« first day (19 days earlier)      last day (4460 days later) » 

12:17 AM
@digitaloday don't mean to draw attention away from cogsci, but ling.SE might better answer your question.
@Josh and @JeromyAnglim thank you for your feedback. I am not worried about my question not receiving an answer, what I am worried is this sudden spike of bad answers. cstheory is the ideal to which I hope cogsci would strive (but people seem to disagree on meta) and there it is perfectly fine for questions to go unanswered for months... because they sometimes become real research questions (in fact, papers had been published off cstheory questions)
@JeromyAnglim I like your edits, but I prefer the older title. This title is definitely more descriptive, but I want to promote a culture of asking "broad" (but still scientific and answerable) questions like my title, but expecting concrete answers. I feel that the body and tags are sufficient to explain that formal models are required.
 
I just wanted to share this bit of positive publicity from one of our users lesswrong.com/r/discussion/lw/9t1/…
2
@ArtemKaznatcheev as with any of my edits, feel free to change back; I don't mind one bit. I edit questions quite a lot and hope that no one takes offence.
 
@digitaloday regardless both are very much on topic on this site :)
@ArtemKaznatcheev Forgot about that site...where does linguistics fall within our scope? I think it was mentioned on Area 51 that linguistics would be in the scope but I don't know if ling.se existed then
I would think that computational lingustics and NLP would be better suited for cog sci though, perhaps...was never one for linguistics
2
Q: The neuroscience and neurobiology tags

Artem KaznatcheevThe neuroscience tag was recently introduced in this question. We already have a neurobiology tag that is actively used. However, most of those questions could/hold be tagged neuroscience. In fact, the early questions I asked were only tagged as neurobiology because there was not a general neuros...

Per that question I suggested a tag synonym for neurobiology but I don't know which way tag synonyms work...
I assume @Josh knows better than I do on that
 
12:52 AM
I +1'd the synonym
@Kaj_Sotala thanks for sharing the link on lesswrong.
 
1:10 AM
@BenBrocka linguistics is significantly ahead of us in development. But it is suffering form relatively low number of questions per day. I would not try to steal questions from them (full disclosure: I have more rep on ling.SE, but I expect to contribute more to cogsci in the longrun) and topics like computational ling and NLP are definitely more ontopic at LING than at CogSci....
NLP in particular, is basically statistics and machine learning now-a-days and has nothing to do with cognition or cognitive modeling
the particular question that was asked @digitaloday is probably more ontopic for cogsci since it deals with inferring a mental state from text, however it would be on topic at linguistics as well.
I think it is healthy for cogsci and linguistics to try to share a userbase and interact a lot with each other. It can kind of be like the pairing between the two research-only stackexchanges: cstheory and theoreticalphysics.
For instance, since my question on models of child learning has not generated much traction here, and since one of the comments was to restrict the scope. I might post a restricted scope (restrict from general learning to language-acquisition) and repost on ling.SE with links back to the original here.
 
1:26 AM
@JeromyAnglim I made some changes to your changes to the neuroplasticity question. I reverted the title, but kept your bolding and tried to make it more clear that the critical-period references were not meant to retrict my scope to language-acquisition.
Thanks for the feedback, everybody!
 
Posted by Joel Spolsky on February 6th, 2012

Stack Exchange co-founder Jeff Atwood announced that he is leaving the company to spend more time with his family, including his twin daughters Maisie Jane (5lb2oz) and June Adeline (5lb 7oz) who celebrated their 0th birthday (and joined Twitter) last Friday, to the great joy of their parents.

It has been a great honor for all of us to have worked with Jeff over the last four years as Stack Exchange grew from absolutely nothing to a world-changing resource with over 30 million monthly visitors.

When I first met Jeff, I told him that when Stack Overflow was built, it would become a standard part  …

 
1:50 AM
@BenBrocka sorry was AFK
And am on the phone now
will check in later
also WTF? JEFF
 
2:25 AM
1
Q: How to engage with graduate students and academics in psychology and cognitive science?

Jeromy AnglimWe already have this general question on How to promote the site and reach out to industry experts?. However, the question is fairly general. Most of the experts in psychology and cognitive science are academics. And there would also be an large group of students completing a PhD or masters the...

 
2:50 AM
0
Q: How can we encourage existing users of cogsci.se to announce the site on Google+?

Jeromy AnglimI shared a little information about the site on Google+. I made sure the link was "Public". However, I imagine other existing users of this site may have both better and different social networks to me. @ArtemKaznatcheev also suggested to me that there may be quite a few academics on Google+. Obv...

 
3:38 AM
0
Q: What are cost effective strategies for site promotion at an academic conference?

Jeromy AnglimI previously asked How to engage with graduate students and academics in psychology and cognitive science?. One suggestion was to promote the site at an academic conference. I've read that there is a potential budget for promotion. We can come up with budgets and promotions but the means and...

 
4:33 AM
@Josh I know way too many people who have stopped doing awesome things once they have kids...I'm happy for him, but, damn
 
 
4 hours later…
8:17 AM
0
Q: NLP tag "dispute" resolution

jonscaRecently, two questions were tagged nlp for Neuro-Linguistic Programming. Today, there was a question asked about natural language processing, and the user assigned the nlp tag as well. In a brief exchange in the comments of today's question, zergeylord brought up the point that this should be ...

 
9:13 AM
@BenBrocka He'll be back in a couple of years I suppose. :)
 
 
3 hours later…
12:30 PM
@BenBrocka I was AFK last night shortly after reading that. I'd say having kids and being there to raise them is a pretty awesome thing in and of itself. But I agree with @StevenJeuris, I seriously doubt we've seen the last of Jeff :-)
@BenBrocka I know how tag synonyms work, it's a mod ability. Which tag we should choose is something I don't feel comfortable deciding here, because I don't know if all of falls within . I would assume so, but I am not sure
@JeromyAnglim that's great! Have a star
@ArtemKaznatcheev If your question receives a large number of bad answers, downvote them and leave a comment explaining why they're bad. This should hopefully discourage others from posting bad answers. If it keeps happening you may need to edit your question to make it more clear what you're looking for. If you've done all that, as a last resort you can flag your question and a mod can protect it for you
17
Q: Protect a question from being spaminated, but allow legitimate users to carry on

randomLet moderators offer a protection status visa to certain questions under constant non-answer fire. Much like Wikipedia, this would allow users above a certain level of reputation (but at what level?) to be able to continue on answering/editing questions, which would normally have been locked. ...

 
 
1 hour later…
2:10 PM
@Josh On non-SO sites it's something decided by people who can vote on tag synonyms :P
Especially ones that don't have dedicated moderators at all
And neuroscience is often seen as a branch of * biology*, so I certainly think all neurobio falls under neuroscience (which is why some people mistakenly think it's "real science" and will thus "replace" psychology)
@ArtemKaznatcheev that sounds reasonable
 
2:41 PM
@BenBrocka huh, I did not know this actually!
Tags are one area of StackExchange which I am not as familiar with as the others, hence my weak tag wikis and unfamiliarity with tag synonyms
 
Ah. Well anyone with...1k? rep can vote on them on this site
...come to think of that, that means me and jeromy are the only ones who can vote so that doesn't really help
 
lol
Ofri is close
 
our vistitors/day is up by 60 in 2 days
 
@BenBrocka Good job! Those links seemed to help.
 
Awesome
 
2:51 PM
:( I still miss the publicist badge I could have gotten. Shared the wrong damn link
 
@BenBrocka I tried hard to get the publicist badge, but was denied
 
I think that incubation question would have done it for me...our most popular/highest voted question
it's at 2k views now
 
yeah that's the one I was trying to share, I posted it to facebook but I guess I don't have enough friends :-)
@Ben just saw this:
> An interesting contribution from a redditor:psych.lancs.ac.uk/people/uploads/tomormerod20090227T152723.pdf – Ben Brocka Feb 3 at 22:26
Reading now
 
3:07 PM
@Josh FB is useless for single link shares, you need hundreds of them. with reddit a good link can get thousands of views
 
yes, very true
I've never been a reddit user...
 
Neither have I, I just know some of it's power
 
3:21 PM
btw relatively our links to reddit got crazy high votes
 
3:47 PM
Cool .... my virtual desktop is working! Almost time to call it a day.
 
4:05 PM
The true Stack Exchange policy, revealed:
31
A: If Jeff Atwood's moving on, who will be "in charge"?

Joel SpolskyBecause we massively overworked Jeff as a matter of corporate policy, and because he wore big clown-like shoes, it will take at least three people to fill his shoes completely. Josh "Shog9" Heyer, Stack Overflow user #811, has overall responsibility for Q&A quality and will be the public fac...

> As always, it's a community, and authority derives from you, our users, even if you're sometimes idiots and we have to overrule you.
XD
 
1
Q: NLP tag "dispute" resolution

jonscaRecently, two questions were tagged nlp for Neuro-Linguistic Programming. Today, there was a question asked about natural language processing, and the user assigned the nlp tag as well. In a brief exchange in the comments of today's question, zergylord (to credit him, not to single him out) bro...

Thoughts? I had never heard of neuro linguistic programming as "NLP". IMO super tiny acronyms should only be used where 100% unambiguous, apparently this isn't the case for NLP?
 
4:25 PM
Yeah I always thought NLP was "Natural language Process[or|ing]"
Although not according to Wikipedia:
Neuro-linguistic programming (NLP) is an approach to communication, personal development, and psychotherapy created in the 1970s. The title refers to a stated connection between the neurological processes ("neuro"), language ("linguistic") and behavioral patterns that have been learned through experience ("programming") and can be organized to achieve specific goals in life. According to certain neuroscientists, and linguists, NLP is unsupported by current scientific evidence, and uses incorrect and misleading terms and concepts. The founders of NLP, Richard Bandler and John Grinder, sa...
I vote two separate tags, and ban the tag
Actually both are NLP in Wikipedia:
NLP may refer to: * Natural language processing, a field of computer science and linguistics concerned with the interactions between computers and human (natural) languages * Nonlinear programming * Neuro-linguistic programming, a method applied to interpersonal communication, organizational change and psychotherapy used to help people change their behaviors ** Neuro-linguistic psychotherapy, an approach to psychotherapy based on Neuro-linguistic programming * No light perception, or total blindness, the complete lack of form and visual light perception * Ninein-like protein, a human gen...
0
A: NLP tag "dispute" resolution

JoshMy thinking is, if it's ambiguous, we should blacklist nlp and enforce the use of neuro-linguistic-prog and natural-language-processing. I suppose a well worded tag wiki for nlp could do the trick but it is my experience that most people don't read, so, we should have two clear tags and avoid th...

 
4:41 PM
Someone in linguistics says their NLP tag never caused confusion, they use it for nat lang proc
 
4:59 PM
@BenBrocka Is Neuro-linguistic programming on topic on that site though?
 
It's linguistic...I'm not sure if they've gotten and Qs on it though
NL programming is largely discredited
I'm kind of surprised we got 2 questions in a row about it
 
 
4 hours later…
8:36 PM
we seem to have more visitors per day (189 vs 172) than linguistics right now, I wonder why that is... is that just the novelty bump?
 
@ArtemKaznatcheev possibly
We have had a lot of users trying hard to publicize our site, which is great!
 
Hopefully we can keep it up :D
We are also ahead of Philosophy (which has been around for 8 months!) and History and Economics (both for 4 months)
 
8:56 PM
woo hoo!
 
9:34 PM
@ArtemKaznatcheev Don't think so, we've been in public beta long enough that the visitors per day metric should be reasonably representative
it's a median, not a mean, so spikes don't affect it unless we have several in a week? period
 
9:45 PM
@BenBrocka That means we're doing really well!
 
Just a quick question. Anyone of you knows whether a PhD always follows the same format?
E.g. is it possible to orient yourself more on implementation issues and less research issues?
(As long as the implementation issues address at least a research issue of course.)
 
I assume implimentaqtion would be acceptable
 
There might be a PhD position opening up related to the thesis I am currently doing. It's really a topic which interests me 100%, but I'm afraid doing a PhD just isn't my thing.
... if I would even make it through the application. :) Well, I could if it's implementation oriented. I'm not that proficient in research so then I might not.
 
@StevenJeuris it all depends on the university and your prof. However, at any decent school it is expected that a PhD (and even just a Masters) thesis contribute significant original research.
However, it is possible to end up doing research while primarily focused on implementation questions
 
@ArtemKaznatcheev Well, the implementation is definitely meant to address original research, as is my current thesis as far as I checked existing research.
It's just the evaluation of the implementation (statistics and such) which I can't be bothered with too much.
 
9:58 PM
If you are a good programmer, decent mathematician, and mildly personable, then it is easy to "do research" by helping Cognitive Scientists with their computational projects
 
3
Q: How to efficiently locate existing psychology and social science measures?

IndoleringWhenever I start a new research project, I typically have to find questions and scales that have proven validity in measuring constructs of interest (e.g., psychological / political science / social science scales). This can often be quite time consuming, and often there are issues of cost and co...

srnk "pretty rad of you" in the comments
 
It is hard to say how an implementation without an analysis of its performance and it doing the right thing is useful
however, if you are strong in one area, but weaker in another, the best way to keep working is by finding co-authors that compliment you
and cover the area you are weaker in
so in your case, it might be worthwhile to work with a statistician or mathematician (?) that can help evaluate your implementation
but really, it all depends on your prof :D
 
Hmm, I am doing a pretty basic evaluation of a UI prototype using a SuS questionnaire and Raw TLX.
But the prototype is only tested during one full day, followed by the SuS questions. And to measure performance only a 2 hour test followed by Raw TLX after each hour. (comparative study)
 
@StevenJeuris I don't know enough about the field you work in to comment on your methedology. I am a theoretical computer science that specializes in quantum computing and evolutionary game theory :P, I can only comment on general mindset of academia.
 
Fine for a masters, but just setting that up took me a lot of time, and interests me greatly less than doing the actual implementation.
@ArtemKaznatcheev Your input is appreciated though. ;p Thanks.
 
10:06 PM
I dunno, for me the funnest part of research is designing and coming up with ideas for new things
so I would be bored implementing other people's ideas
and kind of view those things as the "you have to do it sometimes" parts of the job
while the main focus is still coming up with new ideas or at least synthesizing existing ideas in new ways
*kind of view those (implementation) things
but it probably varies a lot from field to field
 
@ArtemKaznatcheev Oh, of course I am implementing my own ideas ... But it turned out a lot of those ideas were already explored, so I had to choose a few specific ones and work them out in detail.
It's the design/implement/experiment combo I like.
 
What's everyone's thinking on the feeds I added to the room? Do we like them? Do we hate them? Do we want them tweaked? Just let me know...
 
Just justifying all the design decisions in such great detail sometimes bothers me. Especially with HCI research. You're intuition/own user experience tells you a certain design decision could be preferable. But then you have to justify/clarify work procedures/objective measurements/ .... And don't get me wrong, I know those can be essential to doing good research, but often I feel they are just 'extra', and it's actually intuition which I get most merit from.
 
@StevenJeuris I agree that intuition is very important for everything, but sometimes it is fun (and worthwhile) to try to formalize your intuition and find ways to measure it.
 
10:26 PM
2
Q: Should we have a specific tag for questions that ask about the name of a study? And if so, what?

Jeromy AnglimWe had a question: What was the the study called that was on commitment and follow-through and involved college students attending a follow-up event? The essential features: the OP wants to know the name or reference to a study the OP knows that the study exists and has some details about t...

@ArtemKaznatcheev You should make your comment an answer. I'd up vote it.
Haha, funny how I originally wrote 'make your answer a comment'. Somehow that must be ingrained in my mind as I say that a whole lot more often.
 
@StevenJeuris I agree with . Better than !
 
@Josh Where did you get ? I prefer as in his comment.
 
@StevenJeuris it's a joke ;-)
 
@Josh Oh, ... sorry. In that case: "ha ha". ;p
 
is a joke that is :-)
A bad joke, clearly ;-)
 
10:30 PM
You can blame the lack of thorough communication through the internet. ;p
 
morning everyone :-)
 
Morning @Jeromy!
 
@JeromyAnglim Strange australians .... I should almost be getting into bed.
 
@StevenJeuris what time zone are you in again? UTC+5 or something?
 
Just a feeew ... more ... lines of code. :)
+2 I believe. It's 23:32 here.
 
10:33 PM
Cool, thanks. Trying to get a feel for what time it is for all our chatters
 
Oh Jeromy, we were interested in your timezone too. Don't feel offended and start removing posts now. :)
 
picking up on old discussion:
@BenBrocka you certainly deserved the publicist badge for the incubation question; I'm sure the reddit shares have increased the rate of increase of users.
I've noticed that there are quite a few reddit communities in addition to psychology that are relevant to our site. I've started a meta question: meta.cogsci.stackexchange.com/questions/132/…
You can see a list of existing shares here: reddit.com/…
 
@JeromyAnglim advertising on reddit is great for attracting more programmers, but I doubt we have problems with that since we are part of the SE network... is there anything to suggest that academics in cogsci use reddit?
@StevenJeuris okay, I will make it into an answer.
 
@JeromyAnglim yeah, that reddit question was huge...my other 2 attempts have yielded meh results.
 
@ArtemKaznatcheev I know the programmer subreddit is massive; however, there are niche communities specifically targeting topics relevant to this site: e.g., mathpsych is pretty academic and has around 1000 subscribers.
 
10:38 PM
@ArtemKaznatcheev there are psych subreddits, reddit has lots of communities including academic psychology people
 
@ArtemKaznatcheev see the list I've developed here: meta.cogsci.stackexchange.com/questions/132
 
@JeromyAnglim I'm missing HCI people on this site. That's the stuff for which I originally committed. :/ Are there any subreddits on that topic?
 
@StevenJeuris not really. There's a fairly dead /CHI subreddit
very little UX stuff on reddit either
 
@JeromyAnglim I agree, I wish we could get someone to assign it somehow. Alas I seriously doubt it
 
and if you want more HCI people, ask more HCI questions :)
 
10:42 PM
E.g. I would love to have CogSci input on questions like these:
 
@StevenJeuris he shared it yesterday...
 
21
Q: Adaptive UI vs. recognizable UI

Steven JeurisI'm reading a thesis which discusses Task-focused user interfaces. They discuss an application which adapts/filters views based on relevant items determined by an interaction history. They prove their interface (Eclipse Mylyn) results in higher productivity for programmers. Although I don't doub...

 
23 hours ago, by Jeromy Anglim
it's 10.33am; Melbourne: I guess +10 (althought it's daylight savings at the moment) or something
 
@StevenJeuris okay, I made the comment into an answer feel free to edit and refine it more.
 
@StevenJeuris I haven't looked. I wonder where they hang out on the web? Perhaps you could ask some seed questions on HCI. I know "seed" can be a dirty word on this site; but I think if you're honestly curious about the answer, it might get the ball rolling.
 
10:45 PM
Definitely, you're honestly curious about the answer, ask away!
 
@JeromyAnglim I know there are subreddits for our topics. But the topics are of interest to lots of non-academics. Is there a culture of using reddit in cogsci? (Kind of how there is a culture of blogging in math and theoretical computer science)
is what I meant to ask, haha
 
@StevenJeuris then ask them :P I should try and ask more of them
 
like, I can't imagine any of the people I work with hanging out on reddit
 
It's not a massive following of all cog sci people, but I don't know how well the community has embraced online/social things like that
 
@BenBrocka Okay, might do that tomorrow. For now I'm too tired to write proper english.
 
10:48 PM
Psych profs were certainly much less tech savvy on average than math,CS or HCI people
 
@ArtemKaznatcheev it's hard to gauge academic / graduate involvement because people generally post on reddit in quasi-anonymous ways.
 
@BenBrocka people tend to overestimate the tech savvy-ness of mathematicians and theoretical computer scientists
@JeromyAnglim do you have friends and colleagues that use reddit? That is an easy way to estimate.
although I guess it is a biased estimate
 
@ArtemKaznatcheev How can you do theoretical computer sciences and not be tech-savvy?
 
I haven't met many theoretical CS people really, I mean comp sci comp sci people, our university's CS dept was more down to earth, real world problems
even using computers and programming, almost like it was their job :P
 
@StevenJeuris TCS doesn't deal with tech, it is just a special flavor of math.
 
10:51 PM
@ArtemKaznatcheev I'm fairly well connected in the R / statistical computing world; a number of those people have reddit accounts; but I don't really talk to colleagues about Reddit.
 
@ArtemKaznatcheev Well, it does have to do all with algorithms no?
 
@JeromyAnglim okay. I am just being paranoid about not being able to attract academics, still. haha.
 
I always figured interest for Theoretical CS grew out of proving a certain algorithm is more effecient than another.
 
@StevenJeuris algorithms have nothing to do with tech, and algorithms are only a part of TCS, a huge part is lower-bounds (the dual/opposite of algorithms)
@StevenJeuris it is quiet the opposite
 
Older academics will certianly be a challenge. IMO we should reach out to the younger generation like grad/PHD students, people raised with computers
 
10:52 PM
@ArtemKaznatcheev fair enough; I guess I work on the numbers and the percentages. If you get 1000 hits, you might get 10 users and 1 meaningful contributor.
 
I was almost impressed that some of my psych professors used their email...
 
@StevenJeuris computer science started FROM TCS which was called mathematical logic and jut math at the time :P
*just
 
@BenBrocka I agree 100%
 
@BenBrocka Certain people also seem to be just more internet-savvy; I tend to see the same academics popping up on Google+, Twitter, and so on; they have blogs; they're more likely to contribute to sites like Stack Exchange.
 
Was never a TCS fan, hence why I'm in HCI
 
10:54 PM
@ArtemKaznatcheev Interesting, never realized that. But then again, my world view is primarily programming oriented. :)
 
TCS is so beautiful... it is like the natural continuation of analytic philosophy
you get to ask deep questions, but the arrive at concrete answers :D
*then
I should probably get off my soap box
 
@ArtemKaznatcheev You can edit previous chat messages by pressing the "up" key after you sent it.
If you're lucky nobody will notice the mistake. ;p
 
Interesting
 
We can get the older generations to use the site also, but we will draw them when we have a larger userbase and lots of interesting content for them to peruse
 
(except for the edit marker right next to the message)
 
10:56 PM
thanks for letting me know
 
our early adopters will likely be younger, more tech savvy users
 
@Josh the difficulty is in generating that interesting content :P
 
They probably had to, given the tendency of over rapidly submitting answers which SE causes. :)
 
there are plenty of young profs and graduate students
 
@StevenJeuris you can also use the icon to the left of any of your posts (when you hover over them) to edit them
 
10:58 PM
@Josh Give it some time. Older people are starting to amass on facebook. It won't take too long before they see SE is much more interesting. :) Especially with all the new SE sites.
Speaking of which, ... I never figured I would ask this again on chat. Such old memories. asl? :) Actually I'm just interested in the 'age' part.
 
@StevenJeuris Oh I agree, and I think a well defined site will draw them in
 
@StevenJeuris lmao I haven't been asked "asl" since I was probably 16. I could just link you to my profile but since you went all retro: 30/m/Asheville, NC USA
 
@Josh Hehe awesome. :) Could have figured you had a background in I presume, irc?
I'm 25.
 
@StevenJeuris sorry, not sure what you're asking... a background in... what now?
 
11:02 PM
22 at McGill in Montreal, QC right now
 
Sorry @Steven I'm an idiot. You're asking if I used to use irc. yes
And AOL and AIM and many other chat systems when I was a kid/teen
This is what I get for trying to write code and chat simultaneously :-p
 
:) P.s. @Josh, I started watching those Charlie Rose Brain videos. They are pretty good!
I still believe the site could benefit from a poll question like that.
It's perfect to spark interesting questions.
 
@StevenJeuris Awesome, glad you're enjoying those. I love that series
Charlie Rose is a great interviewer, too
 
@Josh I like how it doesn't focus on fancy magic bajumbo with flashy graphics, but actual informative recent content.
 
@StevenJeuris I do too. Even if that makes me an idiot :-)
> As always, it's a community, and authority derives from you, our users, even if you're sometimes idiots and we have to overrule you
@StevenJeuris yeah! And the same plain table he's been using forever. No fancy set, just great content
 
11:15 PM
Well, as long as they don't overrule us. We can attempt to fight back and attempt to create high quality poll questions can't we? :)
 
We can; Apple.SE did it
 
14
Q: Online data repository of research in the cognitive sciences

Jeromy AnglimSharing data is an important part of science (e.g., see APA discussion). It's also often useful to be able to have access to datasets when teaching students how to analyse data in the cognitive sciences. What online data repositories for research are available in the cognitive sciences (e.g., p...

This one still didn't get closed down.
 
@StevenJeuris I think the team's being a bit hands-off so far, letting us find our own way. (They could be busy with Jeff's leaving too however)
 
11:50 PM
writing out questions often helps me think out the answer before I post them
was trying to think of the name of this effect:
The Forer effect (also called the Barnum Effect after P.T. Barnum's observation that "we've got something for everyone") is the observation that individuals will give high accuracy ratings to descriptions of their personality that supposedly are tailored specifically for them, but are in fact vague and general enough to apply to a wide range of people. This effect can provide a partial explanation for the widespread acceptance of some beliefs and practices, such as astrology, fortune telling, and some types of personality tests. A related and more general phenomenon is that of subjective ...
No need to ask now though
 
stats.se has quite a few interesting list style questions. I also think there are ways of framing the question so that it's not technically a poll question; (e.g., you can put more than one per answer). I.e., the question becomes "what are the best" or some such.
 
List questions aren't strictly, 100% banned on SE btw, Apple.SE has accepted them when they're good
 
The two that have been closed: one was asking for a list of psychology video lectures and another was asking for best papers in cognitive science; the first was arguably not very academic in focus, and the second was a bit broad.
 
I think they'll let us decide the validity of those sorts of questions unless they're reaaaally bad
 
Here's stats.se with the [big-list] tag: stats.stackexchange.com/questions/tagged/…
 
11:54 PM
Oh wow, you have more than Apple I think
Hm, how about something like a CW list of cognitive biases?
 
@BenBrocka I'm not sure what the voting mechanism would mean for a list of cognitive biases. I can imagine a question: "What is a complete list of cognitive biases?" where someone could link to or provide a complete list.
 

« first day (19 days earlier)      last day (4460 days later) »