Human behaviour is a key source of uncertainty in fisheries management [1] and management institutions are therefore increasingly interested in understanding human behaviour. The International Council for the Exploration of the Seas (ICES) is a science and advice-giving inter-governmental organization that informs evidence-based fisheries management. The ICES expert Working Group on Maritime Systems (WGMARS) produces research about human behaviour in fisheries that is relevant for governing bodies. WGMARS has set their focus for the 2020–2022 period on methodological, operational, contextual, and management aspects that enable successful ecosystem-based marine management and governance. As part of their terms of reference for this period, WGMARS aims to analyse how the use of behavioural economics can support integrated ecosystem assessment and ecosystem-based management [2]. Behavioural economics studies the cognitive biases and other drivers/mechanisms in decision-making processes. In contrast to traditional economics, which typically neglects the social complexity and context-dependency of human decision-making, behavioural economics makes use of a richer toolbox to explain human behaviour. In fisheries, a top-down management approach under the presumption of rational profit-maximising decision making by fishers is widely employed. While this is effective to an extent, such an approach omits other important mechanisms that are part of fisher behaviour. This may lead to unintended outcomes and raises uncertainties around the success of fisheries-management approaches [1, 3]. Some initial studies have explored the potential of using insights from behavioural economics in a fisheries context (e.g., [4, 5]). The proposed systematic review aims to consolidate knowledge from diverse literature on the use of behavioural-economics mechanisms in a fisheries context and provide an overview of relevant findings. Such an overview will be crucial for guiding the actions and recommendations within the ICES WGMARS and for others interested in understanding human behaviour in a fisheries management context.

The need for evidence on the use of behavioural economics in fisheries was identified by the behavioural-economics task group of WGMARS (BE task group hereafter). This group commissioned the review study by advertising the work to early-career researchers through their various networks. Three suitable researchers (lead authors hereafter) with heterogeneous backgrounds were identified:

The three lead authors dedicate their time out of interest in the topic, to gain experience in conducting such a study and to closely interact with and learn from the experts in the WGMARS.

First, the lead authors established the scope of the review (Table 1, Fig 1). For this study, stakeholders are defined as anyone who would be interested to understand, and/or potentially apply or recommend, the application of, behavioural-economics insights into fisheries management. Thus, as a key link between scientific research and policymakers, the key stakeholders for the scoping and design development of the systematic review are members of WGMARS (S2 Table). The group comprises a mix of natural scientists, economists, and other social scientists. The BE task group was consulted closely while WGMARS members were consulted through a webinar and with a follow-up feedback questionnaire on the scope and design of the systematic review (Table 1, Fig 1, S1 and S2 Files).

Of the 55 WGMARS stakeholders who were invited to give feedback through the questionnaire (S2 Table) 11 responded, most of which (n = 9) belonged to the BE task group. As an interdisciplinary group of stakeholders, five classified themselves as economists, four as other social scientists and three as natural scientists, with some selecting more than one scientific specialty.

When asked about the primary and secondary research question (Q7 and Q8 S2 File), stakeholders were divided about whether the research question should focus on fisheries-management strategies specifically or whether the focus should be on providing a broader overview of behavioural-economics mechanisms in fisheries. Nearly all stakeholders recommended to narrow down the focus in one way or another (e.g, only marine and not freshwater fishing) and to focus on fisher behavior. This is because of many differences in the regulatory, institutional, environmental and economic contexts between freshwater and marine fisheries. Specific suggestions in regard to the secondary questions were to investigate the instigating party, any unintended effects and to look at the relationships between the fishers and managers in the studies (Q8 and Q28 S2 File).

In this questionnaire the stakeholders were also asked to comment on a first overview table of behavioural-economics mechanisms which was previously compiled by the lead authors and the BE task group (Q9 S2 File). Input from this was used to draw up a more advanced version of the table, which will be used by the authors as a tool for the full-text screening (S3 Table). The behavioural-economics mechanism table and the comments provided on it were also used to draw up keywords, in addition to specific suggestions made by stakeholders (Q 10–19 S2 File). The stakeholders were presented with several options for databases from which to draw the literature, but most did not see themselves as experienced enough with systematic reviews to comment. When asked about the eligibility criteria (Q 27 S2 File), the group had a varied understanding of which studies should be included, with some disagreement about to what degree classical economics papers should be included in the review study. Where possible, the lead authors have systematically incorporated the feedback of the stakeholders, attempting to account for the diverse perspectives and priorities of the stakeholder group as well as the feasibility of this systematic review.

Primary question.

Secondary questions.

Drawing upon other review protocols (e.g., [6, 7]), we created a comprehensive list of potential databases and specialised websites. These were listed within the questionnaire (see page 17, S2 File) and stakeholders were asked for feedback. Additionally, the practicality and feasibility of the number of databases and specificity of the keywords were explored by ‘test-driving’ selected keywords. The feedback received from the stakeholder questionnaire and this test-driving process revealed that the amount of literature would be too large to be feasibly addressed systematically. It was therefore decided to focus on three databases: Web of Science (Core Collection), ProQuest (sub-selection of Social Science Core Collection), and EconLit (Table 2). The reason for this selection was the high quality of literature that these three databases return and because of the breath of coverage which is ensured by using a general (Web of Science) and two more research field specific databases (ProQuest for social sciences and EconLit for economics). The three databases will be searched with the keywords presented in Table 3. For each of the databases a specific Boolean search string was constructed from these keywords, for details of these please refer to S3 File. The keywords within the Boolean search string were the results of a pre-scoping exercise on relevant behavioural-economics mechanisms conducted by the lead authors and the BE task group (see “Stakeholder Engagement” section and S3 Table). The keywords were further expanded and commented on through the stakeholder questionnaire before being assessed for practicality (e.g., ambiguity, number of retrieved results in Web of Science).

In addition to extracting studies from web-based databases, stakeholders were given the opportunity to suggest peer-reviewed literature themselves (see stakeholder questionnaire: S2 File and S5 Table).

The search will be conducted in English only. The authors acknowledge that this means that valuable literature and research conducted in other languages will be missed. Being unable to include such literature can cause geographical biases and lack of important different perspectives or elements in the study [8] which result in creating barriers when trying to transfer knowledge internationally [9]. The implied limitations will be acknowledged in the manuscript by citing associated literature such as work by Amano et al. [8], who estimated that up to 35% of relevant literature may be missed if focusing on English language only (in their case in a biodiversity conservation context).

The comprehensiveness of the search was tested by applying the Boolean search string and matching the results with a list of 20 benchmark articles (S4 Table) (as suggested by authors such as Livoreil et al. [10]). The list of benchmark articles was drawn from the references given by Andrews et al. [5] and Battista et al. [4]–two highly relevant and related review studies. From these benchmark articles, AMW made an initial selection of 51 relevant articles, who screened the titles for relevance to the primary research question of this study. AMW, JKH and AS then separately screened the selected articles for relevance using abstracts as well as titles and each selected and ranked the 20 articles they deemed most relevant. From the 51 pre-selected articles, five articles were ranked among the 20 most relevant articles by all three lead authors, 12 by two lead authors and 18 articles were selected by only one lead author. Based on this ranking, the 20 overall highest scoring articles were selected to be used for this benchmark list.

Using this benchmark list, we found that the comprehensiveness of the search strategy (i.e. the number of benchmark articles that our search string retrieves on Web of Science) is only 40%. This appears to be a low number, but during the testing of the search string the lead authors determined that if the Boolean search string was adapted to capture more benchmark articles, the overall search would return a high amount of non-eligible articles. These articles would need to be manually screened and excluded during title, abstract and full-text screening, and the number of eligible articles would not be feasible to include in the systematic review. The lead authors propose that this issue could be caused by the complexity and diversity of (and occasional imprecision in) the terminology related to the field, as well as the accuracy of the indexing of the database [11]. Often, a combination of databases is recommended to ensure coverage of the search, with the Ebsco (of which EconLit is a part) and social-science databases such as PsychINFO being among those which seem to supplement more general ones such as the Web of Science Core Collection [11]. The lead authors have also followed the recommendation to consult experts in the field and ask for specific literature suggestions [12]. We therefore acknowledge that, despite best efforts, the performance of the Boolean search string (S4 Table) indicates that only approximately 40% of articles that were deemed relevant by experts and a qualitative review of the literature will be captured by this search.

We will follow the ROSES flow diagram during article screening, applying the consistency checks and eligibility criteria [13]. All articles returned by the search procedure will be divided between AMW and JKH, who will conduct title and abstract screening of the articles and decide on their inclusion or exclusion. Articles that pass through to full-text screening will be assessed for eligibility based on the entire article content. Articles that fulfill the eligibility criteria will then be reviewed using the synthesis form (S4 File). Articles will be assigned across the lead authors and to volunteers from the BE task team as well as Mary Mackay. First, however, two articles will be selected at random and all reviewers will independently review these articles according to the inclusion criteria below and populate the synthesis form independently. The authors will then meet to discuss and compare their decisions and the collated information by each author and check for consistency before the articles are assigned. This exercise will be repeated once all the reviewers have screened several papers.

The aim of the review study is to provide a narrative overview, based on a systematic review of the literature, of behavioural-economics mechanisms that drive fisher behavior. We are relying on peer-reviewed publications. As the work has already been peer reviewed in the publication process, we do not aim to appraise individual studies or the validity of the body of evidence ourselves.

Information on each of the studies will be collected via a password-protected form (S4 File) made available to all text-screening authors through the General Data Protection Regulation (GDPR)—compliant UK online survey platform (www.onlinesurveys.ac.uk) [14]. The form collects the following information, if available, for each article under review:

Meta-data extraction.

Data extraction.

Narrative synthesis.

The authors have no conflict of interest in the nature of the results. The lead authors are a heterogeneous group and have consulted multiple stakeholders to define the purpose of the review. There is no risk of publication bias, beyond the language bias mentioned previously, because this is simply an overview of past research on behavioural- economics mechanisms in fisheries contexts.

Together with the stakeholders the lead authors have designed the review to target knowledge clusters and knowledge gaps with the aim of synthesising them. Descriptive statistics will be used for the data synthesis to report the distribution of research in terms of mechanism, method, scale, sector, etc. in order to provide a descriptive impression of the knowledge clusters.

Most of the stakeholders work in the field of fisheries social science, and the lead authors have incorporated much of their feedback in designing this protocol. It is a varied group in terms of discipline and region, and all were given equal opportunity to provide written feedback on the study design. It is inevitable that some of the published work of the stakeholders (and co-authors of this work) will be retrieved as part of the review, but the reviewers will process and review each of the papers according to the protocol described above. Currently, none of the lead authors have published work that will be included in the review. However, AS has an article accepted which may be included in the final search. The lead authors have asked the stakeholders of the study for literature suggestions which will be added to the review, some of which may not be retrieved by the search terms. This may introduce a bias. For transparency, we will include the list of the literature that was suggested by the stakeholders in the supplementary information of the manuscript and highlight which of these were and were not captured by the systematic review search.

This systematic review protocol has been written following ROSES pro forma reporting guidelines [13] (for completed ROSES checklist refer to S1 Table). This systematic review study is on peer-reviewed literature and we therefore will not obtain ethics clearance or documentation. The individuals noted in the supplementary material of this manuscript have given written informed consent (as outlined in PLOS consent form) to publish these case details. We expect to publish the systematic review in a peer-reviewed journal.