Techniques for overcoming difficult interdisciplinary dialogue in expert panels: lessons for interactional expertise

Interdisciplinary expert panels face the Kuhnian problem of incommensurability across paradigms: each discipline has its own practices, language, and representations, making dialogue non-trivial. Building on the trading zones framework (Galison; Gorman; Collins & Evans), the paper focuses on fractionated trading zones—voluntary collaboration with heterogeneous outcomes—typical of interdisciplinary policy advice processes. The research question asks which techniques interactional experts (panel managers) can use to overcome obstacles to interdisciplinary dialogue in expert panels formed to produce systematic expert reviews for public authorities. The study situates the work within Gorman’s threefold agenda to refine trading zone taxonomy, detail acquisition of interactional expertise, and identify best collaborative practices, noting that the third strand is underexplored. The article examines French INSERM/INRA systematic expert reviews as a relevant setting where difficult interdisciplinary dialogue intersects with scientists’ reluctance to take public positions on policy problems, and outlines the plan to identify techniques used by panel managers to open, construct, and close debates and the skills these techniques require.

A targeted review of peer‑reviewed literature on trading zones and interactional expertise was conducted using Scopus (preferred over Web of Science and Google Scholar due to coverage and replicability). Approximately 90 works citing Gorman and containing “trading zone*” were identified; about 25 also addressed interactional expertise. The review highlights: (1) substantial work on refining trading zone taxonomy and on acquiring interactional expertise; (2) thinner, fragmented evidence on concrete techniques/best practices for collaboration in trading zones; (3) emerging claims that diversity of epistemic and cognitive perspectives in groups can mitigate groupthink and orthodoxies and foster shared understanding (e.g., Mormina; Moosavi & Browne), and that balance across disciplines and openness to including new disciplines support problem exploration (Norris et al.). The literature diverges on internal/external validity of observed practices: some propose robust, generalizable best practices; others argue for context-dependence and call for further empirical corroboration. The review concludes that techniques enabling interdisciplinary dialogue in panels remain insufficiently integrated and empirically tested, and that implications for measuring and teaching interactional expertise also warrant attention.

Case-based comparative study of five French systematic expert reviews (INSERM/INRA): Stress in the Workplace and Health (INSERM, 2011); Dietary Behaviours and Practices (INRA, 2010); Harm Reduction among Drug Users (INSERM, 2010); Animal Pain (INRA, 2009); Fruits and Vegetables in Eating Behaviours (INRA, 2007). These cases share a common procedure and time-space frame while varying in topics, disciplines, and commissioning bodies. Data collection (2013–2016) employed: (1) semi-structured interviews with panel managers (N=7; all women; mostly research engineers/expertise officers without contributory expertise in panel disciplines; mixed prior management experience) and INRA panel leads (N=3); (2) an online questionnaire to 119 panel members (46 responses; RR=39%); (3) analysis of documentary records (final reports, specifications, meeting minutes, etc.). Interviews and the questionnaire used the critical incident technique to elicit successes and failures in interdisciplinary collaboration. The INSERM/INRA procedure entails commissioning by a public authority; joint scoping and specifications; systematic literature search (with documentalists); assembling an ad hoc interdisciplinary panel; disciplinary literature analyses and presentations by members; drafting an integrative executive summary by managers with the panel; optional stakeholder inputs for controversial topics; and a public presentation. Data were inductively analysed to identify obstacles and manager-deployed solutions; triangulation across methods mitigated single‑method bias.

• Panels routinely faced two obstacles: (1) difficult interdisciplinary dialogue due to divergent disciplinary practices, languages, evidentiary standards, and conceptualizations (e.g., definitions of stress; standards of evidence across epidemiology, economics, sociology, biology); (2) some scientists’ reluctance to take public positions on policy-relevant issues, including self-censorship due to potential politicization or stakeholder pressure.
• Selection techniques (upstream): Managers applied formal and informal criteria beyond chartered principles. Core criteria included: (a) demonstrated scientific competence in the sub-field (publication track record); (b) projected engagement/commitment to the review workload; (c) predicted open-mindedness and capacity to compromise; (d) plurality and balance across disciplines and schools of thought (identified via experience and scientometric mapping), and attention to institutional viewpoints. Negative experiences underscored risks of underqualified contributors, single-school dominance, and missing disciplines.
• Techniques to open and construct debate: (a) structured disciplinary presentations followed by group discussion; (b) sharing public declarations of interests (PDIs) with the group to render affiliations visible and prompt self-restraint; (c) providing a common analytic frame for appraising publications to standardize evidence appraisal; (d) requiring justifications for excluding studies to surface biases and positions; (e) managers drafting and circulating an interdisciplinary summary that juxtaposes positions, which often catalysed transparent debate.
• Techniques to close debate and secure approval: (a) active moderation to temper vehement attitudes; (b) privileging positions supported by peer‑reviewed evidence over intuitions; (c) intra-panel peer rereads (one close, one distant discipline) with iterative revisions; (d) leveraging external authoritative reviewers to validate contentious sections; (e) reputational nudging to dissuade late resignations; (f) documenting persistent disagreements by presenting competing hypotheses in the final report when consensus is unattainable.
• Skill and knowledge conditions: Effective use of the above techniques depended on managers’ accumulated knowledge and know‑how developed through trial‑and‑error across trading zones: meta-disciplinary scientific literacy (biomedical/agricultural), bibliographic/database skills, facilitation and group management, writing/editing, and project management. These capabilities helped managers overcome status/gender prejudices and gain recognition from senior scientists.
• Sample/context: Five case studies; interviews with 7 managers and 3 panel leads; questionnaire RR=39% (46/119).

The findings directly address the research question by identifying concrete, replicable-in-context techniques that interactional experts can deploy to enable interdisciplinary dialogue in fractionated trading zones producing policy advice. Upstream selection for competence, commitment, open-mindedness, and plural schools/disciplines sets conditions for productive engagement. During panel work, structured turn‑taking, transparent conflict-of-interest visibility, standardized analytic frames, and manager-authored integrative summaries expose and clarify disagreements early. To close debates, moderation, evidence-based adjudication, peer and external review, reputational incentives, and explicit documentation of dissent help panels converge sufficiently to deliver reports without erasing legitimate disagreement. These practices have broader significance: they operationalize interactional expertise as a composite of cognitive (meta-disciplinary knowledge), interpersonal (facilitation, diplomacy), and procedural (project management, editorial) components, refining the concept beyond immersion in a single technical field. They also show that interactional expertise can be acquired through iterative participation and reflective learning across panels. Validity considerations are explicitly addressed: internal validity is supported by triangulated methods and detailed process description; external validity is bounded—practices are effective in the INSERM/INRA context and depend on manager capabilities and institutional procedures rather than constituting universal best practices.

The study advances Gorman’s third agenda strand by specifying techniques that interactional experts can use to overcome obstacles to interdisciplinary dialogue in expert panels: targeted selection criteria; structured presentation and discussion processes; transparency via PDIs; common analytic frames; manager-drafted integrative summaries; moderation and evidence-based closure practices; peer/external review; reputational incentives; and explicit reporting of unresolved dissent. It clarifies constituent elements of interactional expertise as knowledge, interpersonal skills, and procedural know‑how, and argues for a trial‑and‑error, reflective mode of acquisition beyond immersion in a single field. Implications include guidance for forming and managing interdisciplinary expert groups and for designing training curricula in interactional expertise (including analysis of failed collaborations). Future research should examine the effects of missing disciplines on outcomes, test these techniques in other institutional settings and issue domains, and evaluate pedagogy that incorporates failures and negative lessons in short university programmes.

• Context specificity: Findings derive from French INSERM/INRA systematic expert reviews and may not generalize to other institutional cultures or procedures.
• Sample and timing: Data collected 2013–2016 and focused on five panels; practices and norms may have evolved; recall bias is possible.
• Self-report bias: Interviews and questionnaires rely on participant recollections and perceptions; nonresponse bias (RR=39%) may affect representativeness.
• Researcher access and selection: Case selection balanced comparability and access, which may limit external validity.
• Measurement constraints: Effectiveness of techniques was inferred from qualitative accounts and document analysis rather than experimental or quantitative outcome measures.