The review explains why the rapid action of ketamine excites so many researchers:
The discovery that ketamine rapidly increases the number and function of synaptic connections has focused attention on synaptogenesis as a fundamental process for the treatment of depressive symptoms and also suggests that disruption of synaptogenesis and loss of connections underlies the pathophysiology of depression.
This excites researchers not because ketamine itself would be used to combat depression, but because depression is still extremely symptomatically defined, making it difficult to design treatments for. That's roughly how it's diagnosed in the diagnostic manual used by most psychiatrists: check off a list of symptoms, if you have enough, you're depressed. It's like going to the doctor and explaining that your stomach hurts and they say "well, looks like you have abdominal pain, here's some Advil." Treating the symptoms would be great if only there were a happiness dial in the brain. Indeed, the effect of most anti-depressants is often demonstrated prior to a mechanistic understanding of why they make many patients feel better.
Recently there has been substantial evidence of "synaptogenesis" - the formation of new potential connections between neurons - from multiple treatments, including ketamine. So now we have this new picture emerging: depressed patients tend to have atrophied and "less-connected" neurons in some brain areas, and some drugs can reverse it, in particular ketamine can reverse it quite rapidly, and it works in rodents as well as humans.
That makes it very amenable to study. The way this often works in the lab is the following. Take some rodents, subject them to unpredictable stress to get them depressed, then give some ketamine. It makes them better. Euthanize the rodents, slice the brains, and note that the non-ketamine ones have less dendritic spines in certain areas ("potential input points to a neuron"), but remarkably, the ketamine ones have more in those areas.
The most important step comes next, where you try to find out what ketamine is actually doing, since, again, there's no happiness dial in the brain. Create strains of "knock-out" rodents, where you block the production of certain chemicals or proteins you think ketamine might affect by altering their genetic composition. This step is crucial, because it allows you to find out which effect of ketamine is providing the benefit, because there are many. You can do this both by observing both behavior (does the ketamine not improve mood in the genetically altered rodents?) and in physiology (does the ketamine still increase synaptogenesis in the altered rodents?).
In the end you can kind of work out a map of sorts: ketamine does X things to the brain and Y in X are the ones that are important, sometimes in certain combinations. Then you can start creating intelligent drugs that pinpoint those important processes, to avoid the unfortunate side effects of drugs like ketamine. Moreover, you now have a better physiological understanding of depression, instead of just a symptomatic one.
To put it in machine-learning language, it's like going from ideal observer analysis like mutual information, to an actual parametric model where you understand the distributions themselves.
This blew me away, but I only now realized why. It was because I knew depression was symptomatically defined, and she had probably found a way to define it physically. I thought that the "lesser" cases she couldn't detect weren't actually physically depressed, and that her discovery was amazing. Other students didn't share my enthusiasm, but maybe that was because it wasn't very interesting from a machine learning perspective.
I wonder what happened with that. I can't remember her name now.
EDIT: I found the paper: http://discovery.ucl.ac.uk/1316862/
The key finding seemed to be predicting whether or not treatment would be successful. This is quite powerful as it would help guide clinicians in figuring out whether or not a specific treatment regimen would be advisable or just a waste of time (and time is critical in treating depression).
You manage to infer, as well as the author of the OP, that depression decreases the number of synapses or eventually induce a sort fatigue in synapses because of stress. That depression is not only correlated but also induced by synapse depletion, that increasing the number of synapse is what makes depressed people feel better, that depression (the symptom) has one cause, here stress inducing synapse fatigue or depletion, etc.
It takes just common sense thinking to understand that most of this reasoning is just crap. Suggesting to put that in a machine learning language is is even more depressing.
Could you be so kind as to elaborate on some of this common sense thinking? I must lack the same common sense you possess, because it seems to me that the parent comment makes a compelling case in favor of further study of the synaptogenesis hypothesis. I did not get the impression that kevinalexbrown was claiming that depression is caused by depletion of synapses, but rather summarizing for the record that this is the model suggested by the observed effectiveness of ketamine.
I'm confused. As I read the post, it contains the bits "The most important step comes next, where you try to find out what ketamine is actually doing, since, again, there's no happiness dial in the brain. ... This step is crucial, because it allows you to find out which effect of ketamine is providing the benefit, because there are many."
While the possibility you take as 'assumed' is the jumping off point for the hypothesis, it is explicitly a testable hypothesis for which the necessary test to (in)validate the original assumption is explicitly defined.
So what exactly is the problem?
In religion and politics, people are judged based on their origins and their "authority". Not in science.
The OP should have posted sources or links, but apart from that, his remark needs to be evaluated on its merits, not his merits.
"Science is the organized skepticism in the reliability of expert opinion." — Richard Feynman
I for one am not about to go chase down every potential citation about a subject I don't know anything about just in order to fully evaluate his remark on its merits just because the subject is somewhat interesting to read about.
In that situation, authority and expertise is important - you are perfectly justified that on average you will do better if you take medical advice from a doctor than a random person, for example.
DO you really think science only takes place between the covers of a journal?
> Authority or expertise is not evidence of the truth of what someone writes, but for most of us, in the absence of the time or expertise to evaluate the evidence, it often serves as a useful shortcut to determine what amount of trust to assign to a statement.
Yes, understood. But not for a scientist.
> I for one am not about to go chase down every potential citation about a subject I don't know anything about just in order to fully evaluate his remark on its merits just because the subject is somewhat interesting to read about.
So, because you won't do research, someone has to prove themselves to you? You're going about this all wrong -- when you decide to be intellectually lazy, the worst thing to do as an encore is to demand that someone prove themselves to you, and then accept claims of authority. Jim Jones' followers died because they were unwilling to think for themselves, do their own research. David Koresh. The list goes on.
> In that situation, authority and expertise is important ...
The bottom line in science, and in any scientific discussion, is that authority and expertise are never important. The greatest amount of scientific eminence is trumped by the smallest amount of scientific evidence.
> you are perfectly justified that on average you will do better if you take medical advice from a doctor than a random person, for example.
That's not a scientific example, because doctors aren't scientists. I'm beginning to realize you can't distinguish between science and anecdote. As they say, the plural of anecdote is not evidence.
DO you really think we are doing science in this discussion? Seriously?
If you genuinely think so, then we can end the discussion right here, as if that is the case, we are arguing on an entirely different basis.
Just to be clear, my argument is based on my expectation of engaging in a discussion that is not particularly scientific, but is a casual debate about a scientific result where most of the participants are not scientists, nor have sufficient knowledge or interest in the field in question to have read much actual research in the field. That fits _my_ observation of this debate. I see little to no evidence that there's much "science" going on in this debate.
> Yes, understood. But not for a scientist.
This is blatantly false. Scientists frequently engage in discussions about fields they only have a superficial interest in too, and where they don't have the time nor the interest to probe deeper, and I have personally seen plenty of examples of scientists deferring to authority on subjects they do not personally know for all kinds of purposes. Yes, that is not "science". But we're not all engaging in science every moment of our waking lives.
> So, because you won't do research, someone has to prove themselves to you?
No. Because I am here reading about something I don't have sufficient interest in to research, I appreciate an indicator of authority as a a contributing factor to judge what degree of trust to assign certain claims.
We all do that pretty much constantly, and largely automatically. Your username, for example, means I assign you a certain amount of trust in matters I have reason to believe you know something about. E.g. if you were to make statements about AppleWrite in conversation on HN, I'd likely take it on trust. Does that mean I'd be unwilling to do research about it in certain cases?
Of course not. If I was writing a scholarly article about the history of word processing, for example, of course I would obtain multiple sources and track down other evidence.
But I am not here conducting science, neither I'd gather are the vast majority of other people here. And then trust and authority does matter as shortcuts. I have plenty of other things to spend my time on. Things I _care_ about, which is not something I can say about neuroscience.
> Jim Jones' followers died because they were unwilling to think for themselves, do their own research.
Did you seriously compare someone asking for indicators of authority from someone who made statements that appeared to be a bit controversial to not being willing to think for oneself before joining a religious cult and participating in mass suicide? Are you for real?
Deferring to authority and assigning trust are not inherently bad just because there are nasty examples of the consequences of taking it too far.
For someone who harp about science, you've made some quite astounding leaps of logic.
> That's not a scientific example, because doctors aren't scientists.
Irrelevant. The issue here is not science, but whether trust and authority have value in determining whether or not it is necessary to do your own basic research of everything.
> The bottom line in science, and in any scientific discussion, is that authority and expertise are never important. The greatest amount of scientific eminence is trumped by the smallest amount of scientific evidence.
If this was a "scientific discussion" perhaps you'd have a point.
The reality is that in a casual discussion, - which is what this is to me, and clearly to a substantial number of the other participants, given the widespread lack of citations and evidence in this discussion - where a large number of the participants are either not scientists and/or not deeply familiar with the subject in question, nor invested enough to find it worthwhile to chase down source material, often it is not clear what the evidence - if presented - actually is and what it actually means to a lay person.
As such, authority and expertise is an important shortcut to get an overview for the purpose of the discussion without bogging the discussion down to the point where the participants lose interest.
> I'm beginning to realize you can't distinguish between science and anecdote. As they say, the plural of anecdote is not evidence.
I'm beginning to realize that you don't understand that in a social conversation, rather than, say, a scientific conference, evidence is not paramount, and often isn't even that important. That's not to say that some relevant evidence and citations are not appreciated, but most of us will not read them. That's not why people participate in these discussions. If we wanted in depth neuro-science, for example, we'd be busy reading the research not discussing a press-release on HN.
Good luck proving that your breakfast isn't poisonous before you eat it tomorrow. I hope you don't starve to death first.
Here is a quick rewrite of the snarkiest bit of the post that I think is not only drastically more civil but just as effective:
"I fear the reasoning here is incorrect as it conflates correlation with causation. Being too hasty in using this conclusion with machine learning would be a mistake."
I know there is a segment here that doesn't believe tone/communication style should matter, that it is, or should be, entirely about the underlying verisimilitude of their statements; unfortunately, that is not how 99% of human beings actually operate. So while you are free to rail against reality it will make you drastically less effective than you might otherwise be.
To my defense, I could say that English is not my mother language and that the conflation of correlation with causation error, as you say, is a source of strong irritation to me. ;)
Thank you for your clarifying and kind comment.
Okay, but first, locate and circle the words you have just assigned to me in my original post. What? Can't find those words? Know the expression "straw man"?
I ... never ... said .. any ... such ... thing.
Circle the words on the screen in front of you, where I ever uttered these words, anywhere, on God's green Earth.
You need to start reading what people say, not what your overactive imagination believes they said.
> ... seems to also imply you view this as a debate instead of a discussion ...
When someone invents a position to conveniently argue against, of course it's a debate -- a debate between fantasy and reality.
Notice how I have in each and every case replied directly to your words, which I quote in full. Notice how you don't have time to check in with reality before ascending your soapbox.
Can you then perhaps elaborate how else we should take this quote from you:
> I suggest that you resist posting this kind of inquiry in a scientific discussion. In science, only evidence counts, not authority or expertise. In science, the quality of evidence is all that matters, not its source.
As well as this in your reply to me:
Of course you might argue that this is not a "scientific discussion", but that would make both of these messages meaningless or intentionally obtuse in the context they were posted.
Getting back to the facts, there is nothing allowing us to think that depressed people feel better when taking ketamine BECAUSE the number of synapses increases when taking ketamin.
A possible correlation is presented as a cause relation. Was it verified on the same people ? If not we can't even consider this to be a true correlation.
The OP goes even beyond that, but this should be enough to justify my critic.
(pseudo) scientific PR are leveraging this common logical error for their promotion to such point that people get totally blind to it. This is debilitating and should be classified as crap information.
This seems to be a strange use of the word "allowing". I would say that the evidence presents an interesting and testable hypothesis about the role of synaptic connections in depression. The obvious next step would be to find other methods of inducing similar effects on synapses by means other than ketamine and see if they have similar effects on depression.
Out of interest, what do you think might be some of the biological mechanisms associated with chronic depression and how would you seek to investigate them?