Wednesday, May 14, 2008

Is Depression Adaptive? And Does It Matter?

Evolutionary biology theorists have hypothesized in recent years that depression may be an adaptive trait, that is, one that confers an evolutionary benefit. Watson and Andrews, in their paper "Toward a revised evolutionary adaptationist analysis of depression: the social navigation hypothesis," published in the Journal of Affective Disorders, have hypothesized as follows:
First, depression induces cognitive changes that focus and enhance capacities for the accurate analysis and solution of key social problems, suggesting a social rumination function. Second, the costs associated with the anhedonia and psychomotor perturbation of depression can persuade reluctant social partners to provide help or make concessions via two possible mechanisms, namely, honest signaling and passive, unintentional fitness extortion. Thus it may also have a social motivation function.

As with many evolutionary psychology hypotheses, there is little empirical support. Observers note that depression is highly prevalent, that it occurs with high frequency even among individuals with high reproductive value, and that it may provide a fitness advantage in certain conditions, such as providing a more accurate assessment of the circumstances (see depressive realism) or providing a convincing signal that the individual needs more support from the group. (Of course, depression is often detrimental to fitness, including reproductive fitness.) The theory that depression is adaptive is interesting; however, it is generous to claim that the evidence for this hypothesis is even weak. It is almost nonexistent.

The more interesting question for my purposes is, if we concluded that depression is an adaptive trait and provided a fitness benefit in our environment of evolutionary adaptation, should this influence our view of the right to suicide? It seems clear that it should not.

If depression is adaptive, that means that the depressed state is a normal, expected part of human life, which conferred a benefit on our ancestors. This hypothesis says nothing about whether people defined as depressed should be forcibly prevented from committing suicide. In fact, if we suppose that depression confers an increase in rationality, we should be even more inclined to respect the depressed person's analysis of his situation, and his decision to commit suicide.

Aside from rationality, the fact that depression may be evolutionarily adaptive would imply that much of our suffering is "by design" - that is to say, part of the evolutionary package. It would be yet another biologically-determined limit on human happiness, and entitle us all the more to refuse to participate in the project of life.

The interesting implication of one hypothesis for the evolutionary adaptiveness of depression has to do with treatment: the idea that people become depressed as a way to credibly signal that they need more support from their group than they are getting. An implication of this would be that depression could be treated by providing depressed people with more "support" (whatever that might mean). It is an interesting hypothesis, but, again, does not provide a compelling reason for prohibiting suicide. First, there is no evidence that any method of treating severe depression is widely effective. Even placebos are much less effective at treating severe depression than at treating moderate or mild depression (meaning severe depression rarely spontaneously improves, and does not respond to sham treatment). Widely prescribed anti-depressants are almost as useless. If some form of "group support" were shown to reliably treat severe depression, that would be wonderful news - but a right to suicide would still be warranted for those who still wish to die despite having submitted to this treatment (or, I would argue, despite refusing this treatment).

I hope to address the question of when, if ever, paternalism is appropriate when dealing with depression and suicide in a later post. I thank reader Mike Kenny for the interesting questions that I'm addressing here.


  1. Whether depression is adaptive or not does matter as far as how it can and, I would say, should be treated. I have just added a page to my web site on the social navigation hypothesis of depression to do with suicide prevention. It was sparked by reading Scott Anderson's July 6th 2008 article in the New York Times Magazine. See: Posting #2 (07/19/2008): Remarks on Suicide Prevention based on the SNH and Scott Anderson's 6 July 2008 NYT Magazine article, The Urge to End It All. While I am not for "forced life" I am in favor of developing and testing ideas that may help many very unhappy people wish to live.

    1. "While I am not for "forced life" I am in favor of developing and testing ideas that may help many very unhappy people wish to live."

      I believe this is a fine endeavor. Indeed, when I give a friend a hug, am I not engaged in the same thing in part?

      I don't mean to diminish the work you do, merely to point out that it is non-coercive, kind, and decent.

  2. Hi there, and thanks for your comment - yes, I saw the Scott Anderson article and have been meaning to respond to it. I found it hard to pay attention to it after noting that he credulously quoted the statistic that 90% of suicides have a mental illness, using the NIMH as his source, when the NIMH's source on that is an uncontrolled study using the suspicious "psychological autopsy" method. Also, he seems to live in a magical world where people like me - unsuccessful suicides who still want to die all the time - do not exist. In Anderson's world, a suicide attempt magically makes people feel better and makes all their problems go away, and if we can just rescue them from themselves, they will thank us in the end.

    This is coming out a bit venomous, sorry. I think there are serious flaws in the Anderson piece, and haven't yet felt objective and unemotional enough to respond to it in detail.

    I think we agree that it would be a great thing to cure depression and make people happier. But I am very suspicious of lines of thinking that allow for coercion of suicidal people, with the idea that they'll thank us later. Anderson's piece seems to me to communicate that message in a very dangerous way.

  3. er... isn't a strong causal factor of suicide something that would be the opposite of adaptive by definition? Not only that but an illness associated with sexual dysfunction, social isolation, and lack of motivation/energy seems strongly inhibitory to reproduction.

    Seems like a load of nonsense to me.

    1. What if it was mostly the least successful people who felt this way?

      Would not their suicide allow for a higher percentage of their genes to be removed from the gene pool, but their more successful close relatives genes' to go on?

      Thus removing the suicidal person's unsuccessful genes, but preserving his successful genes, to be passed on?

      It's a thought.

  4. There are many traits that are adaptive in mild or moderate forms but become maladaptive in extreme forms. Think of things like immune response - very adaptive in a mild or moderate form, but take it to an extreme and you have asthma, severe allergies, etc. Some people have postulated that depression is such a response - adaptive as a response to certain stresses, specifically valuable as a signal to the group because it imposes a significant cost on the organism so signaling, but of course maladaptive when it leads to the self-destruction of the organism.

    It does stink of a bit of naive group selection to me, if that's what you mean. But just because a trait is maladaptive in an extreme form doesn't mean the trait isn't adaptive in general. Some have suggested, for instance, that the maladaptive behavior of familicide (killing one's wife and children) is an extreme form of the adaptive (but of course repugnant) traits of possessiveness and "proprietary-ness." Same thing with suicide/depression, but the case is much weaker.

    1. "But just because a trait is maladaptive in an extreme form doesn't mean the trait isn't adaptive in general."

      That too.

  5. I agree that it is naive to think that once a person has decided, consciously or not, that suicidal behavior could yield adaptive, e.g. social benefits, that simply stopping the attempt at one location or via one method would reliably do much to stop that permanently. (But, Anderson's article presents data that suggests that at least in some cases it does; the people it works for may not be as sub-consciously committed to the high risk suicide attempts as others.) However, the physical delay of successful suicide might be much MORE effective in preserving many more lives if, during the time that has been gained, a meaningful intervention that addresses the underlying fitness-related problems and goals fueling such behavior is offered. The Social Navigation or Bargaining Hypotheses of depression suggest what such an intervention might have to entail. I would urge all, including skeptics, to read carefully the explanation of the Social Navigation Hypothesis of Depression on my University of New Mexico web site. It certainly is not a group selectionist theory. And, it must be remembered that natural selection has caused the evolution of some extremely costly behaviors. Note, all of them are contingently expressed, unless they are being generated by a maladaptive psychopathological form of the widespread, species-typical, "normal" underlying cognitive adaptation (which form, it is true, mental adaptations can take if not developed properly, or if negatively altered in their operation by drugs, in individual cases). As an example of contingently expressed costly behavior, honey bees are much more likely to commit suicidal attacks on humans or other mammals if bothered very near their hive as opposed to far off in some flower field. As far as our hypothesis of depression and suicide goes, it is the very costliness and risk that moves the person's social network to help the person make very substantial fitness-enhancing life changes. The high stakes context for subjecting yourself, and thus simultaneously your dependents, to such costs, would be when standard forms of social negotiation have repeatedly failed to ameliorate social constraints that severely limit one's fitness.

  6. Actually, I completely concede the point that coercive suicide prevention is effective at "saving lives" - more coercive suicide prevention seems to equal fewer suicides.

    What I propose is that saving lives is not necessarily the only ethical goal. Else there would be a great deal more forcible organ transplants from unwilling, healthy "donors" - hey, it would save lives.

    My thinking is that, while it might save lives to engage in all kinds of coercive anti-suicide measures, and while some people thereby "saved" might claim to be grateful, it is not necessarily morally right to engage in such coercive tactics. To me, suicide itself is an important right. Few people seem to be willing to even consider this as a possibility, the way we consider the right to continue living or the right to control over our bodies in other areas to be important rights.

    My main problem with Anderson and his thinking, and honestly every piece of mainstream journalism I've ever read on suicide, is that the idea of moral considerations other than saving lives does not seem to enter his mind.

    1. This.


      Oh, and incidentally, I don't even know that I could possibly succeed in the world anymore. I'm tired of hiding my beliefs on this and other controversial subjects. How the heck could I do such things as get a job, acquire friends, and lovers, while being both intelligent and truthful?

      It seems an insurmountable challenge -- well, at least the economics. There are a few sane, rational people, and some irrational but especially tolerant ones.

  7. There is more evidence than you think for the adaptation hypothesis. Everyone should review these more recent links from Andrews:


Tweets by @TheViewFromHell