Last time, I wrote a bit about the process of submitting research to a psychological conference. The basic idea is that this is a quick way of communicating one’s most recent work. Unfortunately, it has some flaws. The most notable of these is that the work isn’t as rigorously reviewed as it should be, and one can present work which doesn’t live up to the usual standards of the field. Furthermore, there’s no record of what one presented, aside from the title and abstract.

I’ve recently been taking my research in increasingly cross-disciplinary ways, and have been struck at how ideas are communicated in other fields. My initial impression is that there are a lot of advantages to these alternative, though I’m sure part of this is a grass-is-greener kind of situation. I have no doubts that flaws will be revealed as I spend more time with a foot in these fields.

For instance, my collaborators and I have recently submitted some work to a few computer science conferences. Just as in psychology, submitting to these conferences is a quick way of communicating one’s work. However, unlike psychology, in computer science, conferences are the primary way of communicating work! Indeed, journals are viewed as places to find yesterday’s news. So much so that the flagship journal for the subfield I’ve been working in, Computational Linguistics has an impact rating of 1.47 (2013 figure). This is on par with relatively niche journals in psychology. Our flagship journals feature impact factors which are two or three times as high as that, at least (e.g. JPSP = 5.51)

The disappointing features of psych conferences that I highlighted previously are not present in this other discipline. Largely, I think this is driven by two features:

#####1. The paper is written before submitting##### Hard to believe from a psychologists perspective, but yes, you must write the entire paper before submitting to the conference. The paper (rather than just an abstract) is then subjected to the same kind of peer-review that it would receive if it were submitted to a traditional journal.

#####2. Everything at the conference is published##### Everything that is presented is published as part of the conference proceedings. This is a peer-reviewed publication on par with any journal publication.

####How does this change things?#### The problems associated with psych conferences are that the talks are often disappointing, in part because of the extraordinary freedom alloted speakers to take liberties with what they had said they would present. This freedom is due to the peer review of these presentations being less than rigorous, and the lack of accountability due to no archiving of talks.

If researchers are obligated to write the paper before submission, and this paper is reviewed and published, then this obliges them to have a clear set of ideas, which are already well-flushed out (or not! It should also be acceptable to have a mess and communicate that mess - provided there’s something about it that’s exciting or interesting).

Does this make our communication system better? I’m not sure. I think there are certainly some problems with the way we communicate our research in psychology. I’m still a bit of a newbie when it comes to communicating in CS, and so while, on the face of it, this other method seems promising, maybe I’m still in the honeymoon phase, and haven’t yet seen the warts.