Table 1. Previously suggested solutions for improving the peer-review process.

A list of suggested solutions and innovations in previous studies. Paying professional reviewers can be implemented in conjunction with any of the solutions below.

Proposed solution	Rationale and description
Implement peer review as an open, continuous, interactive forum involving all stakeholders	This solution is based on two assumptions: (1) blinding reviewers to the identity of authors does not improve the quality of peer review, 80 and (2) peer review should promote engagement, not judgement. 81 Practically, the idea is to transform the peer review into an open and interactive scientific discussion 3 , 7 that operates in real time where concerned parties (authors, reviewers, editors) can interact via an open platform. This could involve diverse stakeholders 8 through multiple platforms such as blogs and social media. 1 This could be seen as one model of the open peer review framework 82 ). Ultimately, this model would promote constructive discussion to enable transparent editorial decisions that aggregate diverse opinions. 83
Put in place a hybrid two-tier system where only impactful research is sent for peer review	This solution posits that not all submissions are ‘worth’ going through the prepublication peer review process. Instead, a hybrid model with a two-tier system is applied 79 : all submissions can be published and subjected to a post-publication peer review, but the prepublication peer review process needs only to evaluate manuscripts that are expected to have a significant impact on the field. 79 This is similar to some extent to desk-rejection practices where only submissions with high novelty and significance are sent for review. The main concern is of course how to make the initial evaluation about impact as objective as possible. This model, without safeguards, can end up biased toward top labs and renowned researchers. Moreover, it is not clear how citations from the unreviewed papers should be considered in the different research metrics.
Offer a non-selective review model to speed up the peer review process	Reviewers are often asked to make judgments about other aspects of the paper beyond its scientific merit, including for instance rating its novelty, significance, impact, originality, readability, and even whether or not the paper will be highly cited if accepted. The non-selective review model mainly focuses on papers’ scientific quality rather than their perceived importance and novelty. 8 Therefore, peer reviewers only have to evaluate the methodological soundness of papers. It is also possible to go further along this model by asking reviewers only to review the methods and results section. The other sections (including the authors’ interpretations of the results in the Discussion section) can be open for discussion after publication.
Involve independent peer review platforms	An initial evaluation of quality of papers is performed by other players. It is not unusual to hear peer reviewers complaining about the fact that they are often asked to review papers of poor quality. They believe it is not their job to improve papers as this should be the task of the original authors. In this era of publishing as many and as fast as possible, some submissions poorly report the undertaken work, which might make the work of reviewers even harder. Here the idea is to involve other parties to help the authors improve their submissions before submitting to a journal. This has been tested with the creation of independent peer review platforms, 3 helping researchers get their papers in good shape (for example, see platforms Rubriq and Editage). The authors pay the independent reviewers for their contribution to improving their papers. This model might add extra costs to the preparation process of papers.
Adopt a portable peer review across journals and publishers	This idea has been implemented by different initiatives (e.g. the Neuroscience Peer Review Consortium, the Review Commons). It is based on having a ‘portable’ peer review across publishers. 84 In this model, a paper goes through a journal-independent peer review, then the authors revise their paper accordingly and submit both their revised paper and the review reports to any journal that is taking part in the portable peer-review initiative. The revised submission can be moved among journals until a final editorial decision is made. 35 This model might help editors use reviewers time more efficiently but might delay the publication of papers (e.g. a paper might remain stuck in the process until it is selected by a suitable journal).
Create communities of researchers reviewing preprints	This model, promoted by “Peer Community in” (PCI), aims to organize the peer review process outside traditional journals. PCI is a non-profit scientific organization that creates communities of researchers to review and recommend preprints. It is based on an open-access system, a bottom-up process, and an active engagement of the scientific community. 85 It has strong support from many research institutions and funding agencies. As in the case of many not-for-profit initiatives, scalability would be an issue without sufficient resources.
Merge journals into mega-journals with large editorial boards	A mega-journal (like PLOS One and Nature’s Scientific Reports) has a very broad scope, accepting articles in almost all domains. It typically includes a large editorial board and a very large pool of reviewers. Having relatively low rejection rates, mega-journals usually publish thousands of articles per year. They tend to have an efficient (fast turnaround time) peer review process that can handle large submission volumes. 3 They sometimes play a similar role as a cascade journal, publishing papers rejected by more selective top journals. 86 Because they rely on a large number of academic editors across multiple domains, they can efficiently involve peer reviewers. However, this mega-journal model might not serve the interests of all domains, as specialized journals are needed.
Build an online reviewer registry accessible by all journals	The idea here is to build a large online registry of volunteer reviewers 87 who can be invited by any journal that has access to the registry. This registry can be created in a collaborative way by different journals or publishers, including for instance consortia of journals with overlapping scope or domains. 87 Such registries can complement the reviewers databases that journals hold. The registry can include reviewers from diverse geographical locations, demographics and expertise. To sustain such initiatives, a cross-publisher partnership has to oversee the creation and maintenance of the registry, with the mission to support it with the right resources.
Augment the review process with technology	Technology has already modernised the peer review process 88 ; e.g. automated detection of flawed methods or unethical practices like inappropriate statistical analysis, figure manipulation, or data fabrication. One can note the growing interest in the development of powerful and versatile software and platforms to streamline the peer-review process (e.g. Scholastica). It is expected that the adoption of generative AI will have many ramifications on the peer review process in the near future. 89 , 90
You review as much as you publish	As the number of prolific (very productive) researchers is growing very fast, 77 this model requires active involvement from such frequent users of the system. Therefore, researchers who publish more should contribute more to the peer review process. 91 It is not unusual to see many senior authors who publish a lot (and hence who are putting too much pressure on the system) are not involved at all in the peer review process. Some have even suggested that senior authors must provide evidence of their contribution to the peer review process as a requirement for considering their manuscripts for review. 91 This idea, albeit attractive, might not be feasible as it ignores the fact that senior authors are also very often involved in other academic review duties (doctoral theses, grant applications).
Offer incentives and rewards to reviewers	There is already a variety of rewards offered to reviewers as a recognition of their contribution to the peer review process. For instance, this includes prizes, discounts on publication fees, free access to the publisher’s products (e.g. databases), vouchers for book purchases, reduced registration fees for conferences, recognition in databases, and many others. Moreover, active reviewers who are submitting quality reports can be added to the editorial board of the concerned journals. Despite these many incentives, one cannot deny that the number of available reviewers is still declining.
Link peer review involvement to the researcher’s academic benefits	Other incentives argue for linking promotion applications (e.g. for tenured positions) to how active the researcher is in the peer review process. 92 Other suggestions include the possibility of earning continuing education credits upon completion of a review or waiving publication fees in the journal. 93 Others have even suggested the idea of offering co-authorship to reviewers, 94 in particular when the reviewers’ suggestions had significantly improved the original submission. However, this model might exclude those reviewers who are not ‎tenured academics.
Abandon the prepublication peer review model and adopt the post-publication peer-review (PPPR) model	The post-publication peer-review (PPPR) model 95 , 96 can ease the burden on journals with large submission volumes. It does not consider rejection after peer review. There is already rich literature about this model, 82 and it has been tested experimentally by different initiatives (e.g. F1000Research, ScienceOpen, The Winnower). Despite its many merits, this model has some limitations as it might let all papers be available in the public domain, regardless of their original quality and impact. Without safeguards and an initial thorough check of paper quality, this model might end up mimicking the work of predatory journals that publish all submissions without (prepublication) peer review. Hence, PPPR must allow for continuous discussion and improvement of the quality of publications. 95 However, there has to be a limit on when a paper can remain open for discussion as progress in science will lead to new experimental protocols and novel conceptual frameworks that will make ‘old’ publications look erroneous or obsolete.
Abandon the peer review process altogether	The (prepublication) peer review process should be abandoned as there is no strong evidence about its usefulness (e.g. see discussion in Ref. 14). The alternative is to gradually move toward expanding the role of preprint servers 8 , 26 or allowing universities and research institutions to manage their researchers’ publications. It is the researchers who create knowledge and it is thus the responsibility of their institutions to disseminate their work in the format they deem fit.