The correlation/causation conundrum is a particularly frustrating one in the social sciences due to the complex interaction of variables related to human experience.
I've found looking at time-order and thinking of variables-as-events is a helpful way to simplify experimental design seeking to get at causal mechanisms in my behavioral research.
Take the smoking example:
I would consider measuring changes in strength of correlation at various points in an ongoing experiment.
Once a baseline measurement is obtained from those already smoking subjects/participants, we measure the correlation between avg. number of cigarettes smoked per weak and lung capacity. This way one doesn't have to randomize or control, unethically asking people to smoke if they don't already. We already have a hypothesis based on the prior that volume of cigarettes smoked has a strong positive correlation with lung damage, and so reducing the number of cigarettes smoked would improve lung functioning in smokers.
But here we assume that the lifestyles of the smokers studied are relatively stable across the span of the experiment.
The researcher must take into account mediating factors that could impact lung functioning outside of smoking - i.e Intermittent exercise and lifestyle improvements.
In any case, following the same group of people over time is a lot easier than matching comparison groups by race/age/gender/education, or any of the other million human variables.
I would add than a narrow focus on quantifiable data can be limiting, especially when you are researching culture and doing content analysis. Coding is a way to convert that content into numbers - counting mentions of words or themes - but that requires a lot of qualitative analysis to begin with and certain aspects are often lost in translation. Any social scientist worth their salt will take into account as many biases as they can and devise an experimental design to control for them as much as possible. But again, you have to be prepared to defend why you did what you did and human nature is complicated.
Having said that, I think a mixed qual/ quant designs are pretty great, especially when you have meta analyses to back you up.
RationalWiki is a troll site, but that could be a good thing - it gives LW the opportunity to show exactly why it isn't a cult. Yvain makes very convincing counter-arguments and other posters have done a superb job of keeping their cool while answering various inquiries. Some of the questions directed at SIAI and Eliezer are not without merit and can't be dismissed so easily.
Anecdotally speaking, I've been around a few unimaginative 'bro-grammer' types fixated on SIAI - and particularly Eliezer. The individuals I'm thinking of know just enough about computers to feign credibility on reddit while making fun of a group similar to them, yet more eccentric. Mocking Eliezer, a successful nerd with more celebrity than they could hope for, gives these guys a snarky sense of superiority. Kinda dumb.
This is just my impression of where some of this might be coming from.
I really like the way you phrased "meta-suffering" as a term for the many cognitive self-defeating cognitions. The "rumination" symptom commonly observed in people with mood and anxiety disorders (a.k.a "dwelling") seems to be a related condition. Some Buddhists call it addiction, or attachment, to suffering.
The diathesis-stress model is a my favorite way to analyze to mental illness, including depression. In other words, I think depression is a heritable, biological phenomenon and the correlated cognitive biases create a feedback cycle - especially when you factor in the influence of environment and life experience. The cognitive biases on their own aren't enough to cause a depressive episode, and a depressed person may not hold these same biases when their condition subsides for a period.
That being said, cognitive biases accompanying depression have been studied quite a bit. This would save you the trouble of going through an IRB to create your own study, Shannon ; ) You could always do a meta-analysis, though!
Based on Beck's Cognitive Model, 6 primary biases emerge in depression/cognition literature.
These constructs are not definitive, as there are many theoretical models even within cognitive psych. Meta-suffering could incorporate many of these biases.
(please excuse the examples, some of them were taken directly from articles and others were re-worded to make more sense - at least to me)
A) ARBITRARY INFERENCE Drawing specific conclusions in the absence of relevant evidence (i.e. "The bus driver was driving like that because he's taking drugs".)
B) SELECTIVE ABSTRACTION Drawing conclusions on the basis of isolated details of an event, even if it requires ignoring other contradictory evidence. (i.e "She said she had a really good time, but didn't like the gift I got her. She won't to go out with me again")
c) OVERGENERALIZATION Holding extreme beliefs based on a specific event and inappropriately applying them to dissimilar occasions or settings. (*i.e " He was scared by that lizard. He must be afraid of animals.")
d) MAGNIFICATION Overestimation of the significance of events (i.e. A friend of mine got robbed, this world is a dangerous place).
e) PERSONALIZATION Relating events to oneself despite their being no apparent connection (i.e He got into a car accident last week. It's my fault I didn't come over to hang out with him that night.)
f) DICHOTOMOUS THINKING Thinking in all-or-nothing terms, categorizing experiences only in one of two extremes rather than acknowledging grey areas. ( i.e " I don't play sports with my kid. I'm a terrible parent.")
---- some of the above was taken from the articles cited below. for a more complete review of cognitive therapy for depression, check out the work of Aaron Beck.
White J, Davison GC, Haaga DA, White K. (1991) Cognitive bias in the articulated thoughts of depressed and nondepressed psychiatric patients. Journal of Mental and Nervous Disease, 180, 77-81
Krantz SE, Gallagher-Thompson D (1990) Depression and information valence influence depressive cognition. Cognitive Therapy Research14,95-108.
Beck AT (1987) Cognitive models of depression. Journal of Cognitive Psychotherapy 1, 5-37.
Environment matters a lot. Bars (for example) are loud, dark, sometimes crowded, and filled with inebriated people.
I CANNOT STRESS THIS POINT ENOUGH:
Thinking in terms of "picking up women" is the first problem. One should take the approach that they are "MEETING women". The conceptual framing is important here because it will influence intentionality and outcome.
A "meeting" mindset implies equal footing and good intentions, which should be the foundation for any kind of positive human interaction. Many women are turned off by the sense that they are speaking to a man who wants to "pick them up", perhaps sensing that you are nervous about adding them to your dating resume. It's hard to relate to that.
Isn't the goal to engage romantically with a peer, maybe learn something about relationships.
With that little rant out of the way, I think its important to think of where you are best able to have a relaxed and genuine conversation - even with a friend.
If you see a woman at the bar that is especially attractive and worthy of YOUR attention, perhaps admit to her candidly that the location is not your milieu and ask inquisitively if she normally has good conversations at bars. If she says yes and stops at that, chances are she's not interested in talking more with you or simply is not a good conversationalist.