objective existance).

I'd be interested in how you can successfully argue for the concept of morality without first implicitly accepting the idea that people can "extra-causally" 'choose' to do other than they actually do as determined by the laws of physics.

To put it another way, I can't objectively debate the properties of a unicorn's horn without first acknowledging the existance of the unicorn.

I suppose you could redefine morality without the concept of 'culpability'. But it would be missing the element that most people recognize as core.

@ David Montgomery #10: *thumbs up*

I never understood the argument that determinism and free will are mutually exclusive, though it seems to be widely accepted. The issue to me is that free will seems to me to fundamentally an issue of control (self-control, to be exact), and it seems strange to claim that one would have more control over oneself if cause and effect did not hold universally.

Here's again when somebody somehow brings in the magic words "free will": people will discuss their own views about free will instead of what has been stated in the post (or, for that matter, in the paper that is reviewed).

To take it to that question: I don't think the study has anything to do with whether free will exists or not. Instead of giving it the catchy label one might consider to view the results from the perspective of research on automatic vs. controlled processing --- the set-up of the experiment reminds strongly of experiments from the Bargh group on stereotype activation. A typical experiment in that paradigm would for example require participants to read something about the elderly; afterwards, participants tend to walk more slowly.
Now if you view the above "free will" experiment as an example of stereotype activation --- would you really maintain that it is about free will? If Bargh and colleagues had stated outright that they wanted to invoke an automatic stereotype of the elderly, there wouldn't have been an effect. Just that a lot of everyday behavior is not under conscious but automatic control has no relevance at all for free will. If one argued like that, every single psychological experiment would count against free will, because then every single psychological experiment would evidence that people can be influenced.

I've heard back from Kathleen Vohs on the "super-cheater" question brought up by David in the first comment. Here's what she said:

Let me respond to your readers' question -- which is an excellent one!

We have no reason to believe that one person took all the money, or even that a few did this. Recall that in each of the cheating-possible conditions, there were present 2-5 subjects (we abbr. this "Ss", so please note that's what I am referring to when I write it). Let me lay out two possibilities of how such super cheating would have occurred, and why I doubt it did.

If one S wanted to take a lot of the money, one plan would be for him or her to take it soon after they were left alone, meaning before the other Ss were done with their task and done self-scoring their sheet. In this case, if the one S did take a lot of the money, then the other Ss in the group would find very little money leftover, and to the extent they even wanted to pay themselves veridically for their correct test scores, they would not have had enough money to do so. Recall that once Ss leave the room, the experimenter encountered them and gave them a full debriefing. At this point, if I know my undergraduates, I suspect these Ss would have complained vociferously that there was not even enough money to allow them to get their fair payment.

The other possibility is that the super cheater would wait until the others are done, and then even if she or he wanted to take away much money, this would not be as possible since others would have already taken their due, which would meant that a big amount of money would not still be available, but only a moderate amount. So, in short, the structure of the situation made it such that a) we would likely have known if one person aggressively got up and was the first to leave the room, grossly over-paying himself or herself because other Ss would have complained - and we heard none of that; b) it was less possible to grossly overcheat if the person waited later in the experiment to get up and leave, because the takings of the other Ss would have not allowed for enough money to be left to enable such super cheating

" 'Morality' is a dependant concept of 'Free Will'. If there is no objective existance of 'Free Will', then 'Morality' is a null concept (meaning that it in turn describes something that also has no objective existance)."

No this is not true. A robot with no free will, for instance, can be programmed to optimize behavior to avoid punishment, or social disapproval, or simply to not harm others. We have moral instincts, which are essentially pre programmed (by evolution) social behavior schemas which work quite well.

Of course, there is the additional problem of actually testabley defining 'free will'. "non deterministic" keeps coming up, but is that sufficient? What if we have a deterministic with a quantum randomizer that can change some outputs? Is this free will? I think most advocates of free will would say no. But then, how can we define a form of 'free will' that is testably different from this definition?

I appreciate the difficulty of designing a simple experiment to test the effects of belief in free will, but it seems to me that this experiment is a little convoluted. I think there is a distinct possibility that the "determinism" group realized that something was going on. Perhaps the combination of reading the determinism statements, along with the experimenter "unexpectedly" leaving the room, was enough to make them realize that they would not be punished for taking more than their share. In that, one could take these results and present a case that "reading statements about determinism and the possibility of the lack of existence of free will increases awareness of the true nature of situations."

#20 cont'd....

The point being that once they realize the true nature of the situation, taking extra money is no longer cheating. If the students see that it's part of the game, then it can no longer be realistically labeled as cheating. Or, as another commenter suggested, maybe the statements made them think that the _experimenters_ lacked belief in free will, in which case it would still not be cheating; if they thought that the experimenters thought that the students lacked free will, then taking extra money, in the experimenter's eyes, could be compulsory for some individuals. My brain is spinning....

A lot of the problems listed in the experiment are valid questions, and great places to design additional experiments. I think what the experimenters set out to do was find if there was any correlation at all between the literature read and the behavior in the cheating possible setting. It is undeniable that the variation in the final group is significant. The way I read the article, however, they were pretty explicit in saying they had no way to determine causality or to what degree anything affected their results. My suggestion would be then, that someone with a hypothesis about the source of the increased money dispensed should frame a question that someone could test. In a perfect world, even propose a procedure to test their explanation. The whole purpose of the experiment, as I read it, was not to answer the question of free will or if the belief in free will affected behavior, but only whether influencing a subject's belief in it had any affect on that person's behavior at all.

While this study presents an interesting perspective on attitude-behavior consistency research, I'm not sure it clearly adds insight into free will (at least based on my conceptualization of the concept).

The study is consistent with attitude-behavior consistency research to the extent that statements read by participants in the various conditions likely resulted in the activation of one of many attitudes held by participants--deterministic statements may have primed attitudes associated with an external locus of control (e.g., my actions are governed by factors external to me); free will statements may have primed attitudes associated with an internal locus of control (e.g., my actions are governed by factors within me). So the results concerning determinism make sense if participants are primed to think about factors outside the self which may control their behavior; in this case, who will mark the test. In the condition where they mark their own work, a salient external factor is removed (i.e., person who marks their test), which reduces their chance of being caught, and therefore, enhancing their motivation to take money.The rest of the conditions likely activate (default; likely socialized) self-concepts associated with not cheating and fairness.

This interpretation is consistent with another study showing that when trick-or-treaters came to an experimenter's house on Halloween, they were more likely to take extra candy when they believed nobody was watching, and, if no mirror was present. However, when a mirror was present (with the mirror presumably serving as a cue to remind them of implicit self-concepts of fairness) they took less candy. An interesting twist would be to see what happens to the determinism/self-scored condition result when a mirror is placed in the room with the participants.

My intuitive notion of free will is consonant with cognitive (e.g., Stroop effect) and behavioral (e.g., effortful control) flexibility, whereby greater amounts of both constitute more free will. So this implies that free will exists on a continuum and is not an all-or-none phenomenon. Furthermore, in contrast to the implicit assumption of the study that a reduction in free will belief (increase in determinism) results in reduced morality (i.e., increased cheating), it's possible to contend that an increase in free will belief can also result in reduced morality when free will is defined in the context of behavioral and cognitive flexibility.

A case in point is the story of two airline incidents involving DC-10 planes as reported by the show Mayday on the discovery channel. The first incident involved a rear cargo door blowout when the plane reached altitude, which forced the pilot to use all of his skills to land the plane safely. After the NTSB investigated the accident, they determined that the DC-10 cargo door had an engineering defect that could result in a disaster and thus recommended a redesign. Likely due to a profit making incentive, it was discovered a few carrier companies did not follow the NTSB recommendation; another DC-10 crashed approximately 2 years later, killing all 346 people on board with the cause of the accident being exactly the same as the first incident. The free will situation here concerns the choice between making a profit or knowingly jeapordizing the lives of passengers. The powers-that-be did not make the seemingly morally obvious choice and ended up with a lawsuit that probably cost more than it would take to have fixed the defect.

The DC-10 story represents a real-world situation that touches on a study I think is closer to capturing the essence of free will (Leslie, Knobe, & Cohen, 2006). The authors in that study examined the development of a phenomenon called the side-effect effect (i.e., people judge a person's action as intentional when it causes harm and unintentional when it results in something positive). They showed that the side-effect effect (a moral reasoning phenomenon) develops by age 4 and emerges when a particular theory-of-mind emerges. That is, it shows up when children begin to understand that a person can have a mental state in which they don't care about the consequences of their actions.

So a good question might involve determining whether the state of not caring about the consequences of one's actions even though one knows the actions are likely to cause harm, constitutes a free will or deterministic state. An interesting follow up study that would add to the current one, would be to identify what happens to the side-effect effect when exposed to free will or deterministic statements.

Reference

Leslie, A.M., Knobe, J., & Cohen, A. (2006). Acting intentionally and the side-effect effect: Theory of mind and moral judgement. Psychological Science, 17, 421-427.

Interesting point touched on by #12-

If cheating-averse behavior is regulated by feelings of guilt regarding one's actions causing injustice, is this effect caused by the "numbing" of those feelings of guilt? That is to say, was the determinism group temporarily inoculated against the guilt of cheating by being told they are not responsible for their actions? Further, do prior beliefs have a confounding effect? If a "free-willer" read "determinist" literature, does that predispose them to cheating? These results would seem to suggest it, but I think there's more here that could be studied.

Perhaps another test would be sorting subjects into groups who had prior beliefs about determinism or free will and seeing how they do on the test.

Interesting results, though.

I find interesting the suggestion of one of the commenters that reading the pro-determinism texts may make one think the experimenter doesn't believe in free will. One possible rational mechanism: reading the pro-determinism texts may make one think that scientists don't believe in free will, and hence that the experimenter doesn't. One might, then, conclude: "If the experimenter doesn't believe in free will, the experimenter can't blame me for taking money." It's bad moral reasoning (that x can't blame y for doing A is no justification for doing A), but I bet this kind of reasoning is not uncommon.

In my opinion, this debate over free will and predetermined acts only makes people measure every little thing that humans do and makes people in general neurotic. I.e. did I want to do that or did something make me do that? Who cares? Apparently this is for those who feel that they want to make a moral statement about one or the other. This question is geared to say that determinists cheat, is it objective? The stimulus says hey you dont control anything its all determined, what do you expect when you provide for the impulse of greed? The question has suggested to succumb to selfishness because its predetermined and controls so thus they take when no one is looking.

Benjamin Franz said:

"Let's be really clear here: There is NO evidence that the concept of 'Free Will' has any objective existance what-so-ever."

You're assuming it has to be an objective concept, but it doesn't. In my estimation it's of the same class of idea as color labels. We all "know" what red "looks like", even though not one single person in the entire history or humanity has ever experienced what red "looks like" to another person. Now admittedly, comparing the understanding of free will to the concept of qualia isn't the best way to explain what I mean, for now it's the best thing I could come up with. I'm not saying that 'free will' is itself the same kind of thing as qualia, although I will admit that 'free will' certainly could be associated with qualia.

In any case, it could certainly be the case that what constitutes "free will" is dependent upon the person, their context, their history etc. Of course this doesn't disprove the claim there is no such thing as free will, but that's not the point.

The point is that if free will is subjective, there is no way to prove or disprove it. It's beyond the bounds of measurement. Of course I'm assuming the stance of Nagel and Merleau-Ponty here, but I agree with them (and Thomas Kuhn) about the nature of the logic of the scientific method. By it's very nature it excludes the subjective so it's logically incompatible with the idea of subjectivity, qualia, etc. so there will never be a scientific disproof of the existence of free will, and there will never be a scientific proof of its existence for that matter.

The point is, that I think the idea of "free will" is subjective and not objective, despite philosophers and psychologists attempts to make it objective, and because of that, we'll never be able to say with certainty that it does or does not exist.

Interesting stuff. I think it's possible to believe something for five minutes. I do it every time I watch a soap opera. I believe in it just enough to satisfy my needs. The circumstances I am in affect my decision making. A 'test' in a 'laboratory' will precondition me in a certain way.

If you tell a child 'If you eat this bar of chocolate while I am out of the room, then you are very naughty', you are likely to get a different response than if you had said 'If you eat this bar of chocolate while I am out of the room, then I it will not really be your fault, because i left it there unattended'.

What is free will? Can it exist? Is it just a matter of observing the process of making our predetermined choices, without really being aware that they are predetermined? What is predetermination? Is it made of causal factors? Is it made of history, of the past? Is it as simple as billiards? If X hits Y, then Z will happen?

This instant, the here and now, is the cutting edge of time, slicing its way into the future. The immediate future is foggy but not invisible. If I press the K key in 5 seconds' time, there is an extremely high likelihood that the letter K will appear. So to some extent we can see into the future. This probability may be higher than whether I can recall certain recent factual events accurately.

I guess I'm saying I disagree with the suggestion that these experiments make the existence of free will less likely.

the last thing having something to do with 'free will' is the conditioning for human belief system mechanism - as for the rest, though rarely intuitive, 'free will' draws most from the 'ego' and they both gradually fade away with maturity -- one would argue that the external forces are everything that is, and bear in mind that mortals are the only medium available to it - as for the commercial religions thriving on abusing basic sense of existence, its nothing more than filling the empty space reserved for next wave of evolution