ORGANIZING THOUGHTS AND CONNECTING BRAINS: MATERIAL...

® Academy of Management Journal2012. Vol. 55. No. 5, 1232-1259.http://cb!.doi.org/10.5465/amj.2010.0890

ORGANIZING THOUGHTS AND CONNECTING BRAINS:MATERIAL PRACTICES AND THE TRANSITION FROM

INDIVIDUAL TO GROUP-LEVEL PROSPECTIVE SENSEMAKING

ILEANA STIGLIANIImperial College London

DAVIDE RAVASIBocconi University

A growing body of research is drawing attention to the material practices that supportverbal exchanges and cognitive processes in collective sensemaking. In this study,building on an ethnographic study of a design consulting firm, we develop a processmodel that accounts for the interplay between conversational and material practices inthe transition from individual to group-level sensemaking, and we begin to unpackhow the "materialization" of cognitive work supports the collective construction ofnew shared understandings.

Sensemaking is commonly understood as a pro-cess in which individuals or groups attempt to in-terpret novel and ambiguous situations (Weick,1995). The process begins when people confrontevents or tasks they cannot readily interpret usingavailable mental structures (Kiesler & Sproull,1982). Collective sensemaking occurs as individu-als exchange provisional understandings and try toagree on consensual interpretations and a course ofaction (Weick, Sutcliffe, & Obstfeld, 2005).

Scholars generally agree that individual andgroup-level sensemaking processes are related, in

This cirticle is the result of a thoroughly collaborativecoauthoring process. We thank all managers and employ-ees at Continuum, and in particular Gianfranco Zaccai,president and chief design officer. We gratefully ac-knowledge the continuous and encouraging support andinsightful feedback of Kimberly D. Elsbach. We gratefullyacknowledge the helpful comments of Daved Berry,Charlotte Gloutier, Denny Gioia, Andrew Hargadon,Mary Jo Hatch, Gerard Hodgkinson, Sally Maitlis, Ste-phen Mezias, Mike Pratt, Anat Rafaeli, Anna Rylander,the participants in the 2011 Davis Conference on Quali-tative Research, and the participants in the 2009 "Publi-cation Bootcamp," Imperial College Business School, onearlier versions of this work, as well as the guidanceprovided by John Van Maanen to the first author duringher ethnographic study at Continuum. We are especiallythankful to Claus Rerup and Viviane Sergi for their pre-cious suggestions, and to AMJ Associate Editor KevinCorley and to our three anonymous reviewers for thevaluable comments and suggestions provided during thereview process. Davide Ravasi gratefully acknowledgesthe support of the Claudio Demattè Research Division ofthe SDA Bocconi School of Management.

that individual interpretive efforts feed collectiveones (Weick, 1995; Weick et al., 2005). However,although sophisticated theories of individual (e.g.,Weick, 1993, 1995) and collective sensemaking(e.g., Balogun & Johnson, 2004; Maitlis, 2005) areavailable, a comprehensive empirical and theoreti-cal account of the social practices and cognitivework that underpin the transition from one level ofanalysis to tbe other is still missing.

Consistently with a prevailing view of sensemak-ing as "an issue of language, talk, and communica-tion" (Weick et al., 2005: 409), past research oncollective sensemaking has placed emphasis onconversational practices such as argumentation(Weick, 1995), metaphorical communication (Cor-nelissen, 2012), and the exchange of narratives(Sonensheim, 2010) and accounts (Maitlis, 2005)that support convergence around a common inter-pretation of unexpected or ambiguous events, suchas those occurring in times of crises or change(Maitlis & Sonenshein, 2010).

In the last few years, however, a series of studieshas highlighted how individuals rely on a variety ofmaterial practices and artifacts, such as drawingsand prototypes (Bechky, 2003; Carlile, 2002; Sutton& Hargadon, 1996), slide presentations (Kaplan,2011), visual maps (Doyle & Sims, 2002), and evenLego bricks (Oliver & Roos, 2007; Heracleous &Jacobs, 2008), to support the conversational prac-tices through which they exchange, combine, andconstruct interpretations as they collectively en-gage in the less investigated form of prospectivesensemaking (Gioia, Thomas, Clark, & Chittipeddi,1994; Cioia & Mehra, 1996) underpinning future-oriented group processes, such as strategy making.

1232Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed. posted to a listserv, or otherwise transmitted without the copyright holder's expresswritten permission. Users may print, download, or email articles for individual use only.

2012 Stigliani and Ravasi 1233

new product development, and the planning of or-ganizational change.

Research along this line is gaining momentumand, consistently with a broader "material turn" inthe social sciences (Hicks & Beaudry, 2010), it sug-gests how understanding collective sensemaking inpurely linguistic terms may provide only a partialaccount of the process. However, though studentsof sensemaking acknowledge the possibility thatmaterial artifacts may support the construction ofnew shared knowledge structures (e.g.. Rouleau,2005; Balogun & Johnson, 2005; Rouleau & Balo-gun, 2011), scholars still know little about howthey do it. Past studies have focused on specificmaterial practices and artifacts in isolation, but anintegrated theoretical framework accounting for theinterplay between conversational and materialpractices in the transition from individual to col-lective sensemaking is still missing.

To improve understanding of this aspect of thesensemaking process, we carried out an ethno-graphic study of concept development in a designconsulting firm. We considered this setting appro-priate to our research purpose, because productdesigners often face ambiguity regarding both thesolution to the problem they address and the con-text within which this solution will be imple-mented (Clark, 1985; Lawson, 2005), and they usevarious types of artifacts, such as drawings,sketches, and models, to support their interpretiveprocesses (Boland & CoUopy, 2004).

Following a rising perspective in managementstudies, we investigated artifacts as constitutive el-ements of the broader sociomaterial practicesthrough which organizational processes are accom-plished (Orlikowsky & Scott, 2008). Our focus onsocial practices resonates with renewed interest inthe microfoundations of organizational processesin different fields of research, such as strategy (Jar-zabkowski, Balogun, & Seidl, 2007), institutionaltheory (Greenwood, Oliver, Sahlin, & Suddaby,2008), and organizational capabilities (Salvato,2009). In this respect, addressing our research ques-tion is important to deepen understanding of thesociomaterial underpinnings of a fundamental cog-nitive process—prospective sensemaking—that un-derlies all activities associated with planning andinitiating change in organizations.

From our observations, we develop a groundedmodel that advances understanding of this lessinvestigated and undertheorized form of sense-making by unpacking social practices and cogni-tive processes that underpin the transition fromthe individual to the collective level, linking in-dividual generation of early ideas (Hill & Leven-hagen, 1995) with collective engagement in the

negotiation of emerging interpretations (Gioia &Chittipeddi, 1991).

Our model describes prospective collectivesensemaking as based on three interrelated cyclesof retrospective cognitive work occurring as mem-bers of groups go back and forth between the ten-tative organization of selected material cues andthe refinement of corresponding categories, em-body provisional interpretations in material form,and engage in retrospective reflection to establishthe plausibility of emerging accounts. By doing so,our model offers a way to reconcile the future-oriented nature of the collective process with theretrospective nature of individual cognitive work.

Our insights also begin to shed light on how the"materialization of cognitive work," which occursas individuals and groups engage in material prac-tices, supports the construction of new shared un-derstandings. By doing so, we point to materialartifacts as important "sensemaking resources"(Gephart, 1993) that facilitate transitions from in-dividual to collective future-oriented sensemaking.

THEORETICAL BACKGROUND

Sensemaking Theory

Retrospective models of sensemaking. Earlyempirical applications of sensemaking theory fo-cused on discrepancies between a current and anexpected state of the world (e.g., Weick, 1988,1993). Research in this line of inquiry investigatedindividual and group-level responses to unfamiliarevents that occur when people confront circum-stances that do not fit available knowledge struc-tures. According to models of sensemaking arisingfrom these studies, individuals respond to cues thatdisrupt the ordinary, predictable flow of experi-ence and suggest a gap between the reality as itseems to be and how they expected it to be (Barr,1998). These cues trigger conscious attempts to in-terpret unexpected occurrences retrospectively andto bring order into ambiguous realities open to mul-tiple interpretations.

Scholars' understandings of transitions from theindividual to the collective level vary dependingon the focus of their study. On the one hand, re-search on sensemaking in organizational crises(e.g., Weick, 1988, 1993; Weick & Roberts, 1993;Whiteman & Cooper, 2011) has described fast-paced processes in wbich a group's early attemptsto make sense of unexpected events are followed byrapid action aimed at quickly "testing" provisionalinterpretations. Under these circumstances, the ca-pacity of a group to collectively make sense ofchanges and respond to them depends on the ro-

1234 Academy of Management Journal October

bustness of the role system (Weick, 1993) and thequality of social interaction in the group (Weick &Roberts, 1993).

Research on collective sensemaking during or-ganizational change (e.g., Balogun & Johnson, 2004,2005; Rouleau & Balogun, 2010; Sonenshein, 2010),on the other hand, emphasizes the conversationalpractices through which individuals attempt toconstruct a common understanding of a situationrather than their actions to test their understand-ings of the environment. Collective construction ofmeanings is generally described as arising from thespread and sharing of accounts—descriptive con-structions of reality embodying possible interpreta-tions of events and situations (Maitlis, 2005; Mait-lis & Lawrence, 2007).

Prospective models of sensemaking. Anotherrelevant line of inquiry has explored circumstancesunder which individuals and groups cope with am-biguous situations that require them to developnovel understandings and engage in forward-look-ing thinking to "structure the future by imaginingsome desirable (albeit ill-defined) state" (Cioia &Mehra, 1996: 1229). This different type of sense-making has been referred to as "prospective"(Cioia, 1986) or "future-oriented" sensemaking(Cephart, Topal, & Zhang, 2010). Research in thistradition has applied sensemaking theory to strat-egy making (Cioia et al., 1994; Gioia & Thomas,1996), entrepreneurship (Cornelissen & Clarke,2010; Hill & Levenhagen, 1995), and innovation(Rafaeli, Ravid, & Cheshin, 2009; Ravasi & Turati,2005) to investigate the construction of new under-standings of an environment and how to relate to it.Compared to research on crises, these studies showa relatively slow-paced process in which the refine-ment of emerging interpretations results from cy-cles of sensemaking and sensegiving, as groupmembers attempt to infiuence other actors' inter-pretations (Cioia & Chittipeddi, 1991; Hill & Lev-enhagen, 1995).

Despite the fact that prospective sensemakingunderpins fundamental organizational processes,such as those mentioned above, this process is un-derresearched and undertheorized. Available mod-els provide an insigbtful but incomplete conceptu-alization, as little is known of the social interactionand cognitive work that underpin the transitionbetween individual development of new interpre-tations (Hill & Levenhagen, 1995) and collectiveengagement in giving a sense of emerging interpre-tations to relevant stakeholders (Cioia & Chitti-peddi, 1991).

In summary, a review of past literature indicatesthat, with the partial exception of studies of organ-izational crises, collective sensemaking is usually

described as based mainly on conversational prac-tices. Recent research, however, suggests that un-derstanding sensemaking mainly in rhetorical andlinguistic terms may lead one to overlook the im-portant ways in which material practices and arti-facts affect the process, as discussed in the nextparagraph.

Materiality and Conversation inCollective Sensemaking

In management studies, a growing body of re-search on organizational artifacts (see Bechky,2008) has highlighted the symbolic properties ofmaterial items in shaping how individuals makesense of and give sense to an organization (e.g.,Rafaeli & Vilnai-Yavetz, 2004) or their position inits social structure (Pratt & Rafaeli, 2001).

More recent studies, however, suggest that indi-viduals also use various types of artifacts to supportthe construction of new understandings as theyengage in prospective sensemaking. Strategists, forinstance, use PowerPoint presentations (Kaplan,2011) and other visual and textual artifacts (Denis,Langley, & Rouleau, 2006) as they collectively de-fine new courses of action. Drawings, models, andprototypes assist product developers as they eval-uate and refine new ideas (Ewenstein & Whyte,2009; Sutton & Hargadon, 1996) and exchange un-derstandings across professional communities(Bechky, 2003; Carlile, 2002). Finally, the collec-tive production of tridimensional representationsof an organization's strategy (Buergi, Jacobs, &Roos, 2004; Buergi & Roos, 2003; Heracleous &Jacobs, 2008) or identity (Oliver & Roos, 2007) hasbeen shown to facilitate the emergence and thearticulation of tacit assumptions and beliefs.

Findings from these studies resonate with re-search in the sociology of science that shows howscientists use a variety of tools, documents, andinstruments to support scientific inquiry and ar-gues that the "openness" (the interpretive ambigu-ity) of these "epistemic objects" facilitates the col-lective production of new belief structures (Knorr-Cetina, 1981, 1999; Rheinberger, 1997). Researchon the social construction of technology similarlypoints to the role of visual artifacts, such as whiteboards (Suchman, 1988), sketches, and drawings asboth "interactive communication tools" and "indi-vidual thinking tools" (Henderson, 1991: 459).

More generally, research in cognitive psychologysuggests that cognition is "distributed": it does notconsist only of individuals' mental representationsand operations but rather interacts with a materialenvironment "rich in organizing resources"(Hutchins, 1995: 2). Therefore, to fully understand

2012 Stigiiani and Ravasi 1235

cognitive work, one should look at the interactionbetween individuals and the various artifacts thatthey build, use, or surround themselves with. Cen-tral to this perspective is the notion of "cognitiveartifacts" (Norman, 1991: 17)—such as calendars,to-do lists, or computational devices—that facili-tate various mental processes by extending the ca-pacity of the brain to store and process information(Clark, 2008; Clark & Chalmers, 1998),

Taken together, these studies suggest that mate-rial artifacts support cognitive work, at both theindividual and collective levels, as well as the tran-sition ftom one level to the other, and that they maydo so differently ftom discursive artifacts. How-ever, although the influence of material artifactsand practices on cognition is acknowledgedthroughout the social sciences and well docu-mented in management studies, how these artifactsand practices enable individuals and groups to con-struct new understandings is largely missing ftomtheories of collective sensemaking.

METHODS

Our study combined participant observationwith grounded-theory building to investigate prac-tices of collection, production, manipulation, anduse of material artifacts in product design.

Research Setting

Research site. Continuum is an international de-sign consultancy headquartered in Boston (www.continuuminnovation.com). At the time of ourstudy, the Boston office employed over 130 em-ployees, including product designers, engineers,graphic designers, and model makers. Employeeswere organized into three overall groups: Strategy,Product, and Brand, responsible for formulatingguidelines for product design based on consumeranalysis, for executing engineering and product de-sign, and for brand building and communication,respectively. To reduce potential biases associatedwith the specific professional background of groupmembers or an intended project outcome, we stud-ied one project ftom each group: Strategy's ProjectTransport, Product's Project Health, and Brand'sProject Window. Table 1 presents an overview ofthe projects. Only in the case of Health were de-signers asked to produce technical specifications ofa physical object.

Design as sensemaking. The essence of designhas been described as "making sense of things"(Krippendorff, 2006: i). Particularly in the initialstage of concept development, design can be con-sidered as open-ended problem solving, character-

ized by a high level of ambiguity with respect toboth final solution and context (Clark, 1985). Asone of our informants observed, at the beginning ofa project "things are not defined at all": multiplepossible directions may be taken, and multiple in-terpretations about the final solution seem plausi-ble. Accordingly, our informants described thestructure of projects as a "funnel": an iterative pro-cess based on the continuous refinement of emerg-ing ideas.

The design process at Continuum. Most litera-ture on how material artifacts affect collectivesensemaking in product development has focusedon cross-community interactions in the advancedstages of prototyping and testing (e.g., Bechky,2003; Carlile, 2002). Our study, instead, focused onthe "concept development" stage, when designersattempt to produce new understandings of usersand their needs, and of products and their designattributes (see Table 1).

Conceptually, the process can be interpreted asthe gradual construction of new shared under-standings in the form of new "mental models"(Johnson-Laird, 1983): shared mental representa-tions of key elements of groups' tasks and environ-ments (Klimoski & Mohammed, 1994). In an initialphase, designers in each studied group interactedwith clients to build a common understanding ofthe goal oftheir project and, in particular, of how tointerpret the client's "brand pillars" (a commonterm at Continuum and among its clients). A re-search phase followed, in which designers engagedin various activities (collecting objects, interactingwith users, etc.) to inform the reconceptualizationof elements oftheir task (goals, users, uses, etc.). Inthe following phase, designers individually devel-oped new interpretations of design attributes, inlight of mental representations of other elements ofthe task produced earlier. Later, they exchanged,compared, and integrated provisional interpretationsin multiple iterations, unül they converged on whatthey referred to as a "big idea"—a new concept basedon the integration of emerging understandings ofvarious elements of the task.

It could be argued that the inherently visual na-ture of design, as a discipline aimed at the creationof tangible deliverables, and designers' specifictraining in processing information visually ratherthan verbally, might have led our informants tomake a particularly intense use of material artifactsto support sensemaking. We argue, however, thatthese features actually made our setting an "ex-treme case"—an ideal setting in which the phe-nomenon of interest is "transparently observable"(Pettigrew, 1990: 275). Such a setting facilitatesinvestigation of dynamics that characterize, albeit

u

O

•S

-0 'S

• i l

iiÖ O c

CD 03 _D mU do 00 o> 03 ^ Dp

Í W (D ß

; G) T3 ^

"S 'S S •"Ci. CDp3

tí .M _ CO .O CO S =CJ 03

J : ' " 00=^..1^ T3 fi «

g? e S «S.O g go CO . f i '.g T3

T 3 O CO 2 CO sg fi ") s _ 2

« S3 » I »- gS p . . . . CL 5

s J > ", g .2 'SÖ Q) ^ >> (D CO Q3dJ ^ 03 - ^ ' O "M 'TÎ

> -a fid* ^ -s« -a 03 a fi CIH-Î;

3 S

o

I

T3 gS 2

PH m

Vi(-(mtí000H3

H

•S0303fi

203CO

D

S o

CO * ^

S-3a

tí00CO

leva

nt

03

Pi

(0CO03

attr

ibut

ig i

dea

m

T3

I00

e


less visibly, other settings, such as strategy making(e.g., Cioia et al., 1994) or new product develop-ment (e.g., Rafaeli et al, 2006), in which groupscollectively construct new mental models.

Data Collection

Data collection followed common recommen-dations for ethnographic work (e.g., Jorgensen,1989; Spradley, 1980; Van Maanen, 1979) andcombined archival search, participant observa-tion, formal semistructured interviews, and in-formal talks. Table 2 describes our data sourcesand how we used them.

Over ten months, the first author spent from fourto five days per week and from six to eight hours

per day in the field, participating in project meet-ings and conducting formal interviews, but alsoattending social events and having informal andimpromptu conversations. Initially, she tookquick notes at different points during each day toclosely document Continuum's structures andwork practices as she observed them (Emerson,Fretz, & Shaw, 1995; Fetterman, 1998; Lofland,Snow, Anderson, & Lofland, 2006). After a fewweeks in the field, she was granted full access toand participation in the three projects. Duringproject meetings, she would constantly jot fieldnotes in a small pad focusing on the interactionsbetween informants and on the material tracesthat these interactions left (on boards, walls, etc.)and trying to capture as much conversation as

TABLE 2Data Sources and Use

Data Source Type of Data Use in the Analysis

Archival data Company-related documents: Guide fornewcomers, internal presentations of the threegroups, guidelines for project managers, maps ofthe development process.

Project-related documents: Design briefs, meetingminutes, project updates, correspondence withstakeholders, clients' presentations.

Observations Field notes from meeting attendance (79 meetings).(617 pages Detailed record of social interaction,double-spaced) conversations, and use of artifacts observed in

the three projects from the early stages untilfinal presentations.

Informal conversations. Informal talk withmanagers, designers, engineers, and supportstaff, ranging from brief exchanges to longer talksbefore and after meetings and during workbreaks.

Pictures. Visual documentation of material andtextual artifacts produced during work meetings(boards, thumbnails, sketches, etc.).

Interviews Preliminary interviews (17) with all senior and(598 pages middle managers, to investigate Continuum'sdouble-spaced) history, culture, and work processes.

Debriefing interviews (15) with 12 team members(out of 15), to discuss insights or observationsgathered during project meetings; weinterviewed key informants for each projecttwice, at the beginning and at the end of theprojects.

Focused interviews (24) on how the use of materialartifacts supports informants' work; thecomposition of informants (13 for Product, 6 forStrategy, and 5 for Brand) reflected the relativeweight of each group within the company.

Familiarize with the organizational context.

Support the reconstruction of the set ofconcepts produced in each project (Table 1).

Support, integrate, and triangulate evidencefrom observations and interviews.

Produce a map of the material practices and theartifacts that were used or produced asmembers engaged in these practices (Table 3),and link material practices with the outcomeof project work.

"Triangulate" interpretations emerging frominterviews.

Familiarize with the organizational context,gain trust of informants, discuss insights fromobservation, clarify uncertainties regardingproject-related decisions, and supportemerging interpretations.

Keep record of the outcome of practices thatmembers engaged in during the projects (e.g.,group sketching, material assemblage,bucketing, etc.), and share it with the secondauthor.

Familiarize with the organizational context.

Integrate observations with informants'accounts, to improve our understanding ofsocial dynamics and project-related decisions.

Investigate cognitive subprocesses supportedand/or enabled by different material practicesand artifacts (Table 4).


possible verbatim. She would then expand thesenotes every night recounting "what happened" indetail while memories were süU fresh (Emerson et al.,1995: 14), eventually producing 617 double-spacedpage of extended field notes.

In the beginning, design teams would frequentlypeer at her pad and ask questions about her obser-vations. After a few weeks, however, they becameacclimated to her presence, stopped paying atten-tion to what she was "discovering" and writing,and increasingly involved her in project work.Lacking a design background, her participation inthe projects consisted in the performance of mar-ginal roles (e.g., gathering information, taking min-utes, taking pictures). Drawing upon overt involve-ment and not performing the role of a team memberin a strict sense, she managed to balance the roles ofparticipant ("going native") and observer (remain-ing objective) (Spradley, 1980).

Ethnographic observation was supported by 56formal interviews: 17 preliminary interviews withsenior managers on Continuum's history, culture,and work processes, 15 debriefing interviews witbmembers of the project teams, and 24 interviewsfocused on the use of objects, carried out with teammembers. Preliminary and focused interviews fol-lowed different protocols, and debriefing inter-views had a more open format because tbey wereaimed at discussing specific insights or observa-tions gathered during project meetings. Recordingand transcribing all the interviews yielded 598double-spaced pages. Theoretical considerationsled the selection of informants and differed for thethree types of interviews. Preliminary interviewsincluded all senior and middle managers at Con-tinuum. Debriefing interviews aimed at capturingmembers' accounts of project work included 12 outof 15 members of the three project teams. (All mem-bers were contacted, but only 12 of them agreed tobe interviewed.) Key informants for each projectwere interviewed twice, at the beginning and at theend of the projects. Finally, for the interviews fo-cused on artifacts, the first author sat with thecoordinators of the three main company groups(Strategy, Brand, and Product) and together theyidentified a list of members to be contacted for eachgroup, using a mix of seniority, background, andother characteristics as criteria. Most informants(30 out of 45 contacted) agreed to be interviewed.The number of people interviewed for each group(13 for Product, 6 for Strategy, and 5 for Brand)reflected the three groups' relative weights—interms of number of employees—inside the com-pany. Our series of interviews terminated when wefelt we had reached "theoretical saturation" (Glaser& Strauss, 1967).

Data Analysis

Step 1. Tracing individual and group-levelpractices of sensemaking. Dvying the three proj-ects, conversations among members and tentativeverbal articulation of emerging interpretations wereobviously essential to members' moving from earlyindividual ideas to a collectively agreed upon re-conceptualization of users, uses, and product fea-tures. As tbey conversed, however, group membersfrequently used or produced material artifacts tosupport forms of engagement in cognitive work andpatterns of social interaction that informants de-scribed as distinctive practices and labeled accord-ingly ("bucketing," "thumbnailing," etc.). In anearly stage of analysis, consistently with our re-search question, we combined field notes with in-terviews to carefully map the material artifacts thatmembers used and the material practices—under-stood as patterns of behavior involving the collec-tion, production, manipulation, or use of one ormore types of material artifacts—that they engagedin during the projects. Table 3 characterizes thematerial artifacts and presents the labels infor-mants gave tbem.^ As we did so, we also kept traceof the conversational practices they were usuallyassociated with. As conversational practices havealready been amply discussed in previous research,however, for the sake of simplicity, we acknowl-edge them as part of the process in both text andfigures, but we do not present fine-grained obser-vations about them.

Step 2. Tracing cognitive subprocesses of sense-making. In a second step of analysis, we used in-terview data to investigate the cognitive subpro-cesses that, according to informants, materialpractices and artifacts supported and enabled. Fol-lowing past research on sensemaking (e.g., Corley &Gioia, 2004; Maitlis, 2005), data analysis relied oncommon procedures for grounded-theory building(Locke, 2001). We used interview transcripts asprimary data for the analysis, and we used fieldnotes to support and refine the interpretation ofemerging categories and to guide the integration ofcategories into an overall framework.

^ Disentangling conversational practices from materialones could be accomplished in theory, but less so inpractice. Few practices were purely conversational orpurely material. On the one hand, conversational prac-tices were frequently associated with the use of materialartifacts, as team members wrote on bocirds and/or per-sonal notes, used visual imagery to illustrate their inter-pretations, etc. On the other, the production, assembly,or elaboration of material artifacts eventually resulted inverbal articulation.


TABLE 3Material Practices and Artifacts Collected, Used, and Produced during Conceptual Design at Continuum

Aggregate MaterialPractices Micro Material Practices" Artifacts Produced

Material classification

Visual integration

Material memory

Material assemblage. In the research phase, designers usedpictures to produce understandings of fundamental elements ofthe task ("visual library meetings"); they collectively createdboards gathering pictures expressing desired design attributeseither in form of formal features ("image boards") or emotionalresponses ("mood boards"). At times, they collected artifactsrelated to a particular user into special user rooms to helpmembers grasp particular systems of users' meanings.

Bucketing. Designers grouped and regrouped cards and Post-its,searching for patterns that helped them produce a taxonomicalunderstanding of elements of their task (e.g., users, forms ofuser interaction). Later, they "bucketed" thumbnails embodyingearly ideas, to move from initial intuitions to provisionalunderstandings of design attributes.

Thumbnailing. During the creative exploration phase, designersindividually produced a high number of rough and simpledrawings. These "thumbnails" were the very first visualmanifestation of ideas about the design attributes.

Group sketching. During meetings and "creative reviews,"designers would often sketch together on white boards orcombine individual thumbnails into more refined drawingsderivative of the entire group's inputs, but not intended asfinished work

Frameworking. Later in the process, members, individually orcollectively, produced visual representations that tracedconnections (causality, association, sequences, etc.) amongunderstandings of different elements of the task.

Storybuilding. Designers conceived the presentation of the "bigideas" as a story, illustrated through the display of previouslyproduced material [pictures, boards, sketches, and frameworks]organized into a PowerPoint presentation.

Browsing and collecting. During user research, designers perusedbooks, magazines, collections of images, and personalcollections of objects; they leafed through catalogues andlooked at store windows for "inspiration" in task-related andunrelated domains. They took pictures, and videotapedinterviews and users' interactions with objects, and created"user cards"—pieces of cardboard reporting textual and visualinformation—to collect insights, pictures, and excerpts frominterviews.

Brain dumping. Team meetings were regularly held withinproject rooms. During early meetings, designers would pin onthe wall pictures and boards illustrating all the cues that theyhad gathered in the research phase [pictures, objects, cards,boards). Later they regularly added sketches, and frameworksresulting from individual and group work. Often, designersstayed in project rooms even outside meetings, when workingon the project.

Boards and user rooms. Collectionsof visuals and text resulting fromthe research phase.

Boards. Collections of visuals andtext from research or preliminaryreflections

Thumbnails. Small freehanddrawings on paper

Sketches. Relatively detailedfreehand drawings.

Frameworks. Visual representationsof linkages between elements ofthe task.

Slides. Combination of text andimages

Pictures. Visual imagery, snapshotsand films

Cards & Post-itsObjects. Samples and other task-

related objects.

Project rooms. Physical spacescollecting all project-relatedartifacts

The names of artifacts are in italic.

In a preliminary stage, we both engaged in opencoding of interviews, searching for relevant textsegments—phrases and passages that referred tohow and why artifacts were used in the develop-ment process. We initially labeled these segmentswith "in vivo" terms and phrases used by the in-formants (Locke, 2001: 65). Following multiple re-readings of data, we gradually combined in vivo

codes that, although varying in specifics, were sim-ilar in essence, into first-order categories (Locke,2001). At the end of this stage, we compared theseparate coding structures, and we resolved dis-crepancies through discussion and occasional re-coding of data. To ensure the faithfulness of ouremerging account to our informants' vocabularyand interpretation, we labeled most first-order


codes with metaphorical expressions alluding tophysical engagement with abstract cognitive struc-tures ("capturing ideas," "connecting brains," etc.)used by informants to explain how material arti-facts and practices supported cognitive work.

In a further round of coding, we tentatively com-bined first-order categories—describing, in infor-mants' terms, various ways in which material prac-tices and artifacts supported cognitive work—intofewer, broader, and theoretically relevant second-order categories associated with more general cog-nitive processes supported by these practices. Fi-nally, drawing on the content of interviews andfield notes, we associated material practices withcognitive subprocesses. Using this analysis, wegathered material practices into three groups ac-cording to the type of cognitive work that theyenabled. Following past research adopting a similaranalytical approach (e.g. Corley & Gioia, 2004; Har-rison & Corley, 2011), we present the resulting datastructure^ in Figure 1.

^ The dotted boxes and lines indicate parts of the pro-cess not discussed in detail in this article. In particular,the dotted lines at the bottom of the ñgure indicate prac-tices and cognitive processes that marked the transitionbetween intragroup sensemaking and intergroup sense-giving (influence).

Step 3. Building a grounded theoretical frame-work. As core categories emerged ftom the analysisof material practices and cognitive processes, weturned to axial coding (Strauss & Corbin, 1990:123)to uncover relationships among our observations.The observation that different material practiceswere used at different times and that different prac-tices were associated with individual and/or group-level cognitive work led us to organize our emerg-ing interpretations into a multiphase, multilevelprocess model accounting for how material prac-tices interact with conversational ones and supportcognitive work in collective sensemaking. Combin-ing a map of organizational practices resulting ftomethnographic observations with informants' ac-counts of how these practices supported cognitivework, we produced a grounded model of how ma-terial and conversational practices support collec-tive, future-oriented sensemaking efforts. Follow-ing Locke (2001: 76), we tested alternativeconceptual ftameworks until we assembled our cat-egories into an overarching model fitting our evi-dence. To increase the reliability of our overallinterpretation, we routinely submitted provisionalinterpretations—at various stages of the analy-sis—to some informants for feedback. We presentthe interpretative ftamework that emerged in thenext section.

Micro material Practices Aggregate material Practices(First-Order Codes) (Second-Order Codes)

Browsing andcollecting

MATERIALMEMORY ( C U E S )

L H

FIGURE 1Data Structure

Macrophasesoj Sensemaking

NOTICING AND

BRACKETING

Cognitive Processes(Second-Order Codes)

EXTRACTING CUES ITÎOM

FLOW OF EXPERIENCES

Cognitive subprocesses(I'irst-Order Codes)

^ -, .'"Recreating experiences"

Material assemblageBucketing

Brain dumping I i ^

Thumbnailing

Frameworking

Group sketching

Storybuilding

MATERIAL MEMORY

(PROVISIONAL

INTERPRETATIONS)

INTERACTIVE

TALK

VISUAL

INTEGRATION

INFLUENCE

LINKING MATERIAL CUES

AND ABSTRAGT

GATEGORIES

STORING, SHARING, AND

RETRIEVING MENTAL

CONTENT

INTEGRATING AND

REFINING EMERGING

MENTAL STRUGTURES

"Visual referencing""Sorting things out"

'Parking ideas"'Connecting brains"'Getting in the right frameof mind"

'Keeping the breadcrumbs"

"Capturing ideas""Organizing thoughts""Building on each other'sideas"

"Walking the clientthrough"


FINDINGS

Collective sensemaking at Continuum rested onthe interplay between conversational and materialpractices. Informants traced connections betweenthe production and use of material artifacts andspecific cognitive subprocesses that allowed themto gradually combine cues into tentative under-standings of their task and to integrate and refineprovisional interpretations into a more complex setof interrelated mental structures.

In Figure 2 we describe the process as it unfoldedin the three projects we observed. On the left-handside, the figure impacks the material practices thatsupported collective sensemaking and the cogni-tive subprocesses that, according to informants,these practices facilitated. On the right-hand side,the figure shows how these material practices, com-bined with conversational ones, enabled differentstages of the sensemaking process. As the dottedfeedback lines indicate, the process proceededthrough multiple iterations, and there was no exactcorrespondence between the cognitive macro-phases and the levels at which the process un-folded. During the process, members alternated in-dividual work with group meetings, dtiring which

their tentative ideas were shared, debated, dis-carded, or refined. The individual and the collec-tive levels blended into each other, as people artic-ulated individual ideas partly in conversationalpractices with the rest of the group and, conversely,group interaction generated cues that stimulatedfurther individual cognitive work.

In the remainder of this section, we intertwine adetailed narrative of our ethnographic observationswith theoretical insights generated by interviews(Eisenhardt & Graebner, 2007). To provide a com-prehensive overview of the process, we describe allfour phases. However, as the initial phase, "notic-ing and bracketing," and the final phase, "influ-ence," are well known in the literature, we focusour analysis -on the less explored phases, "articu-lating" and "elaborating." Selected quotes support-ing our emerging interpretation are displayed inTable 4.

Phase 1: Noticing and Bracketing

An argument of sensemaking theory is that indi-viduals are constantly immersed in a flow of stim-uli, only few of which are attended to (Weick,

FIGURE 2Micropractices and Macrophases of Collective Sensemaking at Continuum

Micropractices(Material)

Gognitive Subprocesses Aggregate Aggregate MacrophasesMaterial Practices Conversational Practices of Sensemaking

Browsing andcollecting

Material assemblage

Bucketing

Brain dumping

Tbumbnailing

Frameworking

Group sketching

Storybuilding

"Recreating experiences"

"Visual referencing"

"Sorting things out"

"Parking ideas"

"Keeping the bread crumbs"

"Connecting brains"

"Getting in the rightframe of mind"

"Gapturing ideas"

"Organizing thoughts"

"Building on eachother's ideas"

"Walking tbe clientthrough"

MATERIAL MEMORY

(CUES)

MATERIAL

CLASSIFICATION

MATERIAL MEMORY

(PROVISIONAL

INTERPRETATIONS)

NOTICING AND

BRACKETING

CARTICULATING V

. -^r1 VERBAL [>

I ARTICULATION I

VINTERACTIVE

TALKELABORATING

VISUAL INTEGRATION

INFLUENCE


TABLE 4Material Practices and Cognitive Subprocesses: Additional Evidence

Second-Order Codes First-Order Codes Representative Quotes

Linking material cuesand abstractcategories

Visual referencing

Sorting things out

Integrating andrefining emergingmental structin-es

Capturing ideas

Organizingthoughts

Building on eachother's ideas

Walking the clientthrough

Often, the words our clients use do not say much, unless you translate theminto images. That is why we build mood boards. Images help give meaningto words. . . . They link values to the specific object [of the project]. (EG,Product)

The mood board was on the word "inclusive," and it really helped me to seehow differently each person interpret a word and how they visually see theword or translate this word into images. It opens up a lot of opportunity forme to visualize the feeling of inclusion. (CY, Brand)

If we design, for example, a car for teenagers, we would make the projectroom into a teenager's room. We would put posters of Avril Lavigne, painton the walls, or a sofa, just so that you're in the mindset of a teenager.You're no longer a 27 year-old designer: you're an 18 year-old teenager.(BW, Strategy Group)

I lay them out, look at all of them together, and see if there are recurringideas, groups (GF, Product)

Right now we're trying to put everything on little cards or post-its, so you canput them up on the wall. Then we'll move them around, and group them.QS, Strategy)

We have all these drawings and sketches up on a white board, and we look atthem, we move them around, we group them. It's really, really helpful tosort things out. (EY, Brand)

What I try to do first, and this happens in the thumbnails, is, I try to geteverything I have right now in my head out of it. (GY, Brand)

Sometimes I just sketch on paper just as a reminder to myself of somethingthat I thought of; you know, just to capture ideas. (KH, Strategy)

Thumbnails happen all the time, basically to quick capture ideas. . . . They'resmall little ideas that you see and put down. (MA, Product)

When I have these ideas down, and I'm really trying to organize my thoughts,especially when I have a presentation coming up, I'll draw a littlethumbnail of what I anticipate the content is going to be . . . I make a littlerough sketch of what the layout would be." (DV, Product)

Frameworks are a way of thinking, a way of looking at part of what we sawand trying to see similarities and connections that we might not have seenduring the research phase. (PB, Strategy)

If you have three or four people on a project, everybody's sketching,everybody's feeding off each other, you're building off... you might takeone person's idea and then build off those sketches and build other thingsfrom that. (JW, Product)

So, I get feedback on sketches, my own sketching—sketches for me aregenerated from a group giving me input, while thumbnails are typicallygenerated by me just thinking through something. (GB, Brand Group)

This page is an example of a conversation I had with somehody that wherewe're trying to figure out how we're going to build something and I drewsomething, they drew something, I drew something, so we draw on eachother. (RC, Product)

If I consider this as a ladder, then [artifacts] become. . . . This [thumbnail] ispersonal. This [sketch] is more social. . . . It's like a pyramid, as you go onmore and more and more people have reviewed and been involved. (GB,Brand).

So we often show thumbnail sketches step by step. . . . It helps the clients seethe different processes and steps (JS, Product)

It facilitates their understanding of what's going on, of how and why we'vecome to those conclusions and those results. Having pictures and models infront of them helps them to understand the process we went through,what's happened. . . . —It just makes it clearer (GC, Brand)

All the frameworks up there are like pieces of the story. And some of them, inthe end, end up working to tell the story, and some of them are like piecesof another story that we didn't pick. But by just sort of rearranging themyou could support how we developed a whole bunch of ideas. (MAF,Strategy)

Continued


TABLE 4(Continued)

Second-Order Codes First-Order Codes Representative Quotes

Storing, sharing, andretrieving mentalcontent

Connecting brains

Parking ideas

Keeping thebreadcrumbs

Getting in theright frame ofmind

Just putting everything up so you can also see what other people are thinking.(JS, Product)

Being surrounded by it, it's almost like your brain a little bit. . . . It's likepieces and sometimes you just haven't. . . it's sort of serendipity butsometimes you just notice some connections that you wouldn't havenoticed if they weren't all together. (HR, Strategy)

It's very hard to work at your desk by yourself, because it's like designing in avacuum. Project rooms are the anti—vacuum. It's a sharing of ideas. I couldput something up in here that could totally change everybody's attitudetowards something. (GB, Brand)

Whenever I start a project, I always make a parking lot. While I'm doingresearch and an idea occurs to me, I put it in the parking lot, becauseyou're not ready to design yet but you don't want to forget about that thing.(JS, Product)

It's pinned to the wall right over there, and you can immediately make thatconnection between that thing and this other thought that you hadtwo weeks later that is on the wall over there. (KH, Strategy)

It's nice to just have a kind of war room, as I call it, when you just leaveeverything out, and when you need some inspiration you just go downthere and you have everything that you need for the project right in front ofyou. (CC, Product)

[The project room] might make you recall things that you forgot. It mightmake you see something in a different way. It might make you catchsomething you didn't see before. (GB, Brand)

The project room ends up being like a little bit of the timeline of. . . there's alittle bit of chronologicalness to it. (HR, Strategy)

I think it's important to keep like the breadcrumbs so you can find the wayback; it represents the end of a phase. This is where we were, this is whatwe thought, and it captures all our ideas. (GF, Product)

Even though when you write stuff down it's not permanent, at least it's like astake in the ground, it's a marker saying "Remember? We talked about this,and we decided this," and I feel it when we talk about things, and we don'thave those little reminders. (KH, Strategy)

It's important to wrap yourself around what it is you're designing, wrapyourself around images that inspire you, wrap yourself around the currentproducts that are on the market that need improvement. . . and also to bearound each other to exchange those ideas. (DV, Product)

It's nice to have a project room, because, it puts you into that frame of mind,into that experience (GB, Brand)

[Working in a project room] is more like how to almost immerse yourself witheverything that that project is about and nothing else. So, it helps you focusmore into that. (MA, Product)

1995). Senseinaking starts wben tbese stimuli, or"cues," are extracted from the flux of experience("bracketed") for further cognitive work (Weick etal., 2005). Gonsistently with this idea, in the threeprojects we observed, designers initially immersedthemselves in task-related experiences intended tofeed reconceptualization of task elements. They in-terviewed relevant stakeholders and filmed or pho-tographed how they interacted with relevant ob-jects. They collected samples of products and cutbotb product-related and more general images outfrom magazines. All these artifacts represented"embodied experiences" that would be made per-manently available for the interpretive process that

occurred later, as members—first individually,then in groups—produced new provisional inter-pretations of elements of the task, in the form ofemerging mental models.

Phase 2: Articulating

The early sign of members' attempts to bringorder to this flow of experience was the combina-tion of bracketed cues into tentative and ill-definednew understandings of various elements of thetask—a phase we refer to as articulating. In sense-making theory (e.g. Weick, 1995; Weick et al.,2005), articulation generally refers to verbal exprès-


sion of tentative interpretations. In fact, as dis-played in Figure 2, at this stage, members relied onvarious conversational practices that helped themverbally articulate tentative understandings. Thesepractices involved the production and use of longlists of features (e.g., the desired aesthetic featuresof a new hospital chair); metaphors (Cioia et al.,1994) (e.g., the hospital chair as a "blanket," orSprint as a "hero to the rescue of its customers"); ortaxonomical classifications of competitors (see Po-rac, Thomas, Wilson, Paton, & Kanfer, 1995) orproducts (see Rosa, Porac, Runser-Spanjol, &Saxon, 1999).

As they attempted to verbalize new understand-ings of elements of their task, however, group mem-bers also systematically engaged in practices thatinvolved the production and/or manipulation ofmaterial artifacts to support concept formation in anonverbal way. These practices supported the in-fusion of abstract categories with new meanings bylinking them to groupings of material cues. Wecollectively refer to these practices as materialclassification.

When confronting abstract concepts or unfamil-iar objects, members would frequently support ver-balization by assembling visual imagery and previ-ously collected objects (a practice we refer to asmaterial assemblage] or by grouping and regroup-ing cards reporting cues from the research phase(what they referred to as bucketing]. Both thesepractices were characterized by members' physi-cally engaging with cues embodied in visible andtangible form (photographs, cards, other objects) toconstruct new mental categories and imbue themwith meaning.

In the beginning of the project, designers usedso-called image boards or mood boards (see Table3) to build new understandings of elements of thetask, such as brand pillars, users' needs or life-styles, and desired product features. Project mem-bers would browse collections of images, picturesfrom the preliminary phase, or personal files forvisual imagery that they associated analogicallywith the focal concepts and assemble them intolarge posters. Designers eventually combined ma-terial assemblage with lists of features to articulatenew understandings emerging from the researchphase. In an early meeting of Project Health, forinstance, designers used an assemblage of images ofhospital chairs and pictures from their hospital vis-its to elicit a list of words like "discouraging,""depressing," "sterile," and "cumbersome" thatthey associated with the "sense of illness and pain"that, in their view, current products conveyed.

Informants considered the assembled images es-sential to the construction of new understandings.

as these images allowed them to support the verbalarticulation of ambiguous abstract concepts withmultiple "visual references"—concrete cues thattapped into preexisting meaning structures, stimu-lating analogical association. As an informant^observed:

You can make these comparisons or similarities tothings that exist out there already, that people arefamiliar with. And I think it helps. . . . Like if some-one says "rugged," you might not really know whatthey mean, but if you show a picture of a Jeep, thenyou know what they mean. . . . You show a pictureof Mel Gibson. . . . It's all about coming at the wordat every possible angle, so that we get this whole,well-rounded visual description of it. (CC,Brand Group)

Occasionally, in other projects in the firm, al-though not in the three projects we observed, ma-terial assemblages took tbe form of "user rooms,"where members collected various objects that theyassociated with particular users to acquire an un-derstanding of their general needs and lifestyle, orof more abstract concepts such as "girls' friend-ships" or "retirement," and to develop ideasaccordingly.

During each project, group members would alsoengage in tbe tentative grouping and regrouping ofsimple artifacts (Post-its or cards) representing cuesor, in some cases, preliminary ideas, to facilitatethe development of broader categories. For in-stance, during the initial phase of Project Trans-port, members tried to understand what explainedthe relative willingness of target users to accept orreject a minivan as a preferred vehicle. To do that,they used cards representing the mothers they hadinterviewed, including pictures, demographic in-formation (age, number of kids, owned cars), sig-nificant quotes, and interviewers' observations.Grouping these cards on the basis of variables suchas daily schedules, lifestyles, aspirations, and emo-tional needs initially led members to identify threemain groups—"stay at home," "part-time/new ca-reer," and "full-time job"—later reduced to two, asthe team realized that cards did not show signifi-cant differences in self-image, daily needs, and pur-chasing bebavior between the last two groups (seeTables 1 and 4). The same practice applied also totentative visual representations of early ideas em-bodied in tbe "tbumbnails" described in tbe nextsubsection (see Table 3), as illustrated by an indus-trial designer working on Project Health:

^ Initials in parentheses refer to the name of the infor-mant who is quoted.


When you look at all [the thumbnails), you sort ofstart seeing groups, and that's what Mike did there.He grouped them into these groups driven hy thethree concepts of pillows, soft slab, and blanket.(GF, Product Group)

According to informants, the physical manipula-tion of experiences and ideas embodied in tangibleform (pictures, cards. Post-its, thumbnails) helped"sort things out"—that is, organized these experi-ences and ideas on the basis of patterns of differ-ence and similarity. They observed that the possi-bility to "see" their thoughts and ideas right in ftontof them and to physically "move them around"facilitated the detection of commonalities, emer-gence of themes, and their assembly into broadergroups or categories that the informants would laterlabel and define verbally. Informants observed thatusing cards facilitated the intuitive discovery ofpatterns among features that were not necessarilydefined a priori, as comparisons could tap into aricher memory of each encounter. In fact, categori-zation was often revised, as members recalled dif-ferent memories of interviews that would makethem change the initial categorization.

In summary, material assemblage and bucketinghelped members move ftom a preliminary bracket-ing of their experience (as reflected in what theyfilmed, taped, photographed, collected, etc.) to ten-tative new understandings of users, products, andso forth. It did so by linking groupings of materialcues with tentative abstract categories, using thecognitive structures associated with these cues toproduce more general categorizations of elementsof the task. Physical engagement with material ar-tifacts often preceded the verbalization of emergingideas and supported the collective refinement ofearly verbal accounts (or the emergence of newones). Cognitive theory describes the formation ofnew mental models as resulting ftom the sorting ofspecific experiences (objects, events) into moregeneral classes on the basis of perceived similari-ties and differences (Porac & Thomas, 1990). In thisrespect, the embodiment of cues in material formfacilitated the organization of the experiences thatthese artifacts embodied, by allowing members toconsciously and collectively combine visual cuesto imbue ambiguous categories with meanings ("vi-sual referencing") and construct new categoriesftom the detection of similarities and differencesamong cues ("sorting things out"). Only later wouldthey articulate the distinctive features in words. Wereturn to the influence of materiality on cognitionat the individual and group level in the Discussionsection.

Phase 3: Elaborating

As new understandings emerged (articulating),cognitive work increasingly shifted to the gradualintegration of these understandings into more com-plex mental structures linking various elements ofthe task environment (elaborating). Research onshared mental models in teams describes the devel-opment of collective mental structures as the grad-ual assembly of simple components through theforging of links among them (Fiske & Dyer, 1984).In this respect, by "articulating" we refer to thecognitive work that underlies the construction of"components" (manifested, for instance, in the re-conceptualization of task elements), whereas by"elaborating" we refer to the tracing of connectionsamong them. Although conceptually distinct, artic-ulating and elaborating were in fact intertwined, asthe attempt to link emerging understandings of el-ements of a task occasionally triggered the need fora team to revise their interpretations over thecourse of multiple iterations between the twophases of the process.

Elaborating occurred mostly during group meet-ings, as members engaged in interactive talk—ver-bal exchanges of tentative understandings and dis-cussions of possible linkages between them. Evenduring this phase, the conversation was assisted bythe material practices that members engaged in tosupport the sharing, integration, refinement, pres-ervation and recovery of emerging structures.

Visual integration. During the elaborating phase,designers would engage in various material prac-tices—which we refer to as thumbnailing, frame-working, group sketching, and storybuilding—thatrelied on the visualization of tentative linkagesamong emerging mental structures.

As the research phase neared its end, team mem-bers began working individually to produce newinterpretations of design attributes that were con-sistent with the emerging reconceptualization ofusers, use, brand pillars, and so forth. These inter-pretations first manifested themselves as small vi-sual representations that designers referred to as"thumbnails" (see Table 3). They ranged ftom for-mal and functional features to "conceptual andiconic ideas" that represented early attempts to en-vision new interpretations of product features orcommunication th.emes that would address emerg-ing understandings of user needs, brand pillars, etc.In Project Window, for instance, a map of theUnited States with two dots at the opposite coastsand a phone ftamed in a heart in the middle wasused to express the idea of "phone as the extensionof yovir love relationship."


Our informants considered the production ofthumbnails—or thumbnailing—an early step in theelaboration of ill-formed ideas, preceding the at-tempt to give these ideas verbal expression, andused them as "building blocks" to develop morerefined interpretations. As an informant insight-fully put it, thumbnails allowed them to combinecues, which had been "sitting in the back of theirminds," into tentative ideas* and helped them"capture" these ideas before they were "lost" in theflow of thoughts (see Table 4). By bringing ideas"out of the mind," thumbnailing provided design-ers with "physical handles": something tangiblethat could support the conscious examination andelaboration of emerging interpretations. As an in-formant observed: "How can I know what I think, ifI don't see what I think? Or if I don't express whatI think I see?" (CB, Brand Group). By acquiringmaterial form, ill-defined initial insights "became"ideas to be subjected to retrospective assessment.

According to informants, although thumbnailsallowed them to "capture ideas," more complexartifacts helped them "organize thoughts"—that is,draw connections between early ideas and inte-grate them into more complex mental representa-tions. These artifacts involved a broad range ofvisual representations (matrixes, graphs, diagrams,etc.) that informants collectively referred to as"frameworks." During Project Transport, for in-stance, the group produced a framework integratinga classification of consumers' needs with insightsabout possible new product featvu-es. During anearly bucketing session, members had organizedgeneric needs of new mothers resulting from inter-views into five areas ("nurture," "peace of mind,""love/bond," "integration," and "affirmation") andarranged these areas over a pyramidal representa-tion, borrowed from Abraham Maslow's hierarchyof needs. Later, they decided to use the same frame-work to organize a list of over a hundred needsspecifically related to a vehicle, also captured inthe research phase, into five main categories ("Itworks," "I'm safe," "I belong," "I am successful,""It's still me"), and to use these categories to orga-nize "areas of innovation" for the new vehicle,visually connected to the needs by usingcolor codes.

* This observation is consistent with research on cre-ative cognition describing creative outcomes as the resultof the progressive refinement of "pre-inventive struc-tures," defined as incomplete and untested early ideas;mental representations holding some promise of produc-ing novel and useful concepts, to be "generated, regener-ated, and modified throughout the course of creativeexploration" (Finke, Ward, & Smith, 1992).

During a debriefing interview, an informant ad-mitted how important such frameworking had beenfor the project because, at that stage, the group "hada lot of ideas" but "didn't have a way to organizethem in a way that made sense" (HR, StrategyGroup). Both individually and during group dis-cussions, members had attempted to build severalframeworks combining and recombining earlyideas. The process ended when they found a com-bination that gathered the consensus of the entiregroup; or, in other words, the process lasted untilthey achieved a "feeling of order" required forsensemaking to stop (Weick, 1995: 29).

Group sketching was another way to stimulatethe integration of early individual ideas into morerefined interpretations resulting from collective in-teraction (see Tables 3 and 5). Sketches resultedfrom free-hand drawings and/or the physical juxta-position of early ideas embodied in thumbnails. Asan informant remarked:

Sometimes we cut them up and move them arovmd,making a sort of collage. So, it's really important tohave paper print-outs because that is a part of ourprocess, that we combine different ideas into one bycutting them and pasting onto each other. (GY,Brand group)

In Project Window, for instance, sketching ses-sions ended with the walls of the meeting roomcovered with individual sketches and other typesof free-hand drawings, combined and displayed topresent a more elaborated version of the ideas theywere initially intended to convey.

Informants considered group sketching crucial inthe gradual integration of early individual ideasand in the collective refinement of the result, aprocess they referred to as "building on each oth-er's ideas." They perceived the increasing sophis-tication of the artifact as reflecting the gradual in-corporation of different individual ideas andobserved how the physical presence of materialartifacts facilitated the exchange of feedback, byproviding a common visual referent to lead andstructure the discussion (Table 4).

As projects entered a more advanced stage, theconversation gradually shifted toward the arrange-ment of intuitive insights, often based on "gut feel-ings" into a "defensible story"—a practice we referto as storybuilding. In essence, storybuilding con-sisted of collective preparation of a set of slides thatwould be used to present the big idea to the client.Storybuilding, however, was important to bothgroup-level sensemaking and intergroup sensegiv-ing. As sensemaking overlapped with sensegiving,the need to reassure clients about the appropriate-ness, reasonableness, and coherence of emerging


concepts forced members to retrospectively reas-sess the cognitive work leading to final solutions.Consistently with the nature of the sensemakingprocess they were engaged in, team membersseemed to consider plausibility as more importantthan accuracy (Weick, 1995). As an informantobserved:

In the end, there are a lot of good ideas, and I thinkthere's a lot of ways that we could have solved thisproblem. We could have probably picked any ofthose big ideas and done a different versions of it,but to me the idea isn't as important to the client orto me as the whole story of "Why that idea?" (HR,Strategy Group)

While crafting the "story," team members selec-tively used artifacts produced throughout the pro-cess (boards, thumbnails, sketches, etc.) to recon-struct a plausible line of interpretation, among theseveral ideas they had pursued, assessed, cind dis-carded (see Table 4)—a process they referred to as"walking the client through."

In summary, in the elaboration phase, the visualintegration, connection, juxtaposition, or sequenc-ing of representations of tentative understandingsof elements of a task seemed to facilitate cognitivework aimed at forging links between these under-standings at both individual (capturing ideas, orga-nizing thoughts) and group (organizing thoughts,building on each other's ideas) levels. The physicalrearrangement of visual representations of emerg-ing ideas—as manifested in the revision of the se-quence of a set of slides, in the redrawing of aframework, or in the recombination of differentsketches—supported the evaluation of emerging in-terpretations, as members searched for a sense oforder among multiple ill-defined understandingsand provisional connections that would reassurethem of the plausibility of their emerging interpre-tation (walking the client through).

Material memory. Informants observed how theembodiment of cues and ideas in material form alsosupported conversational practices and cognitivework by extending the capacity of members tostore, retrieve, and share mental content. To someextent, practices of material memory supportedcognition in all phases of the process. As men-tioned earlier, in the bracketing phase, collecting,taping, and filming fixed the flow of experienceand made cues permanently available for later cog-nitive work. The relevance of these practices in-creased, however, as the process progressedthrough the articulation and elaboration phases asmembers set up a so-called project room—a prac-tice they referred to as "brain dumping"—and theyused it as a common workspace throughout the

process, to facilitate the sharing of mental contentacross members and its retrieval over time.

At Continuum, every project was assigned to a"project room," where members would collect allproject-related material, gather for project meet-ings, and frequently linger even when working in-dividually. As projects progressed, group memberswould gradually pin things on the walls: pictures,thumbnails, boards, frameworks, and other visualor material artifacts that would help members sharetheir experiences and their thoughts with the rest ofthe team and keep track of the products of collec-tive work. Informants occasionally referred to thispractice as "dumping" the content of their brainsonto the walls of the project rooms (see Table 4).

Individually, brain dumping supported cognitivework by recording early ideas and making themavailable for later cognitive work. During groupdiscussions, members recorded tentative ideas onboards hanging on the walls as soon as these ideaswere verbalized. Informants referred to this func-tion of project rooms as "parking ideas," and theymentioned how being constantly exposed to theseartifacts allowed them to "notice connections thatyou wouldn't have noticed if they weren't all to-gether" (HR) or to identify linkages that they hadnot previously thought of (see Table 4).

Informants observed how the relative opennessof the visual representations pinned on the walls,which were rarely associated with detailed descrip-tions, allowed their reinterpretation over time inlight of evolving cognitive work to produce newtentative elaborations. During Project Transport, forinstance, the big idea eventually revolved aroundthe notion of "It's still me." "Still me/no compro-mises on style" initially emerged from an earlybucketing of a mother's car-related needs. The ideawas written on a Post-it and hung on a board named"Big Ideas." In the following meetings, many otherideas were produced, discussed, and discarded.Members eventually retrieved the notion of "stillme" while talking about the importance of the ve-hicle's preserving its users' identity. They refined"still me" as "It's still me" and added to stylisticpreferences other sides of "me," such as family,social, and professional sides.

"Dumping" everyone's "brain" on the walls alsomade individual experiences and ideas perma-nently accessible to the rest of a group. Projectrooms acted as a shared cognitive space, facilitatingthe trace of connections among individual ideas.Informants mentioned how the ability to "look atother people's thoughts and ideas" allowed them tomore easily and quickly combine early ideas, tak-ing the first step toward the development of a com-


mon interpretation. An informant aptly describedthis function as "connecting brains."

According to informants, exposure to artifactsgathered in a project room also influenced the"frame of mind" with which they approached theirtasks by stimulating rapid and often unconscious"retrieval from memory" of emerging understand-ings. As an informant explained:

You might be attracted to speciñc [artifacts onboards] that are just he kind of sitting there, in yoursubconscious and in your conscious. And you re-spond to it as you sketch. . . . You look at it, it's inyour memory, and you think about it while you'redrawing. (AM, Product Group)

Artifacts included in a project room embodiedthe result of earlier interpretive processes and hadacquired project-specific meanings that membersdrew upon as they engaged in further cogni-tive work.

Finally, informants observed how having alltheir ideas pinned up on the walls helped them"keep the bread crumbs" throughout the process.Designers were often engaged in more than oneproject, and project rooms also served as "stakes inthe ground" supporting the recollection of evolvinggroup interpretations: the cues that triggered reflec-tions, the provisional articulation of new under-standings, and the multiple tentative linkagesamong them (see Table 4). As members engaged instory building, having "dumped their brains" reg-ularly during the project eventually facilitated theretrospective reconstruction of the cognitive pathleading to the selected interpretations.

In summary, the embodiment of experiences andideas in material form supported sensemaking bymaking the mental content they represented perma-nently accessible to team members (connectingbrains), for reuse in prospective (getting in the rightframe of mind, parking ideas) and retrospective(keeping the bread crumbs) cognitive work. By con-veying relationships to mental structures alreadyavailable to team members, embodied cues andideas facilitated the retrieval and use of these struc-tures in the interpretive process (Gioia, 1986).

Phase 4: Influence

Tentative connections between provisional un-derstandings encouraged furtber revisions and, oc-casionally, exposure to new experiences until thegroups felt confident enough in their emerging in-terpretations to present them to their clients. In thephase we refer to as infiuence, group-level sense-making eventually blended into intergroup sense-

giving, as group members met with clients to gainapproval on the emerging ideas.

In this final phase, sensemaking and sensegivingwere inextricably linked. On the one hand, the urgeto provide a convincing account pushed teammembers to refine and revise their interpretationsseveral times. On the other hand, consistently withmodels of future-oriented sensemaking that de-scribe the process as an iterative cycle betweensensemaking and sensegiving (Gioia & Chittipeddi,1991), as members engaged in sensegiving, theyoccasionally collected feedback that promptedthem to reconsider their emerging interpretations,triggering a new phase of elaboration (and possiblyarticulation and bracketing).

A GROUNDED MODEL OF COLLECTIVEFUTURE-ORIENTED SENSEMAKING

Before discussing the theoretical implications ofour observations, let us recapitulate the cognitivework involved in the transition from the individualto the collective level of analysis in the future-oriented, collective sensemaking process we ob-served. The process began with the purposefulexposure of team members to experiences (inter-views, field observations, casual browsing in mag-azines, etc.) that were expected to feed the con-structions of new understandings of elements of atask—still largely undetermined at this stage. Teammembers deliberately attempted to record relevant"chunks of experience" [noticing and bracketing)in material form, such as pictures, images, objects,to share them with the rest of the team and makethem permanently available for cognitive work.Team members would occasionally carry out someof these activities (e.g., filming or interviewing) insmall groups, but the cognitive process wouldmainly unfold separately for each individual, asmembers preselected relevant sets of cues and/orintuitively began to (retrospectively) loosely relatesome of them to one another.

In the second phase [articulating), team membersattempted to more consciously organize these em-bodied experiences into emerging new understand-ings. They used material cues to retrospectively tapinto individual experiences and bring them to bearon the process. Linking abstract categories withassemblages of material cues belped members fillthese categories with meanings resulting from per-ceived overlaps, similarities, and differences in theconcrete experiences embodied in these cues. Thearticulating phase was partly carried out individu-ally and partly in group. Members used prelimi-nary assemblages to tentatively organize cues intointuitive ill-defined concepts, which were made


available to the rest of their group through theassembled artifacts. The refinement of these arti-facts—and the mental structures they embodied—occurred as members jointly engaged in the manip-ulation and rearrangement of assembled cuesthrough successive iterations. At this stage, verbalarticulation and material assemblage were inter-twined. Individuals used combinations of wordsand images to express their tentative understand-ings, by producing retrospective verbal descrip-tions of intuitively assembled cues. The gradualconvergence of group members occurred as groupsmerged, enriched, pruned, and rearranged prelim-inary assemblages, until they produced a satisfac-tory order in the organization of material cues (andthe meanings that these cues evoked).

In the third phase [elaborating], provisional un-derstandings of elements of the task were tenta-tively integrated to produce a more complex inter-pretation of a possible, prospective redefinition ofthe relationship between objects, producers, andusers. This interpretation was conceived in narra-tive form, and in this respect it can be consideredan emerging collective account (Maitlis, 2005). Theconstruction of this account largely occurred asgroups engaged in interactive talk. Even at thisstage, however, material and conversational prac-tices were intertwined, as the embodiment of indi-vidual tentative connections in material form(rough thumbnails, early versions of frameworks,etc.)—a practice we refer to as visual integration—facilitated exchange of provisional understandingsand collective engagement in the assessment,merger, and refinement of these understandings.The process continued until the group felt that theyhad produced a plausible interpretation of theirtask; that is, they had made retrospective order outof the massive amount of cues that they had gath-ered at the beginning of the project. Embodiment ofcues and emerging ideas in material form [materialmemory] also facilitated their storage, retrieval, ex-change, and integration over time and amongmembers.

In the final phase [influence], sensemaking wasreplaced by sensegiving, as the groups attempted topersuade clients about their "preferred interpreta-tion" (Cioia & Chittipeddi, 1991) of the task. Fu-ture-oriented accounts were outlined through a ret-rospective reconstruction of underlying cognitiveprocess and the disclosure of cues and fragments ofinterpretations produced along the way. At thisstage, clients occasionally challenged a group's in-terpretation, pushing members to reconsider theiremerging interpretations and engage in a newround of elaboration and articulation (as expressedby the dotted feedback lines in Figure 3).

DISCUSSION AND CONCLUSIONS

Our study of designers' work helped us producea fine-grained account of the cognitive processes aswell as the social and material practices involvedin the gradual organization of individual experi-ences and in the group-level integration of ill-de-fined early ideas into new and more refined sharedmental structures that occurs when people engagein futuje-oriented collective sensemaking. By un-packing the social practices that the process restsupon, our study improves understanding of theinfrastructure that supports social interaction (Jar-zabkowski, 2003: 24) and the resources (Jarzab-kowski et al., 2007: 9) available to individuals astbey engage in prospective sensemaking.

For the sake of simplicity. Figure 3 portrays thesensemaking process we observed as a linear se-quence of four macrophases (noticing and bracket-ing, articulation, elaboration, and influence) un-folding at three different levels (individual, group,and intergroup). However, as illustrated in the pre-vious section, the process is likely to proceed inmultiple iterations both across levels (as indicatedby the overlap between phases across levels) andacross phases (as indicated by feedback dot-ted lines).

In the remainder of this section, we discuss theimplications of our observations for the theory andpractice of collective sensemaking.

Extending Theories of Fnture-OrientedSensemaking

Our study suggests how extant models provideonly a partial account of prospective sensemaking,as they tend to overlook the processes that underliethe mindful collective construction and refinementof new interpretations. By doing so, our study an-swers a call for "an expansion to the domain ofsensemaking to include both prospective and ret-rospective elements" (Cioia & Mehra, 1996: 1230).Current representations of the cognitive underpin-nings of sensemaking (e.g., noticing, bracketing, la-beling, and other named subprocesses) largely re-flect the traditional retrospective form (e.g. Weick,1995; Weick et al., 2005). Weickian theories ofsensemaking describe interpretation as the tenta-tive "attachment of meaning" to a cue (Weick,1995). The novelty or ambiguity of tbe circum-stances that trigger sensemaking would suggest thatavailable mental structures are not appropriate toexplain tbe situation those circumstances com-prise: new structures need to be constructed or newlinkages need to be traced among available struc-tures. Our model begins to unpack this fundamen-


FIGURE 3A Process Model of Collective Future-Oriented Sensemaking

Noticing and Bracketing

tal phase in prospective sensemaking, by shiftingattention to the articulation and the elaboration ofnew mental structures (as opposed to the labelingof unfamiliar events based on currently availableones) that occur between the individual bracketingof experience and tentative collective action.

Weickian theories of sensemaking tend to skiparticulating and elaborating, compressing them be-tween tentative labeling and action. During crises,time pressure may hamper verbal articulation andthe elaboration of provisional interpretations. Pre-cise agreement is not required for action, and peo-ple eventually reach consensus by adjusting toevolving events (Weick & Roberts, 1993). Duringfuture-oriented sensemaking, instead, more relaxedtime pressure provides the opportunity for the pro-longed and conscious articulation and elaborationof tentative interpretations. Past research, however,provides only a partial and incomplete accovmt ofthese phases. Hill and Levenhagen (1995) focusedon individual cognitive work, proposing how met-aphors support the articulation of future-orientedmental representations. Gioia and Ghittipeddi(1991), instead, highlighted the iterative cycle ofsensemaking and sensegiving that occurs as actorsnegotiate new interpretations with other actors, but

these authors remain silent about how these inter-pretations come to be in the first place. Our modelfills the gap between these theories, by outliningcognitive processes and social practices that under-pin transition ftom individual cognitive work tointergroup cycles of sensemaking and sensegiving,providing a more comprehensive representation ofcollective future-oriented sensemaking.

The lack of fully fledged theoretical accounts ofprospective sensemaking can be ascribed to thedifficulty of questioning the fundamentally' retro-spective nature of interpretations. Building onSchutz (1967), Weick (1979, 1995) convincinglyargued that people can only assign meanings toevents that have already transpired. Accordingly,prospective cognitive work happens as individualsengage in "future perfect thinking" (Weick, 1979:199)—that is, as they "envision a desired or ex-pected future event and then act as if that event hadalready transpired" (Gioia, Corley, & Fabbri, 2002:623). Accounts of prospective sensemaking havedocumented instances of this approach in the pro-jection of aspirational organizational images, aimedat aligning members' beliefs with the ambitions oforganizational leaders (Gioia & Thomas, 1996).Thinking in future perfect tense, however, seems


less appropriate to explain prospective cognitivework when expectations or aspirations about thefuture are ambiguous or unclear, as often occurs inproduct development or strategy making. We be-lieve that our model reconciles the forward-lookingnature of the collective effort with the retrospectivenature of its individual cognitive underpinnings,by describing prospective collective sensemakingas based on three interrelated cycles of retrospec-tive cognitive work.

As outlined in Figure 3, the first cycle of retro-spection occurs as group members deliberately im-merse themselves in task-related experiences(rather than casually being exposed to them), toproduce novel understandings of the environment[noticing and bracketing]. New retrospective inter-pretations begin to arise as members go back andforth between the tentative organization of selectedcues and the refinement of corresponding labels[articulation], eventually producing new under-standings of elements of their task environment.

A second cycle is associated to the gradual estab-lishment of linkages among emerging understand-ings [elaboration]. Members produce and reflectupon multiple and incomplete new mental struc-tures in parallel, gradually refining them into morecomplex provisional interpretations of their task.The expression of tentative linkages in verbal ormaterial form allows members to bracket their flowof thought and to make these provisional interpre-tations available for retrospective assessment.

Finally, the need to give sense of emerging inter-pretations to external actors [influence] inducesgroup members to engage in retrospective reflec-tion to establish the plausibility of prospective ac-counts by reconstructing the chain of thought lead-ing to them. It is in this phase that the multiplepossibilities embodied in the various incompleteand emerging structvires are eventually brought toclosure and linked in one set of recommendationsfor a new course of action.

The Materialization of Cognitive Work

Current conceptualizations of sensemaking em-phasize the centrality of language, talk, and com-munication in the development of individual andcollective interpretations. According to Weick,"Sense is generated by words that are combinedinto the sentences of conversation to convey some-thing about our ongoing experience" (1995: 106).Our study extends and enriches the prevailing por-trayal of sensemaking as a preeminently linguisticactivity by drawing attention to how material prac-tices influence individual and group-level cogni-tive work.

Our observations resonate with recent work incognitive psychology maintaining that the manip-ulation of their physical environment allows indi-viduals to amplify their cognitive capacity(Hutchins, 1995). In line with these ideas, insightsfrom our study suggest that the physicality of ma-terial artifacts extends the capacity of individualsand groups to process mental content, as it allowsthese artifacts to serve as external repositories ofmental structures and to support more consciousengagement in cognitive operations (storage andretrieval of mental structures, categorization of ex-perience, integration of mental structures) that areusually carried out below the threshold ofconsciousness.

Our informants repeatedly alluded to this phe-nomenon—which we refer to as the materializationof cognitive work—as they used several metaphors("capturing" or "parking" ideas, "organizingthoughts," "dumping" and "connecting" brains,etc.) that pointed to how material artifacts allowedthem to physically and consciously engage with themental structures that they embodied.

Material artifacts and individual sensemaking.Making sense of events requires "stepping outsideone's lived experience and analyzing it retrospec-tively" (Gioia, 1986: 61) (see also Weick, 1995: 26).In this respect, by acquiring material form, experi-ences and ideas become "separated" from individ-uals' minds and "available to more conscious pro-cessing." As is implicit in Weick's observationabout how one cannot really know what one thinks,until one sees what one says (Weick, 1995), verbalutterances perform a similar function. Unlike talk,however, material artifacts are intrinsically dvu-able(Pratt & Rafaeli, 2006). Although subject to naturaldecay, they constitute a relatively permanent em-bodiment of experiences and ideas and, as such,they facilitate team members' prolonged retrospec-tive reflection and organization of these experi-ences and ideas (Buergi et al., 2003). Extracted fromthe flow of experiences or "captured" from the flowof thoughts and embodied in material form, cuesand ideas become "sensemaking resources"(Gephart, 1993) team members can store, retrieve,manipulate, and—as discussed later—share andcombine more easily with other team members.

The benefits of the conscious engagement withthe production, manipulation or use of materialartifacts are partly explained by the fact that the"externalization" of cognitive work partly compen-sates for inherent limits in the capacity of a humanbrain to store, retrieve, and process mental content(Clark & Chalmers, 1998; Larkin & Simon, 1987).During meetings, for instance, material artifactsserved as repositories of less refined, task-specific


mental content to support relatively short-termmnemonic processes underpinning the construc-tion of new mental structures. Practices of materialmemory made bracketed cues (and the experiencesthat they are directly or indirectly associated with)and tentative fragments of interpretations perma-nently available and salient for cognitive work in-volved in the categorization of experience (articu-lating) or in the construction of linkages betweenemerging mental structures (elaborating). By doingso, they helped compensate for the restrictionposed by attentive memory to designers' capacity toprocess all the information regarding their task(Kolko, 2010). In the long term, as the frequent useof expressions such as "parking ideas" and "keep-ing the breadcrumbs" suggest, images, boards,sketches, frameworks pinned on the walls con-stantly exposed members to reminders of earliercognitive work, facilitating the (often unconscious)retrieval and re-use of the cues and ideas that theyembodied. Members could then more easily bringthese ideas "back into the conversation" and effec-tively integrate them in the emerging interpreta-tions, as discussed later.

Obviously, practices of material memory couldonly reproduce part of the original experience oridea. Similarly, visual imagery alluded to experi-ences of engagements with objects, places or people(and therefore evoked meanings) that could be inpart personal and unique. In this respect, as high-lighted in the findings, the collective engagementin practices of material assemblage or grouping ofindividual experiences facilitated the emergence ofpossibly different recollections or interpretations.Nevertheless, as discussed later, these differenceswere reconciled as the "visual order" embodied inemerging boards, sketches, groupings, etc., re-flected gradual convergence around a "conceptualorder" in collective interpretations (Henderson,1991; Suchman, 1988).

Prolonged visual and material engagement alsosupported cognitive work by taking advantage ofperceptual inferences associated with the visualprocessing of information, such as the detection ofconceptual linkages between pieces of informationthat closely located in visual representations (Lar-kin & Simon, 1987). Similarly to how, as describedby Clark and Chalmers (1998), Scrabble tiles areused to facilitate the organization of unrelated let-ters into a word, the embodiment of experiencesand ideas in material artifacts (pictures, cards,slides, etc.) facilitated the classification of theseexperiences and ideas into emerging mental cate-gories or their arrangement into sequences, by al-lowing designers to move them around, groupthem, rearrange them, etc. In the same way dia-

grams support scientific discoveries by spatiallygrouping pieces of information to be used together(Cheng & Simon, 1995), the physical juxtapositionof visual or material cues facilitated the perceptionof similarities or causal linkages among the chunksof experiences or ideas that these cues embodied.As outlined in Figure 3, the visualization of provi-sional mental structures through practices of mate-rial assemblage and visual integration supportedthe ongoing articulation and elaboration of emerg-ing interpretations—as manifested in the revisionof assemblages or sequences of slides, in the re-drawing of a framework, or in the recombination ofdifferent sketches—and facilitated the retrospec-tive assessment of the coherence and plausibility ofthe tentative groupings, linkages, and sequencesunderpinning the emerging conceptual order.

These observations are consistent with researchin both strategy and design studies. Buergi andRoos (2003), for instance, observed that the use ofthree-dimensional metaphorical representation ofan organization's strategy allows for a "physicalexperience" of the relatedness of different conceptsor different viewpoints. Similarly, research on de-sign traces a link between physical engagementwith embodied experiences and ideas and a searchfor relationships among these experiences andideas, as the following description of designers'work suggests:

The designer begins to move content around, phys-ically placing items that are related next to eachother.. . . This process is less about finding "right"relationships and more about finding "good" rela-tionships. . . . Labeling makes obvious the meaningthat has been created through the process of organ-ization. (Kolko, 2010: 19)

Informants pointed to this process by using met-aphors such as "capturing ideas" or "organizingthoughts" that alluded to team members' physicalengagement with mental content. They referred tothese artifacts as their "visual support" or "physicalhandle," suggesting how their materiality was es-sential to sustain tbe construction of provisionalinterpretations. In fact, some informants reportedhow at times ideas emerged through relatively freeand nonpurposeful manipulation of matter (draw-ing, molding, assembling images, etc.), which gen-erated cues that were captvired and then gave rise toearly "fragments of interpretation" to be refined at alater time.

These observations point to a relationship be-tween materiality and cognition qualitatively dif-ferent from current understandings in sensemakingtheory. Sensemaking theory maintains that inter-pretation occurs as the abstract (a preexisting men-


tal structure) is linked to a material stimulus (a cue)to produce meaning (Weick, 1995: 110). Weick andcolleagues (2005) refer to this act as "labeling." Asmentioned earlier, however, this conceptualizationhardly explains the construction of new mentalcategories. In the teams we investigated, it was themanipulation of the material that supported theproduction of new abstract concepts and their in-fusion with meaning, by enabling the transitionbetween (empty) abstract categories to concrete ex-periences, and then again from concrete to abstract.In some cases, tentative labels predated interpre-tive work; in otbers, new labels arose from theorganization of cues and ideas. In both cases, theinfusion of labels with meanings occurred as teammembers gradually outlined conceptual boundar-ies for these abstract categories—that is, they de-fined attributes that they perceived to be associatedto the category (Mervis & Rosch, 1981)—throughthe conscious organization and manipulation ofconcrete cues, based on perceived patterns of dif-ference and similarity.

Materiality and tbe transition from individualto collective sensemaking. Current understand-ings of collective sensemaking place emphasis onthe multiple sensegiving acts the process restsupon (e.g., Maitlis, 2005), and the "discursive prac-tices" that underpin them (e.g., Balogun & Johnson,2004; Sonensheim, 2010). The diffusion of newinterpretations tends to be explained in terms of theability of some actors to produce persuasive ac-counts (Maitlis & Lawrence, 2007; Rouleau & Balo-gun, 2011). Consistently with this notion, materialartifacts are generally considered to be nonverbalsymbolic resources (Balogun & Johnson, 2005; Rou-leau, 2005) that individuals use to influence theinterpretations of others. Our observations extendthis notion by showing how material artifacts sup-port practices of collaborative construction of newinterpretations and enable members to "make sensetogether" (rather than, or in addition to, "givingsense to one another"). By doing so, we extendscholarly understanding of how distributed indi-vidual attempts at sensemaking eventually result inthe collective construction of new interpretationsin future-oriented, collective sensemaking.

Implicit in current understandings of collectivesensemaking is the notion that experiences andideas are shared in the form of accounts (Maitlis,2005) through which different actors propose ten-tative interpretations of ambiguous occurrences. Inthe absence of material records, however, the influ-ence of these accounts on collective cognitive workis tied to perceptual and mnemonic processes thatmay cause people to filter out or forget parts ofverbal exchanges they do not perceive as relevant at

the time they are encountered. Collective sense-making, then, may really occur as individuals re-late limited and imperfect recollections of otherindividuals' accounts.

As discussed earlier, instead, embodiment ofcues and ideas in material form helped compensatefor limits to individual memory and the fleetingnature of conversation. As the notions of "parkingideas" and "connecting brains" suggest, informantsconsidered the material embodiment of experi-ences and ideas important to make these experi-ences and ideas permanently accessible and reus-able, directly and independently from the initialaccounts they were associated with. Group mem-bers could then more easily pick up others' expe-riences and ideas and link them to their own—possibly reinterpreting them in light of currentcognitive work. Pratt and Rafaeli (2006: 286) ob-served how materiality facilitates "asynchronous"interaction between the producer and the user of anartifact. In this respect, the "materialization" of ex-periences and ideas supported temporally distrib-uted, multilevel cognitive work, by enabling mem-bers to escape the current flow of the conversationand to freely recover and recombine fragments ofearlier individual accounts into the emerging col-lective narrative.

Research on "boundary objects" shows how ma-terial artifacts act as "tangible explanations" facili-tating the transfer of understandings across differ-ent communities (Bechky, 2003; Garlile, 2002).Insights from our study extend this notion by point-ing to the role of material artifacts as "interactivetools" supporting the transition from individual tocollective sensemaking by facilitating the emer-gence and the resolution of "representational gaps"among team members (Cronin & Weingart, 2007)and their collective convergence around newinterpretations.

In the three projects we observed, pooling indi-vidual mental content and making it physicallyavailable to the rest of their team enabled membersto collectively engage in the tentative organization,integration, and refinement of this "raw material"(through material assembly, bucketing, groupsketching, etc.). The provisional artifacts created bydesigners—tentative boards, temporary groupingsof cards, intermediate versions of sets of slides,etc.—served as evolving material representations ofemerging collective interpretations. They repre-sented the group's "negotiated ideas" and, by serv-ing as "receptacles for knowledge created and ad-justed through group interaction" (Henderson,1991: 458), they helped organize socially distrib-uted cognition within the teams.


According to informants, these artifacts provided"common reference" for collective conversationsand, by doing so, helped bring out potential incon-sistencies among members' provisional under-standings of relevant concepts and of tbe emergingrelationships among them (Heracleous & Jacobs,2008). Materiality per se did not entirely removethe ambiguity inherent in verbal exchanges(Bechky, 2003), yet the visualization of provisionalinterpretations anchored the conversation aroundelements and relationships outlined in the evolvingrepresentations. Boards, frameworks, and slidesserved as "shared interactional spaces" (Suchman,1988) in which the search for a conceptual orderprogressed in iterations of verbal exchanges andalterations of visual representations (as outlined inFigure 3).

As the recurrent expression "building on eachother's ideas" suggests, informants considered thepossibility of collectively producing or reconfigur-ing material artifacts important for engaging in cog-nitive work as a group, rather than individually.The visualization of the tentative structures thatevolved as people collectively regrouped cards, re-arranged slides, or worked on each other's sketchesmade the tentative processing and generation ofmental content by each individual more visible tothe rest of his/her team, who could then more eas-ily participate in the use, organization, and refine-ment of this content. Tentative additions fromother members could be directly incorporated inemerging representations (boards, sketches, etc.),and these representations often preserved a visibletrace of these tentative additions (or deletions) overtime. This observation is consistent with past eth-nographic accounts of engineers' work, pointing tothe coordinating role of visual representations as"interactive tools":

In their early draft stages, drawings are used bydesigners as an interactive tool, in that they may bealtered or corrected by someone other than the per-son who drew them. . . . As an interactive tool,sketches are the most direct way for an engineer tohelp form a concept in the mind of a colleague bygiving form to concepts pictured in her or his ownmind. (Henderson, 1991: 459)

By providing a common referent that all mem-bers could relate to and engage with at the sametime, then, a material embodiment enabled thesharing of cognitive work (categorization of experi-ence, construction of new mental models, etc.) thatwould otherwise occur individually (and, as dis-cussed earlier, less consciously), facilitating thetransition from individual to group-levelsensemaking.

Transferability of Insights to Other Settings

Although some of the material practices we de-scribed in the previous section (e.g., thumbnailing,sketching, bucketing) are typical of the professionalpractice of designers, we believe that our insightsabout the sensemaking process and the materializa-tion of cognitive work can be transferred (Lincoln &Cuba, 1985) from our empirical setting to similarcontexts in which individuals and groups engage inprospective sensemaking. In essence, design isabout making new sense of an object, its potentialuses, and contexts of use, and "forging connec-tions" between these elements (Kolko, 2010: 22).Central to design, then, is a process of "meaningmaking," manifested in the production of newmental models (Kazmierczak, 2003; Krippendorff,2006). Consistently with this notion, the outcomeof the projects we observed was not expressed interms of formal and technical specifications for aphysical object, but as a set of interrelated mentalstructiu-es proposing a new conceptualization ofproducts and consumers. The interpretive work un-derlying these projects, therefore, was not unlikeother forms of prospective sensemaking, such asentrepreneurship (Cornelissen & Clarke, 2010; Hill& Levenhagen, 1995) and strategy making (Cioia &Chittipeddi, 1991), in which individuals or groupsdevelop new mental models (of the market, a prod-uct, an organization) and attempt to convince rele-vant stakeholders to accept them.

In this respect, our observations appear corrobo-rated by researcb on strategy making and organiza-tional change describing similar processes in theaccomplishment of less "visually oriented" tasks.The use of Lego bricks in organizational develop-ment programs, for instance, can be considered as aform of material assemblage aimed at collectivelyconstructing new understandings of organizationalstrategy (Buergi & Roos, 2003; Heracleous & Jacobs,2008) or identity (Oliver & Roos, 2007). Kaplan'sstudy of PowerPoint presentations shows howslides-in-the-making serve as a form of materialmemory (in which to "park" individual ideas as theprocess unfolds) and visual integration (to facilitatethe exchange and merger of ideas) in strategy-mak-ing teams (Kaplan, 2011). Also, research on cogni-tive maps suggests how visualizing managers' un-derstandings of the relationships among elementsof their strategy and competitive environmenthelps them reach convergence around collectivecognitive structures (Langfield-Smith, 1992) andevaluate the coherence of the cognitive structuresthat these maps represent (Fiol & Huff, 1992;Huff, 1990).


In a different setting, Simon and colleagues(Cheng & Simon, 1995; Larkin & Simon, 1997) ob-served how the production and use of diagrams (apractice of visual representation and integration ofdifferent pieces of information) facilitates scientificdiscovery. Similarly Knorr-Cetina's research onlaboratory work in various fields of the naturalscience suggests how various forms of visual rep-resentation produced in the course of experimentalresearch support collective interaction in the pro-duction of new scientific knowledge (Knorr-Cetina,1981, 1999).

Collectively, these studies suggest that our in-sights may be transferable to other settings, settingsin which less visually oriented persons (e.g., strat-egists, scientists) engage in less visually orientedtasks (e.g., strategic planning, scientific research).Our emerging ftamework brings together these dis-persed observations by providing a unifying theo-retical account of how material and conversationalpractices support cognitive work and facilitate thetransition ftom individual to group-level prospec-tive sensemaking.

Implications for Future Research

We expect comparative replication of our analy-sis in more traditional (e.g., strategy making) aswell as less conventional (e.g., criminal investiga-tors, medical teams) settings to increase under-standing of how different contextual conditionsmay lead to different patterns of interaction and useof artifacts. At Continuum, for instance, it is possi-ble that the apparent absence of diverging interestswithin a group influenced the relative prominenceof practices of intragroup understanding (sense-making) rather than intragroup influence (sensegiv-ing). All members shared the same concern withpresenting a story that their client would find cred-ible and plausible. As sensemaking eventuallyblended into sensegiving, we could see artifactssuch as sketches and ftameworks, previously em-ployed as epistemic objects to support understand-ing, being used as discursive resources to supportpersuasion and influence. Future research maypurposefully select settings characterized by intrin-sic divergence of interests among group members toinvestigate in more depth the interplay betweenmaterial practices and political processes.

Future studies may also build on our insights andinvestigate more systematically the extent to whichengaging in material and conversational practicesand/or the types of artifacts used affect the qualityof the process. Researchers interested in more mi-crolevel processes may try to isolate the influenceof materiality on sensemaking using experimental

research to compare the quality of the processand/or the outcome of tasks carried out by groupsusing different combinations of material and/orconversational artifacts. Alternatively, researchersmore interested in the influence of collective prac-tices on later stages of the process may search fornaturally occurring experiments allowing them tocompare the organizational outcome of product de-velopment or strategy-making initiatives makinguse of different combinations of material and dis-cursive practices and artifacts.

Regardless of the design that researchers adopt,the fact that sensemaking is more about "plausibil-ity" than "accuracy" (Weick, 1995) requires them topay particular attention to defining a proper mea-sure of quality. Insights ftom our study suggest thatfuture research might focus on either the relative"efficiency" or "effectiveness" of the process. In thefirst case, researchers may focus on measures suchas the number of ideas elicited and/or how quicklythey are processed, or the rapidity with which con-sensus (on meanings) is reached in a group. In thesecond case, researchers may focus on later stagesof the process and look at how the accounts pro-duced with or without the use of material artifactsproduce a more persuasive story (securing the con-sensus of stakeholders in the sensegiving phase) ora more actionable one (facilitating implementationand producing consequences that are coherent withthe initial goals).

Implications for Management Practice

In conclusion, our findings cast new light on thegrowing debate on the application of design meth-ods and tools to managerial processes (e.g., busi-ness innovation, strategy making) (e.g. Boland &Collopy, 2004; Brown, 2008; Martin, 2009). Advo-cates of "design thinking" argue that managersshould learn to "think like a designer" (Brown,2008: 85) and emphasize designers' mental pro-cesses (Martin, 2009), learning styles (Beckman &Berry, 2007), attitudes and dispositions (Brown,2008), and intellectual skills (Clark & Smith, 2008).Our findings suggest instead that design thinkingcannot be decoupled ftom "design practicing." En-couraging managers to embrace abductive think-ing (Dunne & Martin, 2006) or take a systemsview (Brown, 2008), may indeed produce limitedresults if managers are not trained in the materialpractices that complement and substantiate thesedifferent approaches to problem solving.

Taking design thinking seriously, then, meanstraining managers to complement traditional ana-lytical tools, largely based on the elaboration ofquantitative information along predetermined cat-


egories, with practices of material classificationand visual integration that stimulate open recon-ceptualization of products, clients, and competi-tors, and their integration into new strategies. Ourfindings encourage managers to consider these ac-tivities as important epistemic practices that takeadvantage of distinctive properties of material arti-facts to facilitate the collective reconsiderations ofthe fundamental assumptions driving strategymaking.

REFERENCES

Amann, K., & Knorr-Getina, K. 1988. The fixation ofvisual evidence. Human Studies, 11: 133-169.

Balogun, J., & )ohnson, G. 2004. Organizational restruc-tiuing and middle manager sensemaking. Academyof Management Journal, 47: 523-549.

Balogiui, J., & lohnson, G. 2005. From intended strategiesto unintended outcomes: The impact of change .re-cipient sensemaking. Organization Studies, 26:1573-1601.

Barr, P. S. 1998. Adapting to unfamilicU' environmentalevents: A look at the evolution of interpretation andits role in strategic change. Organization Science, 9:644-669.

Bechky, B. A. 2003. Sharing meaning across occupationalcommunities: The transformation of understandingon a production floor. Organization Science, 14:312-330.

Bechky, B. A. 2008. Analyzing artifacts: Material meth-ods for understanding identity, status, and knowl-edge in organizational life. In D. Barry & H. Hansen(Eds.), The Sage handbook of the new and emerg-ing in management and organization: 98-109.London: Sage.

Beckman, S. L., & Berry, M. 2007. Innovation as a learn-ing process: Embedding design thinking. CaliforniaManagement Review, 50(1): 25-56.

Boland, R. J., & Gollopy, F. 2004. Managing as design-ing. Stanford, GA: Stanford University Press.

Brown, T. 2008. Design thinking. Harvard Business Re-view, 86(6): 84-92, 141.

Buergi, P., lacobs, G. D., & Roos, ). 2004. From metaphorto practice in the crafting of strategy. Journal ofManagement Inquiry, 14: 78-94.

Buergi, P., & Roos, J. 2003. Images of strategy. EuropeanManagement Journal, 21: 69-78.

Garlile, P. R. 2002. A pragmatic view of knowledge andboundaries: Boundary objects in new product devel-opment. Organization Science, 13: 442-455.

Gheng, P. G. H., & Simon, H. A. 1995. Scientific discoveryand creative reasoning with diagrams. In S. M.

Smith, T. B. Ward, & R. A. Finke (Eds.), The creativecognition approach: 205-228. Gambridge, MA: MITPress.

Glark, A. 2008. Supersizing the mind: Embodiment, ac-tion, and cognitive extension. New York: OxfordUniversity Press.

Glark, A., & Ghalmers, D. 1998. The extended mind.Analysis, 58(1): 7-19.

Glark, K. B. 1985. The interaction of design hierarchiesand meirket concepts in technological evolution. Re-search Policy, 14: 235-251.

Glark, K., & Smith, R. 2008. Unleashing the power ofdesign thinking. Design Management Review,19(3): 8-15.

Gorley, K. G., & Gioia, D. A. 2004. Identity ambiguity andchange in the wake of a corporate spin-off. Admin-istrative Science Quarterly, 49: 173-208.

Gornelissen, J. P., & Glarke, I. S. 2010. Imagining andrationalizing opportunities: Inductive reasoning andthe creation and justification of new ventures. Acad-emy of Management Review, 35: 539-557.

Gornelissen, J. P. 2012. Sensemaking under pressxu'e: Theinfluence of professional roles and social account-ability on the creation of sense. Organization Sci-ence, 23: 118-137

Gronin, M. A., & Weingart, L. R. 2007. Representationalgaps, information processing, and conflict in func-tionally diverse teams. Academy of ManagementReview, 32: 761-773.

Denis, I. L., Langley, A., & Rouleau, L. 2006. The power ofnumbers in strategizing. Strategic Organization, 4:349-377.

Doyle, J., & Sims, D. 2002. Enabling strategic metaphor inconversation: A technique of cognitive sculpting forexplicating knowledge. In A. Huff & M. lenkins(Eds.), Mapping strategic knowledge. London:Sage.

Dimne, D., & Martin, R. 2006. Design thinking and how itwill change management education: An interviewand discussion. Academy of Management Learn-ing and Education, 5: 512-523.

Eisenhardt, K., & Graebner, M. E. 2007. Theory buildingfrom cases: Opportunities and challenges. Academyof Management Journal, 50: 25-32.

Emerson, R. M., Fretz, R. I., & Shaw, L. L. 1995. Writingethnographic fieldnotes. Ghicago: University ofGhicago Press.

Ewenstein, B., & Whyte, J. 2009. Knowledge practices indesign: The role of visual representations as "epi-stemic objects". Organization Studies, 30: 7-30.

Fetterman, D. M. 1998. Ethnography. Thousand Oaks,GA: Sage.


Finke, R. A., Ward, T. B., & Smith, S. M. 1992. Creativecognition: Theory, research, and applications.Cambridge, MA: MIT Press.

Fiol, M. C, & Huff, A. S. 1992. Maps for managers: Whereare we? Where do we go from here? Journal of Man-agement Studies, 29: 267-285.

Fiske, S. T., & Dyer, L. M. 1984. Structure and develop-ment of social schemata: Evidence from positive andnegative transfer effects. Journal of Personality andSocial Psychology, 48: 839-852.

Gephait, R. P. 1993. The textual approach—Risk andblame in disaster sensemaking. Academy of Man-agement Journal, 36: 1465-1514.

Gephart, R. P., Topal, C., & Zhang, Z. 2010. Future-ori-ented sensemaking: Temporalities and institutionallegitimation. In T. Hernes & S. Maitlis (Eds.), Pro-cess, sensemaking &• organizing: XXX. Oxford,U.K.: Oxford University Press.

Gioia, D. 1986. Symbols, scripts, and sensemaking: Gre-ating meaning in the organizational experience. InH. P. Sims & D. Gioia (Eds.), The thinking organi-zation: 49-74. San Francisco: Jossey-Bass.

Gioia, D. A., & Chittipeddi, K. 1991. Sensemaking andsensegiving in strategic change initiation. StrategicManagement Journal, 12: 443-448.

Gioia, D. A., Corley, K. G., & Fabbri, T. 2002. Revising thepast (while thinking in the future perfect tense).Journal of Organizational Change Management,15: 622-634.

Gioia, D. A., & Mehra, A. 1996. Sensemaking in organi-zations—^Weick, KE. Academy of Management Re-view, 21: 1226-1230.

Gioia, D. A., Thomas, J. B., & Clark, S. M. 1994. Symbol-ism and strategic change in academia: The dynamicsof sensemaking and influence. Organization Sci-ence, 5: 363-383.

Glaser, B. G., & Strauss, A. L. 1967. The discovery ofgrounded theory. New York: Aldine.

Greenwood, R., Oliver, C, Sahlin, K., & Suddaby, R.2008. Introduction. In R. Greenwood, C. Oliver, K.Sahlin, & R. Suddaby (Eds.), The Sage handbook oforganizational institutionalism: 1-46. BeverlyHills, GA: Sage.

Harrison, S. H., & Corley, K. G. 2011. Clean climbing,carabiners, and cultural cultivation: Developing anopen-systems perspective of culture. OrganizationScience, 22: 391-412.

Henderson, K. 1991. Flexible sketches and inflexible databases. Science, Technology & Human Values, 16:448-473.

Heracleous, L., & Jacobs, C. D. 2008. Crafting strategy:The role of embodied metaphors. Long Range Plan-ning, 41: 309-325.

Hicks, D., & Beaudry, M. C. (Eds.). 2010. The Oxfordhandbook of material culture studies. Oxford, U.K.:Oxford University Press.

Hill, R. C, & Levenhagen, M. 1995. Metaphors and men-tal models: Sensemaking and sensegiving in innova-tive and entrepreneurial activities. Journal of Man-agement, 21: 1057-1074.

Huff, A. S. 1990. Mapping strategic thought. Chichester,U.K.: Wiley.

Hutchins, E. 1995. Cognition in the wild. Cambridge,MA: MIT Press.

Jarzabkowski, P. 2003. Strategic practices: An activitytheory perspective on continuity and change. Jour-nal of Management Studies, 40: 23-55.

Jarzabkowski, P., Balogun, J., & Seidl, D. 2007. Strategiz-ing: The challenges of a practice perspective. HumanRelations, 60: 5-27.

Johnson-Laird, P. N. 1983. Mental models. Cambridge,U.K.: Cambridge University Press.

Jorgensen, D. L. 1989. Participant observation: Amethodology for human studies (Applied social re-search methods series, vol. 15). Thousand Oaks, CA:Sage.

Kaplcin, S. 2011. Strategy and Power Point: An inquiryinto the epistemic culture and machinery of strategymaking. Organization Science, 22: 320-346.

Kazmierczak, E. 2003. Design as meaning making: Frommaking things to the design of thinking. Design Is-sues, 19(2): 45-59.

Kiesler, S., & Sproull, L. 1982. Managerial response tochanging environments—Perspectives on problemsensing from social cognition. Administrative Sci-ence Quarterly, 27: 548-570.

Klimoski, R., & Mohammed, S. 1994. Team mental mod-el: construct or metaphor? Journal of Management,20: 403-437.

Knorr Cetina, K. 1981. The manufacture of Knowledge:An essay on the constructivist and contextual na-ture of science. Oxford, U.K.: Pergamon.

Knorr Cetina, K. 1999. Epistemic cultures: How the sci-ences make knowledge. Cambridge, MA: HarvardUniversity Press.

Kolko, J. 2010. Abductive thinking and sensemaking: Thedrivers of design synthesis. Design Issues, 26(1):15-28.

Krippendorff, K. 2006. The semantic turn: A new foun-dation for design. Boca Raton, FL: CRC.

Langfield-Smith, K. 1992. Exploring the need for a sharedcognitive map. Journal of Management Studies, 29:349-368.

Larkin, J., & Simon, H. 1987. Why a diagram is (some-


times) worth ten thousand words. Cognitive Sci-ence, 11: 65-99.

Lawson, B. 2005. How designers think: The design pro-cess demystified. Oxford, U.K.: Architectural.

Lincoln, Y., & Guba, E. 1985. Naturalistic inquiry. New-bury Park, CA: Sage.

Locke, K. 2001. Grounded theory in management re-search. London: Sage.

Lofland, J., Snow, D., Anderson, L., & Lofland, L. H. 2006.Analyzing social settings: A guide to qualitativeobservation and analysis. Toronto, ON: Wads-worth.

Maitlis, S. 2005. The social process of organizationalsensemaking. Academy of Management Journal,48: 21-49.

Maitlis, S., & Lawrence, T. B. 2007. Triggers and enablersof sensegiving in organizations. Academy of Man-agement Journal, 50: 57-84.

Maitlis, S., & Sonenshein, S. 2010. Sensemaking in crisisand change: Inspiration and insights from Weick(1988). Journal of Management Studies, 47: 551-580.

Martin, R. L. 2009. The design of business: Why designthinking is the next competitive advantage. Boston:Harvard Business School Press.

Mervis, G. B., & Rosch, E. 1981. Categorization of natiu-alobjects. In M. R. Rosenzweig & L. W. Porter (Eds.).Annual review of psychology, vol. 32: 89-113. PaloAlto, GA: Annual Reviews.

Norman, D. A. 1991. Gognitive artifacts. In J. Garroll(Ed.), Designing interaction: 17-38. Gambridge,U.K.: Gambridge University Press.

Oliver, D., & Roos, J. 2007. Beyond text: Gonstructingorganizational identity multimodally. British Jour-nal of Management, 18: 342-358.

Orlikowski, W. J., & Scott, S. V. 2008. Sociomateriality:Ghallenging the separation of technology, work andorganization. In J. P. Walsh & A. P. Brief (Eds.),Academy of Management annals, vol. 2: 433-474.Essex, U.K.: Routledge.

Pettigrew, A. M. 1990. Longitudinal field research onchange, theory, & practise. Organization Science, 1:267-292.

Porac, J. F., & Thomas, H. 1990. Taxonomical mentalmodels in competitor definition. Academy of Man-agement Review, 15: 224-240.

Porac, J. F., Thomas, H., Wilson, F., Paton, D., & Kanfer,A. 1995. Rivalry and the industry model of Scottishknitwear producers. Administrative Science Quar-terly, 40: 203-227.

Pratt, M. G., & Rafaeli, A. 2001. Symbols as a language oforganizational relationships. In B. M. Staw & R. Sut-

ton (Eds.), Research in organizational behavior,vol. 23: 93-132. Greenwich, GT: JAI.

Pratt, M. G., & Rafaeli, A. (Eds.). 2006. Artifacts andorganizations: Beyond mere symbolism. Mahwah,NJ: Erlhaum.

Rafaeli, A., Ravid, S., & Gheshin, A. 2009. Sensemakingin virtual teams: The impact of emotions and sup-port tools on team mental models and team perfor-mance. International Review of Industrial and Or-ganizational Psychology, 24: 151-182.

Rafaeli, A., & Vilnai-Yavetz, I. 2004. Emotions as a con-nection of physical artifacts and organizations. Or-ganization Science, 15: 671-686.

Ravasi, D., & Tuiati, G. 2005. Exploring entrepreneuriallearning: A comparative study of technology devel-opment projects. Journal of Business Venturing, 20:137-164.

Rheinberger, H. J. 1997. Toward a history of epistemicthings: Synthesizing proteins in the test tube. Stan-ford, GA: Stanford University Press.

Rosa, J. A., Porac, J. F., Runser-Spanjol, J., & Saxon,M. S. 1999. Sociocognitive dynamics in a productmarket. Journal of Marketing, 63(special issue):64-77.

Rouleau, L. 2005. Micro-practices of strategic senseméik-ing and sensegiving: How middle managers interpretand sell change every day. Journal of ManagementStudies, 42: 1413-1441.

Rouleau, L., & Balogun, J. 2011. Middle managers, stra-tegic sensemaking, and discursive competence. Jour-nal of Management Studies, 48: 953-983.

Salvato, G. 2009. Gapahilities unveiled: The role ofordinary activities in the evolution of product de-velopment process. Organization Science, 20:384-409.

Schutz, A. 1967. The phenomenology of the socialworld. Evanston, IL: Northwestern University Press.

Sonenshein, S. 2010. We're changing—or are we? Un-tangling the role of progressive, regressive, andstability narratives during strategic change imple-mentation. Academy of Management Journal, 53:477-512.

Spradley, J. P. 1980. Participant observation. Orlando,FL: Harcourt Brace Jovanovich.

Suchman, L. A. 1988. Representing practice in cognitivescience. Human Studies, 11: 305-326.

Sutton, R. I., & Hargadon, A. 1996. Brainstorming groupsin context: Effectiveness in a product design firm.Administrative Science Quarterly, 41: 685-718.

Van Maanen, J. 1979. The fact of fiction in organizationalethnography. Administrative Science Quarterly,24: 539-550.


Weick, K. E. 1979. The social psychology of organizing.Reading, MA: Addison-Wesley.

Weick, K. E. 1988. Enacted sensemaking in crisis situa-tions. Journal of Management Studies, 25: 305-317.

Weick, K. E. 1993. The collapse of sensemaking in organ-izations: The Mann Gulch disaster. AdministrativeScience Quarterly, 38: 628-652.

Weick, K. E. 1995. Sensemaking in organizations.Thousand Oaks, GA: Sage.

Weick, K. E., & Roberts, K. H. 1993. GoUective mind inorganizations—Heedful interrelating in on flightdecks. Administrative Science Quarterly, 38: 357-381.

Weick, K. E., Sutcliffe, K. M., & Obstfeld, D. 2005. Orga-nizing and the process of sensemaking. Organiza-tion Science, 16: 409-421.

Whiteman, G., & Cooper, W. H. 2011. Ecological sense-making. Academy of Management Journal, 54:889-911.

Ileana Stigliani ([email protected]) is assistantprofessor of design and innovation at the Innovation andEntrepreneurship Group of Imperial College London. Hercturent research focuses on how material artifacts andpractices influence cognitive processes—includingsensemaking and sensegiving, and perceptions of organ-izational and professional identities—within organiza-tions. She received her Ph.D. in management from Boc-coni University.

Davide Ravasi ([email protected]) is an asso-ciate professor of management at the Management &Technology Department of Bocconi University. His re-search addresses the sociocognitive processes that under-lie the interrelations among organizational identity, cul-ture, and strategy, and the production and use oforganizational artifacts in different contexts. He receivedhis Ph.D. in management from Bocconi University.

Copyright of Academy of Management Journal is the property of Academy of Management and its content may

not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.

Date post:	22-Mar-2020
Category:	Documents
Upload:	others
View:	4 times
Download:	0 times

ORGANIZING THOUGHTS AND CONNECTING BRAINS: MATERIAL...

Documents