After 18 years (sic), Mark D'Esposito has stepped down as Editor-in-Chief (EIC) of the Journal of Cognitive Neuroscience (JoCN). Our community owes a large debt of gratitude to Mark and, before him, to Mike Gazzaniga, for creating and shepherding this journal that has played such a vital role in the growth and success of our field.
By the time you're reading this, I will have fully transitioned into the EIC role and am grateful that many of the journal's highly talented and dedicated Associate Editors will be staying on and are helping to smooth the transition: Jeff Binder, Randy McIntosh, Earl Miller, and Jordan Taylor. We will miss the expert contributions of the departing Roshan Cools and Kia Nobre but are thrilled that Morgan Barense, Heleen Slagter, and Virginie van Wassenhove have joined our ranks! Although we don't have grand ambitions to “fix what ain't broken” at the journal, we are planning some modifications to how it operates, in the hopes of accommodating the increased importance that our field is putting on robustness in our science and on transparency and equity in our practices.
Our decisions about what to keep at JoCN, and what to change, are guided by the fundamental conviction that there is a special role in our science for the society journal. Society journals, like JoCN, are run by us, for us. Editorial policy is guided by what's best for the field, with minimal influence from commercial considerations. Editorial decisions are made by colleagues who are, like us, active members of our community. At JoCN, there are no layers of professional staff between authors and reviewers as well as no staff generating surveys or promotional messages to clog our inboxes. Along with this special status comes a special responsibility to lead by example, and we hope that the changes that I'll summarize here will be welcomed in that spirit.
If you've read this far, you have almost surely reviewed for the journal. Thank you. We are instituting a “consultation” stage in the peer review process whereby an editor, upon receiving the requisite two reviews of a manuscript, will have the option of opening a three-way dialogue between themselves and the two reviewers. This will entail the reviewers reading each other's review and helping the editor come to a consensus decision about the manuscript. Most often, we find that the two reviews of a manuscript are broadly in concordance, and so for the modal manuscript, such consultations won't be necessary. Sometimes, however, two reviewers will have discordant takes on a manuscript or will differ in their overall recommendations (typically one “revise-and-resubmit” and one “reject”). More rarely, the editor may suspect that one of the reviews lacked good-faith substance (perhaps it was “phoned in”?; perhaps a reviewer has an unfair bias?). Thus, peer review consultations are intended to accomplish two goals: give reviewers more direct involvement in editorial decisions and provide a check against unfair reviews.
Since its inception, one challenge for JoCN has been the fact that we receive many more high-quality manuscripts than can be reviewed and published, necessitating the triage that occurs between the receipt of a manuscript and the send-out for peer review. The decision to editorially reject a manuscript is necessarily a subjective one, and one that's made based on a limited amount of information, so it's unavoidable that many authors will feel that such decisions are unfair, if not wrong. In an effort to improve this aspect of our editorial practices, JoCN is adding a new category of editor, the Consulting Editor. The modal Consulting Editor is a relatively early-stage independent investigator who has been selected because of their reputation as an excellent scientist within a particular subdomain of cognitive neuroscience where the journal needs coverage. Each “editorial reject” decision contemplated by the EIC or an Associate Editor will go to a Consulting Editor, who will then decide either to corroborate that decision or to disagree with it, in which case the “rescued” manuscript will go out for peer review. The roster of Consulting Editors is too large to list here, but their names can be found at www.mitpressjournals.org/journals/jocn/editorial. When you encounter any of these colleagues, be sure to thank them for this valued contribution!
GENDER BIAS IN CITATION PRACTICES
Recently, Jordan Dworkin, Perry Zurn, Danielle Bassett, and colleagues undertook a monumental study of citation patterns in five broad-scope neuroscience journals and discovered that the proportion of articles first- or last-authored by women was markedly lower than what would be expected if gender1 had no role in selecting which articles to cite (Dworkin et al., 2020). (Alarmingly, although the proportion of women publishing in these journals has increased over the past 25 years, the proportion of citations of articles first- and/or last-authored by women has been decreasing over the past several years!) Can you say “h-index”? (See Fairhall and Marder  for a succinct consideration of why this is problematic for our field.)
During the summer of 2020, a team in my laboratory applied the approach of Dworkin et al. (2020) to the 2106 articles published in JoCN in the past 10 years and confirmed that we, too, have this problem. Importantly, this is a systemic problem, because “everybody does it”—the same qualitative pattern of overcitation of “man-first/man-last” articles was seen for reference sections regardless of whether published by a “man-first/man-last,” a “woman-first/man-last,” a “man-first/woman-last,” or a “woman-first/woman-last” team (see Fulvio, Akinnola, & Postle, 2020, this issue). Although we have no illusions that we can remedy this state of affairs by fiat, we hope that by encouraging the generation of a “gender citation balance index” for each paper reviewed by JoCN, we can at least contribute to social norm messaging that can nudge our behavior in the right direction (Murrar, Campbell, & Brauer, 2020).
Because this is a trial run, for the time being, we are hosting the gender citation balance index (GCBI) tool (the “GCBIalyzer”) at postlab.psych.wisc.edu/gcbialyzer/. Authors choosing to participate simply need to paste-in their list of references (be sure to include DOIs) and out will come that list's GCBIs, broken out by the four author-gender categories. If authors choose to include these GCBIs as part of their submission to JoCN, reviewers will be invited to recommend papers from underrepresented author-gender groups that the authors might consider including in their revision. We hope that, within a year or so, we'll have enough cases to allow us to determine whether this modest “nudge” yields GCBI values closer to 0.0 (the value indicating no imbalance relative to the base rate) for published articles relative to when they were first submitted. (Authors will also have an opportunity to include their article's GCBIs in the Diversity in Citation Practices statement that is soon to feature at the end of every article published in JoCN.)
When I was a (relatively) young, impressionable graduate student, our laboratory statistician opined to me that, in a perfect world, papers would be reviewed based on the inherent interest of the hypothesis (Introduction) and the soundness of the design (Methods), with no consideration of the results—if the experiment was worth doing, and if it was carried out well, any outcome would be informative. Now, a few decades later, voilà, we are in an era that has embraced the registered report. Leveler heads than mine have advised that requiring all submissions to have been preregistered would be Procrustean, but we are instituting a policy of accepting “Stage 1” registered reports and a practice of encouraging authors to consider preregistration of their experiments even when they don't go through the formal process of submitting a Stage 1 proposal for peer review before carrying out the research.
Details about the implementation of the new developments that I've summarized here can be found at www.mitpressjournals.org/journals/jocn/sub.
It takes a community to do good science, and JoCN wants to continue to be a valued contributor! We will strive to encourage best practices and to work toward improving the climate in which we work and communicate. It is a privilege for us to be able to work so closely with so many terrifically smart and creative colleagues. I encourage you to be in touch with me if you have ideas about how we can best serve our field.
The results were admittedly imperfect a priori, because they relied on the invalid assumption that gender is a binary variable. This is another realm for which social-equity improvements in data science can't come too soon.