Collaborative Clinical Reasoning in the ICU: How Teams Construct Shared Understanding in Real Time

Collaborative Clinical Reasoning in the ICU: How Teams Construct Shared Understanding in Real Time | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Collaborative Clinical Reasoning in the ICU: How Teams Construct Shared Understanding in Real Time Ching-Yi Lee, Hung-Yi Lai, Ching-Hsin Lee, Mi-Mi Chen, Sze-Yuen Yau This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8666674/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Clinical reasoning research has predominantly conceptualised reasoning as individual cognition, yet contemporary clinical practice distributes reasoning across teams, tools, and time. This study theorises collaborative clinical reasoning (CCR) as a learnable social practice constituted through observable interactional mechanisms. Examining ICU teams as an epistemic extreme case—where high stakes and time pressure amplify reasoning practices present but less visible elsewhere—we ask: through what mechanisms do clinicians construct shared clinical understanding? Video-based microinteractional analysis of 18 ICU encounters (27 hours) identified three mechanisms through which clinicians collaboratively construct shared clinical understanding. We identified three interactional mechanisms through which CCR is accomplished. Progressive clarification is a mechanism of shared problem construction through layered, profession-specific contributions. Conditional alignment is a mechanism of epistemic negotiation enabling hypothesis exploration while preserving team cohesion. Situated coordination is a mechanism of material–interactional coupling whereby spatial positioning and artifacts structure participation. These mechanisms represent learnable practices that can be noticed, supported, and coached. By shifting the unit of analysis from individual cognition to interactional accomplishment, this study provides an empirically grounded foundation for understanding how collaborative reasoning is learned and sustained through participation in clinical work. Collaborative clinical reasoning Critical care Microinteractional analysis Teamwork in healthcare Workplace learning Distributed cognition Figures Figure 1 Figure 2 Introduction Clinical reasoning has been conceptualised predominantly as an individual cognitive process (Croskerry, 2009 ; Eva, 2005 ; Norman, 2005 ). This tradition is grounded in information-processing theory, which models reasoning as a structured progression from data acquisition to problem representation and solution generation, and in dual-process theories, which distinguish between rapid, pattern-based intuitive reasoning and slower, analytic reasoning (Croskerry, 2009 ; Eva & Norman, 2005 ; Norman et al., 2018 ). These frameworks have yielded valuable insights into expert–novice differences and diagnostic error, and they continue to inform teaching strategies and assessment in medical education. However, their underlying assumptions—particularly that clinicians can reason with stable access to relevant information, minimal interruptions, and sufficient cognitive bandwidth—reflect cognitive conditions that are less representative of contemporary clinical environments, where reasoning is frequently distributed, time-pressured, and socially embedded. Modern healthcare is fundamentally team-based, particularly in acute and high-acuity environments where clinical work is distributed across professionals who must integrate heterogeneous expertise to manage rapidly evolving patient conditions (Manser, 2009 ; Endsley, 2017 ; Weaver et al., 2014 ). In such settings, the cognitive resources needed to understand and respond to complex patient trajectories are not held by any single individual but are distributed across team members, technological artifacts, and the physical environment (Reader et al., 2011 ; Salas et al., 2008 ; Hutchins, 1995 , 2000 ; Healey et al., 2006 ; Lingard et al., 2002 ). Despite the central role of collaboration in critical care, research on clinical reasoning has largely remained anchored to individualist cognitive models (Berg, 1999 ; Cox et al., 2017 ; Donchin & Seagull, 2002 ; Lane et al., 2013 ; Reader et al., 2011 ). Although communication frameworks such as TeamSTEPPS and structured handoff protocols address information transfer, they do not provide an account of how reasoning itself is jointly produced in interaction (Hindmarsh & Pilnick, 2007 ; Lingard et al., 2002 ). Theories of sensemaking (Maitlis & Christianson, 2014 ; Weick & Weick, 1995 ) and action-oriented problem solving (Rudolph et al., 2009 ) offer conceptual resources for understanding collective cognition, and distributed cognition (Hazlehurst et al., 2007 ; Hutchins, 1995 , 2000 ) emphasises how cognitive processes extend across people, artifacts, and environments. Yet empirical studies grounded in these perspectives have rarely examined how clinical teams construct shared clinical understanding in real time through moment-to-moment interaction. This gap is consequential for medical education. While calls to situate clinical reasoning within its contextual and social dimensions have grown more prominent (Cook et al., 2013 ; Ilgen et al., 2012 ; Durning et al., 2013 ; Kneebone, 2005 ), much of this work remains theoretical. Simulation-based training offers controlled environments in which to practise team-based reasoning, but without an empirically derived account of how collaborative reasoning unfolds in authentic clinical practice, such training risks reproducing idealised models rather than reflecting the interactional realities of clinical work (Eva, 2005 ; Norman, 2005 ). Given the central role of collaboration in ICU practice and the lack of empirical evidence detailing how CCR unfolds in situ, this study employs video-based microinteractional analysis to theorise collaborative clinical reasoning as a learnable social practice constituted through observable interactional mechanisms. By examining ICU teams as an epistemic extreme case—where high stakes, time pressure, and distributed information structures amplify reasoning practices that are likely present, though less visible, in other clinical education settings—we aim to identify the mechanisms through which clinicians jointly construct shared clinical understanding. This approach offers generalisability that is theoretical and mechanism-based rather than statistical, providing an empirically grounded foundation for understanding how collaborative reasoning is learned and sustained through participation in everyday clinical work. Materials and Methods This study employed video-based microinteractional analysis as a methodological instrument for operationalising abstract theoretical constructs—reasoning, alignment, sensemaking—as observable, sequential practices. This design was selected because CCR is fundamentally an interactional accomplishment, emerging through the moment-to-moment coordination of talk, gesture, gaze, and material artifacts. Retrospective interviews and surveys cannot capture the temporal, embodied, and sequential organisation of reasoning as it unfolds; video-based microanalysis makes these theoretical claims empirically tractable. Our methodological orientation assumes that clinical reasoning becomes empirically visible through the sequential organization of social interaction. By examining how clinicians initiate, respond to, challenge, and align interpretations in real time, microinteractional analysis allows us to describe the mechanisms through which shared clinical understanding is built and sustained under complex, high-stakes conditions. The study was conducted in the adult medical ICU of a tertiary academic medical center in Taiwan, where interdisciplinary teams routinely engage in collaborative assessments and decision making. The ICU was selected as a theoretical and analytic "extreme case" because its time pressure, uncertainty, rapid patient trajectories, and distributed information structures render interactional reasoning processes more visible than in lower-acuity settings. Video-based microinteractional designs require methodological transparency regarding the stages of data generation, selection, and analysis. Accordingly, the study design involved five interlocking components: 1. Naturalistic Video Data Collection We recorded 27 hours of naturally occurring ICU team interactions across 18 discrete encounters, including morning rounds, rapid patient assessments, and emergent decision-making episodes. These encounters involved interdisciplinary participation from attending physicians, fellows, residents, nurses, and respiratory therapists. Video-based observation was conducted in a non-intrusive, naturalistic manner to preserve the ecological fidelity of routine clinical practice. To capture the complexity of team interaction and spatial dynamics, we used two stationary high-definition cameras per event. Cameras were mounted unobtrusively within the patient room or team workspace to provide complementary angles: one oriented toward the main interactional floor (e.g., bedside or rounding cluster), and the other capturing peripheral activity and monitor/tool use. Stationary placement minimized disruption to workflow and enabled continuous recording without operator involvement. This configuration allowed us to document verbal exchanges, gaze and body orientation, use of clinical artifacts (e.g., bedside monitors, ventilators, mobile computers), spatial arrangement and movement within the care environment, and emergent actions during unexpected clinical changes. By capturing clinicians' naturally unfolding communication and coordination, this approach ensured ecologically valid data that reflected how CCR emerges spontaneously, without researcher prompting or simulation constraints. 2. Multimodal Transcription and Interactional Documentation All selected episodes were transcribed using Jeffersonian transcription conventions, enabling fine-grained representation of the sequential organization of talk, including overlaps, micro-pauses, cut-offs, timing, and prosodic contours. Because CCR unfolds through multiple modalities (Mondada, 2018 ), multimodal annotation captured gaze direction and gaze shifts, body orientation and postural alignment, gestures, pointing, and hand movements, object manipulation (e.g., adjusting ventilator settings, touching monitors, flipping charts), interaction with representational artifacts, including electronic health records, bedside monitors, imaging displays, and mobile devices, and spatial configurations of team members during interaction. These multimodal layers were essential for analyzing reasoning as an embodied, spatially distributed, and tool-mediated activity, allowing us to examine how participants coordinated attention, displayed epistemic stance, and constructed shared interpretations through talk-and-action sequences. Transcription and annotation were performed using ELAN (Max Planck Institute for Psycholinguistics), a software platform designed for synchronous, multi-tiered representation of audio-video data. ELAN enabled us to align verbal, prosodic, embodied, and artifact-related actions with millisecond precision, ensuring analytic access to the temporal and interactional contingencies that shape collaborative reasoning in real time. 3. Episode Selection and Theoretical Sampling From the full corpus of 18 recorded ICU encounters, episodes for close analysis were selected through an iterative theoretical sampling process. This process followed interpretive, interaction-focused methodological logic, in which sampling is guided not by representativeness but by the analytic richness of sequences that illuminate the phenomenon of interest—in this case, CCR. During initial rounds of corpus-wide viewing, the research team identified sequences in which clinicians engaged in active interpretive work rather than procedural or purely informational exchanges. Preliminary markers of interest included surfacing or articulating uncertainties, constructing or revising patient trajectories, questioning, refining, or challenging interpretations, coordinating diagnostic or management decisions, and repairing misunderstandings, misalignments, or knowledge gaps. These markers allowed us to locate segments where reasoning was not merely expressed individually but co-constructed interactionally. As analytic patterns emerged, sampling became increasingly focused on episodes that provided maximal insight into the interactional mechanisms of CCR. Out of the 18 recorded encounters, 12 episodes were ultimately selected for full, line-by-line, multimodal microinteractional analysis. These episodes were chosen because they contained clear evidence of collaborative reasoning, rich multimodal interaction, and sequential structures that enabled close examination of how shared understanding was produced and maintained. 4. Sequential, Microinteractional Analysis Data analysis followed the principles of conversation analysis and multimodal interaction analysis, which together make visible the fine-grained practices through which clinicians construct shared clinical reasoning in real time. This approach treats interaction as sequentially organized, meaning that each action (verbal or embodied) derives meaning from how participants position it relative to what came before and anticipate what comes next. Because CCR depends on how clinicians jointly build, contest, and align interpretations, sequential analysis provides a powerful lens for illuminating the underlying mechanisms. Across the selected episodes, we examined turn-taking and participation structure, repair and the management of trouble, epistemic positioning and stance, multimodal coordination of talk, gesture, and artifact use, and response to interruptions and emerging clinical information. Through this sequential, multimodal analysis, we identified recurrent interactional mechanisms through which CCR was achieved, including distributed noticing (how clinicians jointly surfaced and oriented to clinically relevant cues), interpretive alignment (how clinicians converged on shared problem framings or diagnostic narratives), challenge–justification sequences (how disagreements or divergent interpretations were enacted, managed, and resolved), recalibration of shared understanding (how teams updated collective reasoning when new information emerged or prior assumptions were destabilized), and structuring shared problem frames (how participants collaboratively articulated "what is going on," stabilizing the trajectory of diagnostic or management decisions). 5. Reflexive, Multidisciplinary Analytic Review To enhance analytic credibility and ensure that interpretations remained grounded in both clinical reality and interactional theory, the analysis incorporated a reflexive, multidisciplinary review process. The core analytic team consisted of three researchers with complementary expertise in ICU clinical practice, medical education, and ethnomethodology/interaction analysis, supported by one external consultant with a background in psychology and qualitative methodology. This diverse expertise enabled us to examine interactional patterns from multiple epistemic vantage points and to avoid privileging a single disciplinary interpretation. Team meetings occurred iteratively throughout data collection, transcription, and analysis. During these sessions, analysts interrogated preliminary findings, examining alternative explanations for observed interactional phenomena; identified potential analytic blind spots, including the risk of over-interpreting or under-interpreting participants' actions; challenged disciplinary assumptions, ensuring that interpretations did not rely on clinical or educational presumptions unsupported by the data; compared analytic claims against the video record, returning repeatedly to the raw data to validate emerging interpretations; and refined conceptual categories, testing whether identified mechanisms were robust across multiple episodes and contexts. Ethical Considerations This study was approved by the Institutional Review Board of Chang Gung Medical Foundation (Ref: 202002453B0). All participating clinicians were informed of the study purpose, the use of video recording, and their right to withdraw at any time. Written consent was obtained prior to recording. To protect confidentiality, video data were stored on encrypted, access-restricted servers; all identifying information was removed during transcription and analysis. When excerpts were used for reporting or dissemination, faces and identifying features were blurred unless explicit additional consent was obtained. Results Analysis of 12 ICU episodes revealed that CCR emerged not as a discrete cognitive act but as an interactional trajectory, accomplished through the coordinated use of talk, embodied conduct, and material artifacts. Across diverse clinical scenarios—routine rounds, rapid assessments, and emergent deteriorations—three recurrent interactional patterns characterized how teams built, negotiated, and stabilized shared clinical understanding. These three mechanisms—progressive clarification, conditional alignment, and situated coordination—are summarized in Table 1 and illustrated in the conceptual model presented in Fig. 2 . Table 1 Interactional Mechanisms of Collaborative Clinical Reasoning in the ICU Mechanism Core interactional practices Sequential function Multimodal resources Educational significance Progressive clarification (shared problem construction) Layering profession-specific observations; solicitation sequences; repair of ambiguities; conditional framing of epistemic gaps Incrementally assembles coherent clinical picture from distributed fragments; each contribution refines the emerging interpretation Talk; turn-taking; gaze to monitors/charts; pointing gestures Reveals how learners contribute fragments vs premature conclusions; repair sequences expose coachable reasoning gaps; models epistemic humility Conditional alignment (epistemic negotiation) Mitigated proposals; hedged alignments; softened challenges; evidence-grounded justification Enables hypothesis exploration while preserving team cohesion; tests alternatives without premature closure or interpersonal risk Prosody; epistemic markers; gaze patterns; pause placement Mitigated formulations indicate competence, not weakness; stance-taking practices are modelable; hierarchy-crossing contributions become visible Situated coordination (material-interactional coupling) Spatial positioning relative to artifacts; gaze convergence; pointing and tracing; embodied bids for epistemic primacy Structures participation through material and spatial affordances; bodies and tools direct attention and ground claims Body positioning; gaze direction; screen sharing; artifact manipulation Physical arrangement shapes whose knowledge gets uptake; artifact use is coachable reasoning practice; spatial positioning is learnable Note. Each mechanism represents a learnable social practice observable in routine clinical work. Mechanisms operate interdependently rather than as sequential stages. Educational significance describes what supervisors might attend to, without prescribing curricular interventions. Abbreviations: ICU = intensive care unit. 1. Progressive Clarification: A Mechanism of Shared Problem Construction Across all analyzed episodes, CCR did not emerge as a single decisive moment or an individual's formulation of a complete diagnostic frame. Instead, teams constructed reasoning progressively, through a sequence of layered contributions, clarification moves, and epistemic repairs that gradually stabilized a shared understanding of the patient's condition. This pattern—progressive clarification—comprised three recurrent practices: (a) layering partial assessments, (b) refining problem frames through solicitation and repair, and (c) managing epistemic gaps to sustain forward movement in reasoning. Figure 1 illustrates how progressive clarification unfolds as an interactional mechanism of shared problem construction. 1.1 Layering assessments and accumulating fragments of evidence A defining feature of CCR was the way clinicians contributed fragmented, profession-specific observations that, in aggregate, began to form a coherent clinical picture. Nurses frequently offered near-real-time physiologic cues or bedside observations; residents contributed laboratory trends and recent documentation; fellows added ventilator data or imaging interpretations; and attendings synthesized across these informational streams. These contributions were rarely presented as final judgments. Instead, they appeared as tentative fragments—brief observations, trend updates, or short descriptors—positioned sequentially as supplements to one another ("and also...", "I'm seeing...", "it's a bit higher today..."). Through this turn-by-turn accumulation, the team collaboratively constructed a diagnostic storyline that no individual participant held in full. The emerging clinical picture thus reflected a distributed noticing process, shaped by each participant's embodied location in the unit, their access to particular artifacts (monitors, ventilators, electronic charts), and their role-specific responsibilities. 1.2 Refining problem frames through solicitation and repair Once fragments of information began to coalesce into a preliminary frame, clinicians engaged in solicitation and repair practices to refine the accuracy and coherence of the shared interpretation. These moves included requests for clarification ("Was that rising or stable?"), specification probes ("How much higher than yesterday?"), corrections and confirmations ("No, I meant the later result."), and elaborations prompted by misunderstandings or ambiguous phrasing. Importantly, these repair sequences were not mere corrections of error. They played a constitutive role in shaping the diagnostic trajectory, ensuring that assumptions were interrogated and that the shared problem frame rested on mutually confirmed information. Through solicitation and repair, clinicians collaboratively tightened the precision of the emerging narrative, adjusting weight assigned to particular cues and recalibrating interpretations in response to newly clarified details. The multimodal environment also supported this refining process: clinicians frequently oriented to monitors, displayed charts, or gestured toward imaging screens to anchor their clarifications, using material artifacts as external reference points for resolving ambiguity. 1.3 Managing epistemic gaps to sustain forward reasoning Reasoning in the ICU is constrained by incomplete and evolving information. Across episodes, teams routinely encountered missing laboratory results, pending imaging, or contradictory clinical indicators. Rather than halting discussion, clinicians addressed these epistemic gaps by explicitly identifying what was unknown and articulating conditional assumptions to guide interim decision-making ("Let's proceed assuming his CO₂ hasn't worsened and revisit when the ABG returns."). These conditional framings served two functions: (1) Maintaining progress in the face of uncertainty, allowing the team to act without overcommitting to any single interpretation, and (2) Keeping interpretive space open, marking unresolved issues for later verification and preventing premature diagnostic closure. Multimodal cues—such as gestures that circled back to artifacts, pauses that framed conditional statements, or gaze shifts that signaled pending data sources—helped the team collectively track the provisional nature of these assumptions. Through this process, teams enacted epistemic repair, stabilizing the reasoning trajectory while transparently acknowledging uncertainty. 2. Conditional Alignment: A Mechanism of Epistemic Negotiation A second recurrent pattern involved how clinicians aligned around emerging interpretations while preserving flexibility to revise them as new information appeared. Whereas progressive clarification described how teams built a shared clinical picture over time, conditional alignment captured how they managed the epistemic and interpersonal delicacy of proposing and negotiating interpretations in a high-stakes, uncertain environment. Across episodes, clinicians rarely presented diagnostic assessments as categorical claims. Instead, they relied on conditional formulations, mitigated assertions, and softened disagreements that enabled collaborative hypothesis testing without foreclosing alternative framings or jeopardizing team cohesion (Table 1 ). 2.1 Proposals framed as possibilities, not certainties Interpretive trajectories commonly began with hypotheses introduced in mitigated, non-finalized forms—" maybe we're seeing early sepsis ," " it could be fluid-responsive ," " I'm wondering if this is evolving shock. " These proposal formats had two key interactional consequences. First, by avoiding categorical assertions, speakers reduced the interpersonal and epistemic risks associated with potentially inaccurate claims. This was especially salient across hierarchical boundaries, where trainees and junior clinicians routinely employed epistemic downgrades ("maybe," "seems like...," "could be") and embodied displays such as upward glances or softened prosody to position their contributions as tentative rather than authoritative. Such stance markers allowed juniors to contribute interpretively without violating expectations of deference or overclaiming knowledge in front of seniors. Second, mitigated proposals functioned as invitations to joint reasoning. Rather than presenting a hypothesis as a completed cognitive product, these formulations positioned the proposal as open for collective scrutiny and elaboration. Their sequential design—often accompanied by rising intonation, slight pauses, or gaze shifts toward others—projected a responsive next action, signaling that the speaker anticipated additions, modifications, or alternative framings from the team. 2.2 Alignment without premature closure Responses to tentative proposals typically took the form of conditional alignments, rather than unequivocal endorsements. Clinicians frequently responded with turns such as " that might fit given... ," " unless this is more cardiac... ," or " that could explain the drop unless we're missing something ," which simultaneously acknowledged the plausibility of the preceding interpretation while signaling that additional contingencies remained in play. These formulations enabled participants to sustain the forward motion of the reasoning trajectory without committing prematurely to a single explanatory frame. Such alignments were often supported by embodied cues that indexed conditionality: half-nods, brief glances toward monitors or lab displays, or hand movements that traced possible directions of interpretation. These multimodal actions tempered the apparent commitment of the verbal turn, displaying alignment with the proposal while leaving interpretive space open. 2.3 Negotiating competing interpretations Divergent interpretations arose frequently across the dataset, particularly in episodes involving ambiguous presentations or rapidly shifting physiological states. Rather than issuing direct disagreement, clinicians typically introduced alternative framings through softened challenge formats, such as " I wonder if it's actually... " or " Could it instead be...? ". These mitigated formulations allowed speakers to express divergent views without overtly contradicting colleagues, thereby preserving rapport and signaling respect for others' epistemic positions. Once an alternative perspective was introduced, clinicians commonly followed with justification, grounding their reasoning in observable data. Participants oriented toward monitors, gestured toward ventilator readouts, traced trends on charts, or pointed to imaging displays as they marshalled evidence. These embodied and material actions anchored disagreements in shared referential spaces, enabling the team to collectively inspect the basis of the alternative interpretation. Through this interactional work, teams typically moved toward reconciliation or reframing, integrating the newly raised interpretation or clarifying why a prior account remained more plausible. Conditional language, mitigated stance-taking, and multimodally grounded justification enabled teams to negotiate difference productively and safely. 3. Situated Coordination: A Mechanism of Material–Interactional Coupling CCR in the ICU unfolded not only through verbal exchanges but through clinicians' embodied coordination and engagement with the material environment. Spatial positioning, gaze behaviour, and the use of tools such as monitors, ventilators, and charts shaped who contributed, how proposals were taken up, and which interpretations gained momentum. Across episodes, the material and spatial ecology of the ICU acted as an active substrate for reasoning, structuring attention, authority, and interpretive trajectories (Table 1 ). 3.1 Spatial positioning as a resource for participation and authority Across episodes, clinicians' spatial positioning within the patient room played a consequential role in shaping how collaborative reasoning unfolded. Those standing closest to key informational artifacts—particularly the bedside monitor, ventilator screen, or electronic chart—often functioned as epistemic anchors, not by explicit designation but through the affordances of proximity. Their embodied access to real-time physiologic data positioned them to initiate trend commentary or highlight salient cues, such as gesturing toward a dropping pressure or pointing out changes in waveform morphology. These embodied demonstrations routinely drew the team's gaze toward the same artifact, thereby structuring what was treated as relevant evidence in the emerging interpretive trajectory. Physical proximity also influenced participation rights and the flow of contributions. Clinicians adjacent to the monitor or ventilator more frequently assumed the role of "first describer," narrating data trends before others had visual access. Their descriptions often became the reference point for subsequent reasoning moves, with others building upon, questioning, or refining the initial framing. Importantly, spatial positioning did not merely reflect existing hierarchies; it actively shaped them. While attendings often occupied central or mobile positions, nurses and respiratory therapists situated at the bedside or near equipment frequently contributed authoritative assessments anchored in their material proximity and continuous monitoring responsibilities. Thus, spatial arrangement served as a dynamic resource through which authority was enacted, negotiated, and redistributed moment to moment. 3.2 Gaze behavior and the organization of diagnostic attention Gaze behavior played a central role in organizing how clinicians oriented to and evaluated emerging interpretations. Across episodes, proposals were routinely accompanied by predictable gaze shifts—from the speaker to the relevant artifact, from artifact to team members, or among team members themselves—that structured the collective attention of the group. When a clinician advanced a diagnostic or interpretive proposition, others often responded first not with talk but with gaze convergence toward the associated monitor, chart, or ventilator display. This coordinated visual orientation served as an embodied marker of collective evaluation, signaling that the team was inspecting the evidentiary substrate of the proposal. The timing of these gazes shifts contributed to the organization of the reasoning sequence. Early gaze alignment—occurring prior to verbal uptake—often foreshadowed acceptance or elaboration of a proposal, while delayed or hesitant gaze shifts sometimes marked uncertainty or hesitation to commit to the interpretive direction being suggested. Through these intertwined practices, gaze served not simply as a marker of attention but as an interactional mechanism for structuring how reasoning unfolded—highlighting relevant data, cueing participation, and visually synchronizing the team's interpretive efforts. 3.3 Artifacts as co-participants in reasoning Material artifacts—bedside monitors, ventilator interfaces, electronic charts, imaging displays, and laboratory panels—played an active and constitutive role in shaping the trajectory of collaborative clinical reasoning. Across episodes, clinicians treated these tools not merely as repositories of information but as semiotic partners whose displays, alarms, and visual affordances structured how interpretations were formulated, justified, and revised. A common pattern involved clinicians synchronizing talk with embodied engagement with artifacts. Participants frequently gestured toward waveform shifts, traced lab trends with their fingers, or angled screens toward colleagues as they narrated emerging concerns. These multimodal displays transformed artifacts into shared reference points, enabling the team to jointly inspect data and anchor interpretive claims in observable features of the environment. Artifacts also mediated challenge and justification sequences. When questioning a proposal, clinicians often oriented first to the relevant display—scrolling through prior values, enlarging a waveform, or highlighting specific parameters—before articulating an alternative interpretation. This practice grounded disagreements in shared evidentiary terrain, ensuring that divergent perspectives emerged from a common visual foundation. In several episodes, artifacts exerted directional force on the reasoning process by shifting salience as new values appeared or when alarms signaled abrupt physiological change. These events prompted rapid reorientation of team attention, often triggering reframing or recalibration of the working diagnosis. The artifact, in effect, introduced new turns into the reasoning sequence, acting as a participant whose "contributions" shaped what became relevant for immediate discussion. Through these practices, artifacts functioned not as passive background objects but as integral components of the cognitive system within which ICU reasoning unfolded (Fig. 2 ). Their spatial arrangement, visual affordances, and temporal responsiveness actively structured the ways clinicians noticed, evaluated, and coordinated interpretations. Viewing artifacts as co-participants highlights CCR as a deeply distributed, multimodal phenomenon—one in which reasoning emerges from the interaction of people, bodies, and tools in a tightly coupled complex system. Discussion This study examines CCR as it unfolded within the material, spatial, and interactional ecology of the ICU. Rather than treating reasoning as a cognitive process internal to individual clinicians, our findings demonstrate that CCR is an emergent, distributed, and interactionally organized accomplishment. Across 12 video-recorded ICU episodes, teams built and sustained shared clinical understanding through the coordination of talk, embodied conduct, and tool-mediated action. Three mechanisms—progressive clarification, conditional alignment, and situated coordination—capture how clinicians jointly assembled interpretations, managed uncertainty, and oriented their reasoning within the spatial and material affordances of the ICU (Table 1 ; Fig. 2 ). This study extends existing research on clinical reasoning in three ways. First, it shifts the unit of analysis from individual cognition to interactional accomplishment, providing an empirically grounded account of collaborative reasoning as a social practice. Second, it identifies specific interactional mechanisms through which reasoning is constructed, negotiated, and stabilised in practice—mechanisms that represent learnable skills which can be noticed, supported, and coached. Third, it demonstrates how video-based microanalysis can operationalise abstract learning theories, contributing a methodological pathway for future research in health professions education. Although this study does not evaluate an educational intervention, the interactional mechanisms identified here provide a basis for understanding how collaborative reasoning is learned and sustained through participation in everyday clinical work. CCR as Emergent Sensemaking Rather Than Individual Cognition Traditional clinical reasoning frameworks draw heavily on information-processing models of the individual mind (Elstein et al., 1978 ; Croskerry, 2009 ; Norman, 2009 ). Our findings align more closely with distributed cognition perspectives that locate reasoning across people, artifacts, and environments (Cook & Rasmussen, 2005 ; Hutchins, 1995 ; Lingard, 2012 ). The three interactional mechanisms we identified—progressive clarification, conditional alignment, and situated coordination—can be understood as forms of sensemaking: the ongoing, retrospective process through which actors render equivocal situations more comprehensible and actionable (Weick et al., 2005 ; Weick & Weick, 1995 ). To clarify the theoretical boundaries of CCR, we position it as a specific form of sensemaking that is distinct from, though related to, several cognate constructs. Unlike team cognition, which typically examines shared mental representations or transactive memory systems, CCR foregrounds the sequential, turn-by-turn emergence of clinical interpretations through interaction. Unlike shared mental models, which posit stable cognitive structures held by team members, CCR emphasises the dynamic, contingent process through which interpretations are assembled and revised in real time. Unlike team situation awareness, which focuses on the accuracy of shared perceptions about the environment, CCR attends to how such perceptions are interactionally achieved and negotiated. What CCR adds to these frameworks is threefold: first, attention to sequential emergence—how reasoning unfolds through ordered contributions rather than aggregated cognitions; second, a focus on epistemic risk management—how clinicians navigate the personal and professional stakes of proposing and challenging interpretations; and third, emphasis on embodied coordination—how spatial positioning, gaze, and material artifacts constitute rather than merely support reasoning. By specifying these mechanisms, CCR provides an empirically tractable account of collaborative cognition grounded in observable practice. What distinguishes our account from abstract theorizing about sensemaking is its grounding in observable, sequential practices. Drawing on conversation analysis and multimodal interaction analysis (Schegloff, 2007 ; Garfinkel, 2023 ; Heath et al., 2007 ; Hindmarsh & Pilnick, 2007 ; Mondada, 2018 ), we showed how shared clinical understanding is constituted turn by turn, with each contribution shaping—and being shaped by—what came before. Viewing CCR as sensemaking foregrounds how ICU teams construct, sustain, and revise frames of relevance under conditions of uncertainty, rapid physiological change, and heavy information load. This perspective moves beyond cognitive-individualist models and emphasizes CCR as an emergent, collective accomplishment shaped by the social, material, and temporal ecology of clinical work. Interactional Management of Uncertainty and Epistemic Risk Research on team communication has emphasised the importance of psychological safety and speaking up to patient safety (Leonard et al., 2004 ; Sutcliffe et al., 2004 ). Our analysis deepens this work by specifying the interactional resources clinicians use to manage epistemic risk in situ. Central to the mechanism of conditional alignment was the use of mitigated formulations—hedged proposals, partial claims, and tentativeness markers (Heritage, 2012 ; Heritage & Raymond, 2005 )—that enabled clinicians to introduce interpretations without overstating certainty. Junior clinicians, in particular, employed epistemic downgrades (" maybe ," " seems like... ," " I'm wondering if... ") that served dual purposes: they managed the personal and professional risks associated with being wrong, and they positioned their proposals as inviting elaboration rather than asserting conclusions. Such formulations helped to distribute interpretive work across the team, encouraging seniors to contribute refinements without framing the trainee's contribution as an error. Similarly, conditional alignments—responses like " that might fit given... ," " unless this is more cardiac... ," or " that could explain it if... "—allowed the team to move forward collaboratively while explicitly preserving space for revision (Klein et al., 2006 ). Our findings also complicate the assumption that productive disagreement must be explicit or direct. Rather than overt contradiction, clinicians frequently used softened challenge formats (" I wonder if it's actually... ," " Could it instead be...? ") that introduced alternative framings while minimizing interpersonal threat. These patterns are consistent with research on dispreference in social interaction (Kotthoff, 1993 ; Pomerantz, 1984 ) and suggest that effective team reasoning depends on face-sensitive practices that allow divergent views to be aired without escalation. The interactional delicacy involved in surfacing disagreement without damaging rapport (Hutchins, 1995 ) is a learnable skill, one that educational interventions might address through modeling, reflection, and deliberate practice (Edmondson, 1999 ). Materiality and the Spatial Ecology of Reasoning The third mechanism, situated coordination, extends the interactional view of CCR by incorporating the material and spatial dimensions of clinical work. Consistent with distributed cognition (Hazlehurst et al., 2007 ; Hutchins, 1995 ) and professional vision frameworks (Goodwin, 2015 ; Heath et al., 2007 ), we found that artifacts such as monitors, ventilators, and electronic charts were not merely passive information sources. Instead, they functioned as co-participants in reasoning: their alarms prompted reorientation; their visual affordances shaped what was noticed; and their physical placement structured participation opportunities and authority relations. Gaze coordination was central to this process. Drawing on multimodal interaction research (Kendon, 1990 ; Mondada, 2018 ), we observed that clinicians' gaze shifts—toward monitors, charts, or colleagues—served as embodied markers of collective attention, signaling agreement, inviting elaboration, or indicating uncertainty. The synchronization of gaze around shared visual referents created joint attention structures that allowed interpretations to be co-inspected in real time. Similarly, spatial positioning relative to key artifacts influenced who initiated descriptions, whose interpretations gained traction, and how authority was distributed dynamically within the team. This is consistent with distributed cognition's emphasis on how cognition is shaped not only by what individuals know, but by how knowledge is spatially and materially organized (Hutchins, 1995 ). These findings support conceptualizing CCR as a form of ecological cognition, in which reasoning emerges through clinicians' engagement with the spatial, temporal, and material affordances of their environment. The ICU's built ecology—its artifact-rich infrastructure, spatial layout, and real-time data streams—constitutes a scaffold that clinicians continuously draw upon to build, test, and revise interpretations under uncertainty. This perspective highlights that improving clinical reasoning is not solely a matter of training individuals; it also involves designing environments, workflows, and technologies that better support collective sensemaking. By foregrounding the material and spatial dimensions of reasoning, our findings point to new opportunities for enhancing ICU situational awareness and optimizing team performance. On the Learnability of CCR Mechanisms We have characterised the three mechanisms of CCR as “learnable social practices,” and this claim warrants explicit justification. It is important to distinguish learnability from teachability: our study identifies interactional practices that clinicians demonstrably acquire through participation in clinical work, but it does not evaluate instructional interventions or trace individual learning trajectories. The claim of learnability rests on three forms of evidence. First, recurrence: the mechanisms we identified were not idiosyncratic to particular clinicians or episodes but appeared systematically across 12 distinct ICU encounters involving different team compositions and clinical scenarios. This recurrence suggests that these are recognisable, shared practices within the professional community rather than individual cognitive habits. Second, recognisability: participants oriented to these practices as normatively accountable. Mitigated proposals, conditional alignments, and spatially organised contributions were treated as appropriate or expected moves; deviations—such as categorical assertions where hedging was warranted—prompted visible interactional repair. This normative orientation indicates that these practices constitute learnable conventions, not merely spontaneous behaviours. Third, coachability: our analysis revealed moments where senior clinicians modelled practices for juniors, where repair sequences recalibrated inappropriate contributions, and where trainees visibly adapted their participation in response to uptake or non-uptake of prior turns. These interactional features suggest that the mechanisms are amenable to guided development, even though our study does not evaluate formal coaching interventions. Crucially, our claim is that this study identifies what could be learned—the interactional practices through which competent CCR is achieved—rather than documenting how learning trajectories unfold over time. Future research employing longitudinal or intervention-based designs would be needed to specify how these mechanisms are acquired, how expertise develops, and how targeted educational approaches might accelerate competence. Theoretical Implications for Learning and Educational Practice Reconceptualizing CCR as a multimodal, materially distributed, and interactionally organized practice challenges several longstanding assumptions in health professions education. The findings suggest that improving team reasoning requires pedagogical approaches that extend well beyond verbal report, cognitive checklists, or isolated individual performance. Four key implications follow for educators, supervisors, and assessment designers. 1. Reasoning cannot be taught or assessed solely through verbal explanation. Much of CCR unfolds through embodied and spatial practices—gaze coordination, pointing to artifacts, positioning relative to monitors, and synchronizing attention with colleagues—rather than explicit verbal articulation. Educational models that privilege "thinking aloud," structured case presentations, or post-hoc rationales risk obscuring how reasoning is produced in situ. This critique aligns with prior research showing that clinical cognition often becomes visible not through verbal descriptions but through situated action. Simulation design and debriefing practices must therefore expand their focus to include embodied conduct and material engagement, treating them as legitimate components of reasoning rather than peripheral behaviors. 2. Communication training must incorporate embodied and material modalities. Effective CCR requires clinicians to manage far more than verbal exchange. In every episode analyzed, spatial positioning, access to shared displays, direction of gaze, and the coordinated use of artifacts played decisive roles in how interpretations were proposed, taken up, or revised. These competencies extend well beyond the scope of traditional communication curricula, which typically emphasize clarity, assertiveness, or structured formats such as SBAR. What our findings demonstrate is that reasoning is contingent on clinicians' ability to orient their bodies, gestures, and attention in ways that make their interpretations visible and actionable to others. Training programs therefore need to incorporate explicit instruction and guided practice in how to work with the material and spatial affordances of the clinical environment. Learners must be supported in developing an awareness of when to step toward or away from monitors to facilitate shared viewing, how to use gestures or screen orientation to anchor claims in observable evidence, and how gaze coordination signals alignment, invites response, or marks the need for clarification. 3. Assessment systems must evaluate interactional competence, not isolated decision accuracy. Prevailing approaches to assessing clinical reasoning tend to rely on outcome-focused metrics—such as diagnostic accuracy or appropriateness of management decisions—or on post-hoc rationalizations provided through written responses, multiple-choice tests, or oral explanations. These methods assume that reasoning is a discrete cognitive product that can be evaluated after the fact. However, our findings indicate that competence in CCR is enacted in real time through interaction, not merely through the correctness of conclusions reached. Clinicians demonstrated reasoning competence by introducing tentative proposals in ways that invited uptake, collaboratively managing uncertainty as situations evolved, negotiating competing interpretations without undermining team cohesion, and grounding their claims in shared visual or material evidence. Such behaviors are not peripheral to reasoning; they constitute the mechanism through which CCR is achieved. Assessment systems must therefore expand beyond evaluating isolated decisions or retrospective narratives and incorporate methods capable of capturing reasoning as an interactional accomplishment. 4. Interprofessional education should address how epistemic authority is enacted and negotiated. Our findings demonstrate that epistemic authority in the ICU is not a fixed attribute tied to formal role or seniority but an interactional accomplishment that is continually negotiated through stance-taking, mitigated proposals, gaze behaviour, and embodied positioning. These subtle practices regulate who is granted the space to speak, whose interpretations receive uptake, and how uncertainty can be voiced without jeopardizing social cohesion. While many interprofessional training programs encourage "speaking up" as an individual competency, such approaches risk oversimplifying the complex, relational nature of epistemic participation in high-stakes environments. Educational initiatives therefore require a shift toward helping learners recognize how epistemic authority is co-constructed moment to moment. Analysis of authentic clinical footage can make visible the multimodal cues—gaze shifts, bodily alignment with artifacts, prosodic modulation—that signal when authority is being asserted, shared, or ceded. By framing authority as an emergent interactional process rather than a static hierarchical property, interprofessional education can better equip clinicians to participate in and sustain effective CCR. Synthesising across these implications, we suggest that educators, clinical supervisors, and assessment designers attend to three core principles. First, educators should notice that reasoning competence manifests not only in what clinicians say but in how they coordinate talk, gaze, gesture, and positioning with colleagues and artifacts—video review of authentic or simulated practice can make these practices visible in ways that verbal debriefing alone cannot. Second, supervisors should support learners in developing the interactional repertoires that enable effective participation: modelling mitigated proposals, creating space for trainees to contribute tentatively, and explicitly coaching how to ground claims in shared material evidence. Third, assessment designers should reconsider whether current methods—focused on individual diagnostic accuracy or post-hoc verbal justification—capture the collaborative, embodied, and temporally unfolding nature of clinical reasoning; process-oriented assessment using video analysis or structured observation may complement outcome-focused measures. These principles do not prescribe specific interventions but offer a framework for aligning educational practices with the interactional realities of collaborative clinical work. Limitations and Directions for Future Research This study has several limitations that shape the scope and transferability of its findings. First, the analysis was conducted in a single tertiary ICU with a mature interprofessional culture, established workflow patterns, and a high density of technological resources. As interactional and material ecologies vary substantially across clinical environments, the mechanisms identified here may manifest differently in settings such as emergency departments, operating suites, outpatient clinics, or resource-constrained hospitals. Comparative, multi-site work would allow researchers to examine how organizational culture, staffing models, and infrastructural arrangements modulate collaborative reasoning practices. Second, the study focused on naturally occurring team interactions rather than trainee-specific trajectories. Although the data capture moments of junior–senior negotiation, a more developmental perspective is needed to understand how learners acquire competence in multimodal reasoning practices—such as managing spatial positioning, coordinating gaze, or using artifacts to justify claims. Longitudinal or simulation-based methodologies could illuminate how these skills emerge, stabilize, and potentially deteriorate under varying workload and supervisory conditions. Third, because the dataset predates the widespread integration of emerging technologies—remote monitoring, AI-driven decision support, automated trend detection—the interactional architecture of CCR is likely to evolve as such systems become embedded in ICU workflow. Future research should examine how these tools redistribute epistemic authority, reshape attention patterns, or alter the sequence organization of reasoning, building on scholarship in human–technology interaction and distributed cognition. Finally, while this study foregrounds moments of coordination, future work should also investigate episodes where CCR falters: how teams repair breakdowns in alignment, how disagreements escalate or dissipate, and how these ruptures influence patient care. Analyzing breakdowns may deepen understanding of the vulnerabilities in collaborative sensemaking and identify opportunities for targeted educational or organizational interventions. Regarding scope conditions and transferability, we distinguish between context-sensitive and context-robust features of the mechanisms identified. This study makes no claim that all three mechanisms will be equally visible or consequential in non-acute settings. We propose that progressive clarification and conditional alignment are likely to be context-robust—that is, their core interactional logic should operate across clinical settings wherever teams engage in collaborative diagnostic reasoning under uncertainty, though the specific linguistic forms, turn-taking norms, and hierarchical dynamics may vary by culture, specialty, and institutional context. In contrast, situated coordination is more context-sensitive: its specific manifestations depend heavily on the material ecology of the setting—the density and configuration of artifacts, the spatial layout, and the availability of shared visual displays. In an outpatient clinic with limited technology, for example, the role of artifacts in structuring participation may be attenuated, whereas in a technologically sparse environment, verbal and embodied coordination may assume greater salience. Future research could test these hypotheses by examining CCR in lower-acuity settings such as general medical wards, outpatient clinics, or educational simulation environments, where time pressure is reduced and information is less distributed. Such studies would help determine which features of the mechanisms are invariant properties of collaborative clinical reasoning and which are amplified or attenuated by contextual factors. This would advance the theoretical generalisability of the CCR construct while generating actionable insights for educators working across diverse clinical learning environments. Conclusion CCR in the ICU is not a private cognitive act but an emergent, socially, and materially distributed accomplishment. Through progressive clarification, teams incrementally assemble a coherent clinical picture; through conditional alignment, they manage uncertainty and negotiate interpretive risk; and through situated coordination, they leverage spatial, embodied, and material resources to shape the trajectory of reasoning. By conceptualizing CCR as a form of ecological sensemaking—a dynamic interplay among people, artifacts, and environments—this study reframes how reasoning should be understood, taught, assessed, and supported in high-stakes clinical contexts. Our claim of learnability refers to the recognisable, normatively accountable nature of these practices within clinical interaction, rather than to demonstrated instructional effectiveness or accelerated learning through formal intervention. Recognizing CCR as an interactional phenomenon expands possibilities for educational innovation, simulation design, and organizational improvement, offering a foundation for building learning environments that cultivate not only individual cognitive skill but the collective capacities required for safe and effective patient care. Declarations Ethics Approval and Consent to Participate This study was approved by the Institutional Review Board of Chang Gung Medical Foundation (Ref: 202002453B0). All participating clinicians provided informed consent for video recording and research use of clinical interaction data. All procedures were conducted in accordance with the ethical standards of the Declaration of Helsinki and local regulatory requirements. Consent for Publication All participants consented to the publication of anonymized excerpts of interactional data. No identifiable patient information was collected or included in the analysis. All video and transcript materials were de-identified prior to publication. Availability of Data and Materials The datasets generated and analyzed during the current study—including de-identified transcripts and analytic memos—are not publicly available due to the sensitivity of clinical video data and restrictions imposed by the Institutional Review Board. Upon reasonable request, de-identified excerpts relevant to the findings may be shared with qualified researchers, subject to institutional and ethical approval. Competing Interests The authors declare that they have no competing interests. Funding This research was supported by Ministry of Science and Technology (R.O.C.) under Grant Number [NSTC 113-2628-H-182-002-MY2]. The funder had no role in the study design, data collection, analysis, interpretation of data, or preparation of the manuscript. Authors' Contributions CYL led the conception and design of the study; coordinated data collection and video acquisition; conducted the primary interactional, multimodal, and conversation-analytic coding; performed the initial and iterative data analyses; drafted the full manuscript; and integrated co-authors' feedback into the final version. CYL had full access to the data and takes primary responsibility for the accuracy and integrity of the analysis and interpretation. CHL and HYL contributed to the sequential and multimodal analyses and drafted analytic memos that informed the interpretive framework. MMC conducted video data collection, transcription, and multimodal annotation. SYY conceived the study, led the overall design, supervised the microinteractional analysis, and provided methodological consultation in psychology and qualitative inquiry. All authors contributed to manuscript drafting and critical revisions. All authors read and approved the final manuscript. Acknowledgements We thank the clinicians of the Neurosurgery Intensive Care Unit at Linkou Chang Gung Memorial Hospital for their participation and for granting us access to their clinical practice for research purposes. References Berg, M. (1999). Accumulating and coordinating: occasions for information technologies in medical work. Computer Supported Cooperative Work (CSCW), 8(4), 373–401 . https://doi.org/10.1023/A:1008757115404 Cook, D. A., Hamstra, S. J., Brydges, R., Zendejas, B., Szostek, J. H., Wang, A. T., Erwin, P. J., & Hatala, R. (2013). Comparative effectiveness of instructional design features in simulation-based education: systematic review and meta-analysis. Medical Teacher , 35 (1), e867–e898. https://doi.org/10.3109/0142159X.2012.714886 Cook, R., & Rasmussen, J. (2005). Going solid: a model of system dynamics and consequences for patient safety. BMJ Quality & Safety , 14 (2), 130–134. https://doi.org/10.1136/qshc.2003.009530 Cox, E. D., Jacobsohn, G. C., Rajamanickam, V. P., Carayon, P., Kelly, M. M., Wetterneck, T. B., Rathouz, P. J., & Brown, R. L. (2017). A family-centered rounds checklist, family engagement, and patient safety: a randomized trial. Pediatrics , 139 (5), e20161688. https://doi.org/10.1542/peds.2016-1688 Croskerry, P. (2009). Clinical cognition and diagnostic error: applications of a dual process model of reasoning. Advances in Health Sciences Education , 14 (Suppl 1), 27–35. https://doi.org/10.1007/s10459-009-9182-2 Donchin, Y., & Seagull, F. J. (2002). The hostile environment of the intensive care unit. Current Opinion in Critical Care , 8 (4), 316–320. https://doi.org/10.1097/00075198-200208000-00008 Durning, S. J., ArtinoJr, A. R., Schuwirth, L., & Van Der Vleuten, C. (2013). Clarifying assumptions to enhance our understanding and assessment of clinical reasoning. Academic Medicine , 88 (4), 442–448. https://doi.org/10.1097/ACM.0b013e3182851b5b Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly , 44 (2), 350–383. https://doi.org/10.2307/2666999 Elstein, A. S., Shulman, L. S., & Sprafka, S. A. (1978). Medical Problem Solving: An Analysis of Clinical Reasoning . Harvard University Press. Endsley, M. R. (2017). Toward a theory of situation awareness in dynamic systems. In Situational Awareness (pp. 9–42). Routledge. https://doi.org/10.4324/9781315087924-2 Eva, K. W. (2005). What every teacher needs to know about clinical reasoning. Medical Education , 39 (1), 98–106. https://doi.org/10.1111/j.1365-2929.2004.01972.x Eva, K. W., & Norman, G. R. (2005). Heuristics and biases—a biased perspective on clinical reasoning. Medical Education , 39 (9), 870–872. https://doi.org/10.1111/j.1365-2929.2005.02258.x Garfinkel, H. (2023). Studies in ethnomethodology. In Social Theory Re-Wired (pp. 58–66). Routledge. https://doi.org/10.4324/9781003320609-8 Goodwin, C. (2015). Professional vision. Aufmerksamkeit: Geschichte-Theorie-Empirie (pp. 387–425). Springer. Hazlehurst, B., McMullen, C. K., & Gorman, P. N. (2007). Distributed cognition in the heart room: how situation awareness arises from coordinated communications during cardiac surgery. Journal of Biomedical Informatics , 40 (5), 539–551. https://doi.org/10.1016/j.jbi.2007.02.001 Healey, A. N., Undre, S., & Vincent, C. A. (2006). Defining the technical skills of teamwork in surgery. BMJ Quality & Safety , 15 (4), 231–234. https://doi.org/10.1136/qshc.2005.017517 Heath, C., Luff, P., & Sanchez Svensson, M. (2007). Video and qualitative research: analysing medical practice and interaction. Medical Education , 41 (1), 109–116. https://doi.org/10.1111/j.1365-2929.2006.02641.x Heritage, J. (2012). Epistemics in action: Action formation and territories of knowledge. Research on Language & Social Interaction , 45 (1), 1–29. https://doi.org/10.1080/08351813.2012.646684 Heritage, J., & Raymond, G. (2005). The terms of agreement: Indexing epistemic authority and subordination in talk-in-interaction. Social Psychology Quarterly , 68 (1), 15–38. https://doi.org/10.1177/019027250506800103 Hindmarsh, J., & Pilnick, A. (2007). Knowing bodies at work: Embodiment and ephemeral teamwork in anaesthesia. Organization Studies , 28 (9), 1395–1416. https://doi.org/10.1177/0170840607068258 Hutchins, E. (1995). Cognition in the Wild . MIT Press. Hutchins, E. (2000). Distributed cognition. International Encyclopedia of the Social and Behavioral Sciences , 138 (1), 1–10. Ilgen, J. S., Humbert, A. J., Kuhn, G., Hansen, M. L., Norman, G. R., Eva, K. W., Charlin, B., & Sherbino, J. (2012). Assessing diagnostic reasoning: a consensus statement summarizing theory, practice, and future needs. Academic Emergency Medicine , 19 (12), 1454–1461. https://doi.org/10.1111/acem.12034 Kendon, A. (1990). Conducting Interaction: Patterns of Behavior in Focused Encounters (Vol. 7). CUP Archive. Klein, G., Moon, B., & Hoffman, R. R. (2006). Making sense of sensemaking 1: Alternative perspectives. IEEE Intelligent Systems , 21 (4), 70–73. https://doi.org/10.1109/MIS.2006.75 Kneebone, R. (2005). Evaluating clinical simulations for learning procedural skills: a theory-based approach. Academic Medicine , 80 (6), 549–553. https://doi.org/10.1097/00001888-200506000-00006 Kotthoff, H. (1993). Disagreement and concession in disputes: On the context sensitivity of preference structures. Language in Society , 22 (2), 193–216. https://doi.org/10.1017/S0047404500017103 Lane, D., Ferri, M., Lemaire, J., McLaughlin, K., & Stelfox, H. T. (2013). A systematic review of evidence-informed practices for patient care rounds in the ICU. Critical Care Medicine , 41 (8), 2015–2029. https://doi.org/10.1097/CCM.0b013e31828a435f Leonard, M., Graham, S., & Bonacum, D. (2004). The human factor: the critical importance of effective teamwork and communication in providing safe care. BMJ Quality & Safety , 13 (suppl 1), i85–i90. https://doi.org/10.1136/qshc.2004.010033 Lingard, L. (2012). Rethinking competence in the context of teamwork. The Question of Competence: Reconsidering Medical Education in the Twenty-First Century (pp. 42–69). Cornell University Press. Lingard, L., Reznick, R., Espin, S., Regehr, G., & DeVito, I. (2002). Team communications in the operating room: talk patterns, sites of tension, and implications for novices. Academic Medicine , 77 (3), 232–237. https://doi.org/10.1097/00001888-200203000-00010 Maitlis, S., & Christianson, M. (2014). Sensemaking in organizations: Taking stock and moving forward. Academy of Management Annals , 8 (1), 57–125. https://doi.org/10.1080/19416520.2014.873177 Manser, T. (2009). Teamwork and patient safety in dynamic domains of healthcare: a review of the literature. Acta Anaesthesiologica Scandinavica , 53 (2), 143–151. https://doi.org/10.1111/j.1399-6576.2008.01717.x Mondada, L. (2018). Multiple temporalities of language and body in interaction: Challenges for transcribing multimodality. Research on Language and Social Interaction , 51 (1), 85–106. https://doi.org/10.1080/08351813.2018.1413878 Norman, G. (2005). Research in clinical reasoning: past history and current trends. Medical Education , 39 (4), 418–427. https://doi.org/10.1111/j.1365-2929.2005.02127.x Norman, G. (2009). Dual processing and diagnostic errors. Advances in Health Sciences Education , 14 (Suppl 1), 37–49. https://doi.org/10.1007/s10459-009-9179-x Norman, G. R., Grierson, L. E., Sherbino, J., Hamstra, S. J., Schmidt, H. G., & Mamede, S. (2018). Expertise in medicine and surgery. In K. A. Ericsson, R. R. Hoffman, A. Kozbelt, & A. M. Williams (Eds.), The Cambridge Handbook of Expertise and Expert Performance (2nd ed., pp. 331–355). Cambridge University Press. https://doi.org/10.1017/9781316480748.019 Pomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In J. M. Atkinson, & J. Heritage (Eds.), Structures of Social Action: Studies in Conversation Analysis (pp. 57–101). Cambridge University Press. Reader, T. W., Flin, R., Mearns, K., & Cuthbertson, B. H. (2011). Team situation awareness and the anticipation of patient progress during ICU rounds. BMJ Quality & Safety , 20 (12), 1035–1042. https://doi.org/10.1136/bmjqs.2010.048561 Rudolph, J. W., Morrison, J. B., & Carroll, J. S. (2009). The dynamics of action-oriented problem solving: Linking interpretation and choice. Academy of Management Review , 34 (4), 733–756. https://doi.org/10.5465/amr.34.4.zok733 Salas, E., Cooke, N. J., & Rosen, M. A. (2008). On teams, teamwork, and team performance: Discoveries and developments. Human Factors , 50 (3), 540–547. https://doi.org/10.1518/001872008X288457 Schegloff, E. A. (2007). Sequence Organization in Interaction: A Primer in Conversation Analysis I (Vol. 1). Cambridge University Press. https://doi.org/10.1017/CBO9780511791208 Sutcliffe, K. M., Lewton, E., & Rosenthal, M. M. (2004). Communication failures: an insidious contributor to medical mishaps. Academic Medicine , 79 (2), 186–194. https://doi.org/10.1097/00001888-200402000-00019 Weaver, S. J., Dy, S. M., & Rosen, M. A. (2014). Team-training in healthcare: a narrative synthesis of the literature. BMJ Quality & Safety , 23 (5), 359–372. https://doi.org/10.1136/bmjqs-2013-001848 Weick, K. E., Sutcliffe, K. M., & Obstfeld, D. (2005). Organizing and the process of sensemaking. Organization Science , 16 (4), 409–421. https://doi.org/10.1287/orsc.1050.0133 Weick, K. E., & Weick, K. E. (1995). Sensemaking in Organizations (Vol. 3). Sage. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 18 May, 2026 Reviews received at journal 08 May, 2026 Reviewers agreed at journal 26 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 23 Apr, 2026 Reviewers invited by journal 23 Apr, 2026 Editor assigned by journal 23 Jan, 2026 Submission checks completed at journal 23 Jan, 2026 First submitted to journal 22 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8666674","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":631256788,"identity":"8705ca0c-7011-4cd3-a71b-e93223445a93","order_by":0,"name":"Ching-Yi Lee","email":"","orcid":"","institution":"Linkou Chang Gung Memorial Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ching-Yi","middleName":"","lastName":"Lee","suffix":""},{"id":631256789,"identity":"88205ef0-4bb4-4c65-9ac3-99d471ceda97","order_by":1,"name":"Hung-Yi Lai","email":"","orcid":"","institution":"Linkou Chang Gung Memorial Hospital","correspondingAuthor":false,"prefix":"","firstName":"Hung-Yi","middleName":"","lastName":"Lai","suffix":""},{"id":631256790,"identity":"973b1645-4eae-4b56-aec7-4703f984720b","order_by":2,"name":"Ching-Hsin Lee","email":"","orcid":"","institution":"Linkou Chang Gung Memorial Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ching-Hsin","middleName":"","lastName":"Lee","suffix":""},{"id":631256792,"identity":"2f7e30bf-6a4f-4c61-991a-d321f017b144","order_by":3,"name":"Mi-Mi Chen","email":"","orcid":"","institution":"Linkou Chang Gung Memorial Hospital","correspondingAuthor":false,"prefix":"","firstName":"Mi-Mi","middleName":"","lastName":"Chen","suffix":""},{"id":631256794,"identity":"3113a0aa-b740-4ee5-8113-7006893f76e6","order_by":4,"name":"Sze-Yuen Yau","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIiWNgGAWjYHAD5gNAQkKGFC1sCSAtPKRo4TEAkwTVGRw/Y/zxR8XhxH7pns+vbtRY8DCwHz66Aa+WMzlmEhJnDifOnHN2m3XOMaDDeNLSbuDVciDHjMGw7XDihhu524xz2IBaJHjM8Gs5/8b4Q+K/w4n7b+Q8M875R4yWGzkGEgcbgLZI5DA/zm0jQovkjWdlkg3H0o1n3EgzY87tk+BhI+QXvvPJmz/+qLGW7Z+R/Phzzrc6OX72w8fwalE4AKaaQQSbBJjEpxwE5BvAVB2IYP5ASPUoGAWjYBSMTAAA5NdNMJ4af2oAAAAASUVORK5CYII=","orcid":"","institution":"Linkou Chang Gung Memorial Hospital","correspondingAuthor":true,"prefix":"","firstName":"Sze-Yuen","middleName":"","lastName":"Yau","suffix":""}],"badges":[],"createdAt":"2026-01-22 07:42:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8666674/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8666674/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108819562,"identity":"3cec2691-b66e-4eea-8bf2-eda705e41723","added_by":"auto","created_at":"2026-05-08 16:37:54","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2291281,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8666674/v1/3df8af90a722022c42288511.png"},{"id":108820042,"identity":"ec12ce4e-8409-4306-b85d-5d850f087961","added_by":"auto","created_at":"2026-05-08 16:39:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1510345,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8666674/v1/c6d5f20a9b3e059e883adcf9.png"},{"id":108976878,"identity":"15373256-e0e7-46a8-895e-79ba7a7ea392","added_by":"auto","created_at":"2026-05-11 11:29:20","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":523950,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8666674/v1/38698a49-0a07-4089-bf8c-9d95b06d121a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Collaborative Clinical Reasoning in the ICU: How Teams Construct Shared Understanding in Real Time","fulltext":[{"header":"Introduction","content":"\u003cp\u003eClinical reasoning has been conceptualised predominantly as an individual cognitive process (Croskerry, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Eva, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Norman, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). This tradition is grounded in information-processing theory, which models reasoning as a structured progression from data acquisition to problem representation and solution generation, and in dual-process theories, which distinguish between rapid, pattern-based intuitive reasoning and slower, analytic reasoning (Croskerry, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Eva \u0026amp; Norman, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Norman et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). These frameworks have yielded valuable insights into expert\u0026ndash;novice differences and diagnostic error, and they continue to inform teaching strategies and assessment in medical education. However, their underlying assumptions\u0026mdash;particularly that clinicians can reason with stable access to relevant information, minimal interruptions, and sufficient cognitive bandwidth\u0026mdash;reflect cognitive conditions that are less representative of contemporary clinical environments, where reasoning is frequently distributed, time-pressured, and socially embedded.\u003c/p\u003e \u003cp\u003eModern healthcare is fundamentally team-based, particularly in acute and high-acuity environments where clinical work is distributed across professionals who must integrate heterogeneous expertise to manage rapidly evolving patient conditions (Manser, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Endsley, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Weaver et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). In such settings, the cognitive resources needed to understand and respond to complex patient trajectories are not held by any single individual but are distributed across team members, technological artifacts, and the physical environment (Reader et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Salas et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Hutchins, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1995\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Healey et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Lingard et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2002\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite the central role of collaboration in critical care, research on clinical reasoning has largely remained anchored to individualist cognitive models (Berg, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Cox et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Donchin \u0026amp; Seagull, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Lane et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Reader et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Although communication frameworks such as TeamSTEPPS and structured handoff protocols address information transfer, they do not provide an account of how reasoning itself is jointly produced in interaction (Hindmarsh \u0026amp; Pilnick, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Lingard et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Theories of sensemaking (Maitlis \u0026amp; Christianson, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Weick \u0026amp; Weick, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e1995\u003c/span\u003e) and action-oriented problem solving (Rudolph et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) offer conceptual resources for understanding collective cognition, and distributed cognition (Hazlehurst et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Hutchins, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1995\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2000\u003c/span\u003e) emphasises how cognitive processes extend across people, artifacts, and environments. Yet empirical studies grounded in these perspectives have rarely examined how clinical teams construct shared clinical understanding in real time through moment-to-moment interaction.\u003c/p\u003e \u003cp\u003eThis gap is consequential for medical education. While calls to situate clinical reasoning within its contextual and social dimensions have grown more prominent (Cook et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Ilgen et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Durning et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Kneebone, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), much of this work remains theoretical. Simulation-based training offers controlled environments in which to practise team-based reasoning, but without an empirically derived account of how collaborative reasoning unfolds in authentic clinical practice, such training risks reproducing idealised models rather than reflecting the interactional realities of clinical work (Eva, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Norman, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGiven the central role of collaboration in ICU practice and the lack of empirical evidence detailing how CCR unfolds in situ, this study employs video-based microinteractional analysis to theorise collaborative clinical reasoning as a learnable social practice constituted through observable interactional mechanisms. By examining ICU teams as an epistemic extreme case\u0026mdash;where high stakes, time pressure, and distributed information structures amplify reasoning practices that are likely present, though less visible, in other clinical education settings\u0026mdash;we aim to identify the mechanisms through which clinicians jointly construct shared clinical understanding. This approach offers generalisability that is theoretical and mechanism-based rather than statistical, providing an empirically grounded foundation for understanding how collaborative reasoning is learned and sustained through participation in everyday clinical work.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003eThis study employed video-based microinteractional analysis as a methodological instrument for operationalising abstract theoretical constructs\u0026mdash;reasoning, alignment, sensemaking\u0026mdash;as observable, sequential practices. This design was selected because CCR is fundamentally an interactional accomplishment, emerging through the moment-to-moment coordination of talk, gesture, gaze, and material artifacts. Retrospective interviews and surveys cannot capture the temporal, embodied, and sequential organisation of reasoning as it unfolds; video-based microanalysis makes these theoretical claims empirically tractable.\u003c/p\u003e \u003cp\u003eOur methodological orientation assumes that clinical reasoning becomes empirically visible through the sequential organization of social interaction. By examining how clinicians initiate, respond to, challenge, and align interpretations in real time, microinteractional analysis allows us to describe the mechanisms through which shared clinical understanding is built and sustained under complex, high-stakes conditions.\u003c/p\u003e \u003cp\u003eThe study was conducted in the adult medical ICU of a tertiary academic medical center in Taiwan, where interdisciplinary teams routinely engage in collaborative assessments and decision making. The ICU was selected as a theoretical and analytic \"extreme case\" because its time pressure, uncertainty, rapid patient trajectories, and distributed information structures render interactional reasoning processes more visible than in lower-acuity settings.\u003c/p\u003e \u003cp\u003eVideo-based microinteractional designs require methodological transparency regarding the stages of data generation, selection, and analysis. Accordingly, the study design involved five interlocking components:\u003c/p\u003e\n\u003ch3\u003e1. Naturalistic Video Data Collection\u003c/h3\u003e\n\u003cp\u003eWe recorded 27 hours of naturally occurring ICU team interactions across 18 discrete encounters, including morning rounds, rapid patient assessments, and emergent decision-making episodes. These encounters involved interdisciplinary participation from attending physicians, fellows, residents, nurses, and respiratory therapists. Video-based observation was conducted in a non-intrusive, naturalistic manner to preserve the ecological fidelity of routine clinical practice.\u003c/p\u003e \u003cp\u003eTo capture the complexity of team interaction and spatial dynamics, we used two stationary high-definition cameras per event. Cameras were mounted unobtrusively within the patient room or team workspace to provide complementary angles: one oriented toward the main interactional floor (e.g., bedside or rounding cluster), and the other capturing peripheral activity and monitor/tool use. Stationary placement minimized disruption to workflow and enabled continuous recording without operator involvement.\u003c/p\u003e \u003cp\u003eThis configuration allowed us to document verbal exchanges, gaze and body orientation, use of clinical artifacts (e.g., bedside monitors, ventilators, mobile computers), spatial arrangement and movement within the care environment, and emergent actions during unexpected clinical changes. By capturing clinicians' naturally unfolding communication and coordination, this approach ensured ecologically valid data that reflected how CCR emerges spontaneously, without researcher prompting or simulation constraints.\u003c/p\u003e\n\u003ch3\u003e2. Multimodal Transcription and Interactional Documentation\u003c/h3\u003e\n\u003cp\u003eAll selected episodes were transcribed using Jeffersonian transcription conventions, enabling fine-grained representation of the sequential organization of talk, including overlaps, micro-pauses, cut-offs, timing, and prosodic contours. Because CCR unfolds through multiple modalities (Mondada, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), multimodal annotation captured gaze direction and gaze shifts, body orientation and postural alignment, gestures, pointing, and hand movements, object manipulation (e.g., adjusting ventilator settings, touching monitors, flipping charts), interaction with representational artifacts, including electronic health records, bedside monitors, imaging displays, and mobile devices, and spatial configurations of team members during interaction.\u003c/p\u003e \u003cp\u003eThese multimodal layers were essential for analyzing reasoning as an embodied, spatially distributed, and tool-mediated activity, allowing us to examine how participants coordinated attention, displayed epistemic stance, and constructed shared interpretations through talk-and-action sequences.\u003c/p\u003e \u003cp\u003eTranscription and annotation were performed using ELAN (Max Planck Institute for Psycholinguistics), a software platform designed for synchronous, multi-tiered representation of audio-video data. ELAN enabled us to align verbal, prosodic, embodied, and artifact-related actions with millisecond precision, ensuring analytic access to the temporal and interactional contingencies that shape collaborative reasoning in real time.\u003c/p\u003e\n\u003ch3\u003e3. Episode Selection and Theoretical Sampling\u003c/h3\u003e\n\u003cp\u003eFrom the full corpus of 18 recorded ICU encounters, episodes for close analysis were selected through an iterative theoretical sampling process. This process followed interpretive, interaction-focused methodological logic, in which sampling is guided not by representativeness but by the analytic richness of sequences that illuminate the phenomenon of interest\u0026mdash;in this case, CCR.\u003c/p\u003e \u003cp\u003eDuring initial rounds of corpus-wide viewing, the research team identified sequences in which clinicians engaged in active interpretive work rather than procedural or purely informational exchanges. Preliminary markers of interest included surfacing or articulating uncertainties, constructing or revising patient trajectories, questioning, refining, or challenging interpretations, coordinating diagnostic or management decisions, and repairing misunderstandings, misalignments, or knowledge gaps. These markers allowed us to locate segments where reasoning was not merely expressed individually but co-constructed interactionally.\u003c/p\u003e \u003cp\u003eAs analytic patterns emerged, sampling became increasingly focused on episodes that provided maximal insight into the interactional mechanisms of CCR. Out of the 18 recorded encounters, 12 episodes were ultimately selected for full, line-by-line, multimodal microinteractional analysis. These episodes were chosen because they contained clear evidence of collaborative reasoning, rich multimodal interaction, and sequential structures that enabled close examination of how shared understanding was produced and maintained.\u003c/p\u003e\n\u003ch3\u003e4. Sequential, Microinteractional Analysis\u003c/h3\u003e\n\u003cp\u003eData analysis followed the principles of conversation analysis and multimodal interaction analysis, which together make visible the fine-grained practices through which clinicians construct shared clinical reasoning in real time. This approach treats interaction as sequentially organized, meaning that each action (verbal or embodied) derives meaning from how participants position it relative to what came before and anticipate what comes next. Because CCR depends on how clinicians jointly build, contest, and align interpretations, sequential analysis provides a powerful lens for illuminating the underlying mechanisms.\u003c/p\u003e \u003cp\u003e Across the selected episodes, we examined turn-taking and participation structure, repair and the management of trouble, epistemic positioning and stance, multimodal coordination of talk, gesture, and artifact use, and response to interruptions and emerging clinical information.\u003c/p\u003e \u003cp\u003eThrough this sequential, multimodal analysis, we identified recurrent interactional mechanisms through which CCR was achieved, including distributed noticing (how clinicians jointly surfaced and oriented to clinically relevant cues), interpretive alignment (how clinicians converged on shared problem framings or diagnostic narratives), challenge\u0026ndash;justification sequences (how disagreements or divergent interpretations were enacted, managed, and resolved), recalibration of shared understanding (how teams updated collective reasoning when new information emerged or prior assumptions were destabilized), and structuring shared problem frames (how participants collaboratively articulated \"what is going on,\" stabilizing the trajectory of diagnostic or management decisions).\u003c/p\u003e\n\u003ch3\u003e5. Reflexive, Multidisciplinary Analytic Review\u003c/h3\u003e\n\u003cp\u003eTo enhance analytic credibility and ensure that interpretations remained grounded in both clinical reality and interactional theory, the analysis incorporated a reflexive, multidisciplinary review process. The core analytic team consisted of three researchers with complementary expertise in ICU clinical practice, medical education, and ethnomethodology/interaction analysis, supported by one external consultant with a background in psychology and qualitative methodology. This diverse expertise enabled us to examine interactional patterns from multiple epistemic vantage points and to avoid privileging a single disciplinary interpretation.\u003c/p\u003e \u003cp\u003eTeam meetings occurred iteratively throughout data collection, transcription, and analysis. During these sessions, analysts interrogated preliminary findings, examining alternative explanations for observed interactional phenomena; identified potential analytic blind spots, including the risk of over-interpreting or under-interpreting participants' actions; challenged disciplinary assumptions, ensuring that interpretations did not rely on clinical or educational presumptions unsupported by the data; compared analytic claims against the video record, returning repeatedly to the raw data to validate emerging interpretations; and refined conceptual categories, testing whether identified mechanisms were robust across multiple episodes and contexts.\u003c/p\u003e\n\u003ch3\u003eEthical Considerations\u003c/h3\u003e\n\u003cp\u003e This study was approved by the Institutional Review Board of Chang Gung Medical Foundation (Ref: 202002453B0). All participating clinicians were informed of the study purpose, the use of video recording, and their right to withdraw at any time. Written consent was obtained prior to recording. To protect confidentiality, video data were stored on encrypted, access-restricted servers; all identifying information was removed during transcription and analysis. When excerpts were used for reporting or dissemination, faces and identifying features were blurred unless explicit additional consent was obtained.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eAnalysis of 12 ICU episodes revealed that CCR emerged not as a discrete cognitive act but as an interactional trajectory, accomplished through the coordinated use of talk, embodied conduct, and material artifacts. Across diverse clinical scenarios\u0026mdash;routine rounds, rapid assessments, and emergent deteriorations\u0026mdash;three recurrent interactional patterns characterized how teams built, negotiated, and stabilized shared clinical understanding. These three mechanisms\u0026mdash;progressive clarification, conditional alignment, and situated coordination\u0026mdash;are summarized in Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and illustrated in the conceptual model presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eInteractional Mechanisms of Collaborative Clinical Reasoning in the ICU\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"5\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMechanism\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eCore interactional practices\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eSequential function\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eMultimodal resources\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003eEducational significance\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eProgressive clarification\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(shared problem construction)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eLayering profession-specific observations; solicitation sequences; repair of ambiguities; conditional framing of epistemic gaps\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eIncrementally assembles coherent clinical picture from distributed fragments; each contribution refines the emerging interpretation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eTalk; turn-taking; gaze to monitors/charts; pointing gestures\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003eReveals how learners contribute fragments vs premature conclusions; repair sequences expose coachable reasoning gaps; models epistemic humility\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eConditional alignment\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(epistemic negotiation)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eMitigated proposals; hedged alignments; softened challenges; evidence-grounded justification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eEnables hypothesis exploration while preserving team cohesion; tests alternatives without premature closure or interpersonal risk\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eProsody; epistemic markers; gaze patterns; pause placement\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003eMitigated formulations indicate competence, not weakness; stance-taking practices are modelable; hierarchy-crossing contributions become visible\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eSituated coordination\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(material-interactional coupling)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eSpatial positioning relative to artifacts; gaze convergence; pointing and tracing; embodied bids for epistemic primacy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eStructures participation through material and spatial affordances; bodies and tools direct attention and ground claims\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eBody positioning; gaze direction; screen sharing; artifact manipulation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003ePhysical arrangement shapes whose knowledge gets uptake; artifact use is coachable reasoning practice; spatial positioning is learnable\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003ctfoot\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\"\u003e\u003cem\u003eNote.\u003c/em\u003e Each mechanism represents a learnable social practice observable in routine clinical work. Mechanisms operate interdependently rather than as sequential stages. Educational significance describes what supervisors might attend to, without prescribing curricular interventions.\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tfoot\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cem\u003eAbbreviations:\u0026nbsp;\u003c/em\u003eICU = intensive care unit.\u003c/p\u003e\n\u003ch3\u003e1. Progressive Clarification: A Mechanism of Shared Problem Construction\u003c/h3\u003e\n\u003cp\u003eAcross all analyzed episodes, CCR did not emerge as a single decisive moment or an individual\u0026apos;s formulation of a complete diagnostic frame. Instead, teams constructed reasoning progressively, through a sequence of layered contributions, clarification moves, and epistemic repairs that gradually stabilized a shared understanding of the patient\u0026apos;s condition. This pattern\u0026mdash;progressive clarification\u0026mdash;comprised three recurrent practices: (a) layering partial assessments, (b) refining problem frames through solicitation and repair, and (c) managing epistemic gaps to sustain forward movement in reasoning. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e illustrates how progressive clarification unfolds as an interactional mechanism of shared problem construction.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003e1.1 Layering assessments and accumulating fragments of evidence\u003c/h2\u003e\n \u003cp\u003eA defining feature of CCR was the way clinicians contributed fragmented, profession-specific observations that, in aggregate, began to form a coherent clinical picture. Nurses frequently offered near-real-time physiologic cues or bedside observations; residents contributed laboratory trends and recent documentation; fellows added ventilator data or imaging interpretations; and attendings synthesized across these informational streams.\u003c/p\u003e\n \u003cp\u003eThese contributions were rarely presented as final judgments. Instead, they appeared as tentative fragments\u0026mdash;brief observations, trend updates, or short descriptors\u0026mdash;positioned sequentially as supplements to one another (\u0026quot;and also...\u0026quot;, \u0026quot;I\u0026apos;m seeing...\u0026quot;, \u0026quot;it\u0026apos;s a bit higher today...\u0026quot;). Through this turn-by-turn accumulation, the team collaboratively constructed a diagnostic storyline that no individual participant held in full. The emerging clinical picture thus reflected a distributed noticing process, shaped by each participant\u0026apos;s embodied location in the unit, their access to particular artifacts (monitors, ventilators, electronic charts), and their role-specific responsibilities.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003e1.2 Refining problem frames through solicitation and repair\u003c/h2\u003e\n \u003cp\u003eOnce fragments of information began to coalesce into a preliminary frame, clinicians engaged in solicitation and repair practices to refine the accuracy and coherence of the shared interpretation. These moves included requests for clarification (\u0026quot;Was that rising or stable?\u0026quot;), specification probes (\u0026quot;How much higher than yesterday?\u0026quot;), corrections and confirmations (\u0026quot;No, I meant the later result.\u0026quot;), and elaborations prompted by misunderstandings or ambiguous phrasing.\u003c/p\u003e\n \u003cp\u003eImportantly, these repair sequences were not mere corrections of error. They played a constitutive role in shaping the diagnostic trajectory, ensuring that assumptions were interrogated and that the shared problem frame rested on mutually confirmed information. Through solicitation and repair, clinicians collaboratively tightened the precision of the emerging narrative, adjusting weight assigned to particular cues and recalibrating interpretations in response to newly clarified details.\u003c/p\u003e\n \u003cp\u003eThe multimodal environment also supported this refining process: clinicians frequently oriented to monitors, displayed charts, or gestured toward imaging screens to anchor their clarifications, using material artifacts as external reference points for resolving ambiguity.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003e1.3 Managing epistemic gaps to sustain forward reasoning\u003c/h2\u003e\n \u003cp\u003eReasoning in the ICU is constrained by incomplete and evolving information. Across episodes, teams routinely encountered missing laboratory results, pending imaging, or contradictory clinical indicators. Rather than halting discussion, clinicians addressed these epistemic gaps by explicitly identifying what was unknown and articulating conditional assumptions to guide interim decision-making (\u0026quot;Let\u0026apos;s proceed assuming his CO₂ hasn\u0026apos;t worsened and revisit when the ABG returns.\u0026quot;).\u003c/p\u003e\n \u003cp\u003eThese conditional framings served two functions: (1) Maintaining progress in the face of uncertainty, allowing the team to act without overcommitting to any single interpretation, and (2) Keeping interpretive space open, marking unresolved issues for later verification and preventing premature diagnostic closure. Multimodal cues\u0026mdash;such as gestures that circled back to artifacts, pauses that framed conditional statements, or gaze shifts that signaled pending data sources\u0026mdash;helped the team collectively track the provisional nature of these assumptions. Through this process, teams enacted epistemic repair, stabilizing the reasoning trajectory while transparently acknowledging uncertainty.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003e2. Conditional Alignment: A Mechanism of Epistemic Negotiation\u003c/h3\u003e\n\u003cp\u003eA second recurrent pattern involved how clinicians aligned around emerging interpretations while preserving flexibility to revise them as new information appeared. Whereas progressive clarification described how teams built a shared clinical picture over time, conditional alignment captured how they managed the epistemic and interpersonal delicacy of proposing and negotiating interpretations in a high-stakes, uncertain environment. Across episodes, clinicians rarely presented diagnostic assessments as categorical claims. Instead, they relied on conditional formulations, mitigated assertions, and softened disagreements that enabled collaborative hypothesis testing without foreclosing alternative framings or jeopardizing team cohesion (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\n\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\n \u003ch2\u003e2.1 Proposals framed as possibilities, not certainties\u003c/h2\u003e\n \u003cp\u003eInterpretive trajectories commonly began with hypotheses introduced in mitigated, non-finalized forms\u0026mdash;\u0026quot;\u003cem\u003emaybe we\u0026apos;re seeing early sepsis\u003c/em\u003e,\u0026quot; \u0026quot;\u003cem\u003eit could be fluid-responsive\u003c/em\u003e,\u0026quot; \u0026quot;\u003cem\u003eI\u0026apos;m wondering if this is evolving shock.\u003c/em\u003e\u0026quot; These proposal formats had two key interactional consequences.\u003c/p\u003e\n \u003cp\u003eFirst, by avoiding categorical assertions, speakers reduced the interpersonal and epistemic risks associated with potentially inaccurate claims. This was especially salient across hierarchical boundaries, where trainees and junior clinicians routinely employed epistemic downgrades (\u0026quot;maybe,\u0026quot; \u0026quot;seems like...,\u0026quot; \u0026quot;could be\u0026quot;) and embodied displays such as upward glances or softened prosody to position their contributions as tentative rather than authoritative. Such stance markers allowed juniors to contribute interpretively without violating expectations of deference or overclaiming knowledge in front of seniors.\u003c/p\u003e\n \u003cp\u003eSecond, mitigated proposals functioned as invitations to joint reasoning. Rather than presenting a hypothesis as a completed cognitive product, these formulations positioned the proposal as open for collective scrutiny and elaboration. Their sequential design\u0026mdash;often accompanied by rising intonation, slight pauses, or gaze shifts toward others\u0026mdash;projected a responsive next action, signaling that the speaker anticipated additions, modifications, or alternative framings from the team.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003e2.2 Alignment without premature closure\u003c/h2\u003e\n \u003cp\u003eResponses to tentative proposals typically took the form of conditional alignments, rather than unequivocal endorsements. Clinicians frequently responded with turns such as \u0026quot;\u003cem\u003ethat might fit given...\u003c/em\u003e,\u0026quot; \u0026quot;\u003cem\u003eunless this is more cardiac...\u003c/em\u003e,\u0026quot; or \u0026quot;\u003cem\u003ethat could explain the drop unless we\u0026apos;re missing something\u003c/em\u003e,\u0026quot; which simultaneously acknowledged the plausibility of the preceding interpretation while signaling that additional contingencies remained in play. These formulations enabled participants to sustain the forward motion of the reasoning trajectory without committing prematurely to a single explanatory frame.\u003c/p\u003e\n \u003cp\u003eSuch alignments were often supported by embodied cues that indexed conditionality: half-nods, brief glances toward monitors or lab displays, or hand movements that traced possible directions of interpretation. These multimodal actions tempered the apparent commitment of the verbal turn, displaying alignment with the proposal while leaving interpretive space open.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n \u003ch2\u003e2.3 Negotiating competing interpretations\u003c/h2\u003e\n \u003cp\u003eDivergent interpretations arose frequently across the dataset, particularly in episodes involving ambiguous presentations or rapidly shifting physiological states. Rather than issuing direct disagreement, clinicians typically introduced alternative framings through softened challenge formats, such as \u0026quot;\u003cem\u003eI wonder if it\u0026apos;s actually...\u003c/em\u003e\u0026quot; or \u0026quot;\u003cem\u003eCould it instead be...?\u003c/em\u003e\u0026quot;. These mitigated formulations allowed speakers to express divergent views without overtly contradicting colleagues, thereby preserving rapport and signaling respect for others\u0026apos; epistemic positions.\u003c/p\u003e\n \u003cp\u003eOnce an alternative perspective was introduced, clinicians commonly followed with justification, grounding their reasoning in observable data. Participants oriented toward monitors, gestured toward ventilator readouts, traced trends on charts, or pointed to imaging displays as they marshalled evidence. These embodied and material actions anchored disagreements in shared referential spaces, enabling the team to collectively inspect the basis of the alternative interpretation.\u003c/p\u003e\n \u003cp\u003eThrough this interactional work, teams typically moved toward reconciliation or reframing, integrating the newly raised interpretation or clarifying why a prior account remained more plausible. Conditional language, mitigated stance-taking, and multimodally grounded justification enabled teams to negotiate difference productively and safely.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003e3. Situated Coordination: A Mechanism of Material\u0026ndash;Interactional Coupling\u003c/h3\u003e\n\u003cp\u003eCCR in the ICU unfolded not only through verbal exchanges but through clinicians\u0026apos; embodied coordination and engagement with the material environment. Spatial positioning, gaze behaviour, and the use of tools such as monitors, ventilators, and charts shaped who contributed, how proposals were taken up, and which interpretations gained momentum. Across episodes, the material and spatial ecology of the ICU acted as an active substrate for reasoning, structuring attention, authority, and interpretive trajectories (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Spatial positioning as a resource for participation and authority\u003c/h2\u003e\n \u003cp\u003eAcross episodes, clinicians\u0026apos; spatial positioning within the patient room played a consequential role in shaping how collaborative reasoning unfolded. Those standing closest to key informational artifacts\u0026mdash;particularly the bedside monitor, ventilator screen, or electronic chart\u0026mdash;often functioned as epistemic anchors, not by explicit designation but through the affordances of proximity. Their embodied access to real-time physiologic data positioned them to initiate trend commentary or highlight salient cues, such as gesturing toward a dropping pressure or pointing out changes in waveform morphology. These embodied demonstrations routinely drew the team\u0026apos;s gaze toward the same artifact, thereby structuring what was treated as relevant evidence in the emerging interpretive trajectory.\u003c/p\u003e\n \u003cp\u003ePhysical proximity also influenced participation rights and the flow of contributions. Clinicians adjacent to the monitor or ventilator more frequently assumed the role of \u0026quot;first describer,\u0026quot; narrating data trends before others had visual access. Their descriptions often became the reference point for subsequent reasoning moves, with others building upon, questioning, or refining the initial framing.\u003c/p\u003e\n \u003cp\u003eImportantly, spatial positioning did not merely reflect existing hierarchies; it actively shaped them. While attendings often occupied central or mobile positions, nurses and respiratory therapists situated at the bedside or near equipment frequently contributed authoritative assessments anchored in their material proximity and continuous monitoring responsibilities. Thus, spatial arrangement served as a dynamic resource through which authority was enacted, negotiated, and redistributed moment to moment.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Gaze behavior and the organization of diagnostic attention\u003c/h2\u003e\n \u003cp\u003eGaze behavior played a central role in organizing how clinicians oriented to and evaluated emerging interpretations. Across episodes, proposals were routinely accompanied by predictable gaze shifts\u0026mdash;from the speaker to the relevant artifact, from artifact to team members, or among team members themselves\u0026mdash;that structured the collective attention of the group. When a clinician advanced a diagnostic or interpretive proposition, others often responded first not with talk but with gaze convergence toward the associated monitor, chart, or ventilator display. This coordinated visual orientation served as an embodied marker of collective evaluation, signaling that the team was inspecting the evidentiary substrate of the proposal.\u003c/p\u003e\n \u003cp\u003eThe timing of these gazes shifts contributed to the organization of the reasoning sequence. Early gaze alignment\u0026mdash;occurring prior to verbal uptake\u0026mdash;often foreshadowed acceptance or elaboration of a proposal, while delayed or hesitant gaze shifts sometimes marked uncertainty or hesitation to commit to the interpretive direction being suggested.\u003c/p\u003e\n \u003cp\u003eThrough these intertwined practices, gaze served not simply as a marker of attention but as an interactional mechanism for structuring how reasoning unfolded\u0026mdash;highlighting relevant data, cueing participation, and visually synchronizing the team\u0026apos;s interpretive efforts.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 Artifacts as co-participants in reasoning\u003c/h2\u003e\n \u003cp\u003eMaterial artifacts\u0026mdash;bedside monitors, ventilator interfaces, electronic charts, imaging displays, and laboratory panels\u0026mdash;played an active and constitutive role in shaping the trajectory of collaborative clinical reasoning. Across episodes, clinicians treated these tools not merely as repositories of information but as semiotic partners whose displays, alarms, and visual affordances structured how interpretations were formulated, justified, and revised.\u003c/p\u003e\n \u003cp\u003eA common pattern involved clinicians synchronizing talk with embodied engagement with artifacts. Participants frequently gestured toward waveform shifts, traced lab trends with their fingers, or angled screens toward colleagues as they narrated emerging concerns. These multimodal displays transformed artifacts into shared reference points, enabling the team to jointly inspect data and anchor interpretive claims in observable features of the environment.\u003c/p\u003e\n \u003cp\u003eArtifacts also mediated challenge and justification sequences. When questioning a proposal, clinicians often oriented first to the relevant display\u0026mdash;scrolling through prior values, enlarging a waveform, or highlighting specific parameters\u0026mdash;before articulating an alternative interpretation. This practice grounded disagreements in shared evidentiary terrain, ensuring that divergent perspectives emerged from a common visual foundation.\u003c/p\u003e\n \u003cp\u003eIn several episodes, artifacts exerted directional force on the reasoning process by shifting salience as new values appeared or when alarms signaled abrupt physiological change. These events prompted rapid reorientation of team attention, often triggering reframing or recalibration of the working diagnosis. The artifact, in effect, introduced new turns into the reasoning sequence, acting as a participant whose \u0026quot;contributions\u0026quot; shaped what became relevant for immediate discussion.\u003c/p\u003e\n \u003cp\u003eThrough these practices, artifacts functioned not as passive background objects but as integral components of the cognitive system within which ICU reasoning unfolded (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Their spatial arrangement, visual affordances, and temporal responsiveness actively structured the ways clinicians noticed, evaluated, and coordinated interpretations. Viewing artifacts as co-participants highlights CCR as a deeply distributed, multimodal phenomenon\u0026mdash;one in which reasoning emerges from the interaction of people, bodies, and tools in a tightly coupled complex system.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study examines CCR as it unfolded within the material, spatial, and interactional ecology of the ICU. Rather than treating reasoning as a cognitive process internal to individual clinicians, our findings demonstrate that CCR is an emergent, distributed, and interactionally organized accomplishment. Across 12 video-recorded ICU episodes, teams built and sustained shared clinical understanding through the coordination of talk, embodied conduct, and tool-mediated action. Three mechanisms\u0026mdash;progressive clarification, conditional alignment, and situated coordination\u0026mdash;capture how clinicians jointly assembled interpretations, managed uncertainty, and oriented their reasoning within the spatial and material affordances of the ICU (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eThis study extends existing research on clinical reasoning in three ways. First, it shifts the unit of analysis from individual cognition to interactional accomplishment, providing an empirically grounded account of collaborative reasoning as a social practice. Second, it identifies specific interactional mechanisms through which reasoning is constructed, negotiated, and stabilised in practice\u0026mdash;mechanisms that represent learnable skills which can be noticed, supported, and coached. Third, it demonstrates how video-based microanalysis can operationalise abstract learning theories, contributing a methodological pathway for future research in health professions education. Although this study does not evaluate an educational intervention, the interactional mechanisms identified here provide a basis for understanding how collaborative reasoning is learned and sustained through participation in everyday clinical work.\u003c/p\u003e\n\u003ch3\u003eCCR as Emergent Sensemaking Rather Than Individual Cognition\u003c/h3\u003e\n\u003cp\u003eTraditional clinical reasoning frameworks draw heavily on information-processing models of the individual mind (Elstein et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e1978\u003c/span\u003e; Croskerry, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Norman, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). Our findings align more closely with distributed cognition perspectives that locate reasoning across people, artifacts, and environments (Cook \u0026amp; Rasmussen, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Hutchins, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1995\u003c/span\u003e; Lingard, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). The three interactional mechanisms we identified\u0026mdash;progressive clarification, conditional alignment, and situated coordination\u0026mdash;can be understood as forms of sensemaking: the ongoing, retrospective process through which actors render equivocal situations more comprehensible and actionable (Weick et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Weick \u0026amp; Weick, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e1995\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eTo clarify the theoretical boundaries of CCR, we position it as a specific form of sensemaking that is distinct from, though related to, several cognate constructs. Unlike team cognition, which typically examines shared mental representations or transactive memory systems, CCR foregrounds the sequential, turn-by-turn emergence of clinical interpretations through interaction. Unlike shared mental models, which posit stable cognitive structures held by team members, CCR emphasises the dynamic, contingent process through which interpretations are assembled and revised in real time. Unlike team situation awareness, which focuses on the accuracy of shared perceptions about the environment, CCR attends to how such perceptions are interactionally achieved and negotiated. What CCR adds to these frameworks is threefold: first, attention to sequential emergence\u0026mdash;how reasoning unfolds through ordered contributions rather than aggregated cognitions; second, a focus on epistemic risk management\u0026mdash;how clinicians navigate the personal and professional stakes of proposing and challenging interpretations; and third, emphasis on embodied coordination\u0026mdash;how spatial positioning, gaze, and material artifacts constitute rather than merely support reasoning. By specifying these mechanisms, CCR provides an empirically tractable account of collaborative cognition grounded in observable practice.\u003c/p\u003e\n\u003cp\u003eWhat distinguishes our account from abstract theorizing about sensemaking is its grounding in observable, sequential practices. Drawing on conversation analysis and multimodal interaction analysis (Schegloff, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Garfinkel, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Heath et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Hindmarsh \u0026amp; Pilnick, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Mondada, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), we showed how shared clinical understanding is constituted turn by turn, with each contribution shaping\u0026mdash;and being shaped by\u0026mdash;what came before. Viewing CCR as sensemaking foregrounds how ICU teams construct, sustain, and revise frames of relevance under conditions of uncertainty, rapid physiological change, and heavy information load. This perspective moves beyond cognitive-individualist models and emphasizes CCR as an emergent, collective accomplishment shaped by the social, material, and temporal ecology of clinical work.\u003c/p\u003e\n\u003ch3\u003eInteractional Management of Uncertainty and Epistemic Risk\u003c/h3\u003e\n\u003cp\u003eResearch on team communication has emphasised the importance of psychological safety and speaking up to patient safety (Leonard et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Sutcliffe et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Our analysis deepens this work by specifying the interactional resources clinicians use to manage epistemic risk in situ. Central to the mechanism of conditional alignment was the use of mitigated formulations\u0026mdash;hedged proposals, partial claims, and tentativeness markers (Heritage, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Heritage \u0026amp; Raymond, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2005\u003c/span\u003e)\u0026mdash;that enabled clinicians to introduce interpretations without overstating certainty.\u003c/p\u003e\n\u003cp\u003eJunior clinicians, in particular, employed epistemic downgrades (\u0026quot;\u003cem\u003emaybe\u003c/em\u003e,\u0026quot; \u0026quot;\u003cem\u003eseems like...\u003c/em\u003e,\u0026quot; \u0026quot;\u003cem\u003eI\u0026apos;m wondering if...\u003c/em\u003e\u0026quot;) that served dual purposes: they managed the personal and professional risks associated with being wrong, and they positioned their proposals as inviting elaboration rather than asserting conclusions. Such formulations helped to distribute interpretive work across the team, encouraging seniors to contribute refinements without framing the trainee\u0026apos;s contribution as an error. Similarly, conditional alignments\u0026mdash;responses like \u0026quot;\u003cem\u003ethat might fit given...\u003c/em\u003e,\u0026quot; \u0026quot;\u003cem\u003eunless this is more cardiac...\u003c/em\u003e,\u0026quot; or \u0026quot;\u003cem\u003ethat could explain it if...\u003c/em\u003e\u0026quot;\u0026mdash;allowed the team to move forward collaboratively while explicitly preserving space for revision (Klein et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eOur findings also complicate the assumption that productive disagreement must be explicit or direct. Rather than overt contradiction, clinicians frequently used softened challenge formats (\u0026quot;\u003cem\u003eI wonder if it\u0026apos;s actually...\u003c/em\u003e,\u0026quot; \u0026quot;\u003cem\u003eCould it instead be...?\u003c/em\u003e\u0026quot;) that introduced alternative framings while minimizing interpersonal threat. These patterns are consistent with research on dispreference in social interaction (Kotthoff, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e1993\u003c/span\u003e; Pomerantz, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1984\u003c/span\u003e) and suggest that effective team reasoning depends on face-sensitive practices that allow divergent views to be aired without escalation. The interactional delicacy involved in surfacing disagreement without damaging rapport (Hutchins, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1995\u003c/span\u003e) is a learnable skill, one that educational interventions might address through modeling, reflection, and deliberate practice (Edmondson, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e1999\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eMateriality and the Spatial Ecology of Reasoning\u003c/h3\u003e\n\u003cp\u003eThe third mechanism, situated coordination, extends the interactional view of CCR by incorporating the material and spatial dimensions of clinical work. Consistent with distributed cognition (Hazlehurst et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Hutchins, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1995\u003c/span\u003e) and professional vision frameworks (Goodwin, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Heath et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), we found that artifacts such as monitors, ventilators, and electronic charts were not merely passive information sources. Instead, they functioned as co-participants in reasoning: their alarms prompted reorientation; their visual affordances shaped what was noticed; and their physical placement structured participation opportunities and authority relations.\u003c/p\u003e\n\u003cp\u003eGaze coordination was central to this process. Drawing on multimodal interaction research (Kendon, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1990\u003c/span\u003e; Mondada, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), we observed that clinicians\u0026apos; gaze shifts\u0026mdash;toward monitors, charts, or colleagues\u0026mdash;served as embodied markers of collective attention, signaling agreement, inviting elaboration, or indicating uncertainty. The synchronization of gaze around shared visual referents created joint attention structures that allowed interpretations to be co-inspected in real time. Similarly, spatial positioning relative to key artifacts influenced who initiated descriptions, whose interpretations gained traction, and how authority was distributed dynamically within the team. This is consistent with distributed cognition\u0026apos;s emphasis on how cognition is shaped not only by what individuals know, but by how knowledge is spatially and materially organized (Hutchins, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1995\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eThese findings support conceptualizing CCR as a form of ecological cognition, in which reasoning emerges through clinicians\u0026apos; engagement with the spatial, temporal, and material affordances of their environment. The ICU\u0026apos;s built ecology\u0026mdash;its artifact-rich infrastructure, spatial layout, and real-time data streams\u0026mdash;constitutes a scaffold that clinicians continuously draw upon to build, test, and revise interpretations under uncertainty. This perspective highlights that improving clinical reasoning is not solely a matter of training individuals; it also involves designing environments, workflows, and technologies that better support collective sensemaking. By foregrounding the material and spatial dimensions of reasoning, our findings point to new opportunities for enhancing ICU situational awareness and optimizing team performance.\u003c/p\u003e\n\u003ch3\u003eOn the Learnability of CCR Mechanisms\u003c/h3\u003e\n\u003cp\u003eWe have characterised the three mechanisms of CCR as \u0026ldquo;learnable social practices,\u0026rdquo; and this claim warrants explicit justification. It is important to distinguish learnability from teachability: our study identifies interactional practices that clinicians demonstrably acquire through participation in clinical work, but it does not evaluate instructional interventions or trace individual learning trajectories. The claim of learnability rests on three forms of evidence. First, recurrence: the mechanisms we identified were not idiosyncratic to particular clinicians or episodes but appeared systematically across 12 distinct ICU encounters involving different team compositions and clinical scenarios. This recurrence suggests that these are recognisable, shared practices within the professional community rather than individual cognitive habits. Second, recognisability: participants oriented to these practices as normatively accountable. Mitigated proposals, conditional alignments, and spatially organised contributions were treated as appropriate or expected moves; deviations\u0026mdash;such as categorical assertions where hedging was warranted\u0026mdash;prompted visible interactional repair. This normative orientation indicates that these practices constitute learnable conventions, not merely spontaneous behaviours. Third, coachability: our analysis revealed moments where senior clinicians modelled practices for juniors, where repair sequences recalibrated inappropriate contributions, and where trainees visibly adapted their participation in response to uptake or non-uptake of prior turns. These interactional features suggest that the mechanisms are amenable to guided development, even though our study does not evaluate formal coaching interventions. Crucially, our claim is that this study identifies what could be learned\u0026mdash;the interactional practices through which competent CCR is achieved\u0026mdash;rather than documenting how learning trajectories unfold over time. Future research employing longitudinal or intervention-based designs would be needed to specify how these mechanisms are acquired, how expertise develops, and how targeted educational approaches might accelerate competence.\u003c/p\u003e\n\u003ch3\u003eTheoretical Implications for Learning and Educational Practice\u003c/h3\u003e\n\u003cp\u003eReconceptualizing CCR as a multimodal, materially distributed, and interactionally organized practice challenges several longstanding assumptions in health professions education. The findings suggest that improving team reasoning requires pedagogical approaches that extend well beyond verbal report, cognitive checklists, or isolated individual performance. Four key implications follow for educators, supervisors, and assessment designers.\u003c/p\u003e\n\u003cp\u003e\u003cspan\u003e\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e1. Reasoning cannot be taught or assessed solely through verbal explanation.\u003c/strong\u003e Much of CCR unfolds through embodied and spatial practices\u0026mdash;gaze coordination, pointing to artifacts, positioning relative to monitors, and synchronizing attention with colleagues\u0026mdash;rather than explicit verbal articulation. Educational models that privilege \u0026quot;thinking aloud,\u0026quot; structured case presentations, or post-hoc rationales risk obscuring how reasoning is produced in situ. This critique aligns with prior research showing that clinical cognition often becomes visible not through verbal descriptions but through situated action. Simulation design and debriefing practices must therefore expand their focus to include embodied conduct and material engagement, treating them as legitimate components of reasoning rather than peripheral behaviors.\u003c/p\u003e\u003cspan\u003e\n \u003cp\u003e\u003cstrong\u003e2. Communication training must incorporate embodied and material modalities.\u003c/strong\u003e Effective CCR requires clinicians to manage far more than verbal exchange. In every episode analyzed, spatial positioning, access to shared displays, direction of gaze, and the coordinated use of artifacts played decisive roles in how interpretations were proposed, taken up, or revised. These competencies extend well beyond the scope of traditional communication curricula, which typically emphasize clarity, assertiveness, or structured formats such as SBAR. What our findings demonstrate is that reasoning is contingent on clinicians\u0026apos; ability to orient their bodies, gestures, and attention in ways that make their interpretations visible and actionable to others.\u003c/p\u003e\n\u003c/span\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003eTraining programs therefore need to incorporate explicit instruction and guided practice in how to work with the material and spatial affordances of the clinical environment. Learners must be supported in developing an awareness of when to step toward or away from monitors to facilitate shared viewing, how to use gestures or screen orientation to anchor claims in observable evidence, and how gaze coordination signals alignment, invites response, or marks the need for clarification.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3. Assessment systems must evaluate interactional competence, not isolated decision accuracy.\u003c/strong\u003e Prevailing approaches to assessing clinical reasoning tend to rely on outcome-focused metrics\u0026mdash;such as diagnostic accuracy or appropriateness of management decisions\u0026mdash;or on post-hoc rationalizations provided through written responses, multiple-choice tests, or oral explanations. These methods assume that reasoning is a discrete cognitive product that can be evaluated after the fact. However, our findings indicate that competence in CCR is enacted in real time through interaction, not merely through the correctness of conclusions reached.\u003c/p\u003e\n\u003cp\u003eClinicians demonstrated reasoning competence by introducing tentative proposals in ways that invited uptake, collaboratively managing uncertainty as situations evolved, negotiating competing interpretations without undermining team cohesion, and grounding their claims in shared visual or material evidence. Such behaviors are not peripheral to reasoning; they constitute the mechanism through which CCR is achieved. Assessment systems must therefore expand beyond evaluating isolated decisions or retrospective narratives and incorporate methods capable of capturing reasoning as an interactional accomplishment.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4. Interprofessional education should address how epistemic authority is enacted and negotiated.\u003c/strong\u003e Our findings demonstrate that epistemic authority in the ICU is not a fixed attribute tied to formal role or seniority but an interactional accomplishment that is continually negotiated through stance-taking, mitigated proposals, gaze behaviour, and embodied positioning. These subtle practices regulate who is granted the space to speak, whose interpretations receive uptake, and how uncertainty can be voiced without jeopardizing social cohesion. While many interprofessional training programs encourage \u0026quot;speaking up\u0026quot; as an individual competency, such approaches risk oversimplifying the complex, relational nature of epistemic participation in high-stakes environments.\u003c/p\u003e\n\u003cp\u003eEducational initiatives therefore require a shift toward helping learners recognize how epistemic authority is co-constructed moment to moment. Analysis of authentic clinical footage can make visible the multimodal cues\u0026mdash;gaze shifts, bodily alignment with artifacts, prosodic modulation\u0026mdash;that signal when authority is being asserted, shared, or ceded. By framing authority as an emergent interactional process rather than a static hierarchical property, interprofessional education can better equip clinicians to participate in and sustain effective CCR.\u003c/p\u003e\n\u003cp\u003eSynthesising across these implications, we suggest that educators, clinical supervisors, and assessment designers attend to three core principles. First, educators should notice that reasoning competence manifests not only in what clinicians say but in how they coordinate talk, gaze, gesture, and positioning with colleagues and artifacts\u0026mdash;video review of authentic or simulated practice can make these practices visible in ways that verbal debriefing alone cannot. Second, supervisors should support learners in developing the interactional repertoires that enable effective participation: modelling mitigated proposals, creating space for trainees to contribute tentatively, and explicitly coaching how to ground claims in shared material evidence. Third, assessment designers should reconsider whether current methods\u0026mdash;focused on individual diagnostic accuracy or post-hoc verbal justification\u0026mdash;capture the collaborative, embodied, and temporally unfolding nature of clinical reasoning; process-oriented assessment using video analysis or structured observation may complement outcome-focused measures. These principles do not prescribe specific interventions but offer a framework for aligning educational practices with the interactional realities of collaborative clinical work.\u003c/p\u003e\n\u003ch3\u003eLimitations and Directions for Future Research\u003c/h3\u003e\n\u003cp\u003eThis study has several limitations that shape the scope and transferability of its findings. First, the analysis was conducted in a single tertiary ICU with a mature interprofessional culture, established workflow patterns, and a high density of technological resources. As interactional and material ecologies vary substantially across clinical environments, the mechanisms identified here may manifest differently in settings such as emergency departments, operating suites, outpatient clinics, or resource-constrained hospitals. Comparative, multi-site work would allow researchers to examine how organizational culture, staffing models, and infrastructural arrangements modulate collaborative reasoning practices. Second, the study focused on naturally occurring team interactions rather than trainee-specific trajectories. Although the data capture moments of junior\u0026ndash;senior negotiation, a more developmental perspective is needed to understand how learners acquire competence in multimodal reasoning practices\u0026mdash;such as managing spatial positioning, coordinating gaze, or using artifacts to justify claims. Longitudinal or simulation-based methodologies could illuminate how these skills emerge, stabilize, and potentially deteriorate under varying workload and supervisory conditions. Third, because the dataset predates the widespread integration of emerging technologies\u0026mdash;remote monitoring, AI-driven decision support, automated trend detection\u0026mdash;the interactional architecture of CCR is likely to evolve as such systems become embedded in ICU workflow. Future research should examine how these tools redistribute epistemic authority, reshape attention patterns, or alter the sequence organization of reasoning, building on scholarship in human\u0026ndash;technology interaction and distributed cognition. Finally, while this study foregrounds moments of coordination, future work should also investigate episodes where CCR falters: how teams repair breakdowns in alignment, how disagreements escalate or dissipate, and how these ruptures influence patient care. Analyzing breakdowns may deepen understanding of the vulnerabilities in collaborative sensemaking and identify opportunities for targeted educational or organizational interventions.\u003c/p\u003e\n\u003cp\u003eRegarding scope conditions and transferability, we distinguish between context-sensitive and context-robust features of the mechanisms identified. This study makes no claim that all three mechanisms will be equally visible or consequential in non-acute settings. We propose that progressive clarification and conditional alignment are likely to be context-robust\u0026mdash;that is, their core interactional logic should operate across clinical settings wherever teams engage in collaborative diagnostic reasoning under uncertainty, though the specific linguistic forms, turn-taking norms, and hierarchical dynamics may vary by culture, specialty, and institutional context. In contrast, situated coordination is more context-sensitive: its specific manifestations depend heavily on the material ecology of the setting\u0026mdash;the density and configuration of artifacts, the spatial layout, and the availability of shared visual displays. In an outpatient clinic with limited technology, for example, the role of artifacts in structuring participation may be attenuated, whereas in a technologically sparse environment, verbal and embodied coordination may assume greater salience. Future research could test these hypotheses by examining CCR in lower-acuity settings such as general medical wards, outpatient clinics, or educational simulation environments, where time pressure is reduced and information is less distributed. Such studies would help determine which features of the mechanisms are invariant properties of collaborative clinical reasoning and which are amplified or attenuated by contextual factors. This would advance the theoretical generalisability of the CCR construct while generating actionable insights for educators working across diverse clinical learning environments.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eCCR in the ICU is not a private cognitive act but an emergent, socially, and materially distributed accomplishment. Through progressive clarification, teams incrementally assemble a coherent clinical picture; through conditional alignment, they manage uncertainty and negotiate interpretive risk; and through situated coordination, they leverage spatial, embodied, and material resources to shape the trajectory of reasoning. By conceptualizing CCR as a form of ecological sensemaking\u0026mdash;a dynamic interplay among people, artifacts, and environments\u0026mdash;this study reframes how reasoning should be understood, taught, assessed, and supported in high-stakes clinical contexts. Our claim of learnability refers to the recognisable, normatively accountable nature of these practices within clinical interaction, rather than to demonstrated instructional effectiveness or accelerated learning through formal intervention. Recognizing CCR as an interactional phenomenon expands possibilities for educational innovation, simulation design, and organizational improvement, offering a foundation for building learning environments that cultivate not only individual cognitive skill but the collective capacities required for safe and effective patient care.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics Approval and Consent to Participate\u003c/h2\u003e\n\u003cp\u003eThis study was approved by the Institutional Review Board of Chang Gung Medical Foundation (Ref: 202002453B0). All participating clinicians provided informed consent for video recording and research use of clinical interaction data. All procedures were conducted in accordance with the ethical standards of the Declaration of Helsinki and local regulatory requirements.\u003c/p\u003e\n\u003ch2\u003eConsent for Publication\u003c/h2\u003e\n\u003cp\u003eAll participants consented to the publication of anonymized excerpts of interactional data. No identifiable patient information was collected or included in the analysis. All video and transcript materials were de-identified prior to publication.\u003c/p\u003e\n\u003ch2\u003eAvailability of Data and Materials\u003c/h2\u003e\n\u003cp\u003eThe datasets generated and analyzed during the current study\u0026mdash;including de-identified transcripts and analytic memos\u0026mdash;are not publicly available due to the sensitivity of clinical video data and restrictions imposed by the Institutional Review Board. Upon reasonable request, de-identified excerpts relevant to the findings may be shared with qualified researchers, subject to institutional and ethical approval.\u003c/p\u003e\n\u003ch2\u003eCompeting Interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis research was supported by Ministry of Science and Technology (R.O.C.) under Grant Number [NSTC 113-2628-H-182-002-MY2]. The funder had no role in the study design, data collection, analysis, interpretation of data, or preparation of the manuscript.\u003c/p\u003e\n\u003ch2\u003eAuthors\u0026apos; Contributions\u003c/h2\u003e\n\u003cp\u003eCYL led the conception and design of the study; coordinated data collection and video acquisition; conducted the primary interactional, multimodal, and conversation-analytic coding; performed the initial and iterative data analyses; drafted the full manuscript; and integrated co-authors\u0026apos; feedback into the final version. CYL had full access to the data and takes primary responsibility for the accuracy and integrity of the analysis and interpretation. CHL and HYL contributed to the sequential and multimodal analyses and drafted analytic memos that informed the interpretive framework. MMC conducted video data collection, transcription, and multimodal annotation. SYY conceived the study, led the overall design, supervised the microinteractional analysis, and provided methodological consultation in psychology and qualitative inquiry.\u003c/p\u003e\n\u003cp\u003eAll authors contributed to manuscript drafting and critical revisions. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eWe thank the clinicians of the Neurosurgery Intensive Care Unit at Linkou Chang Gung Memorial Hospital for their participation and for granting us access to their clinical practice for research purposes.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBerg, M. (1999). Accumulating and coordinating: occasions for information technologies in medical work. \u003cem\u003eComputer Supported Cooperative Work (CSCW), 8(4), 373\u0026ndash;401\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1023/A:1008757115404\u003c/span\u003e\u003cspan address=\"10.1023/A:1008757115404\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCook, D. A., Hamstra, S. J., Brydges, R., Zendejas, B., Szostek, J. H., Wang, A. T., Erwin, P. J., \u0026amp; Hatala, R. (2013). Comparative effectiveness of instructional design features in simulation-based education: systematic review and meta-analysis. \u003cem\u003eMedical Teacher\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(1), e867\u0026ndash;e898. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3109/0142159X.2012.714886\u003c/span\u003e\u003cspan address=\"10.3109/0142159X.2012.714886\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCook, R., \u0026amp; Rasmussen, J. (2005). Going solid: a model of system dynamics and consequences for patient safety. \u003cem\u003eBMJ Quality \u0026amp; Safety\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(2), 130\u0026ndash;134. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/qshc.2003.009530\u003c/span\u003e\u003cspan address=\"10.1136/qshc.2003.009530\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCox, E. D., Jacobsohn, G. C., Rajamanickam, V. P., Carayon, P., Kelly, M. M., Wetterneck, T. B., Rathouz, P. J., \u0026amp; Brown, R. L. (2017). A family-centered rounds checklist, family engagement, and patient safety: a randomized trial. \u003cem\u003ePediatrics\u003c/em\u003e, \u003cem\u003e139\u003c/em\u003e(5), e20161688. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1542/peds.2016-1688\u003c/span\u003e\u003cspan address=\"10.1542/peds.2016-1688\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCroskerry, P. (2009). Clinical cognition and diagnostic error: applications of a dual process model of reasoning. \u003cem\u003eAdvances in Health Sciences Education\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(Suppl 1), 27\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10459-009-9182-2\u003c/span\u003e\u003cspan address=\"10.1007/s10459-009-9182-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDonchin, Y., \u0026amp; Seagull, F. J. (2002). The hostile environment of the intensive care unit. \u003cem\u003eCurrent Opinion in Critical Care\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(4), 316\u0026ndash;320. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/00075198-200208000-00008\u003c/span\u003e\u003cspan address=\"10.1097/00075198-200208000-00008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurning, S. J., ArtinoJr, A. R., Schuwirth, L., \u0026amp; Van Der Vleuten, C. (2013). Clarifying assumptions to enhance our understanding and assessment of clinical reasoning. \u003cem\u003eAcademic Medicine\u003c/em\u003e, \u003cem\u003e88\u003c/em\u003e(4), 442\u0026ndash;448. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/ACM.0b013e3182851b5b\u003c/span\u003e\u003cspan address=\"10.1097/ACM.0b013e3182851b5b\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdmondson, A. (1999). Psychological safety and learning behavior in work teams. \u003cem\u003eAdministrative Science Quarterly\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(2), 350\u0026ndash;383. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2666999\u003c/span\u003e\u003cspan address=\"10.2307/2666999\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElstein, A. S., Shulman, L. S., \u0026amp; Sprafka, S. A. (1978). \u003cem\u003eMedical Problem Solving: An Analysis of Clinical Reasoning\u003c/em\u003e. Harvard University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEndsley, M. R. (2017). Toward a theory of situation awareness in dynamic systems. In \u003cem\u003eSituational Awareness\u003c/em\u003e (pp. 9\u0026ndash;42). Routledge. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4324/9781315087924-2\u003c/span\u003e\u003cspan address=\"10.4324/9781315087924-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEva, K. W. (2005). What every teacher needs to know about clinical reasoning. \u003cem\u003eMedical Education\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(1), 98\u0026ndash;106. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-2929.2004.01972.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2929.2004.01972.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEva, K. W., \u0026amp; Norman, G. R. (2005). Heuristics and biases\u0026mdash;a biased perspective on clinical reasoning. \u003cem\u003eMedical Education\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(9), 870\u0026ndash;872. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-2929.2005.02258.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2929.2005.02258.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarfinkel, H. (2023). Studies in ethnomethodology. In \u003cem\u003eSocial Theory Re-Wired\u003c/em\u003e (pp. 58\u0026ndash;66). Routledge. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4324/9781003320609-8\u003c/span\u003e\u003cspan address=\"10.4324/9781003320609-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoodwin, C. (2015). Professional vision. \u003cem\u003eAufmerksamkeit: Geschichte-Theorie-Empirie\u003c/em\u003e (pp. 387\u0026ndash;425). Springer.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHazlehurst, B., McMullen, C. K., \u0026amp; Gorman, P. N. (2007). Distributed cognition in the heart room: how situation awareness arises from coordinated communications during cardiac surgery. \u003cem\u003eJournal of Biomedical Informatics\u003c/em\u003e, \u003cem\u003e40\u003c/em\u003e(5), 539\u0026ndash;551. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jbi.2007.02.001\u003c/span\u003e\u003cspan address=\"10.1016/j.jbi.2007.02.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHealey, A. N., Undre, S., \u0026amp; Vincent, C. A. (2006). Defining the technical skills of teamwork in surgery. \u003cem\u003eBMJ Quality \u0026amp; Safety\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(4), 231\u0026ndash;234. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/qshc.2005.017517\u003c/span\u003e\u003cspan address=\"10.1136/qshc.2005.017517\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeath, C., Luff, P., \u0026amp; Sanchez Svensson, M. (2007). Video and qualitative research: analysing medical practice and interaction. \u003cem\u003eMedical Education\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(1), 109\u0026ndash;116. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-2929.2006.02641.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2929.2006.02641.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeritage, J. (2012). Epistemics in action: Action formation and territories of knowledge. \u003cem\u003eResearch on Language \u0026amp; Social Interaction\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e(1), 1\u0026ndash;29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/08351813.2012.646684\u003c/span\u003e\u003cspan address=\"10.1080/08351813.2012.646684\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeritage, J., \u0026amp; Raymond, G. (2005). The terms of agreement: Indexing epistemic authority and subordination in talk-in-interaction. \u003cem\u003eSocial Psychology Quarterly\u003c/em\u003e, \u003cem\u003e68\u003c/em\u003e(1), 15\u0026ndash;38. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/019027250506800103\u003c/span\u003e\u003cspan address=\"10.1177/019027250506800103\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHindmarsh, J., \u0026amp; Pilnick, A. (2007). Knowing bodies at work: Embodiment and ephemeral teamwork in anaesthesia. \u003cem\u003eOrganization Studies\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(9), 1395\u0026ndash;1416. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0170840607068258\u003c/span\u003e\u003cspan address=\"10.1177/0170840607068258\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHutchins, E. (1995). \u003cem\u003eCognition in the Wild\u003c/em\u003e. MIT Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHutchins, E. (2000). Distributed cognition. \u003cem\u003eInternational Encyclopedia of the Social and Behavioral Sciences\u003c/em\u003e, \u003cem\u003e138\u003c/em\u003e(1), 1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIlgen, J. S., Humbert, A. J., Kuhn, G., Hansen, M. L., Norman, G. R., Eva, K. W., Charlin, B., \u0026amp; Sherbino, J. (2012). Assessing diagnostic reasoning: a consensus statement summarizing theory, practice, and future needs. \u003cem\u003eAcademic Emergency Medicine\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(12), 1454\u0026ndash;1461. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/acem.12034\u003c/span\u003e\u003cspan address=\"10.1111/acem.12034\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKendon, A. (1990). \u003cem\u003eConducting Interaction: Patterns of Behavior in Focused Encounters\u003c/em\u003e (Vol. 7). CUP Archive.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKlein, G., Moon, B., \u0026amp; Hoffman, R. R. (2006). Making sense of sensemaking 1: Alternative perspectives. \u003cem\u003eIEEE Intelligent Systems\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(4), 70\u0026ndash;73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/MIS.2006.75\u003c/span\u003e\u003cspan address=\"10.1109/MIS.2006.75\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKneebone, R. (2005). Evaluating clinical simulations for learning procedural skills: a theory-based approach. \u003cem\u003eAcademic Medicine\u003c/em\u003e, \u003cem\u003e80\u003c/em\u003e(6), 549\u0026ndash;553. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/00001888-200506000-00006\u003c/span\u003e\u003cspan address=\"10.1097/00001888-200506000-00006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKotthoff, H. (1993). Disagreement and concession in disputes: On the context sensitivity of preference structures. \u003cem\u003eLanguage in Society\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(2), 193\u0026ndash;216. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S0047404500017103\u003c/span\u003e\u003cspan address=\"10.1017/S0047404500017103\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLane, D., Ferri, M., Lemaire, J., McLaughlin, K., \u0026amp; Stelfox, H. T. (2013). A systematic review of evidence-informed practices for patient care rounds in the ICU. \u003cem\u003eCritical Care Medicine\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(8), 2015\u0026ndash;2029. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/CCM.0b013e31828a435f\u003c/span\u003e\u003cspan address=\"10.1097/CCM.0b013e31828a435f\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeonard, M., Graham, S., \u0026amp; Bonacum, D. (2004). The human factor: the critical importance of effective teamwork and communication in providing safe care. \u003cem\u003eBMJ Quality \u0026amp; Safety\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(suppl 1), i85\u0026ndash;i90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/qshc.2004.010033\u003c/span\u003e\u003cspan address=\"10.1136/qshc.2004.010033\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLingard, L. (2012). Rethinking competence in the context of teamwork. \u003cem\u003eThe Question of Competence: Reconsidering Medical Education in the Twenty-First Century\u003c/em\u003e (pp. 42\u0026ndash;69). Cornell University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLingard, L., Reznick, R., Espin, S., Regehr, G., \u0026amp; DeVito, I. (2002). Team communications in the operating room: talk patterns, sites of tension, and implications for novices. \u003cem\u003eAcademic Medicine\u003c/em\u003e, \u003cem\u003e77\u003c/em\u003e(3), 232\u0026ndash;237. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/00001888-200203000-00010\u003c/span\u003e\u003cspan address=\"10.1097/00001888-200203000-00010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaitlis, S., \u0026amp; Christianson, M. (2014). Sensemaking in organizations: Taking stock and moving forward. \u003cem\u003eAcademy of Management Annals\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(1), 57\u0026ndash;125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/19416520.2014.873177\u003c/span\u003e\u003cspan address=\"10.1080/19416520.2014.873177\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManser, T. (2009). Teamwork and patient safety in dynamic domains of healthcare: a review of the literature. \u003cem\u003eActa Anaesthesiologica Scandinavica\u003c/em\u003e, \u003cem\u003e53\u003c/em\u003e(2), 143\u0026ndash;151. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1399-6576.2008.01717.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1399-6576.2008.01717.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMondada, L. (2018). Multiple temporalities of language and body in interaction: Challenges for transcribing multimodality. \u003cem\u003eResearch on Language and Social Interaction\u003c/em\u003e, \u003cem\u003e51\u003c/em\u003e(1), 85\u0026ndash;106. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/08351813.2018.1413878\u003c/span\u003e\u003cspan address=\"10.1080/08351813.2018.1413878\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNorman, G. (2005). Research in clinical reasoning: past history and current trends. \u003cem\u003eMedical Education\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(4), 418\u0026ndash;427. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-2929.2005.02127.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2929.2005.02127.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNorman, G. (2009). Dual processing and diagnostic errors. \u003cem\u003eAdvances in Health Sciences Education\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(Suppl 1), 37\u0026ndash;49. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10459-009-9179-x\u003c/span\u003e\u003cspan address=\"10.1007/s10459-009-9179-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNorman, G. R., Grierson, L. E., Sherbino, J., Hamstra, S. J., Schmidt, H. G., \u0026amp; Mamede, S. (2018). Expertise in medicine and surgery. In K. A. Ericsson, R. R. Hoffman, A. Kozbelt, \u0026amp; A. M. Williams (Eds.), \u003cem\u003eThe Cambridge Handbook of Expertise and Expert Performance\u003c/em\u003e (2nd ed., pp. 331\u0026ndash;355). Cambridge University Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/9781316480748.019\u003c/span\u003e\u003cspan address=\"10.1017/9781316480748.019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In J. M. Atkinson, \u0026amp; J. Heritage (Eds.), \u003cem\u003eStructures of Social Action: Studies in Conversation Analysis\u003c/em\u003e (pp. 57\u0026ndash;101). Cambridge University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReader, T. W., Flin, R., Mearns, K., \u0026amp; Cuthbertson, B. H. (2011). Team situation awareness and the anticipation of patient progress during ICU rounds. \u003cem\u003eBMJ Quality \u0026amp; Safety\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(12), 1035\u0026ndash;1042. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmjqs.2010.048561\u003c/span\u003e\u003cspan address=\"10.1136/bmjqs.2010.048561\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRudolph, J. W., Morrison, J. B., \u0026amp; Carroll, J. S. (2009). The dynamics of action-oriented problem solving: Linking interpretation and choice. \u003cem\u003eAcademy of Management Review\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(4), 733\u0026ndash;756. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5465/amr.34.4.zok733\u003c/span\u003e\u003cspan address=\"10.5465/amr.34.4.zok733\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSalas, E., Cooke, N. J., \u0026amp; Rosen, M. A. (2008). On teams, teamwork, and team performance: Discoveries and developments. \u003cem\u003eHuman Factors\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e(3), 540\u0026ndash;547. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1518/001872008X288457\u003c/span\u003e\u003cspan address=\"10.1518/001872008X288457\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchegloff, E. A. (2007). \u003cem\u003eSequence Organization in Interaction: A Primer in Conversation Analysis I\u003c/em\u003e (Vol. 1). Cambridge University Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/CBO9780511791208\u003c/span\u003e\u003cspan address=\"10.1017/CBO9780511791208\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSutcliffe, K. M., Lewton, E., \u0026amp; Rosenthal, M. M. (2004). Communication failures: an insidious contributor to medical mishaps. \u003cem\u003eAcademic Medicine\u003c/em\u003e, \u003cem\u003e79\u003c/em\u003e(2), 186\u0026ndash;194. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/00001888-200402000-00019\u003c/span\u003e\u003cspan address=\"10.1097/00001888-200402000-00019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeaver, S. J., Dy, S. M., \u0026amp; Rosen, M. A. (2014). Team-training in healthcare: a narrative synthesis of the literature. \u003cem\u003eBMJ Quality \u0026amp; Safety\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(5), 359\u0026ndash;372. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmjqs-2013-001848\u003c/span\u003e\u003cspan address=\"10.1136/bmjqs-2013-001848\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeick, K. E., Sutcliffe, K. M., \u0026amp; Obstfeld, D. (2005). Organizing and the process of sensemaking. \u003cem\u003eOrganization Science\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e(4), 409\u0026ndash;421. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1287/orsc.1050.0133\u003c/span\u003e\u003cspan address=\"10.1287/orsc.1050.0133\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeick, K. E., \u0026amp; Weick, K. E. (1995). \u003cem\u003eSensemaking in Organizations\u003c/em\u003e (Vol. 3). Sage.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"advances-in-health-sciences-education","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ahse","sideBox":"Learn more about [Advances in Health Sciences Education](http://link.springer.com/journal/10459)","snPcode":"10459","submissionUrl":"https://submission.nature.com/new-submission/10459/3","title":"Advances in Health Sciences Education","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Collaborative clinical reasoning, Critical care, Microinteractional analysis, Teamwork in healthcare, Workplace learning, Distributed cognition","lastPublishedDoi":"10.21203/rs.3.rs-8666674/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8666674/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eClinical reasoning research has predominantly conceptualised reasoning as individual cognition, yet contemporary clinical practice distributes reasoning across teams, tools, and time. This study theorises collaborative clinical reasoning (CCR) as a learnable social practice constituted through observable interactional mechanisms. Examining ICU teams as an epistemic extreme case\u0026mdash;where high stakes and time pressure amplify reasoning practices present but less visible elsewhere\u0026mdash;we ask: through what mechanisms do clinicians construct shared clinical understanding?\u003c/p\u003e \u003cp\u003eVideo-based microinteractional analysis of 18 ICU encounters (27 hours) identified three mechanisms through which clinicians collaboratively construct shared clinical understanding.\u003c/p\u003e \u003cp\u003eWe identified three interactional mechanisms through which CCR is accomplished. \u003cem\u003eProgressive clarification\u003c/em\u003e is a mechanism of shared problem construction through layered, profession-specific contributions. \u003cem\u003eConditional alignment\u003c/em\u003e is a mechanism of epistemic negotiation enabling hypothesis exploration while preserving team cohesion. \u003cem\u003eSituated coordination\u003c/em\u003e is a mechanism of material\u0026ndash;interactional coupling whereby spatial positioning and artifacts structure participation.\u003c/p\u003e \u003cp\u003eThese mechanisms represent learnable practices that can be noticed, supported, and coached. By shifting the unit of analysis from individual cognition to interactional accomplishment, this study provides an empirically grounded foundation for understanding how collaborative reasoning is learned and sustained through participation in clinical work.\u003c/p\u003e","manuscriptTitle":"Collaborative Clinical Reasoning in the ICU: How Teams Construct Shared Understanding in Real Time","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-08 16:16:15","doi":"10.21203/rs.3.rs-8666674/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-19T00:59:00+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-08T12:09:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"182924418260460903415311923819308249929","date":"2026-04-27T01:44:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"180076111847005683674704809269038871477","date":"2026-04-24T15:50:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"49235303803677740966794816930668689097","date":"2026-04-23T17:33:37+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-23T16:13:20+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-24T01:38:37+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-24T01:37:39+00:00","index":"","fulltext":""},{"type":"submitted","content":"Advances in Health Sciences Education","date":"2026-01-22T07:22:11+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"advances-in-health-sciences-education","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ahse","sideBox":"Learn more about [Advances in Health Sciences Education](http://link.springer.com/journal/10459)","snPcode":"10459","submissionUrl":"https://submission.nature.com/new-submission/10459/3","title":"Advances in Health Sciences Education","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d038432a-1831-446c-8df5-69c190be8f7f","owner":[],"postedDate":"May 8th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-19T00:59:00+00:00","index":26,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-08T12:09:51+00:00","index":25,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-08T16:16:21+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-08 16:16:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8666674","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8666674","identity":"rs-8666674","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}