the editors of the Journal of Neurophysiology are committed to the rapid and constructive peer review of manuscripts. The average processing times (from submission to first decision) for new papers submitted in 2016 are indicated in Fig. 1. In this figure, each of the major article types published by the Journal (Innovative Methodology, Neuro Forum, Rapid Reports, Research Articles, and Review Articles) are represented in a separate column. Figure 1 shows that processing times for manuscripts varied tremendously, particularly for Research Articles (the time from submission to first decision ranged from 0 to 63 days). Although the median processing times were relatively short for most papers (25 days for Innovative Methodology, 15 days for Neuro Forum, 14 days for Rapid Reports, 24 days for Research Articles, and 23 days for Review Articles), a few authors had to wait up to two months for an editorial decision. This variability in article processing times was the impetus for an analysis of the root cause of delayed editorial decisions, with the intention of eliminating the problem.
All submissions to the Journal of Neurophysiology are handled through a standard workflow. After the article is submitted through the editorial Web site, the files are reviewed by the Journal’s peer review coordinator to ensure they are in the correct format and follow the prescribed style. This initial review usually occurs in less than one business day after receipt of the manuscript. Review Articles are additionally analyzed through plagiarism software to ensure there is no duplication of text from previous publications. Unless there is a problem with the formatting of an article or plagiarism is detected, the manuscript is transmitted through the editorial system to the Editor-in-Chief (EIC) the same day that the initial screening occurs. The EIC assesses each manuscript to determine whether it is consistent with the Journal’s mission and is of adequate scientific quality to merit peer review. A small fraction of manuscripts (3.8% in 2016) were rejected following this initial review by the EIC, accounting for the articles depicted in Fig. 1 with a processing time of 0 days. At this stage, the EIC decides to handle the manuscript personally or to transmit it to one of the Associate Editors (AEs). The EIC routinely handles some manuscript types, including Rapid Reports and Neuro Forum. The EIC also handles manuscripts on request of the author or when the AE designated by the author is unavailable. In most cases, the editor selected by the author is assigned the manuscript if that editor is available and is not in a position of conflict of interest (e.g., an author of the manuscript is in the same department as the editor or the two have an ongoing collaboration).
When an AE is assigned an article, they are tasked to solicit peer reviewers as quickly as possible; the EIC monitors the process and sends reminders to an AE if reviewers are not secured rapidly. Typically, invitations are sent to potential reviewers the day a manuscript is received. However, securing reviewers is often difficult. In 2016, an average of 10 reviewers were asked to review a manuscript before securing the desired number (usually 3), as indicated in Fig. 2. This represents a sharp rise over the number of reviewers asked to consider manuscripts in previous years (see Fig. 2). Many reviewers (average of 4 per manuscript; see Fig. 2) who were asked to consider a manuscript in 2016 never responded to the invitation, and an average of 3.6 reviewers per manuscript declined the invitation to review. It was often necessary to send several invitations to the same reviewer to solicit a response, which slowed the peer review process appreciably.
Invitation letters to potential reviewers request that they reject the assignment if they cannot complete the review within 14 days (10 days for Rapid Reports). Referees are sent reminders about the due date at 7 and 3 days before a review is due, the day after it is due, and weekly afterward. After a review is a week late, editors typically send additional personal reminders to the tardy referees. However, despite these efforts to secure timely reviews, it is often difficult to obtain the reviewer feedback required to make an editorial decision. Only 58% of referees submitted their reviews by the deadline in 2016; consequently, for 34% of manuscripts, the time from securing referees to receiving all of the requested reviews was over three weeks.
The processing time for 19 papers in 2016 (2% of the manuscripts received by the Journal) was over 50 days. In every case, this long processing time was due to late reviews by one or more referees. For most of these papers, the editor needed to solicit an additional reviewer well after the other referees were secured, to replace one or more reviewers who never completed their assignment.
Editor Strategies to Deal with the New Reality of Peer Review
Since it is difficult to solicit reviewers for manuscripts, and often harder to obtain reviews once referees have agreed to consider a manuscript, most of the Journal’s editors are adopting new strategies to reduce peer review times:
More reviewers are initially being invited to consider a manuscript, with the intention of quickly obtaining the desired number of reviewers.
Four reviewers are secured for some manuscripts, so is possible to dismiss late reviewers when an editor has enough feedback to make an editorial decision.
Certainly, there are drawbacks to these strategies. First, the most appropriate reviewers are not assigned a manuscript if they do not quickly respond to an invitation from an editor. Second, in those cases where all invited referees provide feedback on time, authors can receive four or more reviews. In 2016, 12% of Journal of Neurophysiology manuscripts had four or five reviewer reports.
Potential Author Strategies to Deal with the New Reality Of Peer Review
Like most journals, Journal of Neurophysiology allows authors to suggest potential referees and to disqualify referees based on scientific disagreements or personal conflicts. Suggestions of senior, impartial referees are always helpful to an editor. However, the Journal has a conflict of interest standard related to reviewer selection: a reviewer must be recused if they are a family member or close associate of one of the authors, have an ongoing scientific collaboration with any author, or have a joint publication with an author in the past three years or a mentor-trainee relationship in the past five years. Unfortunately, it is common for authors to suggest conflicted referees. Typically, an editor discovers these conflicts through a literature search or they are reported by a conflicted reviewer when asked to consider a manuscript. In a few cases, however, reviewer conflicts were not discovered until later, forcing the editor to disregard a review and seek additional referees. Introduction of a conflict of interest into the review process can greatly increase the processing time for a manuscript, and thus authors are strongly cautioned to adhere strictly to the Journal’s conflict of interest policy when suggesting referees.
Reviewer comments are often contradictory, which can result in an editor making an incongruous editorial decision. Consequently, the Journal has an appeal procedure to allow authors of rejected manuscripts to respond to reviews that contain inaccurate information and request that the editorial decision on their submission be reconsidered. Such appeals should be sent directly to the EIC. However, authors should be judicious when appealing an editorial decision, as the previous reviewers are typically consulted as the EIC evaluates the request. Authors who provide cogent arguments in an appeal are typically permitted to resubmit their manuscript, and on request new reviewers are solicited.
Our goal is to provide an editorial decision on manuscripts within three weeks after submission (two weeks for Rapid Reports). As indicated in Fig. 1, we meet this goal for most Rapid Reports but not for many other types of manuscripts. Although most AEs are vigilant in pursing tardy reviewers, authors are welcome to contact the AE handling a manuscript, the EIC, or both when there are concerns about processing delays. At the very least, the source of the delay is communicated to the authors following such inquiries.
Improving Peer Review
The peer review process, when it works appropriately, serves as a gatekeeper to prevent flawed information from being incorporated into the scientific literature and provides essential feedback to authors that improves their subsequent work. Although it is impossible to make definitive associations between problems in the peer review process and reliability of published scientific data (Landis et al. 2012; Morris 2012; Wadman 2013), it seems likely that improving peer review will also increase the quality of the literature. Fair, thorough, and rapid peer review is critical for the scientific enterprise to succeed.
Senior scientists are increasingly deferring the peer review of manuscripts to their trainees (Schiermeier 2016). Although we encourage the inclusion of trainees in the peer review process, as this provides essential training to the next generation of scientists, this exercise only is productive if the senior scientist provides oversight. We encourage senior scientists to accept review assignments and work collaboratively with their graduate students and postdoctoral trainees to complete the review, with the senior scientist providing quality control for the final review that is submitted.
With current constraints on grant funding in many countries and increasing demands of academia, it is certainly understandable that scientists often do not have time to review a manuscript. In such cases, they should decline a review request and provide thoughtful suggestions to the editor about alternative reviewers. One of the most time-consuming (and thankless) chores for an editor is soliciting reviews that are late. Referees should always bear in mind that late reviews are unfair to authors, who need rapid feedback on their work to maintain adequate productivity.
The Journal of Neurophysiology tracks the performance of reviewers and invites high-performers to join the editorial board. The nine editorial board members listed in Table 1 completed 10 or more reviews for Journal of Neurophysiology in 2016, all with average turnover times ≤15 days. We also list our best reviewers in our newsletter and provide other nonfinancial recognition of their efforts. However, the scientific community, academic institutions, and publishers need to come together to develop additional strategies that encourage scientists to participate in the peer review process.
Securing timely and objective reviews from scientists is essential to a successful peer review process. Without first-class peer review, the value of the scientific literature will decline, and ultimately public confidence in the scientific process will diminish. This cycle will inevitably lead to a loss of funding for scientific research. Peer review is one of the most worthwhile tasks of any scientist and should be regarded as such. In addition, the strategies used for peer review should be scrutinized scientifically (Rennie 2016), and new standards should be promulgated to improve the process.
B. J. Yates is Editor-in-Chief of the Journal of Neurophysiology and is a member of the American Physiological Society Council.
B.J.Y. analyzed data; B.J.Y. prepared figures; B.J.Y. drafted manuscript; B.J.Y. edited and revised manuscript; B.J.Y. approved final version of manuscript.
I thank Alex Potocki for assistance in collecting the data provided in this Editorial.
- Copyright © 2017 the American Physiological Society