I've received a few emails along this line:
"How does this new theory about depression enhancing problem-solving relate to all the studies that have shown cognitive deficits in people with depression?"
That's a really good question. I tried to address this issue quickly in the article - I referenced the fact that the "cognitive deficits disappear when test subjects are first distracted from their depression and thus better able to focus on the exercise" - but I think it's worth spending a little more time on the scientific literature. The key point here is that the deficits are "unstable," which means they can be made to disappear when subjects are first distracted from their ruminations. (As Andrews told me, "Depressed subjects are trying to cope with a major life issue...It shouldn't be too surprising that they have difficulty memorizing a string of random numbers.") Look, for instance, at a 2002 study which compared clinically depressed subjects to non-depressed controls. In a test of executive function, the deficits of depressed subjects were erased when they were first distracted from their ruminations. Here's their conclusion:
The aspects of executive function involved in random number generation are not fundamentally impaired in depressed patients. In depressed patients, the rumination induction seems to trigger the continued generation of ruminative stimulus independent thoughts, which interferes with concurrent executive processing.
In other words, the emotional pain of depressed subjects consumed scarce mental resources - the mind is a bounded machine - which meant they didn't have enough attention left over to think about anything else, especially some artificial lab task. When we're wracked with pain, everything but the pain is irrelevant.
I also think it's worth pointing out that this latest hypothesis builds (as always in science) on numerous earlier conjectures. One important precursor for Andrews and Thomson's idea was theoretical work done on "psychic pain," by the evolutionary biologists Randy and Nancy Thornhill and by Randy Nesse, a psychiatrist at the University of Michigan. In essence, these scientists argued that emotional pain serves the same biological need as physical pain. When we break a bone in the foot, the discomfort keeps us from walking, which allows the bone to heal. The pain is also a learning signal, teaching us to avoid the dangerous behavior that caused the injury in the first place.
But there has been no shortage of clever conjectures on why depression exists. If you'd like an overview of the literature, I'd suggest reading this paper by Paul Watson, an evolutionary biologist at the University of New Mexico, whose work also influenced Andrews and Thomson. As I explicitly stated in the article, all of these theories remain just that: theoretical. There is very little direct evidence in support of any of them.
Finally, I think it's worth repeating the obvious, which is that depression is an extremely varied mental illness. Although we only have one psychiatric diagnosis - major depressive disorder - that diagnosis covers a tremendous range of symptoms. (It's also in constant flux, and will likely be altered yet again in the next DSM revision.) As I noted yesterday, one of the most cited papers in the field found that MDD exists on a spectrum of severity, ranging from mild (10.4 percent of patients) to very severe (12.9 percent), with the vast majority of patients somewhere in between. I tried to make this heterogeneity clear in the article, because I think it represents a real challenge for any theory that attempts to explain depression, either from an evolutionary perspective or from a neuroscientific perspective.
Although Nesse says he admires the analytic-rumination hypothesis, he adds that it fails to capture the heterogeneity of depressive disorder. Andrews and Thomson compare depression to a fever helping to fight off infection, but Nesse says a more accurate metaphor is chronic pain, which can arise for innumerable reasons. "Sometimes, the pain is going to have an organic source," he says. "Maybe you've slipped a disc or pinched a nerve, in which case you've got to solve that underlying problem. But much of the time there is no origin for the pain. The pain itself is the dysfunction."
Andrews and Thomson respond to such criticisms by acknowledging that depression is a vast continuum, a catch-all term for a spectrum of symptoms. While the analytic-rumination hypothesis might explain those patients reacting to an "acute stressor," it can't account for those whose suffering has no discernible cause or whose sadness refuses to lift for years at a time.
Personally, I think these are the two most important paragraphs in the article. One of the most challenging aspects of studying depression is the vast amount of contradiction in the literature. Virtually every claim comes with a contradictory claim, which is also supported by evidence. I tend to believe this confusion will persist until our definitions of depression become more precise, so that intense sadness and paralyzing, chronic, suicidal despair are no longer lumped together in the same psychiatric category.
Thank you again for all your comments and emails.
A response, more to Andrews and Thomson, than to your article. I wonder how many people who read the NYT piece actually took the time to wade through the paper for themselves.
I'm critical of what seems to be the dominant view - that mental illness, (and in fact all mental phenomena), can be explained via biological evolution. Sociologist Liah Greenfeld has put forth the idea that culture - the symbolic process by which human ways of life are transmitted across time and space - is an emergent phenomenon, logically consistent with the laws of biology but autonomous nonetheless. From this view comes a theory of mind as the individualized process of culture - obviously dependent on brain mechanisms for support but not reducible to or explained by the workings of the brain. I say this to suggest that considering the symbolic process of culture totally changes the game in terms of explanations for mental illness. The possibility that culture, a symbolic process, can cause the mind (and therefore the brain) to malfunction, opens up.
I won't go on and on here, but I hope you'll read my post and consider Greenfeld's position in opposition to the evolutionary view. Glad to have found your blog - by the looks of it there's plenty to read.
One of the most challenging aspects of studying depression is the vast amount of contradiction in the literature.
This is unfortunately true of so many neurological and psychiatric disorders - I'm currently working in a lab studying multiple sclerosis and it is extremely challenging to just get a handle on the literature. For a disease or disorder like depression that doesn't have apparent histopathology or markers detectable by medical imaging it must be even worse.
I broke a bone in my foot two Sundays ago. I am currently walking some on this foot. The doctor told me that it would not hurt the healing process to put some pressure on it, and he said that I should "let my pain be my guide" when it comes to what I can and cannot do.
The problem that I am having with this advice and with the argument that pain is a learning signal is that sometimes it is hard for me to tell when I am actually in pain. I have gotten in the habit of convincing myself that pain itself is irrelevant. I guess the "no pain, no gain" attitude can be taken to an extreme.
This is all very deep and dare I say depressing.
Questions and answers that are no doubt very meaningful, but getting into the detail a bit from what was an interesting and educational article in a mainstream publication. It is good to know that the grey matter brigade are out there paying attention and pointing out the oddities, that is what peer reviewing is all about, I guess.
How about we climb up into the helicopter and take a look around at the broader issues - why is it less depressing to live in one part of the country or the world compared to another; why are rates of suicide constant each year (I've just started reading Durkheim), why do people in the same cities have different life expectancies depending on their social status (irrespective of access to healthcare); etc.
Come on you marshmallow test passing contributors, move on.
Wow, Jonah, you sure get some interesting people commenting on your site.
I just have a quick question. I know that in some circumstances where a patient has depression and a comorbid attention disorder, doctors will not treat the attention disorder until the depression is taken care of. Do you think your observations imply that certain symptoms of depression might, in fact, be managed by treating the attention deficit?
Depression, aside from purely neurological origins, is not a disorder at all. It is a rational behavioral economic function of the brain.
I operationalize depression as involving low absolute mood levels. Mood is the sum of all perceived net reinforcing options, temporally discounted. Or, you can say that it is the net rate of intake of reinforcement.
It is described by a negatively accelerated curve, corresponding to a reinterpretation of Hernnstein's matching law, or less precisely, to marginal utility in economics.
So, a simplified formula might be:
U = N [ g / ( h + g ) - L / ( h - L )], where U = net utility or net motivation (subjective net expected gain or loss, or you can call this simply demand), g = objective gain, L = objective loss or response cost, h = baseline rate of intake of reinforcement, or "have" (mood), and N = required rate of reinforcement (Need).
Some implications include defining depression as a motivational trap for the 80% of people who are risk averse. With this model, as mood decreases, the subjective value of gains and losses both increase, but with that of losses more rapidly. Hence, the seeming paradox that the need for reinforcement increases with decreased intake, but with less motivation to seek more.
There is a shift toward more immediately(decreased delay of gratification) profitable gains, with a particular emphasis on sex, high calorie foods, as examples. This serves the purpose of optimizing inclusive fitness in an environment relatively hostile to this goal. Hence, low moods, with increased risk/loss aversion, serve to maintain behavioral economic homeostasis.
The neurotransmitter proxy for mood obviously seems to be 5-HT serotonin, the behavioral dosage response curve of which fits this model nicely, as do those for GABA, dopamine, and opiods.
Generally, I'll conclude by stating that only behavioral economics can provide the ultimate direction and contexts for behavioral research. Those who wish to learn more are invited to glance at my blog.
@5: Or, in contrast, he writes highly-read articles for the New York Freakin' Times, which drives people to his site.
Regarding the comparison with physical pain, I think it speaks in favor rather than against using antidepressants (in a perfect world, the psychiatric equivalent to painkillers). One does not need to feel intense pain to know to avoid walking on one's broken foot nor does feeling intense pain facilitate healing in any way. Similarly, one does not need to feel intense depression to be able to cope with serious life issues. If anything, mitigating pain (notice I didn't say eliminate pain completely), be it physical or psychological, seems like a good first step, if not a requirement, on the way to healing.
Hmmm. One weakness of your reporting: taking lifetime incidence in western, developed nations as normative. It's possible, isn't it, that depression is a disease of Western modernity, almost as much as say, anorexia is?
My sense of the literature--and my own fieldwork abroad--is that depression is a good bit rarer in other kinds of societies.
That follow up was pretty disingenuous. It admits the central problem with the thesis originally advanced: that the pain itself is detrimental and so would counter-balance any difference made in the frontal cortex, but doesn't face the fact head on that this is all but fatal to the idea that depression somehow has survival value (Its like saying, "yea, but when you're not sitting with a gun in your mouth or a needle in your arm, you can concentrate really well". He all but admits that this objection is valid but ignores the full counter-evidentiary weight of the objection by saying "but there are contradictions all over the place in psychology, so I'll just ignore this and do so allowably". Fail.)
I think the problem is with the whole evolutionary psychological approach. I suggest this article here:
Also, he mentions the need to specify different types of depression more exactly, and even mentions the upcoming DSM revision. I think this article makes a good case that specificity in psychological diagnoses hasn't helped, and in many ways has made matters worse:
Having dealt with what is often referred to colloquially as "grief" - which had some accompanying depression and "adjustment disorder" symptoms; I would say that the "emotion handler" has it's own way of dealing with severe emotional pain. Talk with people in a grief group. They'll talk about the "stages of grief" or the "emotional rollercoaster".
There are periods of time where one goes for days or weeks in a state of emotional numbness. Then, some event or thought or memory triggers something, and one finds oneself experiencing the intense emotional pain, sometimes manifesting as anger, sometimes being dealt with by defense mechanisms of denial, guilt, etc. Some describe this as a cyclic experience, as if the emotional mind is processing the trauma, a little bit at a time, and when it tires, or when it becomes necessary to deal with more urgent life matters, the feelings become dulled - often ALL feelings.
So - as a learning mechanism, emotional pain? I'm not sure it's as simple as that. I don't think it is as simple as it's physical analogue. Maybe I'm confusing depression with emotional trauma, or post-traumatic stress (for which, depression-symptoms are often a side-effect). But there is a primitive survival mechanism at work, and it feels a lot like - loss of security, loss of a familiar feeling of control and safety, and learning to accept the new reality, and training the brain to recognize that the time when everything got turned upside down, whether it was the death of a loved one, loss of a job, physical loss, whatever, often, there are really deep misconnections made to other traumatic memories, other seemingly unrelated things, that tend to bring the pain back to the surface at unwelcome times, cause one to ruminate ceaselessly.
When you read the recovery literature, there's a process of training, to overcome the misconnections, and to accept that the feelings of insecurity and fear are irrational connections of the present to a past trauma. And it sounds simple - but not when you're in the dark. I don't know what function the pain serves there. If there's an evolutionary purpose, it's probably for short-term survival, and long term? You're probably supposed to have reproduced, and now you can just shrivel up and die. Evolution doesn't care after that.
Readers following this may be interested in my post on this whole kerfuffle at my blog, Neuron Culture: Does depression have an upside? It's complicated.