7

Understanding the brain is a great challenge that is all the more interesting because no satisfactory theory has been found yet. For that reason, all brain sciences are, in my opinion, one of the most interesting thing you can study. As a result, there is a lot of scientific literature about it, but the stack-exchange community for cognitive sciences is so small (about 170 questions). Why is that ?

  • 3
    532 questions as of this moment, actually – Ben Brocka Oct 15 '12 at 11:55
10

I can give you an answer from the perspective of a formally active user that is becoming progressively more passive.

Although there is a large amount of activity on the internet about mind & brain sciences, the overwhelming majority of it is at the level of popular science. People who have an understanding of the mind/brain at this level, are often not qualified to provide quality answers or even ask interesting questions. In terms of the academic and professional communities, neuroscience & psychology are actually not well represented on the internet. Compared to mathematics, computer science, and physics, there is very little (from my experience) academic discussion outside of the standard published format. Most that exists seems to be targeted at the lay audience, popularization, and policy (i.e. explaining catchy studies) and not discussion of problems and methods of importance to active researchers.

Neuroscience/psychology seems to be slow on the uptake of open and non-standard publishing techniques and discussions through open platforms like SE & blogs. This can be seen by the lack of widely accepted pre-publication archives in the brain sciences, in contrast to the ArXiv used by mathematicians, physicists, and computer scientists. As such, it has been difficult to build and retain a technical user base on this site. Unfortunately, at this time, I do not see how to overcome this. Although we still have a core group of amazingly dedicated and knowledgeable members, this group does not seem to be growing or expanding to represent the wider scientific community.

We have had many theories on what might be the culprit (my pet theory is the name), but we have not had much success in solving these problems. In the process, we have received very little support and encouragement from the overlords, having to fight needlessly hard to achieve even basic changes (like MathJax being enabled).

I do agree that CogSci.SE could be an amazing resource, and a great community. I sincerely hope that my pessimism will be proven wrong.

  • Thanks for expressing this. – Chuck Sherrington Oct 15 '12 at 10:30
  • I agree largely with this analysis (i.e., psych communities are under-engaged with the internet / open access, etc.; a more inclusive name would help). However, my conclusion is that this highlights the need for a site like cogsci.se. I think that building awareness outside of mainstream StackExchange community will take time. I think we already have a unique site that is generating useful content, and I hope that over time it will get even better. So I hope my optimism will be proven right :-) – Jeromy Anglim Oct 16 '12 at 4:30
  • 1
    Agreed. There's too much popsci on here to attract most cognitive scientists. It's a chicken and egg problem, and we don't have a chicken. I do think lack of awareness plays a big part too though – Jeff Nov 13 '12 at 17:17
  • @Jeff I think it is worse than just chicken-and-egg problem. I tried for a bit to feed the site with close-to-research level questions and promote it to my colleagues, but after a few months of trying I made no progress. So I think we need more than just an enthusiastic initial few researchers. Something needs to change in policy, or elsewhere. My longstanding suspect has been the site-name, but I doubt it is that simple anymore. :( – Artem Kaznatcheev Nov 13 '12 at 22:30
  • 1
    @Artem I don't have a problem with the name, but I suspect that reflects different academic backgrounds more than anything else? (i.e. cognitive neuroscientists often don't like to be called psychologists, even if that's what they are) – Jeff Nov 13 '12 at 23:13
  • 1
    Promoting research-level questions is great, but the proportion of them on this site is too small. Reading the front page Qs often does not give the impression that we are a research-level site, but not allowing those questions would significantly impact our stats (visits, Qs, etc)... I don't really know what the answer is. – Jeff Nov 13 '12 at 23:15
  • There is no reason to avoid popular science-level questions, and in fact I think we should actively encourage them as much as we do research questions. If we focus on the desired end-result of vibrant research question activity right away, we will never actually get there. Popular-level questions are an opportunity to attract valuable undergrad members (even a blind hen sometimes finds a grain of corn) and get the research out of the lab, not an impediment to attracting professionals. – Christian Hummeluhr Mar 29 '13 at 11:59
  • 2
    @ChristianHummeluhr except it actually is an impediment. Every time I show this website to my colleagues, they go to the front page, see the mass of non-research and often non-scientific content, and leave without taking the (considerable amount of) time to find the few research level questions on this site. But I lost the battle for encouraging research questions on this site long ago, and I doubt there is ever a chance of attracting significant numbers of researchers for this site, anymore :(. – Artem Kaznatcheev Mar 29 '13 at 13:09
  • @ArtemKaznatcheev I get the feeling that the site has been fighting this fight for a while, so I apologize if I'm stepping on toes here, but I don't understand why the consensus is that "if only we had more research questions, researchers would come here." If your colleagues are anything like mine, the concern about question ratio was likely a matter of obviousness. They would have rejected the site for some other reason even if it consisted of nothing but research questions, unless you led them by the nose to something you knew they would find obviously relevant or they found it on their own. – Christian Hummeluhr Mar 29 '13 at 13:59
  • @ChristianHummeluhr my thoughts on this come from my experience with the only research-level SE: cstheory. As far as I can tell, that website made it because of very aggressive culling of non-research questions, but also because they had an in-flow of experienced users from the research level MathOverflow. I wasn't around in the early days of MO, to know what they did right. Ofcourse, it might be possible to build a research-friendly site in other ways. – Artem Kaznatcheev Mar 30 '13 at 1:48
  • I have largely given up on my agenda of encouraging this site to be research level. Partially because I don't think it is possible, and partially because I felt that my quest was driven by self-interest: I simply didn't want to read non-research level questions and variants on common misconceptions. – Artem Kaznatcheev Mar 30 '13 at 1:50
  • @ArtemKaznatcheev Right now, that is largely a hypothetical problem. You can not cull what is not there in the first place. When there is a healthy influx of questions, we can worry about what kind of questions they are, but first things first. It's much easier to handle one problem at a time instead of trying to do everything at once. First we get the questions, then we cull them. – Christian Hummeluhr Mar 30 '13 at 22:24
6

First off, the site is still in beta, and at the time of writing only exists for 270 days. Everyone active here wished it would grow quicker, but we aren't doing that bad. We still aren't decreasing in size. P.s. I suppose you were referring to the amount of questions on meta, instead of the amount of questions on the main site, which is much higher. (529 at the time of writing).

Having that little peptalk over with, the topic has been raised before in several different formats:

The main problem in my opinion is we are trying to reach out to an audience which isn't used to the SE format. I do believe we will get there, but we just have to be patient. However, proactively suggesting ways in which to reach out to this audience is more than welcomed. Ideas/suggestions are a better approach than starting a discussion on why.

I'm really glad to have you on board and actively wanting to participate on meta. That's what we need! Be sure to drop in on chat from time to time as well.

  • 1
    I checked the number on meta indeed, shame on me. But 529 is still not a lot. The fact that the audience isn't used to the SE format makes me think of this talk: ted.com/talks/…. I think stackexchange is some kind of media where information exchange is quicker than what science uses traditionally, so it can have a big inpact (ie: books > scientific journals > stackexchange & other internet content) – Benjamin Crouzier Oct 14 '12 at 22:25
  • I realize that I didn't do my research correctly before asking this quesion. But as least now I have a nice reference of topics to my problem. – Benjamin Crouzier Oct 14 '12 at 22:39
  • 1
    I wouldn't say the SE format is a problem so much as awareness of the sites/network. Tons of programmers are familiar with them due to Stack Overflow's massive popularity, and lots of people who program + (other topic of interest) have thus found the network. There's relatively little overlap between psych/etc and programming though. – Ben Brocka Oct 15 '12 at 11:58
  • I happen to know cogsci.se because I program and use stackoverflow. I think we need to find the talented where they are, which is not clear to me (big universities websites, online research websites, coursera ?). I really don't have a clue where they are and how to get those people interested, but it is an interesting subject. – Benjamin Crouzier Oct 15 '12 at 12:10
5

I believe that

(a) Professionals in the field (academics) have so much pressure on them that they simply have no interest in answering questions by laypersons. They barely have time for their students.

(b) Professionals in the field are often extremely specialized and unqualified to answer professional level questions in other areas of their field.

(c) Professionals in the field are well networked [is that English?] and have a lot of pertinent experts to discuss their questions with without coming to the internet.

I also think that

cogsci.SE could be great for students of psychology, cognitive sciences and neighboring fields. We might consider promoting our site among this group. There are lots of Facebook groups where students of one semester of one field of one university collect to discuss questions about the topics they have to learn. We could recruit there. Or we could send an invitation email to relevant student organizations, student self-government etc.

  • Reminds me of this: marciovm.com/i-want-a-github-of-science. I think there it would benefit everyone of if the discussions you are talking about were taking place online – Benjamin Crouzier Mar 11 '13 at 21:53
  • 1
    I guess the hope for this is site is that it gets more research-level questions (rather than lay questions). My feeling is also that the model of open science whereby question asking is encouraged as a mean of creating online resources has been less adopted in psychology than some other fields (e.g., statistics, mathematics, computing, etc.). Furthermore the reward structures at universities do not currently particularly encourage engagement with such sites. – Jeromy Anglim Mar 19 '13 at 1:16
4

I doubt I will fit this into a comment, so I will add an answer.

I was working with search engine optimization for a while, and a page that is human readable would look very different to a web crawler. Until these crawlers start to rank cogsci questions higher, the community would stay small.

Sites like stackoverflow are very useful, and the questions asked there very quickly rise to the top in google. So any time a new programmer searches for an answer online, the programmer is likely to find stackoverflow, get hooked and stay for a while.

I think part of stackoverflow's success is due to a wide variety of tags and keywords that are used for search engine optimization. Some of those tags are "popular" enough and would match more search queries.

For example, the iPhone tag and having iPhone in the question name provides great visibility from google. Having iPhone and iOS means even better visibility. Such specificity means that SO dominates the programmer QA field.

Personally, I don't know how people search for cogsci related info on the web, but if there's little search engine optimization, then other "popular science" sites may rank higher than cogsci, for the same search terms.

I suggest broader use of tags, and better titles\ keyword density, so questions are more visible from search engines

  • I think this is an interesting perspective. I imagine that over time our link related weighting in Google will increase. Our content is often better than what is available elsewhere on the web. I think accurate question titles are also really important (this often requires an edit to achieve). – Jeromy Anglim Nov 14 '12 at 2:33
  • 1
    More content will also translate to more page views and in turn more people joining our community. It is useful to look at the ratio of questions to visits per day on various stack exchange sites: stackexchange.com/sites#traffic ; Stack Overflow is about 1.2 visits per day per question. Cogsci is around 0.7 visits per day per question. Keeping an eye on this ratio I think is a useful statistic for telling us how successful we are in attracting Google's attention. – Jeromy Anglim Nov 14 '12 at 2:38
2

Why do we use SE at all?

When I go to stats.se or StackOverflow, it's usually because I need an answer to some concrete problem. I need to produce or alter some graph, get best practices for reporting or interpreting, consider limitations, etc. I think this sort of structure facilitates the creation of "hook" questions, such as the 10,000 view question about Lumosity. These hooks guarantee a stable influx of people to the site. For us, as long as Lumosity is popular and the question of brain training remains controversial, it will probably continue drawing members. I think the most appropriate way to frame the problem is therefore as one of change in membership. A concrete problem for us is that we don't have enough of these member attractors. Rather than waiting for the right question to roll around naturally laissez-faire style, we might want to consider just designing some.

Long-term

For long-term stable member attractors, there are a number of candidate topics we might consider that people are almost universally interested in, such as academic/sports/artistic expertise acquisition, sex/love/attraction, decision theory/biases and mental health. Well-formulated, broad questions on these topics with well-referenced answers that cover a lot of theoretical ground, sprinkled with a touch of coordinated publicity effort, would guarantee a constant influx of new members. The downside to these questions is that they would take time and effort to design and answer in a satisfactory way, but I think they would be an excellent investment.

Current events

Another option in-between the "big questions" approach and the "laissez faire" approach might be to utilize current events, e.g. taking a few swings at Bad Science. You know the ones I mean: minimal findings about some topic, which was obviously selected for its sensationalist potential in the first place, stretched to their theoretical limit by the magic of science reporting. "Meat eaters are mean, says science", "Sitting on a wobbly chair makes people prefer unstable relationships," etc. By addressing these thoroughly and in a timely manner, we would provide easily digestible answers for interested parties, which would guarantee an influx of members for as long as the topic remains relevant. These would be trivial to design, but still might take some time and effort to actually answer. It wouldn't have to be about bad science, it could just as well be exposition on any cogsci topic currently in the news.

Member attractors

Obviously these two types of member attractors can be implemented independently, and I think the idea of designing and implementing some of these attractors has merit in general. They should attract both lay people and professional academics--at least I know preciously few academics who wouldn't jump at the opportunity to dispel a few popular misconceptions within their field of expertise. And then they're hooked ...

  • 1
    Interesting suggestions. A few reflections: (1) I liked your question on mindfulness; I'd like to see more questions that are built around critically evaluating an existing journal article (2) I think if active users want to ask some interesting or populist questions (and even answer those questions with a good scientific explanation) that'd be great. Even better if you or others want to promote the resulting answers in relevant forums. – Jeromy Anglim Mar 30 '13 at 11:39
  • I think you have some great points here Christian. – Josh Mar 30 '13 at 14:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .