Does vision science need a new open access journal (Pre-discussion for ECVP2015)

I guess the difference in the long run is up to us.

One thing that might sustain a journal (and thereby prevent it being bought out by a publisher) would be to have a fairly aggressive recruitment of younger academics to the editorial board, alongside the established stars. My strong impression is that the younger someone is, the more inclined they are towards the whole open access thing. This, of course, conflicts with a desire to publish in high-impact journals at the beginning of a career...

The nice thing about this would be that it would provide a proof of concept, and perhaps encourage others to try the same thing, or steer papers away from the big publishers towards the journal.

I wonder whether iPerception charges will stay the same with the new publisher, Sage? I notice it does not mention fee waivers for poor authors (unlike PLoS).

I think we need to reflect on what our priorities are for open science and publishing. One desideratum is a completely free-to-author and free-to-reader journal. But there are others, and some may entail a tradeoff between ease of getting those features and ease of having a completely free journal:

• Typesetting. If we're happy to have authors themselves entirely format the PDF that will be published, none is needed. However I have spoken to more than one perception scientist who thinks that professional layout is super-important, the professor I am thinking of did not think that loss of professional layout was worth the exchange for free open-access! Typesetting costs money.
• A continually upgrading publishing platform that keeps up with emerging altmetrics, ORCID linking, and whatever else is in publishing's future
• Automatic archiving by a third party so if the journal dies, loses a host and funding, the articles will live forever. I think this is a definite requirement, and is usually accomplished via CLOCKSS. I don't know how much they charge.
• How long are we willing to devote to this before gaining a steady stream of submissions? The more independent we are of other journals/publishers, the longer it may take to gain traction. Partly because the lack of branding will boost the skeptical thinking that this new thing may not be a goer. Authors will wait to see whether this thing will stick around.
• Open data requirements. No journal in perception currently requires that , but such requirements are gaining traction in other fields.
• Any other desiderata?

As I said on Facebook, my dream would be for the journal to have all the functionality and support of PeerJ, e.g. PeerJ Computer Science https://peerj.com/computer-science/ and create a PeerJ Perception. Currently it costs something like $100/paper but I hope we could bring some institutional funding and work out a deal with PeerJ to reduce that , maybe to zero in some circumstances- although seems doubtful that could be achieved or they'd agree to it. But PeerJ is doing all the right things IMO, with optionally publishing the entire review history of articles, and great technology including image setup, plus commenting and an optional publishing flow from preprint to the journal (which speeds science).

To address Marco's point, often it is the case that one may need to act locally but think globally. Or even think regionally- I'd like to see neighboring journals' editors eventually resign and join the new initative. Especially Elsevier and other outrageous profiteers, including Vision Research. That would be a very worthwhile achievement, to me.

All sounds good to me. I know next to nothing about PeerJ, but are you saying that we could try to achieve our aims within its framework? Obviously, that would make things a lot easier.

Fee waivers for poor authors will raise costs for the rest. And know several people who cheerfully claim poverty in order top publish for free in PLoS, even though they're sitting on top of large grants... but perhaps these are minor issues. (No idea what will happen with Perception / i-Perception.)

I'd like to see some of the Vis Res crowd jumping shop (or J Vis, for that matter), but not necessarily all of them. This is also potentially a chance to ring out the old and ring in the new as far as figureheads go. A balance between the established and the less so looks attractive to me (but perhaps there are differing opinions?).

Regarding clarification on i-Perceptions costs. I've also been wondering why JoV costs so much? Why isn't that already the cheap open-access journal the field wants. Can we demand 'open accounts' from all the journals from our field? Maybe it would be a good start to email all of the major journals in our field for this information prior to our discussion at ECVP.

It's interesting, i-Perception charges less than JOV even though the former is owned by a commercial company and the latter by a non-for-profit organisation. This may of course change now, and it may have to do with Pion being small. Equally frustrating is that JOV does not comply with CC-BY which here in Britain is a big deal. Even though the ESRC funds my research they refuse to pay the page charges for JOV on the ground that the open access is not CC-BY. JOV did say they were going to make changes following complains from people in the UK but nothing has changed yet.

JoV is an American journal. I don't think they're that interested in what anyone outside the USA thinks.

Open accounts is an idea I've floated in the past. My ideal journal is open access, open accounts, and makes no profit.

Surely there are also plenty of US vision scientists who are troubled by the high cost of JOV.

It's board are also scientists, (many from outside the US!), who are also potentially troubled by this.

It is surely not the case that we cannot influence them. The question is, do enough people care, to collectively act to demand something. For example, not submitting unless JOV accounts are open. Even if only ECVPers and some people in the US did this, it would surely already have big impact.

I think it's a great idea to push JoV to provide their cost accounting and loosen their copyright restrictions (they also told me when I queried it, maybe 9 months ago, that they were working on it). I believe they've said in the past that they receive a subsidy from ARVO required in spite of the large amounts they receive from we authors.

I'm guessing the main problem with loosening the copyright and publishing the costs is not the JoV people but rather ARVO. It's a very large organisation including doctors, who tend to be more hierarchical and less receptive to bottom-up input or demands. But we ought to try.

I think Alex's list of desiderata is important to discuss further. In particular, why would anybody want to publish in this new journal and what requirements does that imply? Here is at least what I think about when submitting a paper:

• Perceived importance of journal. Good stuff will always go to the top journals and no new journal can change that in a short run. But that's okay, as the target here should be that anything at the level of JoV and Vision Research are now published with this new journal.
• Impact factor. Hate to say this but many grants are still coupled tightly with IFs. At the very least, any new journal should have an IF (even if a very modest one) so that it's worth publishing there.
• Who reads the journal? Part of the reason I like JoV over any other vision journal is that despite the costs, it is well-read by the community of vision researchers. In contrast, PeerJ is not read by many, so it would be a waste to publish there. But I think the situation could be easily remedied by branding, e.g. PeerJ Perception or PeerJ Vision.
• Typesetting. Not only many people prefer professional-looking pdfs (as Alex points out), but also many researchers lack even rudimentary skills to format anything themselves. E.g., APA double-spaced Times manuscripts with figures ate the end are just not legible on screen yet many people still do this. Some journals try to solve it by providing templates, but cross-references are hard to include in Word, so this will never be a complete solution.
• Publisher reputation. So many scam journals around make it important to be perceived as reliable.

There are of course many other things I would want to have (open access / data, easy submission, fast review etc) but in practice when submitting I just don't care about these. I'm betting most people also don't, so we don't need to address this now.

So altogether, I think PeerJ platform offers the best of the two worlds: it's modern, it's professional and it costs very little. Paying $99 or so for a lifetime subscription sounds a fair amount for the services they provide (modern publishing platform, typesetting, archiving). It's just not possible with the above requirements to have it completely free. Of course, if people were reasonable, they'd just post stuff to biorXiv, end of story. But since we're in this game of showing off, we gotta pay for now.

I'm also coming to be persuaded that working via PeerJ could be the best route. Of course, we could just submit our work there already, but I'm not convinced that is ideal. Who are the editors of perception oriented research in PeerJ? Who sets the standards? I think if it were to become a standard platform for vision science we would need to have some ownership over it. But this seems to be a direction PeerJ are open to, as they have a distinct PeerJComputerScience journal. It would be good to work out what would make them interested in hosting a "PeerJPerception" journal (although honestly I'd prefer if we could just call it Journal of Perception, 'PeerJ...' sounds ood).

I agree with most of the above. I'm not too worried that a bigger journal will try to buy it out if it's successful: that would be a decision to take and would just need the board to contractually protect the things that were important.

Of course, it might also be that the possibility of a new journal might get JoV (which I'm mostly very fond of) moving a bit faster in the right direction.

For a name I'd prefer "Vision" to "Perception". Although perception might be technically more accurate, I've found some people treating 'vision' with more (even unreasonable) respect. Like a 'vision' scientist would have to know maths and "clever stuff", but a perception scientist doesn't. That doesn't stop the perception scientists wanting to go there (notice the attendance at VSS). In fact it makes them feel good and validated that they were able to. I've genuinely heard people that weren't vision scientists bragging that they got their abstract into VSS and now felt bad-ass (I left them with their illusion ;-)

What's the key to being the new JoV, rather than the new i-Perception (beyond the name)? Hard to know, but my guesses are:
- clean start (i-perception was too closely associated with Perception already)
- good set of first articles - we need a few people to take a punt on it being a success
- high standards for acceptance (even if that means a low rate of papers published). The issue here might be whether good reviewers will play

Recruiting big-wigs to the board might be tricky because they either have to leave their current journal (e.g. for Johan to leave Perception/iPerception would be quite a statement that he might not want to make?) or they've decided by now they don't want to be editors. Again, that's just a guess

I also didn't know much about PeerJ but apparently my uni has already paid an institutional fee that means my submissions would be totally free! Happy days!

best wishes all
Jon

I published a few perception-related papers in PeerJ, and I'm also an editor there, as is Stephen Macknik. I thought the journal did a great job. They just announced their first (partial) impact factor, which is presently above 2 (not bad considering it is such a new publication model).

In terms of editorial board, it's not hard to come up with at least a few well-known people (though maybe not exactly big shots yet) who have expressed clear support towards Open Science and might be interested in this initiative, e.g., Hans Op de Beeck, Nikos Kriegeskorte, Chris Baker.

I agree that a PeerJ journal would be good because its not just open access (which generally does not solve the issue of huge profits for publsihers) its also open publishing (a relatively painless annual subscription covers the cost and so this does solve the financial problem).

I have some general concerns with the open-access model. In conventional publishing, the publisher makes money when journal issues, or today mostly downloads, are sold. It's in the publisher's interest to have high-quality content, and to maintain it. In the open-access model all the money is made up front. The publisher has no financial interest in quality or in maintaining the database. What will be retained in 5 or 10 years?

Conflict of interest disclaimer: I am editor-in-chief of a journal that accepts both conventional and open-access papers.

I think this is a valid concern @brucebridgeman but I think the key has to be that we can ensure standards are maintained because we can decide where we submit our work. If a publisher gets a reputation for lowering standards, or designing the peer review process to ensure as much as possible is accepted, we can stop submitting to that publisher (as many have decided to do with Frontiers). For me this is about the academic community taking control of the type of publishing we want to see, and making it clear that we will only submit to journals that maintain high standards (and offer value for money). If we only submit to OA journals that maintain high-quality content and maintain it well - they have the same financial incentives as subscription based publishers.

Lee, good points. Journals like the Journal of Vision can maintain a reputation because they are run by academics who have an interest in quality. The Journal of Eye Movement Research has another model, open access but without a publishing fee. It is supported by a modest Swiss grant, and authors format their own papers from formatting software the journal provides.

So experienced scholars can reward the quality open-access journals. Another concern is with those less well connected - from countries not in the academic mainstream, or from young scholars at institutions without experienced scholars in their fields. They will have a hard time separating the quality open-access journals from the junk journals. Several times per week I get a solicitation from an open-access journal I've never heard of, with an important-sounding name - I'm sure you've had the same.

regards,

Bruce Bridgeman
Editor-in-chief, Consciousness and Cognition
Edward A. Dickson 2013 Professor of Psychology
University of California, Santa Cruz
106 Social Sciences 2 Tel. (831) 459 4005
Santa Cruz, Ca. 95064 Fax (831) 459 3519
http://people.ucsc.edu/~bruceb/ ( http://people.ucsc.edu/~bruceb/ )

The only journal I am aware of that is truly open access (i.e. free to publish
in, and free to read) is the Journal of Machine Learning Research
(http://www.jmlr.org/). It would be great to have something similar for Vision
Science. JMLR has been very successful in attracting high quality submissions and has a very good impact factor. I suspect that this was achieved by it being set up by many of the top people in the field, via a high profile mass defection of almost the entire editorial board from (the Springer journal) Machine Learning. JMLR gets around the issue of typesetting by requiring all authors to submit their work in Latex. This might be more of an impediment in vision research than machine learning. However, this impediment would also exist if we used PeerJ which also seems to require submissions in Latex (or via overleaf which is just a front end to Latex).

PeerJ does not require LateX submission. Pdf and Word are fine.

For a somewhat more radical view.

Open access with authors' fees reverses the economics: before a publisher had a cost to accept a paper now it has a cost in rejecting a paper. To expect that this would not create problems as long as editor/reviewers do their work is naive. The money pressure will find its way into the system and that is what has happened with some recent well-publicised cases (Frontiers). It is actually a triple whammy because, in addition to the reduced quality control, the inflation in number of publications is also a problem as it becomes impossible to read/browse this mountain of papers, and because the author's fee means poor authors/institutions have a disadvantage. The solution was something that everybody could have seen in advance: Keep the publishing of journals in the hands of Academic Institutions and without author's fees. The fact that there is a cost in accepting a paper has a positive side effect, as it forces journals to be selective. Libraries should have led the way on this, they are the ones that benefit most from open access, they have the staff, and they are located next to the academics who do all the work. I fear it may all be too late, once a system develops with so many people milking it for profit it may be hard to reverse the process. But scientists and academic societies could do something if they want, because the actual process of publication online is simple and with very low marginal costs.

PeerJ accepts Word and pdf for the initial submission, but what about the final, accepted, manuscript?

I support most of the comments above. While I'm yet to submit an article there, the PeerJ platform and publishing model seems to be great.

One small complaint I have is that I don't like the single-column pdf formatting. I understand this was done for people who read on handheld screens, but it just doesn't appeal to me. Perhaps once I'm used to it.

I will echo the thoughts above that before the journal launches we want to make sure there is a fairly large and reputable editorial board on hand. Ideally those people would also submit articles for the launch of the new PeerJ section. Perhaps we could start having quiet words to those we know on the boards of JoV and Vis Res about whether there would be interest in "defecting".

Regarding the name: while "Vision" is appealing to me, "Perception" or "Sensation and Perception" are broader, and could encourage submissions from researchers in other sensory modalities (particularly if we had a good auditory or crossmodal person on the board). I feel that the two existing PeerJ versions are very broad (approximately, "biological science" and "computer science") so they might be resistant to creating such a specific section.

Another thing that I think is really important is to encourage open data and code to be an integral part of publishing in the new journal. While I'm not sure that the mandatory route (as for PLoS) is what we want, I would like to have some mechanisms to strongly encourage the open sharing and archiving of data and code. We could integrate the badge system that Alex and others developed to reward people for doing so. A more negative option would be to have articles for which the authors declined to share data / code state this fact in big bold letters prominently at the top of the article ;)

To this end, perhaps while talking with PeerJ about hosting the journal on their platform, we could also be thinking about an integrated archiving service such as http://datadryad.org/ or http://dataverse.org/. Or perhaps PeerJ already has these links?

Ultimately, we should try to streamline as much as possible the process of archiving data and code. When you submit to PeerJ Perception, it's a simple zip file upload (or forking a git repository) to also archive data and code for the paper for all eternity*. Moreover, we strongly encourage authors to do this.

• see Bruce's great comment. This is why I think integrating with a service like one of those linked above will help in this respect – rather than, e.g. relying on an author's university web server.

I strongly support Tom's point about open data and I think this should help with reputation as he suggests. I think a positive rather than a negative encouragement system would be optimal. I think perhaps, with PeerJ, there is less of an incentive to accept rubbish papers because of the different business model ($99 to publish 1 paper a year until you drop off the perch) - so in fact it's less in their interest to publish poor quality articles as it will dilute their impact. But these are all valid and interesting views worth discussing at the forum!

Marco, I agree. The Journal of Eye Movement Research is free both to authors and to readers, and attracts the top people in that rather specialized field. The editor is Rudi Groener, at Bern, Switzerland. The journal provides a template that is relatively easy to use, and doesn't require any additional software from the author's side.

regards,

Bruce Bridgeman
Editor-in-chief, Consciousness and Cognition
Edward A. Dickson 2013 Professor of Psychology
University of California, Santa Cruz
106 Social Sciences 2 Tel. (831) 459 4005
Santa Cruz, Ca. 95064 Fax (831) 459 3519
http://people.ucsc.edu/~bruceb/ ( http://people.ucsc.edu/~bruceb/ )

yes.

regards,

Bruce Bridgeman
Editor-in-chief, Consciousness and Cognition
Edward A. Dickson 2013 Professor of Psychology
University of California, Santa Cruz
106 Social Sciences 2 Tel. (831) 459 4005
Santa Cruz, Ca. 95064 Fax (831) 459 3519
http://people.ucsc.edu/~bruceb/ ( http://people.ucsc.edu/~bruceb/ )

Tom, thanks for your thoughts. But “Sensation and Perception” has always bothered me. Sensation is leftover 19th-century functionalism - we now know that perception doesn't progress through a stage of experiencing unprocessed sensations. I reviewed a “Sensation and Perception” textbook and recommended that “Sensation” be dropped. They agreed, but when the new edition of the book came out sensation was back. Blame conservative publishers.

regards,

Bruce Bridgeman
Editor-in-chief, Consciousness and Cognition
Edward A. Dickson 2013 Professor of Psychology
University of California, Santa Cruz
106 Social Sciences 2 Tel. (831) 459 4005
Santa Cruz, Ca. 95064 Fax (831) 459 3519
http://people.ucsc.edu/~bruceb/ ( http://people.ucsc.edu/~bruceb/ )

I'm seeing two different and potentially conflicting positions emerging here. Some people seem to push for a high quality, rather selective journal. Others seem to have an idea that this journal should accept anything that's technically sound (like PLOS ONE) so that people have a cheap venue to publish their work. Let's be honest here, not every paper we write is grand, yet if this new journal is too selective, where shall I publish my lesser studies? In the traditional journals, of course, and so that's not solving the original problem we had.

A selective journal is important, as others mentioned, as a means to filter stuff. Yet I'd argue that it is not the function of a journal / editors to filter research. I'm all against the file drawer problem. Rather, we need better tools to filter post-publication, something like recommendation systems that Amazon or Netflix have. Not sure if PeerJ has something like that.

Bruce: your point is well made; I'd be happy to drop the "sensation" label. I more wanted to encourage the submission of articles examining other modalities (hence "perception" rather than "vision"). I think this would only work if we could get a few auditory / crossmodal people on the board as well.

Jonas raises a great point. I think this highlights the importance of having a good and motivated editorial board. I 100% support the editorial policy that articles should not be judged on their "potential impact" or similar "sexiness" dimensions. However, there are grades of "technically sound", and it would be possible for a well-chosen, subject-specific editorial board to enforce a strong cutoff on this. I feel that's the place that JoV and Vis Res fill in the current landscape.

I think one important reason why PLOS and Frontiers have become dumping grounds for "lesser studies" is that the editors and reviewers are often not subject matter experts (this is particularly true of Frontiers due to their reviewing / publishing model). Somebody who knows little about the methods of a field, or even a topic within a field, likely has a more loose definition of "technically sound" than a real expert.

We should decide where we want the new journal to stand on this. One avenue is what Jonas suggests: a cheap place to publish peer reviewed studies, but with a relatively low bar and an emphasis on post-publication recommendation and review. I'm intrigued by this model, and PeerJ might be one of the better platforms to support it because members are encouraged to review (including post-publication comments) to maintain their membership. This might provide the critical mass to make post-pub review actually viable.

The other avenue would be to set a relatively strong cutoff for what we mean by "technically sound", and ensure that we have the editors and reviewers on hand to enforce it to the best of their abilities. In my opinion this is the current JoV / Vis Res market.

My current position is the latter. The new journal should not become a dumping ground for technically flawed papers. I feel there are enough of those to serve the "lesser paper" market already (though as Jonas says, they're not as cheap as they could be). In my mind, this new journal is seeking to take the place of Elsevier's Vis Res in terms of community standing (and hopefully some of their editors as well).

I'm happy to be convinced otherwise, if others have strong opinions.

The realities of the REF in the UK (and increasingly similar exercises in other countries) means that we need to be able to publish in two types of journals: strong specialist journals (equiv of Vis Res or JoV) and "fancy" multidisciplinary ones (equiv of Current Biology, Nat Neuro etc). The discussion so far addresses the former but not the latter.

It seems to me that there is a suitable "fancy" journal out there already if we as a community decided to adopt it - eLife. It meets all the concerns re OA and from what I recall its already got a number of vision scientists on the Editorial Board.

Thoughts?

To pick up a different point - I understand all the arguments for OA but there is another model we could consider, society journals. Take QJEP, its run by the EPS (Experimental Psychology Society; http://www.eps.ac.uk/), its subscription rates are very low, I believe it provides free subscriptions to universities in less affluent countries, and the profits go to funding travel awards for students, supporting conferences etc. Its also of course already an established title with an Impact Factor et and there is no fee for publication.

So rather than starting from scratch we could switch our publishing allegiances. To take QJEP as an example, many of us know people involved in the journal and I'd expect they might be quite keen to embrace vision research.

I guess this begs the question: what is the priority? Open access full stop, or stopping publishers profiteering?

How did the ECVP discussion go- what points were raised, what was the general feeling?

Lee chaired the meeting and I'm sure he will write a detailed message for this group.

I think that everybody really appreciated the setting up of the issues that Lee provided, because it was careful and factual, including discussing journals and options that already exist but people may not be aware of (for instance only one person in the audience had published in PeerJ, mainly because they just did not know about it).

Opinions ranged. I had a sense that people were very aware of how much work is involved in running a journal and thus were a bit skeptical about the idea of setting up a new one.

There were approximately 150 people in attendance. I reviewed some of the options for open access, as detailed in my slides (http://www.slideshare.net/leedewit/open-access-ecvp-2015-51874667). At the end I took a number of straw polls of the popularity of the different options. There was an almost unanimous agreement that it would be useful to try to use eLife as the 'high impact' OA alternative for vision science (thanks to @simonrushton for putting this on the agenda). Approximately 65% of those present thought that setting up a 'subject page' within PeerJ would be a good move. There was only a little support for the other alternatives (setting up a journal with Ubiquity, setting up a journal with open software, or supporting Eye and Vision). I would say one of the main criticisms of PeerJ was that it was still a commercial model, and that ideally publications should be free, and directly supported by research funding (hence the support for eLife). I will start two new threads to develop these two options:

PeerJ -
https://www.loomio.org/d/c5QXFDaj/a-vision-science-subject-page-within-peerj

eLife -
https://www.loomio.org/d/xRlPkHe4/elife-as-a-prestigious-publisher-in-vision-science