An interesting story -- about science, what gets published, and what the incentives for scientists are. But really it is about whether you ought to believe published research.

The summary has three parts (I am quoting from the story).

Part 1 : We were inspired by the fast growing literature on embodiment that demonstrates surprising links between body and mind (Markman & Brendl, 2005; Proffitt, 2006) to investigate embodiment of political extremism. Participants from the political left, right and center (N = 1,979) completed a perceptual judgment task in which words were presented in different shades of gray. Participants had to click along a gradient representing grays from near black to near white to select a shade that matched the shade of the word. We calculated accuracy: How close to the actual shade did participants get? The results were stunning. Moderates perceived the shades of gray more accurately than extremists on the left and right (p = .01). Our conclusion: political extremists perceive the world in black-and-white, figuratively and literally. Our design and follow-up analyses ruled out obvious alternative explanations such as time spent on task and a tendency to s... (read more)

Ugh. I am generally in the unsympathetic-to-PUA thinking camp, so I offer the following not to bring up a controversial subject again, but because I think publicly acknowledging when one encounters inconvenient evidence for one's priors is a healthy habit to be in...

Recently I added the following (truthful) text to my OkCupid! profile:

Note, July 2013 -- I can't claim to be in a relationship yet, but I have had a couple of dates with a someone who had me totally enthralled within 30 minutes of meeting her. I'm flattered by the wave of other letters that have come in the past month, but I've put responding to anyone else on hold while I devote myself to worshiping the ground she walks on.

Having noted that I am a)unavailable and b)getting lots of competing offers, a high status combination, the result is... in three days, the number of women rating my profile highly has gone from 61 to 113.

I've noticed a few times how surprisingly easy it is to be in the upper echelon of some narrow area with a relatively small amount of expenditure (for an upper middle class American professional). This is easy to see in various entertainment hobbies- an American professional adult who puts, say, 10% of his salary into Legos will have a massive collection by the standards of most people who own Legos. Similarly, putting 10% of a professional's salary into buying gadgets means that you would be buying a new one or two every month.

I recently came across an article on political donations and saw the same effect- to be in the top .01% of American political donors, it only takes about $11k an election cycle (more in presidential years, less in legislative only years). Again, at 10% of income, that only takes an income of ~$55k a year (since the cycles occur every two years), which is comparable to the median American salary (and lower than the starting salaries for most of my friends who graduated with STEM bachelor's degrees).

It's not clear to me what percentage of people do this. It's the sort of thing that you could only do for a few narrow niches, since buying a ton of Legos impedes ... (read more)

After a short discussion on irc regarding basilisks I declared that if anyone has any basilisks that they consider dangerous or potentially real in anyway to please private message me about them. I am extending this invitation here as well. Furthermore, I will be completely at fault for any harm caused to me by learning about it. Please don't let my potential harm discourage you.

So according to this article a large factor in rising tuition costs in American universities is attributable to increases in administration and overhead costs. For example,

Over the past four decades, though, the number of full-time professors or “full-time equivalents”—that is, slots filled by two or more part-time faculty members whose combined hours equal those of a full-timer—increased slightly more than 50 percent. That percentage is comparable to the growth in student enrollments during the same time period. But the number of administrators and administrative staffers employed by those schools increased by an astonishing 85 percent and 240 percent, respectively.

Certainly some of these increases are attributable to the need for more staff supporting new technological infrastructure such as network/computer administration but those needs don't explain the magnitude of the increases seen.

The author also highlights examples of excess and waste in administrative spending such as large pay hikes for top administrators in the face of budget cuts and the creation of pointless committees. How much these incidents contribute to the cost of tuition is somewhat questionable as the evi... (read more)

Question: where can I upload jailbroken PDFs that is public & Google-visible?

For a job, I compiled ~100MB of lipreading research, some of them extremely obscure & hard to find (I also have some Japanese literature PDFs in a similar situation); while I have no personal interest in the topic and do not want to host indefinitely the PDFs on gwern.net, I feel it would be a massive waste to simply delete them.

I cannot simply put them in a Dropbox public folder because they wouldn't show up in Google, and Scribd is an abomination I despise.

(crosspost from Google+)

Open comment thread, Monday July 29th

If it's worth saying, but not worth its own top-level comment in the open thread, it goes here.

Regarding the obvious recursion, please note that jokes are generally only funny the first time. :)

Does anyone know why GiveWell is registered with the IRS under a different name (Clear Fund)? I am including a link to their recommendation for the AMF on a wedding registry and have already gotten a question about about why their name differs.

Any LW readers living in India? I recently moved here (specifically, New Delhi) from the United States and I'm interested in the possibility of a local meet-up.

I typed up the below message before discovering that the term I was looking for is "data dredging" or "hypothesis fishing." Still decided to post below so others know.

Is there a well-known term for the kind of error that pre-registrations of scientific studies is meant to avoid? I mean the error where an experiment is designed to test something like "This drug cures the common cold," but then when the results show no effect, the researchers repeatedly do the analysis on smaller slices of the data from the experiment, until eventually they have the results "This drug cures the common cold in males aged 40-60, p<.05," when of course that result is just due to random chance (because if you do the statistical tests on 20 subsets of the data, chances are one of them will show an effect with p<.05).

It's similar to the file drawer effect, except it's within a single experiment, not many.

As I understand applying Bayes to science, the aim is to direct research into areas that make sense. However, sometimes valuable discoveries are made by accident.

Is there any way to tell whether your research is over-focused? To improve the odds of noticing valuable anomalies?

In the past, people like Eliezer Yudkowsky and, I think, Luke Meulhauser have argued that MIRI has a medium probability of success. What is this probability estimate based on and how is success defined? I've read standard MIRI literature (like "Evidence and Import" and "Five Theses"), but I may have missed something.

I have a question about the Simulation Argument.

Suppose that it's some point in the future, and we're able to run conscious simulations of our ancestors. We're considering whether or not to run such a simulations.

We are also curious about whether we are in a simulation ourselves, and we know that knowledge that civilizations like ours run ancestor simulations would be evidence for the proposition that we ourselves are in a simulation.

Could the choice at this point whether or not to run a simulation be used as a form of acausal control over the probability that we ourselves are living in a simulation?

Waffled between putting this here and putting this in the Stupid Questions thread:

Why is the default assumption that a superintelligence of any type will populate its light cone?

I can see why any sort of tiling AI would do this - paperclip maximizers and the like. And for obvious reasons there's an inherent problem with predicting the actions of an alien FAI (friendly relative to alien values).

But it certainly seems to me that a human CEV-equivalent wouldn't necessarily support lightspeed expansion. Certainly, humanity has expanded whenever it has the opp... (read more)

If I believe that automation causing mass unemployment is around the corner (10-20 years), what do I do or invest in now to prepare for it?

What is the function of the karma awards page?

Warning: politics, etc., etc.

What do conservative political traditions squabble over?

My upbringing and social circles are moderately left-wing. There's a well-observed failure mode in these circles, not entirely dissimilar to what's discussed in Why Our Kind Can't Cooperate, where participants sabotage cooperation by going out of their way to find things to disagree about, presumably for moral posturing and virtue-signalling reasons.

In recent years I have become fairly sceptical of intrinsic differences between political groups, which leads me to my openi... (read more)