ORGANIZATIONAL STRUCTURE, INFORMATION …agenda for the decision-making implications of orga-nizational structure and information processing. Therefore, the goal of this study is to

r Academy of Management Annals2020, Vol. 14, No. 1, 267–302.https://doi.org/10.5465/annals.2017.0103

ORGANIZATIONAL STRUCTURE, INFORMATIONPROCESSING, AND DECISION-MAKING: A RETROSPECTIVE

AND ROAD MAP FOR RESEARCH

JOHN JOSEPH1

University of California, Irvine

VIBHA GABAINSEAD

Beginning with Simon (1947)—and motivated by an interest in the effect of formal or-ganizational structure on decision-making—a large body of research has examined howorganizations process information. Yet, research in this area is extremely diverse andfragmented. We offer a retrospective of past research to summarize our collectiveknowledge, as well as identify and advance new concerns and questions. In doing so, weidentify three critical issues: a division between an aggregation perspective and a con-straint perspective of structure, little focus on informational sources of conflict, anduneven treatment of various stages of decision-making. We then offer a road map forfuture research that elaborates the role of organizational structure in decision-making.In this endeavor, we offer an ecological perspective of information processing that ad-dresses the issues and provides opportunities to expand research in new directions.

INTRODUCTION

Scholars have long been interested in the effects oforganizational structure and its influence on decision-making. As remarked by Simon (1997: 240), “[i]n apost-industrial society, the key problem in researchrelated to organizational structure is how to organizeto make decisions—that is, to process information.”Hence, the literature addressing these aspects of orga-nization have traditionally relied on some form of in-formation processing: gathering, interpretation, andsynthesisof information (Burton&Obel,1984;Tushman&Nadler, 1978: 614; see also; Galbraith, 1974; Puranam,Raveendran, & Knudsen, 2012; Van Knippenberg,Dahlander, Haas, & George, 2015).

Research on organizational structure, informationprocessing, and decision-making has spanned overseven decades. The areas of the organization theory,strategy, and organizational economics (among others)have concerned themselves with this subject and have

used different theories and methods to examine a va-riety of structural features, causal mechanisms, andoutcomes. There is a considerable amount of recentwork that addresses the decision-making and perfor-mance implications of organizational structure, whichreflects the organization design research agenda pur-sued by an increasing number of researchers (Burton,Obel,&Hakonsson,2015;Gulati,Puranam,&Tushman,2012; Joseph, Baumann, Burton, & Srikanth, 2018;Puranam, 2018).

The growth in scholarly attention, along with ad-vances in adjacent fields, has led tomultiple streams ofresearch on this topic. Each of these streams uses dif-ferent ways to link organizational structure—whichwedefine as the ways in which an organization divides itslabor and integrates their efforts (Mintzberg, 1979)2—todecision-making. Although this trend stems, in part,from different methodological approaches (e.g., mathe-matical models, agent-based models, and empiricalstudies), the research reflects a more fundamental di-vision with regard to its overall focus and to the

We thank Oliver Baumann, Phil Bromiley, Felipe Csaszar,Kathy Eisenhardt, Henrich Greve, Rouslan Koumakhov, DanLevinthal, Steve Postrel, Phanish Puranam, andWillie Ocasiofor discussions on earlier versions of thepaper.Wewould alsolike to thank ourAssociate Editor, J.P. Eggers, for his insightfulfeedback on previous versions of the manuscript. We aregrateful for the research assistance provided by John Kim.Both authors contributed equally to this paper.

1Corresponding author.

2 We recognize that there are many definitions of organi-zational structure. Each of these definitions emphasizesdifferent aspects of structure including interactions (Purnam,2018), configurations (Burton, Obel, & DeSanctis, 2011), andhybrids (Soda & Zaheer, 2012). We purposely draw on afoundational and general definition of formal organizationalstructure.

267

Copyright 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.

https://doi.org/10.5465/annals.2017.0103

theoretical treatment of information processing. Thedownside of such divergence is that further develop-mentwill behindered to theextent that research retreatsinto a regress into respective domains.

Notably absent is a concerted effort to review andassess the literature. Although a few studies have ac-knowledged thegrowing interest in this subject (Josephet al., 2018; Puranam, 2012) and some articles includestructure within their remit (Gavetti, Levinthal, &Ocasio, 2007; Posen, Keil, Kim, & Meissner, 2018),none have either directly surveyed and identified keyissues in this research stream or crafted a researchagenda for the decision-making implications of orga-nizational structure and information processing.

Therefore, the goal of this study is to review the ex-tant literature and summarize our collective knowl-edge, as well as identify and advance new concernsand questions about organizational structure anddecision-making. In this process, we revisit some ofSimon’s original ideas and assess how they are re-flected in contemporary research. We analyze the dif-ferent perspectives of how an organization’s structureaffectsdecision-makingand, insodoing, identify someof the literature’s key issues. We then offer a roadmapfor future research that addresses these issues and apoint of view that couldbring theseperspectivesclosertogether and expand research in new directions.

This endeavor should yield a fresh perspective onthe design of organizations. It is useful, given the pro-liferation of new organizational forms and supra-firmarchitectures (Burton, 2013).Abetterunderstandingofthe decision-making implications of these new orga-nizational forms may depend on a more refined un-derstanding of the information processing propertiesassociated with them. Finally, in terms of practicalimplications, organization structure provides a pow-erful set of levers that are directly accessible to the topmanagement. Hence, a more informed view of in-formation processing may improve the ability of topmanagement to use these levers effectively.

Our review of the literature identified four majorstreams of research: problem-skill matching, screening,adaptation, and cognition. In concert with our catego-rizingefforts,wealso identified threecritical issues.First,existing research is divided in its treatment of the role ofstructure in information processing. Although some ofthe literature concerns itself with how individual de-cisions come together (i.e., on how structure aggregates),other streams focus on how the organizational contextaffects individual decision-making (i.e., on how struc-ture constrains). Because both views typically neglectshared cognition and the constraint view routinely ne-glects interactions,neitherapproachadequately links the

socio-cultural properties of organizational structure toparticular contexts of joint decision-making.

Second, the literature largely overlooks the potentialfor conflict in decision-making. This shortcoming re-flects, inter alia, the belief that conflict results from di-vergent interests and poor incentive design (Gibbons,2003).Because somuch research intentionally abstractsfrom incentives, it avoids discussions about the sourceand consequences of intraorganizational tensions. Thislack of conflict stems also from a lack of focus on theorganization’ssystemofmeaning-makingandattention.Structure’s impact on the variation (and, hence, differ-ences) in the interpretation of information is incidentalto the theory, and hence, omits the possibility that in-formational sources of conflict may arise. Also, withouta solid understanding of the relationship between for-mal structure and attentional processes, we are unableto fully establish the conditions for when conflict isbeneficial for decision-making.

Third, the treatment of various stages of decision-making isuneven.Simon (1947) articulated four steps inthe decision-making process: agenda setting, problemrepresentation, search, and evaluation. Most studies fo-cus on search and (to a lesser degree) evaluation. Theeffects of structural variation on the agenda setting andproblem representation remain relatively unexplored.This omission is consequential in that it ignores the po-tential impact of setting an agenda and representingproblems on search, and for a recursive relationshipbetween alternatives selected and subsequent agendasand problems.

In addressing these issues, we offer a view whichmoves away from the literature’s emphasis on indi-vidual cognition and brings back Simons’ notion ofcommon maps or shared representations (Simon,1952) to enable a more complete view of the role ofstructure on decision-making. In doing so, we in-corporate the concepts of embedded and situated in-teractions to better capture the information-processingproperties of structure, and specifically how sharedrepresentations aremade accessible and activated.Weoffer an ecological perspective which not only bridgesthe aggregation and constraint views but also helpsexplain how conflict arises in the process and howagendas and problem representations are affected.

We place several boundary conditions on thescope of this study, a necessary restriction given thevast literature that references information process-ing. First, we consider only formal organizationalstructure. Thus, we ignore informal organizationalstructures and social networks except when con-sidered in conjunction with formal structure. Sec-ond, we mostly examine information processing in

268 JanuaryAcademy of Management Annals

relation to problem-solving and choice, that is, we areinterested in how organizational structure affectsdecision-making but not vice versa. Hence, we leaveaside those studies that examine (i) the effect of de-composability on organizational structure (Zhou, 2013)and product architecture (e.g., Baldwin & Clark, 2000;Yayavaram & Ahuja 2008); (ii) the effects of formalstructure on the informal organization (Clement &Puranam, 2017; Kleinbaum, Stuart, & Tushman, 2013);or (iii) governance decisions, firm boundaries (Foss &Weber, 2016), or interfirm relationships (Aggarwal,Siggelkow, & Singh, 2011). Neither do we considerstudies in which structure is merely a moderator or asecond-order boundary condition of the primary theo-retical argument. Thus, our review maintains, as itsprimary focus, organization structure and its implica-tions for decision-making.

Finally, because we are interested in conflict thatarises from misunderstandings rather than from mis-aligned incentives, we keep incentive issues in thebackground. Understanding the role of incentivesis an important area of inquiry in divisionalizedfirms (Argyres & Silverman, 2004), and incentives arewidely considered as a credible alternative explanationin studies of organizational structure and decision-making (Gibbons, 1998; Kretschmer & Puranam, 2008).Even so, organizational economists and strategyscholars acknowledge that incentives and informationprocessing involve different sets of causal mecha-nisms; hence,wewill address the former only in termsof their role in information-processing accounts.

Most of our review is limited to articles publishedwithin the last 20 years (2000–2019). Although wediscuss the literature’s foundational articles, we aremainly interested in the trends that have becomeevident over these last two decades. By classifying theextant researchandhighlighting themajor information-processing perspectives, we lay the groundwork—anddevise a road map—for a renewed and hopefullyfruitful program of research into the relationship be-tween organizational structure and decision-making.

FOUNDATIONAL LITERATURE ONINFORMATION PROCESSING

The models of information processing proposed byHerbert Simonhaveprovided,directly or indirectly, theconceptual scaffolding for much of the literature thataddresses the relationship between organizationalstructure anddecision-making.We, therefore, beginourreview by examining these contributions, after whichwe detail the more current research. In this latter task,we categorize previous work into four areas. The

thematic complementarities and differences acrossthese four areas drive the specifics of our proposed re-search agenda.

Herbert Simon and the Origins ofInformation Processing

In the field of organization research, informationprocessing in relation to decision-making traces its lin-eage back toHerbert Simon—a scholarwhowasmainlyconcerned with understanding how people solveproblems andmake decisions. Through his intellectualefforts, Simon brought psychological research intoeconomics and established amore behavioral approachto the study of human decision-making (Augier, 2001).

Simon’s ideas on information processing were ini-tially and most fully developed within two streams ofwork. The first stream dealt with administrative theory,which identified hierarchically ordered decision-making as the key concept underlying “the superstruc-ture of the theory of bounded rationality”—a notion thatwas central to his research that established a behavioralapproach to rational choice (Simon, 1955, 1956). Al-though Simon did not use the term “information pro-cessing” in the original Administrative Behavior (1948)book, the idea itself and its central components wereclearly present. Both March and Simon’s (1958) Orga-nizations and Cyert and March’s (1963) A BehavioralTheory of the Firm were explicit in their view of theorganizationas an information-processing anddecision-rendering system. According to Cyert and March, “weneed more reliable information on where and how or-ganizations secure information, how that information iscommunicated through the organization, and how au-thoritative decisions are reached, and finally how suchdecisions are implemented in the organization” (1963:20). Similarly, March and Simon discuss the communi-cation requirements and processes for effective co-ordination. According to them, “the capacity of anorganization to maintain a complex interdependentpattern of activity is limited in part by its capacity tohandle the communication required for coordination”(1958: 183).

A second stream of Simon’s research, which in-cluded his collaboration with Alan Newell, con-cerned human problem-solving, symbol processing,and heuristic search. Newell and Simon (1956)identified the individual decision-making processas a key unit of analysis. In this theory, there is animportant distinction between the actual task en-vironment and “the way a particular subject repre-sents the task to work on it” (Newell & Simon, 1972:151). Simon argued that “simplifications of the real

2020 269Joseph and Gaba

world for purposes of choice introduce discrep-ancies between the simplified model and the re-ality; and these discrepancies, in turn, explainmanyof the phenomenaof organizational behavior”(1995: 114).

It is worth noting that although Simon’s work re-flects two distinct information-processing fields(which focused, respectively, on organizations’ andindividual psychology), these two streams wereinterconnected. Simon’s research dealing with or-ganizational structure did not neglect individualdecision-making, and key concepts of his work onindividual cognition appeared in the framework ofhis work on organizations (Koumakhov, 2009). Al-though the former category did not always explicitlyconsider organizational structure (e.g., hierarchy), itwas implicit in the idea that structural boundariesand the division of labor reflect how the organizationrepresents its problems and affect how individualsfilter information.3 Simon suggested that for organi-zational members, their corresponding subgroup ororganizational participation influences their com-mon maps and selective perception (Dearborn &Simon, 1958). Common maps or shared representa-tions4 (i.e., perceptual mechanisms) arise in socialcontexts and supply interacting individuals with the“social definition of a situation” (Simon, 1995: 299),which is constructed via a firm’s formal organiza-tional structure (March & Simon, 1958). For Simon,then, the socio-cognitiveproperties of organizationalstructure establish a link—between individual cog-nition and collective decision-making. As we willdiscuss, this aspect of Simon’s work has beenmostlyoverlooked in recent research.

Information Processing Perspective

An information-processing perspective emergedin the field of management during the 1970s, and itreflected a growing interest by organization theoristsand in the question of how organizations are affectedby their external environment. Themost notable andinfluential proponents of this perspective were JayGalbraith, David Nadler, and Michael Tushman,who built on the work of Thompson (1967) andLawrence and Lorsch (1967). Within this stream,information processing was a molar concept relatedprimarily (although not entirely) to knowledge

acquisition and communication among decisionmakers. According to these scholars, the role ofstructure is to increase theorganization’s information-processing capacity to deal with internal complexityand environmental uncertainty (Galbraith, 1977;Gulati, Lawrence, & Puranam, 2005; Tushman &Nadler, 1978).

Among these scholars, therewas general agreementthat no single template for an optimal formal organi-zational structure exists; in other words, the “best”structural solution depends on a variety of contin-gencies.Sorather thanprescribeanidealuniversal typeof organization design, scholars postulate that the re-lationship among strategy, structure, and performancedepends on multiple factors (e.g., Donaldson, 2001).This perspective, which is known as contingencytheory, holds that an organization cannot be effectiveunless there is “fit” between its environment and itsstructure. Fit is achieved by mutually reinforcinginternal activities and by matching an organization’sstructural characteristics to its information-processingdemands (Burton & Obel, 2004).

Although contingency theory lost much of its po-tency in the 1980s and 1990s, interest in informationprocessing continued to grow and became more so-phisticated. Scholars in the field of organization andstrategy sought to model more complex organizations,and agent-based computer modeling techniques led toresearch that could account formultiple design choices(Siggelkow, 2011). This literature rediscovered Simon’sinterest in individual information processing and thenotion of nearly decomposable systems (Simon, 1964).These foundational ideas, when expressed with newmodeling tools, made it possible to explore systemati-cally the trade-offs involved with—and the decision-making implications of—interacting agents across agreater number and variety of information-processingstructures. Researchers adopted these methodolo-gies, which led to a rebirth of organization design instrategy and organization theory.

MAJOR THEMES IN RECENT RESEARCH

Our initial scan of the literature involved identi-fying key foundational works and review articles. Tobuild our set of representative articles, we accessedtheGoogleScholar andSocial ScienceCitation Indexand used a Boolean search string to retrieve articleswhose titles, abstracts, or keywords contained theterms “organization* structure” (or “organization*design”, “organization* form”, “organization* archi-tecture”, “multi-business”, “structure”, “corporat*”,hierarch*, “subunit”), and “information processing”,

3 We thank Phil Bromiley and Rouslan Koumakhov formany of these insights into Herbert Simon’s work.

4 The terminology “common maps” and “shared repre-sentations” areused interchangeably throughout the article.


and “decision making” (or “adapt*”, “search”, “ex-plore”, “evaluat*”). To ensure that our analysiswouldencompassmainly articles ofmore recent vintage, welimited our search to those published—over the last20 years—in the field’s nine leading journals: Acad-emy of Management Journal, Academy of Man-agement Review, Administrative Science Quarterly,Journal of Management, Journal of ManagementStudies,ManagementScience,OrganizationScience,Organization Studies, and Strategic ManagementJournal. We then narrowed our results based on theboundary conditions described in the Introductionsection. This effort yielded 281 articles.

After eliminating those articles that contained onlybrief usage of the terms and topics in question, wewere leftwith a total of 70.We then read these articlesand coded the central focus of the study in terms ofits overall conceptualization of the role of structurein decision-making. We found that these articlesaddressed fourmajor categories of research: problem-skill matching, screening, adaptation, and cognition.These four categories of the literature all echo foun-dationalwork in viewing organizational structure as asolution to the problems associated with informationprocessing and a means for coordination in decision-making. We reviewed the initial articles from eachof the streams and identified some of their majorsimilarities and differences. We sorted the 70 articlesaccording to the number of Google Scholar citationsthey had received. We proceeded to examine thebackward and forward citations of the most influen-tial articles in the list. This exercise led us to botharticles outside of the initial sample thatmayhavenotused verbatim one of the original search terms andrelevant articles in journals in adjacent fields. Wecontinued with the exercise until we felt confidentthat we had identified relevant and representativearticles in each of the four categories published in thelast 20 years.

See Tables 1 and 2, respectively, for illustrativecitations and comparison across streams.

Problem-Skill Matching

Problem-skill matching studies are grounded inorganizational economics.The research in this streamtakes a decision-theoretic approach that focuses onrepresentations of efficient allocation of tasks amongthe members of a multi-agent team. The centralproblem addressed by suchmodels is that, althoughorganizational members need to coordinate, thetasks and skills used to make decisions vary fromonemember to the next (Garicano, 2000). Hierarchy

affects organizational decision-making by ensuringthat people see problems that uniquely require theirparticular level of skill.

This stream originates from the same tradition asteam theory, which was an early economic model ofdecision-making in organizations, and examines thecosts and benefits of decentralized information pro-cessing (Marschak & Radner, 1972; Radner, 1993). Inthese models, the objective is to compute a set of de-cision rules or programs, for each individual of the“team” (i.e., organization), that maximizes the ex-pected payoff in a stochastic environment. Organi-zational members process different information andcommunicate their conclusions up the vertical hier-archy, with the top of the hierarchy making the finaldecisions (Van Zandt, 1999). This theory of teamsultimately served as the foundation for a variety ofsubsequentmodels, including those that elaborate onthe optimal grouping of functions (Cremer, 1980),diversity of information (Cremer, 1983), and alloca-tion of decision rights (Sah & Stiglitz, 1986). Studiesaddressing the last of these have been useful in strat-egy research that seeks to build and test theories ofcentralization and decentralization in organizationaldesign, which we shall describe in detail (cf. Csaszar,2012, 2013a, b).

Among the efforts by economists to model orga-nizational decision-making, the work by Garicanoand colleagues is notable for management scholars(Bloom, Garicano, Sadun, & Van Reenen, 2014;Garicano, 2000; Garicano & Wu, 2012). The goal ofsuch studies is to devise an optimal organizationalstructure, given the costs of communication and in-formation acquisition. The hierarchy’s rank orderingof knowledge serves to manage exceptions and tomatch problems with solutions. More difficult orcomplexproblems are referred up thehierarchy to behandled by more highly skilled problem solvers.

Garicano (2000) developed a formalmodel inwhichthe division of labor increases specialization and, thus,allows lower level specialists to solvesimpleproblems;harder or more complex problems are elevated up thehierarchy.Thecostsofworkersacquiringknowledge tosolve problems (delegation) are weighed against thecosts of elevating those problems to the managementlevel. Garicano’s model indicates that hierarchy mini-mizes the cost of knowledge acquisition and increasesthe specialized use of knowledge—especially whenknowledge is tacit and/or when problems are difficultto identify ex ante.

Bloom et al. (2014) found empirical evidence con-sistentwithGaricano’s (2000) theory.Usingplant-leveldata on information technology investment and a


survey to elicit the structural features of firms, theyfound that lowering the costs of acquiring and com-municating knowledge affects the extent of decen-tralization within a firm. These authors showed thatinvestment in “enterprise resource-planning” sys-tems increases the autonomy of plant managers andalso of workers; by contrast, investment in intranettechnology—which lowers communication costs—reduces both manager and worker autonomy. Wu(2015) extended Bloom et al.’s work by modeling thechoice of organizational structure and products. Hefound that the greater span of control (flatter organi-zation with more intensive vertical interactions)lends itself to producing higher value products and isenhancedwith greater communication of knowledge.

Bridging organizational economics and strategyperspectives, Garicano and Wu (2012) argue that tosupport coordination, the choice of an organizationalstructure is driven by the complexity of tasks faced bythe firm and by the “modifiability” of knowledge re-quired to perform those tasks.Here, variation in costs,which necessarily follows from limited attention, is afunction both of the firm’s specialization in activitiesand of its capacity for vertical information processing

when matching problems with solutions. They alsohighlight the benefit of shared codes for coordinationwhen information is at least partially tacit.

Synthesis. Although the work in this streammodels the actions of economic agents, it deviatesfrom foundational work on team theory. In particu-lar, team theory is concerned with the aggregation ofinformation. These models articulate vertical infor-mation processing in a decentralized organizationas a concern of both the amount of information,the timeliness of processing that information, and, ofcourse, the associated costs.

However, the task assignmentmodels proposed byGaricano and colleagues do not aggregate informationin this sense that multiple organizational memberscollectively contribute to an overall final decision(Gibbons, 2003). Instead, the focus is onhowstructureconstrains decision-making. Organizational structureis modeled as a hierarchy of knowledge (or skill)rather than a hierarchy of authority.5 The central

TABLE 1Representative Studies

Theme Examples

Problem-skill matching Garicano (2000) Bloom et al. (2014)Garicano & Wu (2012) Wu (2015)

Screening Decision rules Psychological mechanismsKnudsen & Levinthal (2007) Reitzig & Sorenson (2013)Christensen & Knudsen (2010) Fang et al. (2014)Csaszar (2012, 2013a, b) Reitzig & Maciejovsky (2015)Csaszar & Eggers (2013) Keum & See (2017)

Adaptation Modularity Coupled multi-level search Integration through knowledge sharingEthiraj & Levinthal (2004) Siggelkow & Rivkin (2009) Karim & Mitchell (2000)Siggelkow (2002) Lee & Puranam (2016) Helfat & Eisenhardt (2004)Billinger et al. (2014) Levinthal & Workiewicz (2018) Williams & Mitchell (2004)

Integration through hierarchy Interdependencies in the M-form Brusoni & Prencipe (2006)Rivkin & Siggelkow (2003) Obloj & Zenger (2017) Fang et al.(2010)Siggelkow & Levinthal (2003) Hu et al. (2017) Karim & Kaul (2015)Sengul & Gimeno (2013) Baumann et al. (2018) Stan & Puranam (2017)Joseph et al. (2016) Tarakci et al. (2018)Eggers & Kaul (2018) Knott & Turner (2019)

Cognition Mental models/frames Attention/cognitive accessibilityGavetti (2005) Rerup (2009)Jacobides (2007) Rhee et al. (2019)Srikanth & Puranam (2011) Gaba & Joseph (2013)Puranam & Swamy (2016) Joseph & Ocasio (2012)Joseph & Wilson (2018) Piezunka & Dahlander (2015)Csaszar & Laureiro-Martinez(2018)

Dutt & Joseph (2019)

5 The two are rarely isomorphic; imagine a divisionmanager being tasked with solving a production processproblem that the shop floor engineer cannot.


TABLE2

Organ

izational

Structure

andDec

ision-M

aking

Literature

Them

eKey

Inform

ation

ProcessingChallenge

Primary

Dec

ision-

Mak

ingFoc

us

Form

ofOrgan

izational

Structure

Roleof

Hierarchy

PrimaryView

ofStructure

and

Inform

ationProcessing

Key

Results

Problem

-skill

match

ing

How

doyo

udev

isean

optimal

orga

nizational

structure,g

iven

the

costsof

communication

andac

quiring

inform

ation?

Eva

luation

Ran

korder

ofskills

(skill

hierarchy)

aligned

withflow

ofproblem

s

Exc

eption

man

agem

ent

Con

straintindecision-

mak

ingco

mes

from

spec

ialize

dkn

owledge

Hierarchyof

skillthrough

whichproblem

sflow

bottom

-upis

optimal

design.

Scree

ning

How

doyo

ustructure

dec

isionrigh

tsin

agrou

pso

that

they

mak

efewer

errors

ofom

ission

andco

mmission

?

Sea

rchan

dev

aluation

Polya

rchyan

dhierarchy;

allc

ombinations

Validationof

proposals

Agg

rega

tion

ofdec

isions

viadec

isionrules

(e.g.,vo

tingrules);

Con

straintc

omes

from

theeffectsof

structure

onindividual

decision-

mak

ingbiases.

Differentd

ecision

structuresproduce

differentratiosof

Typ

e1

toTyp

e2errors.

Adap

tation

How

doe

sorganizational

structure

helpag

ents

evaluatech

oice

sets

insituationsof

coupled

andparallels

earch?

Sea

rch(parallel

andco

upled)

Sym

metric(dya

dsan

dintegrators)

and

asym

metric(authority)

influen

cestructures

Eva

luationof

choice

sets

Agg

rega

tion

ofdec

isions

viainteractionswithin

andam

ongsu

bsystems;

Con

straintindecision-

mak

ingco

mes

from

coupledmultilev

elsearch

Parallels

earch:o

ptimal

clusteringproduces

mostd

iversity

and

preserves

it.

Cou

pledsearch

:the

benefitsof

asym

metric

influen

cean

d“co

mmon

prior”effect/shared

beliefsca

nhelpwith

coordinationev

enwhen

they

arewrong.

Cog

nition

How

doe

sorganizational

structure

influen

cewhichinform

ationan

ddec

isionsare

prioritized

?

Age

ndasetting,

problem

form

ulation

,an

dsearch

Authorityhierarchy

Provision

ofdec

ision

premises;d

irect

attention

Con

straintindecision-

mak

ingco

mes

from

structure-lim

iting

cogn

ition(cog

nitive

structuresan

dattention

)

Problem

form

ulation

and

importance

vary

bythe

loca

tion

oftheproblem

solver

inthestructure.


information-processing problem is an optimal taskallocation problem, given the costs of knowledgeacquisition and communication. The hierarchy is atool for exception management where problems ofincreasing complexity are elevated within the orga-nization. Lower level individuals are, thus, limitedor constrained by their specialized knowledge andhandle problems that match their skill level.

Importantly, the models all assume that memberinterests do not differ. Decision-making reflects im-plications of different individualswhoholddifferentinformation, have different skills, and control dif-ferent decisions, but are working toward the sameend. This crucial assumption allows researchers toavoid the incentive problem and to focus insteadon the information problem. Even so, conflicts thatmight arise from specialized knowledge and/or in-formation sharing in coordinated decision-makingare largely omitted.

Also, Garicano and colleagues generally considerthe differences in know-how between hierarchicallevels to be infinitely “sticky.” But the specific factsto be processed are “mobile,” in that they can bereadily transmitted and the only cost involved isfrom transmission and attention. However, as notedbyPostrel (2002: 304, quotingKogut&Zander, 1996),“this approach assumes away the real difficulties ofcommunication among people, which have to dowith such things as conflicting conceptual categoriesand semantic ambiguities.” Finally, while acknowl-edging the specialization of decision-making at dif-ferent levels of organization, the studies here do notexplicitly articulate different steps (or types) in thedecision-making process.

Screening

The literature here deals with the screening of in-formation by individuals situated in different struc-tures. It includes studies of how different decisionrules affect quality of decision-making and of howhierarchies affect thepsychological biases indecision-making.More practically, the literature on screeningsheds light on how an organization’s structure canbe modified to compensate for its member’s fallibil-ity. This is a diverse set of studies using a variety ofmethods and approaches drawn from economics,social psychology, strategy, and organization theory.

Decision rules.The first set of studies largely drawfrom and build on the basic model proposed by Sahand Stiglitz (1985, 1986, 1988), which compares thescreening properties of hierarchies and polyarchies;those studies reflect (respectively) centralized and

decentralized decision-making structures in the econ-omy. The basic premise underlying these models isthat individuals are prone to errors of judgment andthat those errors are affected by the aggregation, de-cision rules, or voting rules associated with differentstructures. The key insight is that, if decision makersare fallible, then hierarchical (respectively, poly-archical) structures increase errors of omission (re-spectively, errors of commission). Sah and Stiglitz(1991) also showedthat in structures that are relativelycentralized, highly capable decision makers havemore beneficial effects (than in a decentralized struc-ture) on decision quality. This latter result amounts toa salient qualification on the implicit assumption thatguides the problem-skillmatching literature—namelythe structural position and skill level are equivalent.

In applying these insights into organizations andempirically testing their propositions, strategy andorganization scholars have suggested some importantextensions. For example, Christensen and Knudsen(2010) examined the reliability of different structuresas a function of the organization’s number of indi-vidual members. Their analytical model considersnot only the extremes (hierarchy vs. polyarchy) butalso the full range of organizational architectures,enabling the specification of structures that tradeoff Type I and Type II errors (i.e., those of omissionand commission, respectively) as the relative degreeof hierarchy and polyarchy shifts. Csaszar (2012,2013a, b) exploited the stock-picking decisions ofmutual fund managers and found that decentralizedstructures are associatedwith the acceptance of moreprojects—with fewer errors of omission and moreerrors of commission—than are centralized struc-tures. In linking Sah and Stieglitz with the signal de-tection theory, the work suggests that if errors ofomission (respectively, commission)arecostlier, thenthe organization is best served by a decentralized(respectively, centralized or hierarchical) structure.

Knudsen and Levinthal (2007) described howagents within different organizational structuresperceive and search in an NK landscape and foundthat some structures are better than others at promot-ing exploration. Screening ability and organizationalstructure exhibit a high degree of complementarity.The less (respectively more) able are individualevaluators, the more attractive are organizationalforms that tend toward hierarchy (respectively poly-archy). Thus, ahierarchical structure compensates forthe high error rates of less able individual evaluatorswhereas a polyarchy—or, more strictly, the variancethat it induces—compensates for what can be theoverly precise judgments of abler evaluators.


Similarly, Csaszar (2013b) looked at the relationshipbetween the structure of an organization and itsability to explore and exploit; in that study, explo-ration and exploitation are viewed in terms of theerrors of (respectively) omission and commission.He showed that “hybrid” structures not only cantrade off one error against the other but also canachieve a smaller error overall, that is, hybrid struc-tures allow for a simultaneously high degree of ex-ploration and exploitation.

Csaszar andEggers (2013) evaluated the robustnessof various decision-making structures—delegationtoexperts,majorityvoting, andaveragingofopinions—to environmental changes and to differences in theexpertise of decision makers. They found that eachstructure’s performance depends on the breadth ofknowledge within the firm and on changes in theenvironment. Delegation is a common structurefor organizations in relatively stable environments,whereas voting is more common under changingenvironments or memberships.

Research in this stream has also started to addresshow more distributed structures, such as communi-ties and crowds access, process, and screen in-formation (Van Knippenberg, Dahlander, Haas, &George 2015). For example, Retelny et al (2014) andValentine, Retelny, To, Rahmati, Doshi, and Bernstein(2017) introduce the idea of “flash teams” and “flashorganizations”—dynamically assembled online ex-perts from the crowd—to manage complex and inter-dependent tasks. These computationally representedstructures rely on traditional notions of roles, teamsand hierarchy and the roles encode interdepen-dencies, and the nesting of roles encodes hierarchyand decision rights. The information flows up to thehierarchy as a worker submits a task in the system,who then reviews and accepts it or returns it withfeedback for revision. At the same time, the structureadapts continuously by reconfiguring roles, teams,and hierarchy based on both top-down and bottom-up information flows.

Psychological mechanisms. The second set ofstudies introduces the idea that hierarchymay affectthe behavioral tendencies of those sending proposalsup the chain of command for approval. For example,in a study of commercial banks, McNamara, Moon,and Bromiley (2002) found that the propensity ofmanagers to loan money to businesses despite anegative change in their credit worthiness was at-tenuated, when those managers faced hierarchicalevaluation of their decisions. This de-escalation ofcommitment is attributable to the increased moni-toring that comes with hierarchy.

In other cases, structure may exacerbate decision-making biases. Managers may feel threatened by thetop management or by the centralized provision offeedback, a dynamic that has implications for crea-tivity and novel search (Kim & Kim, 2019). For ex-ample, Fang, Kim, andMilliken (2014) reasoned thatan organization’s members may screen informationin their reluctance to inform managers of bad news;hence, lower level staff may “sugarcoat” negative feed-back and, thus, leave those managers with a distortedviewof theorganization’sperformance.Yet they foundthat theremay actually be somepositive consequencesto a moderate amount of systematically distorted neg-ative information: it may create a sense of well-beingthat is sufficient to prevent potentially valuable ex-ploratory efforts from being prematurely abandoned.

Reitzig and Sorenson (2013) reported that the fail-ure to adopt an idea or innovation can arise from thein-group bias among employees of an organizationalsubunit; such bias would result in those individualssystematically undervaluing the ideas proposed byorganization members outside their own subunit.Along similar lines, Reitzig and Maciejovsky (2015)used a data set of innovation ideas submitted bymid-level managers in a large European consumer goodsfirm.They found that a hierarchical structure reducedthe number of ideas that these managers passed up tosuperiors. They offer two explanations to this finding:(i) mid-level managers fear negative feedback fromerrors of commission, in which case structures thatare more hierarchical induce more apprehensionabout evaluation; and (ii) these managers may viewthemselves as lacking control and, hence, would pre-fer to forgo the administrative burden of transmittingnew ideas. These explanations notably run counter tothose implicit in the Sah and Stigliz model, underwhich mid-level managers should be less likely—given that ideas are screened also at higher levels—toexhibit conservative screening behavior and, thus,more likely to send ideas up the chain of command.Building on this empirical work, Keum and See(2017) adopted amixed-method approach to examineboth idea generation (search) and idea selection(evaluation). Combining an experiment with data onapparel launches by a multinational fashion retailer,they hypothesized and demonstrated that hierarchymay impede idea generation owing to the “evaluationapprehension” and lack of control experienced bythose at lower levels. Hierarchy is beneficial in theselection phase, however, because it may reduce thebiasof individual subunits thatareprone to favor theirown ideas and, thus, may encourage the promotion(and acceptance) of proposals made by other units.


Synthesis. Our analysis has identified two sub-streams of screening research: one that builds on theSah and Stiglitz (1986) decision rules model andthe second that highlights the behavioral biases ofmanagers. A key difference across these substreamsis the role of the organizational structure. The de-cision rules models are about information aggrega-tion. This literature documents that the search forand evaluation of alternatives are consequences ofdifferent structure types, which, in turn, affect howthe information is screened and aggregated by bound-edly rational individuals (Csaszar & Eggers, 2013;Christensen&Knudsen, 2010). The behavioral studiesare less about aggregation and more about how thestructural context affects individual decision-makingbiases. Although both streams recognize fallible de-cision makers, the decision rules models emphasizeerrors in judgment, whereas the behavioral modelsemphasize psychological biases.Aswe shall see in thefollowing paragraphs, this relates the screening liter-ature to some of the adaptation literature which is ournext section.

These two substreams are similar in that for both,hierarchy serves as a mechanism for proposal eval-uation. In particular, hierarchy is a tool for validatingalternatives generated at lower levels; thus, pro-posals are either endorsed or rejected at higher levelsafter vetting at lower levels. However, the nature ofthe hierarchy also differs across the two substreams,in that in decision rulesmodels, the termhierarchy isused in a narrow sense, that is, the structures do notfully incorporate issues of authority or power. In thebehavioral models, the role of hierarchical authorityis important (although usually implicit), in that itinfluences the behavior of lower level people.

Much of the decision rules work also assumes thatindividual evaluators are homogeneous in terms oftheir interests and screening abilities.6 The focus ison who makes which decisions using what infor-mation and on how those factors are related to thestructure of communication (March & Simon, 1958,1993). Because it relies on the independence of in-dividual judgments in support of common out-comes, this substreamalso neglects anymotivationaldifferences between decision makers. The potentialfor conflict among decision makers is not yet fullyexplored.

Note that the studies largely focus on search andevaluation of alternatives, and the findings are

broadly consistent. Both sets of studies suggest thatthe impact of hierarchy may vary with the outcomeand stage of decision-making. Although the decisionrulesmodels focus on search,much of the behavioralmodels focus on evaluation (Keum & See, 2017 is anexception). Hierarchy seems to be more detrimentalfor search than for evaluation (Keum & See, 2017;McNamara et al., 2002; Reitzig &Maciejovsky, 2015;Reitzig & Sorenson, 2013), consistent with the mod-eling work that suggests omission errors occur moreinhierarchies than indecentralized structures owingto vetoing ideas as they are elevated (Christensen &Knudsen, 2010; Csaszar, 2012, 2013a, b).

Adaptation

The third vein of research pertains to learning andadaptation. Much as in the problem-skill matchingand screening literatures, information processing inthe adaptation stream of research reflects Simon’snotion of limited human cognition. When presentedwith problems, rather than search for optimal solu-tions, individuals satisfice and choose the first alter-native that meets their aspiration levels. Scholars inthis domain view organizations as adaptive systems(Denrell & March, 2001; Posen & Levinthal, 2012);thus, firms adapt via experiential, trial-and-error, and/or reinforcement learning—as indicated by theirupdated actions in response to performance feedback.

Organizational structure plays several roles in thisliterature. A large share of research in this domainfocuses mainly (although not exclusively) on un-derstanding how organizations enable their bound-edly rational members to adapt collectively to theirinterdependencies (Aggarwal &Wu, 2015; Baumann,2015; Puranam, Stieglitz, Osman, & Pillutla, 2015). Aprimary concern in this literature is how best to bal-ance the interactionswithin and between individualsor units as a change in one aspect of a firm may, inturn, affect its other aspects (Siggelkow, 2001). Giventhe complexity of such interdependencies, it is hardlypossible for a manager to be aware of—much less tocomprehend—all these interactions. Some of thesestudies also share a common methodological ap-proach: agent-based modeling. Such models—whichinclude, inter alia, NK models, bandit models, andcoupled learningmodels (Puranamet al., 2015)—oftencharacterize decision sets as existing on a “perfor-mance landscape”where variations in the interactionsbetween choices result in different performance levelsor “peaks” (Levinthal, 1997). In real-world organiza-tions, these combinations manifest as interconnecteddecisions between upstream and downstream

6 A notable exception is Csaszar and Eggers (2013) who,like the problem-skill matching literature, account forheterogeneity in terms of ability.


functional departments, collaborating business units,adjacent components of the value chain, or inte-gratedproductdevelopment activities (e.g.,Sorenson,2003; Sosa, Eppinger, & Rowles, 2004). Next, weexamine the structural variations that characterizethe efforts of firms to deal with interdependenciesand consider the impact of these efforts on decision-making.

Modularity. Drawing on the notion of modularityfeatured in the literature on design (Baldwin & Clark2000; Sanchez & Mahoney, 1996), several studiesseek to determine the optimal degree of modular-ity considering the underlying internal interdepen-dencies. For example, Ethiraj and Levinthal (2004)modeled the outcomes ofmanagers under- and over-modularizing their organizational designs relativeto the actual structure. They found that excessivemodularization may obfuscate important interac-tions and create significant uncertainty about systemcomplexity, a detriment to search. Along similar lines,Siggelkow (2002) explored the consequences of man-agers who do not fully comprehend the strength ofinteractions between activities. According to the re-sults of that study, misperceptions involving com-plementary activities are costlier than those involvingsubstitute activities because complements (re-spectively, substitutes) tend to amplify (respectively,attenuate) the performance consequences of mis-perception. In a mixed-method study includinga theoretical model and a laboratory experiment,Billinger, Stieglitz, and Schumacher (2014) askedparticipants to design a new product in whichmultiple product features must be combined in aparticular configuration. They established differentlevels of interdependencies among design elementsto manipulate task complexity (i.e., by creatingdifferent landscapes). Studyparticipants in the low-complexity landscape found the global optimum,whereas none of them found it in the high-complexitylandscape. In addition, deviations from local searchwere more strongly associated with complex tasks—especially in the advanced stages of search.

Aggarwal and Wu (2015) used a panel data set offirms in the U.S. defense industry between 1996 and2006 to examine the organization’s interdependencestructures—and its associated coordination needs—toward the end of explaining differences in how firmsadapt to an industry-wide demand shock. They foundthat coordination across product areas creates greateradaptation challenges than does coordination withinproduct areas. They also report that the negative ef-fects of interproduct coordination are enhancedwhen the firm’s products have a greater number

of underlying interactions (i.e., higher productcomplementarity) yet are mitigated when interde-pendences are grouped by the organizational unit.

Integration through hierarchy. In studies address-ing this topic, hierarchy is a key integrating mecha-nismtolimit suboptimalchoices (becauseofunobservedinteractions) and thereby improve overall decisionquality. One group of studies test whether, in thepresence of decision interdependencies or otherboundary conditions, a centralized instead of decen-tralized structure yields more benefits. For example,Rivkin and Siggelkow (2003) found that a hierarchytends to yield better outcomes when interactionsamong decisions are pervasive—but only if there issufficient information flowing up the hierarchy. Thatis, centralized decision makers amid interdepen-dencies are better (than decentralized ones) at vettingproposals but require a robust set of proposals towhich they can respond. A later article establishedenvironmental turbulence and complexity as impor-tant boundary conditions of such centralization anddemonstrated lateral communication (between de-partment heads) as aneffective alternative (Siggelkow& Rivkin, 2005).

In a related article, SiggelkowandLevinthal (2003)examined how decentralization affects the searchfor high-performing combinations of activities. Theyfound, in accord with prior research, that decen-tralization allows for the sufficient exploration ofnew combinations; however, they found also thatreintegration (i.e., centralization) is needed to ensurea complete accounting of all the interdependenciesamong activities. These benefits of temporal shifts instructure were likewise identified by Nickerson andZenger (2002), who argued that regularly switchingbetween efficient and flexible structures often placesthe organization in an intermediate stage that en-hances its performance.

This interest in hierarchy and adaptive behaviorhas not been confined tomodelers. In their empiricalstudy of innovation inmulti-technology firms, EggersandKaul (2018) found thatmanagers responsible for aportfolio of technologies are more willing to takegreater risks (to launch radical innovation projects)than are managers responsible for a single technol-ogy. Centralized control over a portfolio leads theformer to be less concerned about the risk of any onetechnology and to seek expansion of the portfolio, anorientation whose effect is to reduce the portfolio’soverall risk. Joseph, Klingebiel, and Wilson (2016)similarly reported that more centralized managers(i.e., those responsible for a portfolio of products) areless concerned about the fate of any particular


product than are decentralized, single-product man-agers; hence, centralized managers are more willingto pull unsuccessful products from the market andto reinvest those resources in the portfolio’s other,more successful products. In conglomerates withsubsidiaries exposed to multi-market competition,centralized decision makers (i.e., corporate office)may be best off to limit the decision rights and re-sources of constituent subsidiaries—what Senguland Gimeno (2013) call imposing constrained dis-cretion. Constrained discretion limits subsidiaries’competitive actions, which is beneficial becauseaggressive responses can serve as a strong basis forretaliation by rivals.

Finally, Seshadri, Shapira, and Tucci (2019) studythe relationship between degree of hierarchy andR&D and find that there is an “optimal” level of hi-erarchy for the quality of R&D decisions, a findingthat is consistent with other research that has iden-tified the benefits of a moderate degree of structure(Davis, Eisenhardt, & Bingham, 2009).

Coupled search within hierarchies. Another se-ries of studies examine the coupled search processwithin multilevel hierarchies. This dynamic is oftenreflected in efforts by seniormanagers to find superiorcombinations of policy choices and parallel effortsby lower level managers to find superior combina-tionsof activities that conformto thosepolicychoices.For example, Siggelkow and Rivkin (2009) examinedsuch a hierarchical coupled learning problem andestablished that such a search process can, at lowerlevels, obscure the true impact of higher level choices.In particular, lower level organizationalmembers can(owing to“luck”, say)makegoodchoicesdespitepoorhigher level choices or can misattribute their goodchoices to good high-level choices that have beensince discarded. Lee and Puranam (2016) extendedthis line of thinking to examine what happens whenthe organizational member who holds a belief or fa-vors a related strategy (e.g., senior manager) is not thesame person who undertakes action based on thatstrategy (e.g., lower level manager). In this case, per-fect implementation could be beneficial even if exante beliefs were imperfect because the updating ofimperfectmodels and strategymay benefitmore fromaccurate performance feedback.

More recently, Levinthal and Workiewicz (2018)examined a hierarchical form commonly found in“matrix” organizations, where an individual mem-ber reports to two different managers. In their setup,multiple higher level managers search for betterpolicies, whereas multiple lower level managerssearch for better strategies. Such dual authority may

provide lower level managers with more autonomyin decision-making; multi-authority structuresmakeit possible for the manager to negotiate betweensuperiors, who may, in turn, offer greater latitudein light of the dual demands. The Levinthal andWorkiewicz results indicate that a dual reportingstructure performs better when organizational de-mands likewise have a dual focus—in other words,when the organization needs to enable local adap-tation by subunits but still must coordinate acrossthose subunits. The matrix organization’s capacityfor coordination and negotiation has similarly beendemonstrated empirically in the context of alliances(Sytch, Wohlgezogen & Zajac, 2018).

Interdependencies within the multidivisionalfirm. Another group of articles has examined be-havioral interdependencieswithin amultidivisionalfirm or business group. These studies explore thatboth thehorizontal differentiation of subunitswithina larger corporation and the role of internal socialcomparison affect subunit responses to performancefeedback. For example, scholars have usefully ex-amined what happens when multiple units or indi-viduals must assess performance feedback in thecontext of others’ performance, a form of interde-pendence that reflects competition over resources.

Although partly an incentive story, Obloj andZenger’s (2017) study examined the formal and in-formal design elements shaping the structural, geo-graphic, or social distance of advantaged peers at aretail bank. They found that in organizational struc-tures facilitating the development of frequent andclose social interactions between subunits, there is agreater tendency to engage in social comparisons.Hu, He, Blettner, andBettis (2017) called the internalsocial comparisons with other divisions a “politicalreference point” and suggested that because relativeperformance determines the amount of resourcesand attention received from the corporate office, itshapes (in concert with external social comparisons)the firm’s aspirations. These authors found that in-consistent feedback leads to more attention beinggiven to the social reference point relative to whichthe focal division is underperforming. Tarakci, Ates,Floyd, Ahn, and Wooldridge (2018) found thatcomparing performance to peers and other subunitsbetter motivates managers’ divergent strategic behav-ior. Managers who identify more with the organiza-tion pay more attention to organizational rather thanindividual attainment discrepancies. In a computa-tional model, Baumann, Eggers, and Stiglitz (2018)alsoargued that intraorganizational comparisonsare anatural part of the firm’s political landscape. Their


study demonstrates that internal social comparisonsas a function of corporate membership create—moreso than do historic comparisons—winners and losersand, thus, are more likely to result in a better balancebetween exploration and exploitation activities at theorganizational level. In other words, units that areperforming better than their peers engage in exploi-tation, whereas underperforming units explore.

Finally, Knott and Turner (2019) suggest a key rolefor the corporate headquarters in stimulating these in-ternal dynamics. Using an analytical model and a casestudy of Banc One, they argue that headquarters pro-motes both interunit community which affords co-operation, and interunit competition which stimulatessocial comparison and innovation as units attempt tomaintain their favorable position in the corporation.

Integration through knowledge sharing. A dis-tinct subset of work examines how organizationsadapt by adding, redeploying, recombining, or divest-ing knowledge and resources to achieve efficiency,to explore new opportunities, and to innovate (e.g.,Helfat & Eisenhardt, 2004; Karim, 2006; Karim &Mitchell, 2000). Adaptation occurs as units evolve(or “morph”; Rindova & Kotha, 2001) and as corpo-rate executives patch (or “re-architect”) their line ofbusiness portfolios by creating newdivisions (Gilbert,2005), by shifting product market charters from onebusiness unit to another (Galunic&Eisenhardt, 2001),and by eliminating, splitting, or combining extantunits (Karim, 2006). These studies are broadly con-cerned with interactions and movement between indi-vidual decision makers but much less concerned withinternal interdependencies or performance aspirations.

For example, Karim andMitchell (2000) found thatunder certain conditions, structural recombinationserves as a mechanism to recombine intraorganiza-tional knowledge and to disrupt the firm’s ownknowledge base. Similarly, Karim and Kaul (2015)found that structural recombination has a morepositive effect on firm innovation when there areknowledge synergies within the organization, whenthe technology is novel, and when the disruptive ef-fects of structural recombination are contained.

Integration through knowledge sharing may beespecially critical for coupling divisionalized firmsseeking to take advantage of new external opportu-nities. Fang, Lee, and Schilling (2010) pick up thistheme in an agent-based model and argue for thebenefits of improving the information flow acrossindividuals and groups. They discovered that mod-erate levels of information exchange between sub-units are optimal for allowing superior choices todiffuse across groups without reducing the diversity

of those ideas too quickly. In an empirical study ofthe telecom industry, Williams and Mitchell (2004)found that links between business units are benefi-cial in that they provide the cooperating units withnew sets of information.

Martin and Eisenhardt (2010) studied how execu-tives create high-performing cross-business unit col-laborations inmulti-business firms. They find that thehighest performing collaborations occur when busi-ness unit GMs interact. The highest system-level per-formance results from small events that bring lowerlevel managers together (e.g., an industry conference)and from interactions that make it easier to share in-formation and make collaborative decisions.

In another an empirical study of in vitro fertiliza-tion clinics, Stan and Puranam (2017) showed thatintegrators—in this case, healthcare professionalswho set the treatment course and handle idiosyn-cratic patient adjustments—help the organizationavoid “superstitious” learning (i.e., misattribution ofcausal linkages between behavior and outcomes)through active questioning, seeking justification forprocedures, and so forth. In Brusoni and Prencipe’s(2006) in-depth case study of Pirelli, they examinedhow the integration of product and process knowl-edge created a new kind of tire designer: that of anengineer whowas competent in the entire process oftiredesign andproduction.Thekey to the integrationwas modular design rules at the plant level whichlead to the unintended consequence of creating anintegrated body for engineering know-how.

Synthesis. The adaptation stream is a highly di-verse one, incorporating many properties of struc-ture and exploring many kinds of decisions. Themodeling articles are similar to those in the screeningliterature, in that the focus is on information or (moreaccurately) choice aggregation structures. Here, theorganizational structure acts as a mechanism thataggregates individual perceptions into a group-levelevaluation of alternatives. Alternatively, empiricalstudies of hierarchy, the multidivisional firm, cou-pled search, and knowledge sharing are closer to thecognition articles, in that—rather than aggregation—they convey a sense of the constraining and enablingrole of structure. For example, the multidivisionalfirm creates a context for peer unit social comparison.Also, in a firm with coupled search, the higher levelchoices may constrain those made at lower levels.

This research is distinct, in that it explicitly con-sidersways tomanage interdependencesand integratethe organization through limiting the interactionsamong decisions, providing hierarchical oversight,and increasing shared knowledge. The focus is on


how firms can organize to ensure that misunder-standings, misrepresentations, misattributions, orsimply bad luck that arises from uncertainty aboutthe interactions among activities are mitigatedthrough structural choices.Hierarchy, in particular,is a mechanism used to provide an integrated eval-uation of alternatives. Senior managers search andevaluate alternatives with their interdependenciesin mind and so are better able to manage such de-cisions (e.g., Eggers &Kaul, 2018; Joseph et al., 2016;Sengul & Gimeno, 2013).

Research on adaptation also differs frompriorworkin that the information problem is one of uncertaintyabout the consequences of interdependencies—thatis, rather than a skill gap or questionable proposalquality. As a result, costs are not directly modeled;information-processing costs are driven by uncer-tainties in resolving interdependencies not by thedirect costs associated with information acquisitionor those associated with communication.

Althougha common theme in the literature concernssearch, it offers a diverse set of decision outcomes in-cluding imitation (Ethiraj, Levinthal, &Roy, 2008), newproduct development (Kotha & Srikanth, 2013; Sosaet al., 2004), radical innovation (Eggers & Kaul, 2018),product exit (Joseph et al., 2016), alliances (Aggarwalet al., 2011), strategic renewal (Albert, Kreutzer, &Lechner, 2015), and learning (Stan & Puranam, 2017),amongothers. Basedon this variety, it is evident thatnoone structure that is best for search or adaptation. Alsolearning from positive feedback does not imply, a for-tiori guarantee, better performance. Because of in-teractions, individuals may not recognize when theconclusions they draw from feedback are inaccurate.Instances of positive feedback may, therefore, lead tosuperstitious learning, flawed decision making, andperformance-impairing behavior with significant long-term consequences (Levinthal & Posen, 2007) and,hence, suggest a role for centralized decision makers.

Overall, the studies in this stream have effectivelydemonstrated how the structural context can suc-cessfully address not only the environment’s fun-damental complexities but also the difficulties thatthis complexity creates for problem-solving anddecision-making. But it also suggests that structure isfundamentally viewed as a coordination tool andany potential frictions when managing interdepen-dencies are not fully articulated or explored.

Cognition

The fourth stream of research broadly concernsthe relationship between organizational structure,

individual cognition, and decision-making. The cog-nition strand draws from Simon’s notion that man-agers bring a set of simplified models to the problemsthey identify, the feedback they receive, the solutionsthey find, and the decisions they make (Gavetti &Levinthal, 2000; Gavetti, Levinthal, & Rivkin, 2005;Simon, 1991). The idea that managerial cognition isconsequential for organizational behavior andstrategyis well established (Walsh, 1995; Gavetti & Levinthal,2000; Eggers & Kaplan, 2013; Helfat & Peteraf, 2015).Yet inmuch (but not all) of the research covered so far,the simplified models used by individuals are eitherimplicit or neglected (Csaszar, 2018). Of central in-terest here is the idea that these models, derived fromthe structural properties of the organization, assistindividuals in categorizing environmental signals,managing uncertainties, and focusing attention. Thisresearch recognizes that organizational structure cre-ates differentiated contexts that lead to distinct re-sponses to environmental information.

Integration through mutual knowledge. Onesubstream considers how firms may create sharedmentalmodels by increasing themutual knowledgebetween individuals, teams, or units, thereby se-verely reducing the need for direct communicationor hierarchical intervention. Postrel (2002) referredto this as trans-specialist knowledge and suggestedthat it may be especially helpful in the face of“glitches” or potential misunderstandings betweentwo units (Hoopes & Postrel, 1999). Mutual knowl-edge is necessary for coordination, especially whenthere is “epistemic interdependence” (Puranamet al., 2012), that is, mutual knowledge allowsagents to predict what other agents will do amidstinterdependencies, thereby making it easier for in-dividuals to make better choices.

For example, a series of studies examining prob-lems of coordinated exploration considered the roleofcommonground—knowledge that isbothsharedandknown to be shared—in creating shared representa-tions. In a study that examined theprocessof offshoringbusiness processes, Srikanth and Puranam (2011)demonstrated that mutually shared knowledge mayreduce theneed for explicit communicationor forplan-based coordination mechanisms even in situationsof complex interdependence. Puranam, Singh, andChaudhuri (2009) found that interdependence moti-vates the structural integration of an acquired firm butthat pre-existing common ground gives acquirers analternative means to integrate. Knudsen and Srikanth(2014) examined the common knowledge provided byindividuals with T-shaped skills (i.e., deep knowledgein one area combined with adequate knowledge in


other domains). Such individuals serve as effective in-tegrationmechanismswhen the goal is exploration; thereason is that they can search for solutions to problemswhile sufficiently accounting for constraints related tohow their choice is likely to interact with other con-straints that a joint solution must satisfy.

In one of the few empirical studies to link repre-sentations and performance, Csaszar and Laureiro-Martınez (2018) used an experimental researchdesign to explore the relationship between in-dividual’s mental representations and the ability topredict strategic outcomes (what they call strategicforesight). They found that strategic foresight isgreater in individuals whose mental representationsare broad and accurately match the consensus. Fur-thermore, by comparing individual and group per-formance, they also found that groups exhibit greaterstrategic foresight than do individuals. But thiswas mainly because of aggregating group members’predictions than because of aggregating theirrepresentations.

Hierarchy and cognition. A second subset ofstudies explicitly incorporate the role of hierarchy.Gavetti (2005) presented a treatment of mental repre-sentations within multiunit organization structures—and of what these structurally driven representationsimply for informationprocessing.Resultsderived fromGavetti’s model revealed that a fundamental driverof organizational search and, hence, of accumulatedcapabilities, ismanagers’ cognitive representations oftheir strategic decision problem. He also showed thattheaccuracyof amanager’s representationsvaries asafunction of that manager’s position in the organiza-tional hierarchy.

Jacobides (2007) also linked hierarchy and cogni-tion ina study that examinedhowGreek’smilitaryanddiplomatic hierarchical structure failed to preventescalating tensions with Turkey. He found that thegovernment hierarchy failed precisely because dif-ferent governmental divisions focused on differentaspects of the problem and overlooked key infor-mation. In addition, the Greek government made noeffort to overturn routine inappropriate responses,and the divisionalization within the governmentfurther led to events being framed in ways thatencouraged—rather than discouraged—escalation.In this case, the hierarchy failed to uniformly framethe escalation and focus attention on key issues.

Joseph and Wilson (2018) explicitly argued thatboth top-down and bottom-up information process-ing help explain how organizationsmove away fromroutine patterns of behavior and sustain the alloca-tion of cognitive resources to new opportunities.

Using examples drawn from Motorola’s entry intocellular technology, they showed that the corporateoffice can not only provide frames for the organiza-tion but can also intervene directly in divisionaldecision-making. It was only through the corporateoffice’s framing and attention-directing efforts thatthe new technology was allowed to grow in onedivision—despite rising opposition to that technol-ogy within another division.

Attention. Several empirical studies have con-sidered the attention patterns afforded by the loca-tion of a manager or unit within a complexorganization. In a study of Korean business groups,for example, Rhee, Ocasio, and Kim (2019) hy-pothesized that business group membership makesgroup-level issues and solutions more cognitivelyaccessible to managers and their subunits. In otherwords, if particular problems or solutions are viewed asgroup-level issues, then they are more easily retrievedfrom memory, and so managers will direct attentionto them when evaluating performance. They foundthat member units engage in more R&D search whenthere are more member firms performing poorly anda greater number of R&D-intensive firms within thebusiness group.

Gabaand Joseph (2013) linked interactions betweencorporate and business units or among businessunits with the emergent properties of responding tofeedback. They argued that because managers ofcorporate units and those of business units focus ondifferent response repertoires, they have differentideas about what constitutes a “local” search for so-lutions. In their study of new product introductions,they found that business units—when respondingto negative performance feedback—tend to focustheir attention on tactical solutions (e.g., revenueenhancements and efficiency improvements toincrease product output); by contrast, the corporateoffice will focus on reallocating resources and mayeven resort to disruptive firm-level reorganizations(which stifle the introduction of new products).

Several related studies draw more directly on theattention-based view (ABV; Ocasio, 1997). As a mod-ern extension of the Carnegie School tradition, theABV is considered to be an information-processingperspective because it views managerial attention asthe organization’s key constraint. Yet the ABV alsohas expanded information-processing perspectives byrecognizing that (a) the distribution of attention withincomplex organizations is not uniform and (b) the rele-vance, interpretation, and use of particular problemsand solutions may vary (in part) with the structuralposition of individual decision makers (Blettner, He,


Hu, & Bettis, 2015; Bouquet & Birkinshaw, 2008; Gaba& Greve, 2018; Gaba & Joseph, 2013; Joseph & Ocasio,2012; Joseph & Wilson, 2018; Rerup, 2009; Tuggle,Sirmon, Reutzel, & Bierman, 2010).

For example, Barreto and Patient (2013) studiedhow deregulation affected frames (problem repre-sentation) of a single firm in the oil and gas industry.They found that a subunit close to that environ-mental shock’s locus is more likely (than is the cor-porate office) to frame the shock as a threat than asan opportunity. In a related work, Dutt and Joseph(2019) explored how corporate structure affects acorporate agenda in the face of regulatory uncer-tainty in the renewable electricity industry. Theyfound that because corporate managers are likely tobe more sensitive than subsidiary managers to thepreferences and intentions of external stakeholdersand they also have more information about the like-lihood of future regulatory changes, corporate man-agers are less likely (than subsidiary managers) toexhibit uncertainty avoidance and will more likelyrespond to regulatory uncertainty by including re-newable sources of electricity on the firm’s agenda.

Another set of empirical studiesmake thenotionofinteractions more central to examine the agendasetting. In a study of the pharmaceutical companyNovo Nordisk, Rerup (2009) found that absent de-liberate interactions between staff at different levelsof the hierarchy, the attention given to particularissues—and the extent to which those issues are in-fused with specific beliefs and meaning—will vary.Rerupdocumented that adopting anewvalue systemwhich shaped beliefs and instituting managementreview sessions (a key information channel) enabledthe organization to direct its attention to key issuesacross the chain of command and to coordinate ac-tivities in response to those issues.

Even loosely coupled organizational forms such ascommunities and crowds, which bring people to-gether in what Dahlander, Lifshitz-Assaf, Piezunka,Seong, and Stroube (2017) labeled “interstitialspaces,”have unique attention-directing properties.These spaces are an “assembly of actors in a sharedonline space (e.g., social media users, open-sourcecontributors, and crowd-funding lenders) or whatemerges among and between these actors as theycontinue to engage in social interaction.” Piezunkaand Dahlander (2015) found that the aggregation ofcrowd contributions narrows organizational atten-tion and that capturing distant knowledge, which isan often mentioned benefit of such organizationalforms, can instead lead the organization to pay moreattention to alternatives that are familiar.

However, interactions are themselves linked in aformal network: issues and initiatives flow throughcommunication channels, and organizational mem-bers participate in a variety of different firm channels.In a series of articles, Joseph and Ocasio (Joseph &Ocasio, 2012; Ocasio & Joseph, 2005) examined theeffect of a system of communication channels on situ-ated attention patterns within the firm and on thecorporate agenda. They found that at GE, senior man-agers’ perceptions did not always dominate within-channel interactions; in fact, business unit managersoften played a leading role in shaping an emergentcollective perspective. They discovered that the firm’scombination of specialized and integrated channelsincreases its ability to adapt because that combinationhighlights key issues and facilitates the agenda settingacross corporate and business unit managers.

Digital information sources (e.g., intranets and socialmedia) are channels of communication throughwhichindividuals may engage in both direct and indirect in-teractions. The latter, as occurswhen apassive receiveris merely “lurking,” offers decision makers whatscholars refer to as communicationvisibility (Leonardi,2014) Such visibility transforms previously invisiblecommunication between organizational members intovisible knowledge about who knows what and aboutwho knows which third-party individuals. In his anal-ysis of a large financial service firm’s social networkingplatform, Leonardi argued that this visibility enablescoworkers to better accommodate new ideas and soshould result in products that are more innovative.Moreover, greater communication visibility may re-duce the organization’s dependence on meetings orliaison positions for increasing direct lateral com-munication among employees or subunits; it mayalso increase “transactive” memory (Ren & Argote,2011)—which has been shown to improve decisionquality—and reduce the duplication of work.

Synthesis. As the cited articles demonstrate, cog-nitive expression is evident in how organizationalstructure—in particular, hierarchy—provides thedecision-making context and shapes the mentalmodels used for decision-making. The role of the or-ganizational structure in this stream of the literature isto mostly constrain the cognition of individual actors.In particular, hierarchy is a mechanism for providingdecision premises and channeling attention for theentirety of the organization. But whereas modeling ar-ticles tend to rely on different initial beliefs or experi-ence, empirical work places greater emphasis ondifferent locations in complex organizations.

Although these studies offer a simplified form ofcognition and rarelymeasure it, individual cognition


TABLE3

Roa

dMap

forFuture

Resea

rch

Key

Issu

esin

Literature

Exp

lanationof

Key

Issu

esFuture

Resea

rchAge

nda

Exa

mple

Question

s

Therole

ofstructure

indecision-m

aking

divided

into

aggreg

ationvs.con

straint

persp

ective

s

Theag

greg

ationview

reflec

tshow

differenttyp

esof

structure

enab

leindividualsto

cometoge

ther

(i.e.,to

interact)for

thepurposeof

mak

ing

collective

dec

isions.

Nee

dgreaterfocu

son

capturingthe

impacto

fstructuralcon

text,through

whichindividual

cogn

itionsare

crea

tedan

dsh

ared

-toinform

orga

nizational

dec

ision-m

aking.

How

doe

stheorganizational

structure

shap

esh

ared

representation

sin

supporto

fdec

ision-m

aking?

Theco

nstraintv

iew

reflec

tshow

the

context

established

bythe

organizational

structure

enab

lesor

constrainsindividual

decision-

mak

ing,

includingasense

forhow

structure

may

affect

heu

risticsan

dbiases.

What

istheprocess

bywhichthesh

ared

representation

sresu

ltingfrom

social

interactions(jointm

eaning-mak

ing)

and

selectiveattention

affect

dec

ision-m

aking?

Disproportion

atefocu

son

structure

asa

sourceof

coordinationrather

than

conflict

Lackof

conflictreflectsassu

mption

sof

uniformityin

interpretation

ofinform

ation.

Nee

dto

better

understan

d:

When

doe

sorga

nizational

structure

yield

conflictinparticu

lardec

ision-m

aking?

Fails

tofullyac

countfor

variationin

attention

patternsof

dec

isionmak

ers

whichmay

contribute

toco

nflict.

Theeffect

ofstructurallyinduced

interpretative

variationof

inform

ationon

decision-m

aking

Isthesourceof

conflicts

tructurallydrive

naccessibilityor

activa

tion

ofparticu

lar

goals?

Roleof

multiplego

alsan

dva

rieg

ated

attention

ondecision-m

aking.

When

issu

chco

nflictb

eneficialfor

orga

nizations?

Unev

entreatm

ento

fvariousstep

sin

the

dec

ision-m

akingproce

ssDisproportion

ateem

phasis

onsearch

andev

aluation(i.e.,dow

nstream

aspects

ofdecision-m

aking).

Nee

dmoreem

phasis

onHow

doe

sthestructuralc

ontext

affect

upstream

aspec

tsof

dec

ision-m

aking?

Lackof

attention

torelation

ship

betw

een

step

sin

thedec

ision-m

akingprocess.

Age

ndasetting

How

dothedow

nstream

aspects

ofdec

ision-

mak

inginteract

withthestructuralc

ontext

toaffect

theupstream

aspects?

Problem

representation

(That

is,

structure’sim

pacto

nupstream

aspects

ofdec

ision-m

aking)


is the key theoretical apparatus that links structureand decision-making. The cognition literature, suchas the adaptation stream, is also similar in its occa-sional reliance on a performance feedback mecha-nism for shaping attention and cognition—mostly insupport of some kind of search, although this is theonly substream in which some consideration ofagendas and problems is evident.

Also, as in much of the empirical work in the ad-aptation stream, the notion of interactions is mostlyimplicit and underdeveloped. Although feedbackstudies recognize such interactions occur, the focusis again on individuals; shared cognitions are ancil-lary to the theory. Also, the common ground notionutilized is more about mutual knowledge than mu-tual understanding (Bechky, 2003). The qualitativeABV-related articles do provide some accounts ofattentional variation, but we need more research link-ing it with the cognitive representations of managers.Because it is within and across channels that boththe cognitive representations and the emergent prop-erties of the attention-directing interactions maycome together, it follows that such channels are aprime subject for future research. Despite this recog-nition that cognitions and attention vary within thefirm, very little of research considers the potentialfor conflict to arise.

TAKING STOCKOF THE EXISTING LITERATURE

Our cross-sectional review of the literature dem-onstrates that the information-processing propertiesof an organizational structure remain a considerableinterest to a diverse group of scholars interestedin decision-making. Research reviewed here hasadvanced our understanding of the information-processing challenges that organizational structureis designed to solve. The review also showed thatdespite the common use of the term “informationprocessing,” these studies describe different waysthrough which structure—in particular, hierarchy—affects different aspects of decision-making.

From these observations, we can trace three criticalissues. First, extant research is divided—implicitlyfocusing on one of two aspects of organizationalstructure and decision-making: aggregation or con-straint. The aggregation view reflects how differenttypes of structure enable individuals to come together(i.e., to interact) for the purpose of making collectivedecisions. This perspective is dominated by thescreening and adaptation literatures, which focus on(respectively) aggregated voting patterns (rules) andchoice sets.

The constraint view reflects how the contextestablished by the organizational structure enablesor constrains individual decision-making, includinga sense for how structure may affect heuristics andbiases. This view includes problem-skill matching,which addresses the limits of specialized knowl-edge; adaptation, which focuses on structurally im-posed limits on search; and cognition, which showsthe constraints that a problem solver’s context(e.g., location) puts on the choice of problems andsolutions. Most constraint studies abstract from howthe organizational structure aggregates managers’choices and so, in effect, discount the effects of in-dividuals, teams, or units interacting.

In neither set of studies is cognition is especiallysocial. Researchers have yet to account fully for theidea that shared representation is a social construc-tion and how organizational structure shapes thecognitive processes underlying shared understand-ing. At root of this issue is that with a few importantexceptions (e.g., Csaszar & Laureiro-Martınez, 2018;Knudsen & Srikanth, 2014; Puranam & Swamy, 2016),what matters most in these studies are the features ofindividuals’ limited attention and simplified mentalrepresentations (see e.g. Csaszar, 2018; Gavetti &Levinthal, 2000) or heuristics (Gavetti et al., 2005).Although there is interest in organization-level im-plications (Knudsen & Srikanth, 2014; Martignoni,Menon, & Siggelkow, 2016; Puranam & Swamy,2016), the focus is on the individual actor—thestrategist, the manager, and the “cognizer”—whoseown perspective (based on mental representations,beliefs, and experience with the local world) offerssome general guidance for making decisions (Csaszar& Eggers, 2013; Eggers & Suh, 2019; Levinthal, 2011).

Second, the studies we have reviewed almostuniversally ignore the potential for conflict.7 Con-flict, according toMarch and Simon (1958: 132), canresult from the need for joint decision-making anddifferences in attention, or differences in perceptionof reality, or both. Underlying this lacuna is the litera-ture’s assumption of uniformity in how information isinterpreted, shared, and stored. Variation in the in-terpretive aspect of information is important because,if a situation is ambiguous, its meaning must oftenbe adjudicated before decision-making can proceed(Denis, Dompierre, Langley, & Rouleau, 2011). Varia-tion in the sharing and storing aspects of information isimportant because it is through socio-cognitive pro-cesses and communication practices by which infor-mation is transformed into abstract representations,

7 See Gulati et al. (2005)


compared with what is in organizational memory,and ultimately used to make decisions.

Yet, as Daft and Lengel (1986: 554) lamented, amajor problem for managers is ambiguous informa-tion, not a lack of data. Ambiguity implies that thereare multiple interpretations of an organizational situ-ation (Feldman, 1989). Ambiguity, unlike uncertainty,cannot be resolved—at least theoretically—with ad-ditional information. In such cases, the information-processing problem is not how best to manageuncertainty via the availability or distribution ofinformation but rather how to align different mean-ings to reduce ambiguity. Given that environmentalcues are often ambiguous (Rerup, 2009; Weick &Sutcliffe, 2015), the implications of interpretive vari-ation and structure’s role in the (mis)alignment ofmeaningsmay, in turn, have ramifications for conflictin decision-making.

We also observe that most studies in our pur-view assume that to deal with the attentional burden(Simon, 1955) and potential discord (Cyert & March,1963) commonly associated with multiple and oftenconflicting goals, attention to goals is selective. In theempirical adaptation and cognition research, this of-ten translates to fairly strong assumptions that (a)managers or subgroups do not jointly consider multi-ple goals and (b) goal prioritization is both plausibleand uncontested. However, these assumptions do notaccount for goal interdependencies and largely statethat tensions will continue to exist among subgoals(March & Simon, 1958) but that tensions will bemanaged through sequential attention (Greve, 2018)or temporal and structural differentiation of goals(Ethiraj & Levinthal, 2009). In the literature onadaptation, for example, theories of performancefeedback propose that attention to goals is sequentialand basedon the need to resolve pressing problemsorto close gaps between performance and aspirations(e.g., Cyert & March, 1963; Gaba & Bhattacharya,2012; Greve, 2008).

Also, there is an assumption that organizationalgoals are largely agreed on before decision-makingand do not change overtime. Only recently, studieshave begun to account for the decision-making out-comes when multiple goals either are difficult toprioritize or offer inconsistent signals on appropriatecourses of action (Gaba & Greve, 2019; Hu & Bettis,2018). Although it is useful to consider conflictinggoals and goal prioritization as a function of theirimportance and of the performance relative to theaspiration level, other possibilities include formationof temporary coalitions in support of different goals oralternatives and environmental links to the distinct

activities in which organizational members are en-gaged (sales, production, and research and develop-ment). Thus, we suggest that any road map shouldredirect research to how elements of the specific sit-uation affect which goals are activated and commandthe attention patterns of decision makers. Doing soshould provide a window into coordination and intoinformation-based conflict.

Third, recall that Simon (1947) articulated foursteps in the problem-solving or decision-makingprocess: setting anagenda, representing theproblem,searching for alternatives, and evaluating alterna-tives. Yetmodeling and empirical studies so far havefocused primarily on search and, to a lesser extent,on alternative evaluation. This gap originates fromthe fact that much of the prior research presents de-cisionmakers as “modelers” or “updaters” and takesa relatively passive approach to cognition (Gavetti,2005: 614). However, the notion of interpretive var-iation and variegated attention posits a more activeapproach to cognition and decision makers as“interacting selective shapers” who may guide theselection, retention, and reconstitution of such in-formation. As a result, our approach should be es-pecially helpful in providing a window into theagenda setting and problem formulation.

The dearth of research on these aspects of decision-making has implications for our road map. The“downstream” aspects of decision-making (i.e., searchand evaluation) are largely determined by the agendaand problem representation, as they are defined bythe organizational structure—and more specificallysubgroup membership. For a problem or opportunitytobeaddressed, itmustbeon theagenda, andasSimon(1947: 124) argued, “different representations of theproblem will produce different proposals for solu-tions.”Without proper theory about the information-processing mechanisms of the agenda setting andproblem representation, we cannot fully articulate atheory of decision-making.

Also, there is recursive relationship betweensearch and evaluation and subsequent agendas andproblem representations; that is, as certain solutionsbecome familiar, they are more likely to shape theagenda itself (Ocasio & Joseph, 2005). For example,the outcomes of search and evaluation may impactsubsequent goals chosen to pursue (rather than justtheir aspiration levels). Thus, the search and evalu-ation space inwhich the organization explores is notonly bounded by the agenda and problem represen-tation, but there is a process at work wherebysuccessful initiatives emerging from search andevaluation reinforce the existing cognitive models


and attention patterns and so narrow the agenda andthe way the problems are represented.

Because some shared cognitive models used forinterpreting information are more accessible andmore frequently activated than others, it follows thatinformation processing can be viewed as an ecology.8

However, in the focal literature, aggregation is accu-mulative, and cognitions do not vary even with theirdiffusion. Although this approach might be reason-ably accurate if one assumes that higher level aggre-gates truly reflect of lower level elements (Simon,1947; Puranam, 2018), they may not fully reflect theidea that attention is selective and that shared repre-sentations result from social interactions.

A ROAD MAP FOR INTEGRATING ANDADVANCING FUTURE RESEARCH

In this section, we offer a perspective that empha-sizes opportunities for integration and promotes amore comprehensive view of the role of structure indecision-making. To do so, we incorporate researchon embedded agency (Seo & Creed, 2002) and atten-tion in the organizational theory (Ocasio, 1997) aswell as work on social cognition, which argues thatthe availability and accessibility of shared represen-tations and their activation is predictive of individ-uals’ actions (cf. Higgins, 1996). In this endeavor, weemphasizehoworganization’s structural context bothembeds (in the institutional and organizational envi-ronments) and socially situates (in a particular timeand place) decision makers and their collective in-teractions.9 Our perspective emphasizing structure’srole in shaping collective interactions (joint meaningmaking) and selective attention introduces the possi-bility of information-based conflict. It also suggeststhat organizational structure may play especially im-portant role in setting agendas and representing prob-lems. See Table 3 for a summary of key issues in theliterature and our Roadmap.

Embedded and Situated Interactions

Embedded interactions. Embedded interactionsdraw on the embedded agency model of human

behavior (Seo&Creed, 2002),whichholds that socialaction is embedded in the institutional environmentand that shared representations (e.g., cognitive frames,cultural categories, and vocabulary structures) aredecision-making resources expressed in sets, toolkits,or repertoires (cf. Giorgi, Lockwood, & Glynn, 2015).In our model, the external environment makes avail-able sharedrepresentationsaswell as their expressionthrough organizational structure (Simon, 1947: 110).The organizational structure’s socio-cultural proper-ties reflect the shared representations that are acces-sible to decision makers.10 Organizational structureserves as a key mechanism for assembling the reper-toire and establishes that not all available represen-tations in the environment are readily accessible. Forinstance, the language associated with a particularspecialization is a function of the shared representa-tions associated with the social systems in whichindividuals are interacting.11 An engineering de-partment in a cloud services firmmay reflect both anengineering culture (Kunda, 2009) and the repertoiresassociated with the firm’s cloud services division.

The canonical example is the multidivisional firm.The environments and industries in which an multi-divisional firm operates make certain resources cog-nitively available. Because the multidivisional firmoperates at the confluence of multiple industries, itserves as a mechanism to make accessible thoseframes, categories, and vocabularies supplied by eachof its constituent industries (Harrison & Corley, 2011;Dalpiaz, Rindova, & Ravasi, 2016).12 Research hasestablished that firms may proactively integratethese resources to formulate strategies and allocate

8 We thank Willie Ocasio for making this distinctionclear.

9 Establishing a link between embeddedness and sit-uatedness to account for organizational behavior is animportant development in the institutional theory (cf.Thornton et al., 2012); however, it has not yet been appliedby studies to information processing and decision-making.

10 The notionof embeddedbehavior or embedded agencyis a subject important to sociologists (e.g.,Granovetter, 1985)and organization theorists (cf. Thorton et al., 2012). Em-bedded behavior which implies individual agency, albeitsubject to constraints is neither under- nor over-socialized.

11 Simon (1952: 1,138) put it this way: “It is an importantquestion as to how far specialization is determined by con-straints external to the organization . . . and how far it is de-terminedby internal constraints—[that is, ]thepsychologicaland sociological limitations upon rational adaptation.”

12 This set of resources is referred to as an “industryregister.” Weber (2005) found that the pharmaceuticalindustry’s register includes action strategies (e.g., productdevelopment and divestiture) and frames that give the or-ganization a means to view itself and its potential actions.In Rindova’s (2011) analysis of Alessi, a design-forwardproducer of high-end kitchen and bathroom products, hefound that the company was at the confluence of severaldifferent register components: arts, crafts, anthropology,and psychoanalysis.


economic resources (Weber, 2005), build support fornew agendas (Rao & Giorgi, 2006), develop newproducts (Rindova, Dalpiaz, & Ravasi, 2011), or adoptnew practices (Ocasio & Joseph, 2005).

Firms may undertake this integration organically—that is, by creating a new division that creates thenecessary repertoires from corresponding industriesor logics (Gaba&Meyer, 2008; Perkmann,McKelvey,& Phillips, 2018)—or itmay acquire another firm thatis already integrated in thatway (e.g., Canato, Ravasi,& Phillips, 2013). Firms may also do this throughhierarchy or specialization. For example, the hier-archymay impose new frames or vocabularies on thefirm; on the other hand, it might strengthen theexisting complementarities among them (Bertels,Howard-Grenville & Pek, 2016; Raffaelli, Glynn, &Tushman, 2019).

Situated interactions. Situated interactions en-compass social interactions among organizational

members that transform individual acts of meaningconstruction into collective ones. Situated in-teractions reflect the idea that interactions occurwithin a particular social context (Nisbett & Ross,1991). Formal organizational structure provides thecontexts within which any of the various frames,categories, and vocabularies can be activated,thereby shaping the meaning construction of in-coming information. Adopting this approach ac-knowledges the interactive and situated nature ofcognition (Elsbach, Barr, & Hargadon, 2005) andsuggests that activation occurs not at the level of theindividual but rather at the level of the interaction,(Soderstrom & Weber, 2019).

In our approach, there is no assumption that cogni-tion is necessarily internally coherent. So in this way,our view deviates somewhat from the Simon’s strongnotion that the boundaries of common maps—andcorresponding subgroup membership—are fully

FIGURE 1Ecological View of Information Processing

Individual

cognition

Situated

interactions

Institutional, industry, technological environment

Organizational structure

Conflict Coordination

Shared representations

Activation of frames,categories, vocabularies

Accessibility of frames, categories, vocabularies

Agendasetting

Problemrepresentation

Search Evaluation

Availability of frames, categories, vocabularies

Decision making


“reified” (Simon, 1952; March & Simon, 1958). Ourapproach recognizes that organizational membershave multiple—and often loosely coupled or evencontradictory—representations derived from thelogics of internal and external memberships, tech-nologies, and industries (Thornton, Ocasio, &Lounsbury, 2012). Moreover, it is within communi-cation channels (formal and informal) that socialinteractions occur and that individuals (jointly andselectively) attend to information. The activation ofparticular frames, categories, and vocabularies re-sults from a combination of channel characteristics(Ocasio & Joseph, 2008) and the communicationpractices occurring within that channel (Ocasio,Laamanen, & Vaara, 2017). In other words, sharedcognition arises from the regularizedwithin-channel“social interaction that builds on speech, gestures,texts, discourses, and other means” (Cornelissen,Durand, Fiss, Lammers, & Vaara, 2015: 11). Actorsmay flexibly draw on a repertoire of shared cognitiverepresentations when constructing their “strategiesof action” (Swidler, 1986; Weber, 2005; Weber &Dacin, 2011) and do so when interacting in decisionsituations. For example, some categories or vocabu-laries may be collectively attended to more thanothers. Certain ones may be altered so as to conformto the particular properties of the situation in whichthey are introduced (Cornelissen & Durand, 2012;Durand &Paolella, 2013); othersmay simply fall intodisuse. It is, therefore, the organizational structurethat yields the ecological space in which the learn-ing, comparing, blending, or diffusion of sharedrepresentations occurs. See Figure 1 for a stylizedversion of our perspective.

Integrating Aggregation and Constraint Viewsof Structure

Ourapproachaddresses theaggregation–constraintdichotomy because it explains how structure simul-taneously brings together and constrains individualdecision makers, that is, it incorporates the idea thatindividual cognitions are shared and that creationof shared representations can act as a constraint ondecision-making. That way, we move from distinctaggregation and constraint perspectives to an ap-proach that links aspects of both.

Our argument recalls Simon’s (1947) argumentthat humanbehavior is not just boundedly rational; itis intendedly rational or what March (1978: 590)called “behavior with constraints.” The notion ofintended rationality recognizes that informationprocessing is shaped not only by the limits of

individual attention and heuristics (Simon, 1947;Newell & Simon, 1972) but also by common maps.Thesecommonmaps serveas aperceptualmechanismthat supplies interacting individuals with the “socialdefinition of a situation” (Simon, 1952). As notedabove, they include such shared cognitive represen-tations as schemas, frames, categories, classifications,systems of concepts, and vocabularies (March &Simon, 1958, 1993: 184–86). Simon notably arguedthat it is only to the “extent that suchmaps areheld incommon, [that] they must be counted among the in-ternal constraints on rational adaptation” (1952:1135), suggesting that the shared aspect of thesemaps is especially important.

Our perspective suggests that the organizationalstructure makes accessible certain common maps orshared representations from the variety available inthe environment. The structure also shapes the socialinteractions which activate (or draw attention to)particular representations during collective decision-making. The resulting conflict and/or coordinationyielding particular decisions, and the correspondingshared representations and attention patterns whichled to them are reinforced (making themmore acces-sible overtime). Thus, overtime, the organizationalstructure guides the selection, alteration, and re-tention of particular frames, categories, and vocabu-laries. And so our perspective may help usunderstand the structure’s impact on informationprocessing, not as an aggregation/summation processor that of individual constraint, but as an ecology.

Although embedded and situated interactions areneither uniformly nor explicitly conveyed in mostprior research, our review indicates widespreadagreement concerning this claim: it is the joint in-teractions and shared meanings created within aparticular structural context that generate the atten-tion patterns necessary to coordinate activities.Thus, our call is for a more integrated approach thatlinks (a) the common maps that the organizationalstructure makes accessible with (b) the social inter-actions among organizational members, where par-ticular aspects of these maps are activated to processinformation in support of decision-making.

Informational Sources of Conflict

Our perspective offers an avenue for understand-ing how organizational discord manifests and whenconflict over differences in interpretation of infor-mation and attention to goals can impede or improveorganizational decision-making. Here, again we ar-ticulate a role for interactions that are situated and


embedded—in this case, linking organizational struc-ture and conflict in decision-making.

Differences in information interpretation. Wemust bear in mind that information-processing prob-lems are driven by both uncertainty and ambiguousinformation. Recall from our literature review thatdecision-making may be clouded by various situa-tional aspects, which include the tacit nature ofknowledge (Garicano & Wu, 2012), the variablequality of proposals (Csaszar, 2013b), and the com-plexity of the organization (Baumann & Siggelkow,2011) andof the environment (Greenwood,Dıaz, Li, &Lorente, 2010). That information is open to multipleinterpretations is consistentwith the theorizing aboutmechanisms discussed in the feedback and cognitionstreams, where these concerns are more central.

Because decision makers must construct meaningin the presence of ambiguous information, it followsthat the situated and embedded nature of interac-tions will figure prominently in how information isinterpreted. The potential variation in frames, cate-gories, and vocabularies and their activation in situ-ateddecision opportunities leave open the possibilityfor different applications of them in decision-making.Interpretations may diverge either unintentionally orintentionally.

For example, the specialization that results fromstructurally segregated interactions usually leadsan organization’s members to hold different mentalmodels, which, in turn, results in a natural divergentinterpretation of issues. Variations in perceptionsmay fuel debate concerning the best course of actionin response to feedback (Kaplan, 2008) and mayprovidemanagers thechance to“self-enhance” (Jordan& Audia, 2012) through over-favorable interpretationof feedback (Joseph & Gaba, 2015). Divergent in-terpretations may lead to disagreements about thebest course of action or the evaluation of alternatives.For example, it might shape whether new opportu-nities are viewed as threats or opportunities (Gilbert,2005). It may also lead to inaction as organizationalmembers continually undo or reverse decisions al-ready made (Denis et al., 2011).

Organizational members may also purposely dis-tort informationwhich can also amplify the potentialfor conflict in decision-making. As Cyert and March(1992: 67) emphasized, one can expect the informa-tion transmitted among subunits to exhibit somebias, and there may be attempts to manipulate in-formation toward the endof alteringdecisions.A fewstudies (e.g., Fang et al., 2014; Reitzig &Maciejovsky,2015) have explicitly examined the distortion of in-formation inhierarchies, butmore research isneeded.

Most of these studies examine individual distortionand not the possibility that groups of interacting in-dividuals might deliberately distort information tomaintain extant cognitive models. The resulting dy-namic could, in turn, create opportunities for pro-cessing and communicating information inways thatpolarize and support parochial interests and agendas.

In the cases of both inadvertent and intentionaldivergence, the potential for intraorganizationalconflict over problems and solutions will increase.For instance, disagreements among subunits aboutthe firm’s agenda may create conflict related to re-source allocation and control (i.e., autonomy). Re-search has shown that subunits often compete overnew domains (Galunic & Eisenhardt, 2001), espe-cially when the domain is seen as encompassingproblems related to—and offering solutions of rele-vance to—their current operations (Birkinshaw &Lingblad, 2005).

Scholars have described the potential negativeeffects of such conflict, which include reducedinformation-processing efficiency and increased dif-ficulties with coordination (Birkinshaw & Lingblad,2005). That said, benefits have also been observed:such competition may encourage subunits to exploitexisting resources more completely and to developnew resources more thoroughly, to reduce the “timetomarket” for newproducts and to increase the firm’soverall market coverage (Joseph & Wilson, 2018;Bauman et al. 2018). We conclude that more work isneeded to sort out the positive and negative effects ofconflict, whichwould enable a deeper understandingof its informational drivers in the context of multiple(competing) goals and divergent interpretations.

Onepromising line of inquirywouldbe to examinethe source of interpretive differences—whether it isone because of limits of accessibility or one of acti-vation. Recall from our framework that accessibilitydepends on the environment in which the individ-ual is embedded. A stable organizational structure(subgroup membership) reflects the situation’s so-cial definition; this means that only a subset of rep-resentations may be cognitively accessible todecision makers. Gaba and Joseph (2013), who rec-ognized that the corporate office and business unitshave access to different repertoireswhen respondingto performance feedback,offered an illustration.

Activation depends on the situation. Not all pro-posals, information, or feedback receive the samequality of attention, and selective attention to issuesdepends on the decision situation. Within such sit-uations, a variety of factors are at play: the decisionmakers’ previous experiences (Eggers & Kaul, 2018;


Gaba, Lee, Meyer-Doyle, & Zhao, 2019), rules in-voked at the time of the decision (Knudsen &Levinthal, 2007; Csaszar, 2012, 2013a, b), the com-mon ground on which decision makers agree(Puranam et al., 2015), communication acts (Ocasio,Loewenstein, & Nigam, 2015), and other materialproperties of the channels through which decisionsaremade (Joseph&Ocasio, 2012). Someprogress hasbeen made in understanding such attention quality(e.g., Rerup, 2009), but more research is needed toconnect it with structure and decision-making.

For instance, adaptation studies couldmake sharedrepresentations, and the active sharing of those rep-resentations an important parameter in theirmodels.Cognition studies should focus more closely on theactual interactions that shape, for instance, socialcomparisons and responses to feedback. Along theselines, Vissa, Greve, and Chen (2010), and Rhee et al.(2019) each acknowledged the cognitive accessibil-ity provided by interactions between the subunits ofa multidivisional firm; however, none of those arti-cles captures the dynamics of interactions. BothJacobides (2007) and Joseph and Wilson (2018) of-fered some evidence for hierarchy’s role in the pro-vision of frames, but neither study detailed thecircumstances of their activation.

Differences in attention. Organizational scholarshave long noted the attention problems arising fromthe pursuit of multiple goals as well as the conflictthat follows from that goal diversity (e.g., Cohen,1984; Ethiraj & Levinthal, 2009; Greve & Gaba, 2017;March & Simon, 1958; Perrow, 1961; Simon, 1964).Conflict is likely to arise when individuals indifferent roles and situated indifferent decision-makingstructures pursue goals linked to those positions and,therefore, process different pieces of information in thepursuit of those goals. Interdependences and complexlinkages in the task environment can further contributeto a general sense of uncertainty, which not only in-troduces coordination challenges but also creates latentor overt conflict within the organization. For example,organizational subunits with distinct functions are ex-pectedtodeveloptheirownobjectivesandnormsandtocompete for scarce resourceswith other units, althoughthey must cooperate in support of decisions. Many oftheseobjectivesareassumedtobeessential, continuous,and operative, which means that they can poseproblems—in the form of potential conflict—for theorganization.

As mentioned previously, the sharing of task en-vironments entails that actions taken in pursuit ofone goal directly affect the organization’s perfor-mance vis-a-vis other goals. And as the number of

technological goals increases, sowill the complexityof assigning credit to the individual or team re-sponsible for a single goal or component. Thus, thelimitations imposed by a shared task environmenthave the potential to cause conflicts and to impedethe achievement of multiple goals. For example, Huand Bettis (2018) found that feedback interdepen-dency across multiple technological goals (e.g.,speed, fuel efficiency, and reliability) in automobilemanufacturing can lead to severe and misleadingconfusion about learning from the feedback related topursuing thosegoals.GabaandGreve (2019)examinedairlines’ dual focus on safety and profitability as twohigh priority goals on the decisions regarding fleetchanges. They argued and found that the pursuit ofsafety goals cannot be understood in isolation fromprofitability goals, and in fact, responsiveness to safetygoals is strengthened by low profitability becausesafety is associated more closely with survival. Thesestudies are important, in that theymoredirectly examinethe implications of goal conflict and technologicalinterdependencies for decision-making. Still, theprocess throughwhichdecisionmakersmay addressgoal conflict and more broadly interdependenciesamong multiple goals remains an overlooked areaof research.

Our approach suggests that decisionmakers coulddeal with multiple goals in a different manner. Froma situated and embedded perspective, goals are notnecessarily prioritized or agreed on before the con-sideration of alternatives; rather, they are drawnfrom a pool of existing goals as the decision-makingprocess proceeds. Thus, our approachmakes centralthe concern of when multiple goals are made avail-able, accessible, and activated. From this perspective,managers are aware of constellations of organiza-tional goals, yet only a subset of those goals are acti-vated through interactions among agents.

Future research may need to examine cases wheredecision makers and their corresponding subunitsare embedded in different institutional environmentsand, thus, could face a variety of goals. Such “in-stitutional complexity” exists when the firm simulta-neously pursues various goals prescribed by differentindustry logics (Greenwood et al., 2010) and can bespread by the diffusion of rating and ranking systemsas externally imposed goals (Rowley, Shipilov, &Greve, 2017). For instance, it is implicitly assumedthat growth and size goals are central to large publiclytraded corporations—just as safety goals are central toplayers in the airline industry (Gaba & Greve, 2019).The resulting complexity creates competing demandson organizational decision makers.


Whereas the problem-skill matching and screen-ing literature do not directly consider goals per se,the adaptation literature relies primarily on goal ac-tivation through feedback. However, our approachoffers an important alternative under which organi-zational goals are contested overall and in specificdecisions. Questions for future research include—inlight of the competition for managerial attention(Ocasio, 1997)—why some goals receive more atten-tion thanothers andwhether there are somegoals thatcan be ignored (Cyert &March, 1963). Goal activationreflects our proposition that throughout a decision-making process, the organizationwill attend to only asubset of its goals, thereby increasing the likelihoodofthose goals being satisfied (Simon, 1964).

It follows that goal activation may be as much aprocess for prioritizing appropriate goals as forfindingways to achieve them. In this sense, goals canbe viewed as frames andmay be used accordingly insupport of selling the importance of issues to moresenior managers (Dutton & Ashford, 1993), for pur-poses of “sensegiving” to others in the organization(Gioia & Chittipeddi, 1991; Sonenshein, 2010), tobuild coalitions and recruit allies through recogni-tion of shared interests (Zhang & Greve, 2019), toexert political influence (Kaplan 2008; Kaplan &Tripsas 2008), and to engage in the symbolic ma-nipulation of information (Elsbach & Sutton, 1992).Thus, organizational goals are a function of socialinteractions and can be deployed as needed in thedecision-making process.

By extension, much of the literature implicitlytreats attention to goals and their consequentiality fordecisions as invariably linked, although theymay notbe. Theories of “loose coupling” suggest that organi-zationalgoalsneednotaffectdecision-makingdespitebeing used to justify action (Cohen, March, & Olsen,1972; Weick, 1976) rather than provide an explana-tion for purposeful decision-making (Eisenhardt &Zbaracki, 1992). However, no adequate account hasbeen given of exactly how loose coupling explainswhich of multiple organizational goals are conse-quential and under what circumstances. Loose cou-pling also highlights the importance of goals asmotivators and of undertaking alternative evaluations(Keum&Eggers, 2018) andsuggests theneed for amorenuanced understanding of when failure to achieve agoal leads to goal activation, which may not occuruniformly (Gaba & Greve, 2019; Rowley et al., 2017).

Perhapsmost importantly, loose coupling confirmsthe necessity of paying close attention to the role oforganizational structure. The formal hierarchy mayplay a critical role in the matching of organizational

goals and subunit divisions (Galunic & Eisenhardt,2001). In particular, interactions between the corpo-rate office and business units may drive empiricalregularities in the relationships among interactions,goal activation, and decision-making (Gaba & Joseph,2013; Joseph & Wilson, 2018). Our formulation—whereby decision makers can choose which con-straints to satisfy—stands in contrast to other models,most of which assume that goals are either inconse-quential or serve as hard constraints. It would beworthwhile to examine whether any commonly heldgoals serve as implicit constraints ondecision-making.If so, then we should also like to know which of thesegoals emerge from a goal activation process.When oneconsiders theubiquity of goal conflict inorganizations,it is clear that more work is needed to understand thesubtle connections between structure, attention, anddecision-making as well as the trade-offs via whichsuch conflicts are resolved in organizations.

Agenda Setting and Problem Representation

Our viewmay be especially helpful in articulatingthe overlooked aspects of Simon’s decision-makingprocess—the intensive cognitive process of theagenda setting and problem definition and its im-portance for strategy making (Nickerson, Yen, &Mahoney, 2012; Nickerson&Zenger, 2004), problemsolving, and search (Posen et al., 2018). In recogniz-ing the particular aspects of common maps that arelikely to be accessible and activated, we may betterpredict how certain agendas or problem representa-tions may come to dominate within an organization.Because our view transcends Simon’s conception ofreified groups, it acknowledges the potential for dy-namic changes to the commonmaps used, that is, wemight see a recursive relationship between thesearch and evaluation of alternatives and the sub-sequent common maps used in decision-making.The solutions chosen may be recognized as adecision-making pattern (Mintzberg, 1979) and,hence, come to serve as an input to the firm’s agenda.

Successive use overtime may reinforce the selec-tion (and preferred use) of particular categories orvocabularies with those who share similar percep-tual references.Whencertain representations appearfrequently in a single decision situation or acrossmany decision situations, those representations areselectively retained and become an enduring part ofthe system’s information structure.

Those aspects that are frequently activated throughinteractions become more broadly diffused overtime—even as other aspects remain unattended (Ocasio &


Joseph, 2005).13 Throughout this attention selectionprocess, information may be reinterpreted (Ocasioet al., 2015). Thus, differences in the shared represen-tations used within the organization derive from theavailability of various frames, categories, and vocabu-laries within a particular subunit as well as from theemphasiswithwhichelements are activated across thebroader organization (Weber & Glynn, 2006).

Overall, then, an embedded and situated perspec-tive of interactions suggests an ecology whereby theprocessing of information—and the correspondingshared representations that allowsorganizations todoso—follows an evolutionary pattern. As the structureaggregates, some aspects of these representations areaccessible, attended to, and altered in meaning. Se-lective attention results in the retaining of certain in-formation and of the shared representations used tointerpret it.

In pursuing such a program, researchers shouldgive greater emphasis to the content of shared rep-resentations. In particular, they should explore howcommon maps are socially constructed, reinforced,and altered through the situated interactions guidedby the organizational structure—in particular, thefirm’s formal hierarchy. It may also be worth exam-ining how attention may be apportioned among thetasks of setting agendas, formulating problems, andsearching and evaluating alternatives. Little is knownabout the hierarchical properties of such systems.Moreover, hierarchy in the decision-making processmay themselves reflect different stages in the ecologyof information processing (e.g., agenda setting at thetop, problem formulation in the middle, and search-ing andevaluatingat thebottom—which, in turn,mayhave implications for the agenda setting again at thetop). This evolutionary reconstitution of a problemrepresentation and the firm’s agenda is a fruitful av-enue for future research, one that may inform a moredynamic version of Simon’s social definition of thesituation.

Methodologies

We encourage the use of a variety of methods toexplore these various relationships. One promisingoption for expanding the scopeof research in this fieldis to conduct more individual-level experiments ininformation processing (Puranam, 2012). A good

example of such work is the article by Turner andMakhija (2012), who examined the relationshipsamong organizational structure, individual informa-tion processing, and problem-solving. They took anexperimental approach to examining how “organic”and “mechanistic” structures affect the way individ-uals gather, interpret, and synthesize information—and on how those structures affect individuals’problem-solving orientation.

The perspective we have outlined would alsobenefit greatly from case studies of organizations. Asa result of such qualitative approaches, research onculture and organizations (e.g., Rindova et al., 2011)have made great strides in understanding the impli-cations of shared cognition and cultural resources,although their focus is less on organizational struc-ture than on decision-making. However, such anapproach is needed if we are to understand the nu-ances of interactions, the nature of the evolutionaryprocess of attention and interpretation, and theconsequences for behavior (e.g., agenda setting andproblem formulation) of a less public nature. An il-lustrative example is Valentines’s (2019) study,which examines the use of artificial intelligence (AI)at an online clothing retailer. The algorithm used bythe AI group created new categories into which theretailer organized its activities, categories that werenew and that crossed the previous division of labor.Valentine found that the centralization of an AI unitwithin the organization had two notable effects: itcentralized decisions about product portfolios andthreatened the role of planners at the product level,flattening the organization’s hierarchy. In essence,the problems’ definitions changed. More studies inthis vein are in order because we must learn howsuch technology can alter the agenda of a firm byaffecting its structure.

Yet another option is to adopt a micro-structural ap-proach to information-processing research (Puranam,2018)—that is, focusing on the “micro-structures—small groups of interacting individuals who are thecruxof thedesignordecisionproblemtheorganizationfaces.” Several different studies adopt this approach,including laboratory experiments (Raveendran,Puranam, & Warglien, 2015) and agent-based models(Csaszar & Eggers, 2013). The advantage of such re-search over studies that adopt a more macro-structural approach (e.g., Burton & Obel, 1984) isthat decision makers are proximate to the decisionsthey make, which makes it easier to identify causallinkages between structural properties and the vari-ous problem-solving steps. Moreover, if one as-sumes that subsystems exhibit the same pattern of

13 The interactions may also reflect an ecology. AsRivkin and Siggelkow (2007: 1084) pointed out. “Patternsthat improve ‘searchability’ may very well prevail in eco-logical competition among interaction patterns.”


relationships as the systems they constitute, then themicro-structural approach can be applied at any hi-erarchical level in the organization through carefulapplication of aggregation principles such as scalingand recursion (Puranam, 2018).

Finally, recent advances such as big data, machinelearning, and natural language-processing methodol-ogies offer substantial opportunities to more directlycapture the decision-making implicationsof commonmaps. As researchers strive to model more complexorganizations and, in particular, the language (sharedcognitions) used in those organizations, new topicmodeling and text analysis techniques are leading to anew stream of research which can account for theemergence and performance consequences of culture(Srivastava, Goldberg, Manian, & Potts, 2017). Forexample, Corritore, Goldberg, & Srivastava 2019) usecomputational linguistics tools toderive time-varyingmeasures of interpersonal and intrapersonal culturalheterogeneity. They further demonstrate that inter-personal heterogeneity—the extent to which organi-zational members diverge in their understanding offirm culture—is negatively associated with effectivecoordination and execution, whereas intrapersonalheterogeneity—the breadth of cultural beliefs aboutthe organization that are held by members—is posi-tively linked to creativity and the capacity forrecombinant innovation. With these new tools, it isnow possible to explore systematically trade-offs andperformance implications of a variety of organiza-tional structures.

CONCLUSION

Our review of the literature offers both a retro-spective and a road map for new avenues of inquiry.Our efforts identified four streams of structure anddecision-making research: problem-skill matching,screening, adaptation, and cognition. We also iden-tified several limitations andproposedopportunitiesto advance our understanding of structure’s infor-mation processing properties.

First, the bifurcation of the literature into aggrega-tion and constraint perspectives highlights the focuson individual cognition and absence of shared rep-resentations and common maps; as a result, currentdecision-making theories do not fully capture thespirit of Simon’s social definition of the situation andits attendant constraints. Second, the literature haslargely abstracted from interpretive and attentionalprocesses and so has not sufficiently addressed thepotential causes and consequences of intraorganiza-tional conflict. Third, we apply this process to

less studied aspects of decision-making. Accountingfor a closer correspondence between upstream anddownstreamsteps indecision-making and a recursiverelationship between evaluation and agendas, ourapproach reconceives information processing as anecological rather than a summation or combinatorialprocess. Therefore, to advance our understanding oforganizational structure and decision-making, weoutline an evolutionary process of embedded andsituated interactions within the organization.

Through this, we (and other researchers) recognizethat the notion of “interactions” is changing and thatadvances in information processing have led tochanges in the informational regularities withincomplex systems (Van Knippenberg et al., 2015). The“empty world” hypothesis states that reality can beadequately described by accounting for only a smallfraction of the possible interactions, that is, becausethere are only weak connections among most of ourworld’s constituents (Simon, 1962). Yet although ourreality may reflect an empty world, it has evolved interms of what is loosely versus tightly coupled(Bromley & Powell, 2012). It is, therefore, incumbenton organizational scholars to revisit the information-processing properties of complex systems, whichinclude familiar (yet changing) organizational struc-tures and new organizational forms. So despite theprominence, for example, of large multidivisionalorganizations in world economies, there has been aresurgence in the use of functional organizationalstructures and flatter hierarchies. Witness the rise ofthe holocracy, an organizational form without a for-mal hierarchy, job titles, or job descriptions (Puranam& Hakonsson, 2015) and that will almost certainlyrequire novel ways of integrating agents and theiractivities within and across firms.

From our perspective, one research implication ofthis approach is that organizations with businessmodels that span multiple industries, blur industryboundaries, or require highly coupled activitiesshould be especially adept at marshaling and in-tegrating various cultural resources (e.g., in support ofsearch). The combination of institutional complexityand decreasing near-decomposability in ecosystemsand complex businessmodels suggests that firmswillhave amore often andmore urgent need to adjudicateexternal demands. Hence, it is important for futureresearch to consider both embedded and situated in-teractions in decision-making.

We can also observe the increased adoption ofplatforms, ecosystems, and crowds that is meant tohelp solve organizational design problems. Each ofthese approaches reflects a type ofmeta-organization


that encompasses many corporations, communities,or individuals linked not by contracts but rather bytechnology and/or a common goal (Gulati et al.,2012). These new organizational forms clearly differfrom a bureaucratic hierarchy. Many such forms are“new” in the sense that they address core designconcerns by translating—in anovelway—individualefforts into collective action (Puranam, Alexy, &Reitzig, 2014).

Yet meta-organizational forms are distinct notonly from the traditional bureaucratic hierarchy butalso from other organizational forms and relation-ships (Kapoor, 2018). Instead of exhibiting verticalintegration or sequential interdependence, they re-flect systems of “epistemic” interdependence(Puranam et al., 2012), unprecedented availability ofdata (Van Knippenberg et al., 2015), and the hyper-specialization of agents (Malone, Laubacher, &Johns, 2011). These differences add considerablecomplexity both to the shared representations onwhich the firm may draw and to the types of in-teractions that may occur between actors. At thesame time, these diverse features remind us not toassume too much coherence in common values andassumptions (Schein, 2004; Van den Steen, 2010).

As part of these changes, information processingis affected by the growth of social networking sitesas sources of ambient information. Experts inthis field argue that digital information sources(e.g., intranets and social networking) offer decisionmakers access to meta-knowledge. Although the no-tion of ambient information is a recent topic of re-search interest, it may offer information-processingscholars an entirely new lens through which theycan examine arenas for the activation of particularrepresentations.

An area worth mentioning again is the impactof artificial intelligence on decision-making. Mostwould argue that AI should reduce the information-processing requirements demanded of individualsand increase the firm’s overall information capacity.Such advances may result in artificial neural net-works learning to solve complex problems—therebyrealizing Simon’s early efforts in this domain(Newell & Simon, 1972). Of course, AI increases thepotential for more rapid cognitive adjustments,thanks to its nearly instantaneous analysis of BigData. These developments raise difficult questions,however. Do such fundamental changes in technol-ogy, when combined with novel organizationalforms, presage (cf. Puranam, Shrestha, He, & vonKrogh, 2018) something fundamentally new for or-ganizational decision-making? Do these converging

trends suggest that the sun is now setting on theWeberian bureaucracy? And do they, perhaps, her-ald the dawn of a new paradigm? Explicating thecomplex relationship between AI-assisted informa-tion processing and decision-making requires thatwe understand how the organization and its chosentechnology sort through the voluminous informationacquired. Attempting to process all available infor-mation can, paradoxically, result in the firm resortingto a narrower focus (Piezunka & Dahlander, 2015;Sullivan, 2010) if itdoesnot lead tocognitiveoverload(Castellaneta & Zollo, 2014; Laamanen, Maula,Kajanto, & Kunnas, 2018). Therefore, another oppor-tunity for research involves the quality of attentionthat canbedevoted to information inaBigDataworld.

Finally, our ecological approach also suggests thatneither the source nor quantity of information issufficient for understanding how it is processed.Although far more data are available now than be-fore, the common maps that aid in their interpreta-tion and attention are far from unlimited and mayeven bemore constrained than before. We are awashin a sea of information, yet are limited, in that newshared models by which we can navigate it have yetto emerge. This calls for new theories in our field.

In any event, changes in the organizational formand advancements in technology are certainly in-dicative of a shift in informationprocessing, and theystrongly imply a need to the advance current theoryand means of analysis. As certain aspects of organi-zations (e.g., business models) become more tightlycoupled, it may well become more difficult to iden-tify the optimal combination of design choices forachieving desired outcomes. Given that organiza-tions have become increasingly characterized bydistributed decision-making (i.e., across ecosystems,platforms, or communities), we shall require a morecomplete understanding of how organizations canadapt to changing circumstances. The themes ad-vanced in this article emphasize that distributed in-formation processing is increasingly embedded andsituated—and that a more integrated approach toexploring aggregation and constraint in informationprocessing offers the promise of renewed and prof-itable research into organizational structure and de-cision-making.

REFERENCES

Aggarwal, V. A., Siggelkow, N., & Singh, H. 2011. Gov-erning collaborative activity: Interdependence andthe impact of coordination and exploration. StrategicManagement Journal, 32(7): 705–730.


Aggarwal, V. A., &Wu, B. 2015. Organizational constraintsto adaptation: Intrafirm asymmetry in the locus ofcoordination. Organization Science, 26(1): 218–238.

Albert, D., Kreutzer, M., & Lechner, C. 2015. Resolving theparadox of interdependency and strategic renewal inactivity systems. Academy of Management Review,40(2): 210–234.

Augier, M. 2001. Simon says: Bounded rationalitymatters:Introduction and interview. Journal of ManagementInquiry, 10(3): 268–275.

Baldwin, C. Y., & Clark, K. B. 2000. Design rules: Thepower of modularity. Cambridge, MA: MIT Press.

Barreto, I., & Patient, D. L. 2013. Toward a theory of intra-organizational attention based on desirability andfeasibility factors. Strategic Management Journal,34(6): 687–703.

Baumann,O. 2015.Models of complex adaptive systems instrategy and organization research. Mind & Society,14(2): 169–183.

Baumann, O., Eggers, J., & Stieglitz, N. 2018. Colleaguesand competitors: How internal social comparisonsshape organizational search and adaptation. Admin-istrative Science Quarterly, 64(2): 275–309.

Baumann, O., & Siggelkow, N. 2011. Complexity andcompetitive advantage.Handbook of complexity andmanagement: 494–505. London: SAGE.

Bechky, B. A. 2003. Sharing meaning across occupationalcommunities: The transformation of understandingon a production floor. Organization Science, 14(3):312–330.

Bertels, S., Howard-Grenville, J., & Pek, S. 2016. Culturalmolding, shielding, and shoring at oilco: The role ofculture in the integration of routines. OrganizationScience, 27(3): 573–593.

Billinger, S., Stieglitz, N., & Schumacher, T. R. 2014.Search on rugged landscapes: An experimental study.Organization Science, 25(1): 93–108.

Birkinshaw, J., & Lingblad, M. 2005. Intrafirm competitionand charter evolution in the multibusiness firm. Or-ganization Science, 16(6): 674–686.

Blettner, D. P., He, Z., Hu, S., & Bettis, R. A. 2015. Adaptiveaspirations and performance heterogeneity: Attentionallocation amongmultiple reference points.StrategicManagement Journal, 36(7): 987–1005.

Bloom, N., Garicano, L., Sadun, R., & Van Reenen, J. 2014.The distinct effects of information technology andcommunication technology on firm organization.Management Science, 60(12): 2859–2885.

Bouquet, C., & Birkinshaw, J. 2008. Weight versus voice:How foreign subsidiaries gain attention from corpo-rate headquarters. Academy of Management Jour-nal, 51(3): 577–601.

Bromley, P., & Powell, W. W. 2012. From smoke and mir-rors to walking the talk: Decoupling in the contem-porary world. Academy of Management Annals,6(1): 483–530.

Brusoni, S., & Prencipe, A. 2006. Making design rules: Amultidomain perspective. Organization Science,17(2): 179–189.

Burton, R. 2013. The future of organization design: An in-terpretative synthesis in three themes. Journal of Or-ganization Design, 2(1): 42–44.

Burton, R. M., & Obel, B. 1984. Designing efficient orga-nizations: Modelling and experimentation. NorthHolland: Elsevier Science Publishers B.V.

Burton, R. M., & Obel, B. 2004. Strategic organizationaldiagnosis and design: The dynamics of fit (3rd ed.).The Netherlands: Kluwer.

Burton, R. M., Obel, B. M., & DeSanctis, G. 2011. Organi-zational design: A step-by-step approach (2nd ed.).Cambridge, United Kingdom: Cambridge UniversityPress.

Burton, R. M., Obel, B., & Hakonsson, D. D. 2015. Orga-nizational design: A step-by-step approach. Cam-bridge, United Kingdom: Cambridge University Press.

Canato,A., Ravasi, D., &Phillips,N. 2013.Coercedpracticeimplementation in cases of low cultural fit: Culturalchange and practice adaptation during the imple-mentation of six sigma at 3M. Academy of Manage-ment Journal, 56(6): 1724–1753.

Castellaneta, F., & Zollo, M. 2014. The dimensions of ex-periential learning in themanagement of activity load.Organization Science, 26(1): 140–157.

Christensen, M., & Knudsen, T. 2010. Design of decision-making organizations. Management Science, 56(1):71–89.

Clement, J., & Puranam, P. 2017. Searching for structure:Formal organization design as a guide to networkevolution. Management Science, 64(8): 3879–3895.

Cohen,M.D. 1984. Conflict and complexity: Goal diversityand organizational search effectiveness. AmericanPolitical Science Review, 78(2): 435–451.

Cohen, M. D., March, J. G., & Olsen, J. P. 1972. A garbagecan model of organizational choice. AdministrativeScience Quarterly, 17(1): 1–25.

Cornelissen, J., & Durand, R. 2012. More than just novelty:Conceptual blending and causality. Academy ofManagement Review, 37(1): 152–154.

Cornelissen, J. P., Durand, R., Fiss, P. C., Lammers, J. C., &Vaara, E. 2015. Putting communication front andcenter in institutional theory and analysis. Academyof Management Review, 40(1): 10–27.

Corritore, M., Goldberg, A., & Srivastava, S. 2019. Dualityin diversity: How intrapersonal and interpersonal


cultural heterogeneity relate to firm performance.Administrative Science Quarterly, Forthcoming.

Cremer, J. 1980. A partial theory of the optimal organiza-tion of a bureaucracy.The Bell Journal of Economics,11 (Autumn 1980): 683–93.

Cremer, J. 1983. The discrete heal algorithm with in-termediate goods. The Review of Economic Studies,50(2): 383–391.

Csaszar, F. A. 2012. Organizational structure as a determi-nant of performance: Evidence from mutual funds.Strategic Management Journal, 33(6): 611–632.

Csaszar, F. A. 2013a. Strategic decision making. Encyclo-pedia of Management Theory. Thousand Oaks, CASAGE Publications.

Csaszar, F. A. 2013b. An efficient frontier in organizationdesign: Organizational structure as a determinant ofexploration and exploitation. Organization Science,24(4): 1083–1101.

Csaszar, F. A. 2018. What makes a decision strategic?Strategic representations. Strategy Science, 3(4):606–619.

Csaszar, F. A., & Eggers, J. 2013. Organizational decisionmaking: An information aggregation view. Manage-ment Science, 59(10): 2257–2277.

Csaszar, F. A., & Laureiro-Martınez, D. 2018. Individualand organizational antecedents of strategic foresight:A representational approach. Strategy Science, 3(3):513–532.

Cyert, R. M., & March, J. G. 1963. A behavioral theory ofthe firm. Englewood Cliffs, NJ: Prentice-Hall.

Cyert, R. M., & March, J. G. 1992. A behavioral theory ofthe firm (2nd ed.). EnglewoodCliffs,NJ: Prentice-Hall.

Daft, R. L., & Lengel, R. H. 1986. Organizational informa-tion requirements, media richness and structural de-sign. Management Science, 32(5): 554–571.

Dahlander, L., Lifshitz-Assaf, H., Piezunka, H., Seong, S., &Stroube, B. K. 2017. Uncovering crowds as interstitialspaces: Crowds that influence and that ought to beinfluenced. Paper presented at the Academy of Man-agement Proceedings, 2017(1): 10566.

Dalpiaz, E., Rindova, V., & Ravasi, D. 2016. Combininglogics to transform organizational agency. Adminis-trative Science Quarterly, 61(3): 347–392.

Davis, J. P., Eisenhardt, K. M., & Bingham, C. B. 2009. Op-timal structure, market dynamism, and the strategy ofsimple rules. Administrative Science Quarterly,54(3): 413–452.

Dearborn, D. C., & Simon,H.A. 1958. Selective perception:A note on the departmental identifications of execu-tives. Sociometry, 21(2): 140–144.

Denis, J., Dompierre, G., Langley, A., & Rouleau, L. 2011.Escalating indecision: Between reification and strate-gic ambiguity.Organization Science, 22(1): 225–244.

Denrell, J., & March, J. G. 2001. Adaptation as informationrestriction: The hot stove effect. Organization Sci-ence, 12(5): 523–538.

Donaldson, L. 2001. The contingency theory of organi-zations. Thousand Oaks, CA: Sage.

Durand, R., & Paolella, L. 2013. Category stretching: Reor-ienting research on categories in strategy, entrepreneur-ship, and organization theory. Journal of ManagementStudies, 50(6): 1100–1123.

Dutt, N., & Joseph, J. 2019. Regulatory uncertainty, corpo-rate structure, and strategic agendas: Evidence fromthe US renewable electricity industry. Academy ofManagement Journal, 62(3): 800–827.

Dutton, J. E., & Ashford, S. J. 1993. Selling issues to topmanagement. Academy of Management Review,18(3): 397–428.

Eggers, J., & Kaul, A. 2018. Motivation and ability? A be-havioral perspective on the pursuit of radical inven-tion in multi-technology incumbents. Academy ofManagement Journal, 61(1): 67–93.

Eggers, J. P., & Kaplan, S. 2013. Cognition and capabilities:A multi-level perspective.Academy of ManagementAnnals, 7(1): 295–340.

Eggers, J. P., & Suh, J.-H. 2019. Experience and behavior:How negative feedback in new versus experienced do-mains affects firm action and subsequent performance.Academy of Management Journal, 62(2): 309–334.

Eisenhardt, K. M., & Zbaracki, M. J. 1992. Strategic de-cision making. Strategic Management Journal, 13(S2):17–37.

Elsbach, K. D., Barr, P. S., & Hargadon, A. B. 2005. Identi-fying situated cognition in organizations. Organiza-tion Science, 16(4): 422–433.

Elsbach,K.D., &Sutton,R. I. 1992.Acquiringorganizationallegitimacy through illegitimate actions: A marriage ofinstitutional and impression management theories.Academy of Management Journal, 35(4): 699–738.

Ethiraj, S. K., & Levinthal, D. 2004. Modularity and inno-vation in complex systems. Management Science,50(2): 159–173.

Ethiraj, S. K., & Levinthal, D. 2009. Hoping for A to Zwhilerewarding only A: Complex organizations and multi-ple goals. Organization Science, 20(1): 4–21.

Ethiraj, S. K., Levinthal, D., &Roy, R. R. 2008. The dual roleof modularity: Innovation and imitation. Manage-ment Science, 54(5): 939–955.

Fang, C., Kim, J., & Milliken, F. J. 2014. When bad news issugarcoated: Information distortion, organizational


search and the behavioral theory of the firm.StrategicManagement Journal, 35(8): 1186–1201.

Fang, C., Lee, J., & Schilling, M. A. 2010. Balancing ex-ploration and exploitation through structural design:The isolation of subgroups and organizational learn-ing. Organization Science, 21(3): 625–642.

Feldman, M. S. 1989.Order without design: Informationproduction and policy making. Stanford, CA: Stan-ford University Press.

Foss, N. J., & Weber, L. 2016. Moving opportunism to theback seat: Bounded rationality, costly conflict, and hi-erarchical forms. Academy of Management Review,41(1): 61–79.

Gaba, V., Bhattacharya, S. 2012. Aspirations, innovations,and corporate venture capital: A behavioral perspec-tive. Strategic Entrepreneurship Journal, 6: 178–199.

Gaba, V., Greve, H. 2019. Safe or profitable? Pursuing con-flicting goals.Organization Science, 30(4): 647–867.

Gaba, V., & Joseph, J. 2013. Corporate structure and per-formance feedback: Aspirations and adaptation in M-form firms.Organization Science, 24(4): 1102–1119.

Gaba, V., Lee, S., Meyer-Doyle, P., & Zhao, A. 2019. Priorexperiences of managers and responsiveness to per-formance feedback: Evidence from mutual funds.Working Paper, INSEAD.

Gaba, V., Meyer, A. D. 2008. Crossing organizational spe-cies boundaries: How venture capital infiltrated theinformation technology sector. Academy of Man-agement Journal, 51(5): 976–998.

Galbraith, J. R. 1974. Organization design: An informationprocessing view. Interfaces, 4(3): 28–36.

Galbraith, J. R. 1977. Organization design. Reading, MA:Addison-Wesley.

Galunic, D. C., & Eisenhardt, K. M. 2001. Architecturalinnovation and modular corporate forms. Academyof Management Journal, 44(6): 1229–1249.

Garicano, L. 2000. Hierarchies and the organization ofknowledge in production. Journal of Political Econ-omy, 108(5): 874–904.

Garicano, L., & Wu, Y. 2012. Knowledge, communication,and organizational capabilities. Organization Sci-ence, 23(5): 1382–1397.

Gavetti, G. 2005. Cognition and hierarchy: Rethinking themicrofoundations of capabilities’ development. Or-ganization Science, 16(6): 599–617.

Gavetti, G., & Levinthal, D. 2000. Looking forward andlooking backward: Cognitive and experiential search.Administrative Science Quarterly, 45(1): 113–137.

Gavetti, G., Levinthal, D., & Ocasio, W. 2007. Perspective—Neo-carnegie: The carnegie school’s past, present, and

reconstructing for the future. Organization Science,18(3): 523–536.

Gavetti, G., Levinthal, D. A., & Rivkin, J. W. 2005. Strategymaking in novel and complex worlds: The power ofanalogy.StrategicManagementJournal,26(8):691–712.

Gibbons, R. 1998. Incentives in organizations. Journal ofEconomic Perspectives, 12(4): 115–132.

Gibbons, R. 2003. Team theory, garbage cans and real or-ganizations: Some history and prospects of economicresearch on decision‐making in organizations. In-dustrial and Corporate Change, 12(4): 753–787.

Gilbert, C. G. 2005. Unbundling the structure of inertia:Resource versus routine rigidity. Academy of Man-agement Journal, 48(5): 741–763.

Gioia, D. A., & Chittipeddi, K. 1991. Sensemaking andsensegiving in strategic change initiation. StrategicManagement Journal, 12(6): 433–448.

Giorgi, S., Lockwood, C., & Glynn, M. A. 2015. The manyfaces of culture: Making sense of 30 years of researchon culture in organization studies. The Academy ofManagement Annals, 9(1): 1–54.

Granovetter,M. 1985. Economic action and social structure:The problem of embeddedness. American Journal ofSociology, 91(3): 481–510.

Greenwood, R., Dıaz, A. M., Li, S. X., & Lorente, J. C. 2010.The multiplicity of institutional logics and the het-erogeneity of organizational responses.OrganizationScience, 21(2): 521–539.

Greve, H. R. 2008. A behavioral theory of firm growth:Sequential attention to size and performance goals.Academy of Management Journal, 51(3): 476–494.

Greve, H. 2018. “Where to Search?”. Behavioral Strategyin Perspective (Advances in StrategicManagement,Vol. 39), pp. 91–10. Bingley, UK: Emerald Publishing.

Greve, H. R., & Gaba, V. 2017. Performance feedbackin organizations and groups: Common themes. Thehandbook of group and organizational learning.Oxford: Oxford University Press.

Gulati, R., Lawrence, P. R., & Puranam, P. 2005. Adap-tation in vertical relationships: Beyond incentive con-flict. Strategic Management Journal, 26(5): 415–440.

Gulati, R., Puranam, P., & Tushman, M. 2012. Meta‐organization design: Rethinking design in interor-ganizational and community contexts. StrategicManagement Journal, 33(6): 571–586.

Harrison, S. H., & Corley, K. G. 2011. Clean climbing, cara-biners, and cultural cultivation: Developing an open-systems perspective of culture.Organization Science,22(2): 391–412.

Helfat, C. E., & Eisenhardt, K. M. 2004. Inter‐temporaleconomies of scope, organizational modularity, and


the dynamics of diversification. Strategic Manage-ment Journal, 25(13): 1217–1232.

Helfat, C. E., & Peteraf, M. A. 2015. Managerial cognitive ca-pabilities and the microfoundations of dynamic capa-bilities.StrategicManagement Journal, 36(6): 831–850.

Higgins, E. T. 1996.Activation:Accessibility, and salience.Social psychology: Handbook of basic principles:133–168 New York: The Guilford Press.

Hoopes, D. G., & Postrel, S. 1999. Shared knowledge,“glitches,” and product development performance.Strategic Management Journal, 20(9): 837–865.

Hu, S., & Bettis, R. A. 2018. Multiple organization goalswith feedback from shared technological task envi-ronments. Organization Science, 29(5): 873–889.

Hu, S., He, Z., Blettner, D. P., & Bettis, R. A. 2017. Conflictinside and outside: Social comparisons and attentionshifts in multidivisional firms. Strategic Manage-ment Journal, 38(7): 1435–1454.

Jacobides, M. G. 2007. The inherent limits of organizationalstructure and the unfulfilled role of hierarchy: Lessonsfromanear-war.OrganizationScience, 18(3): 455–477.

Jordan, A. H., & Audia, P. G. 2012. Self-enhancement andlearning from performance feedback. Academy ofManagement Review, 37(2): 211–231.

Joseph, J., Baumann, O., Burton, R., & Srikanth, K. 2018.Reviewing, revisiting, and renewing the foundationsof organization design. Advances in Strategic Man-agement: Organization Design: 1–23. Bingley, UK:Emerald Publishing.

Joseph, J., & Gaba, V. 2015. The fog of feedback: Ambiguityand firm responses to multiple aspiration levels.Strategic Management Journal, 36(13): 1960–1978.

Joseph, J., Klingebiel, R., &Wilson, A. J. 2016. Organizationalstructure and performance feedback: Centralization,aspirations, and termination decisions. OrganizationScience, 27(5): 1065–1083.

Joseph, J., & Ocasio, W. 2012. Architecture, attention, and ad-aptation in the multibusiness firm: General electric from1951 to 2001. Strategic Management Journal, 33(6):633–660.

Joseph, J., & Wilson, A. J. 2018. The growth of the firm: Anattention‐based view. Strategic Management Jour-nal, 39(6): 1779–1800.

Kaplan, S. 2008. Cognition, capabilities, and incentives:Assessing firm response to the fiber-optic revolution.Academy of Management Journal, 51(4): 672–695.

Kaplan, S., & Tripsas,M. 2008. Thinking about technology:Applying a cognitive lens to technical change. Re-search Policy, 37(5): 790–805.

Kapoor, R. 2018. Ecosystems: Broadening the locus of valuecreation. Journal of Organization Design, 7(1): 12.

Karim, S. 2006. Modularity in organizational structure:The reconfiguration of internally developed and ac-quired business units. Strategic Management Jour-nal, 27(9): 799–823.

Karim, S., & Kaul, A. 2015. Structural recombination andinnovation: Unlocking intraorganizational knowledgesynergy through structural change. Organization Sci-ence, 26(2): 439–455.

Karim, S., & Mitchell, W. 2000. Path‐dependent and path‐breaking change: Reconfiguring business resources fol-lowing acquisitions in theU.S.Medical sector, 1978–1995.StrategicManagement Journal, 21(1011): 1061–1081.

Keum, D. D., & Eggers, J. 2018. Setting the bar: The evalu-ative and allocative roles of organizational aspira-tions. Organization Science, 29(6): 1170–1186.

Keum, D. D., & See, K. E. 2017. The influence of hierarchyon idea generation and selection in the innovationprocess. Organization Science, 28(4): 653–669.

Kim, Y. J., & Kim, J. 2019. Does negative feedback benefit(or harm) recipient creativity? The role of thedirectionof feedback flow. Academy of Management Journal.In Press.

Kleinbaum, A. M., Stuart, T. E., & Tushman, M. L. 2013.Discretion within constraint: Homophily and struc-ture in a formal organization. Organization Science,24(5): 1316–1336.

Knott, A. M., & Turner, S. F. 2019. An innovation theory ofheadquarters value in multibusiness firms. Organi-zation Science, 30(1): 19–39.

Knudsen, T., & Levinthal, D. A. 2007. Two faces of search:Alternative generation and alternative evaluation.Organization Science, 18(1): 39–54.

Knudsen, T., & Srikanth, K. 2014. Coordinated explora-tion: Organizing joint search bymultiple specialists toovercome mutual confusion and joint myopia. Ad-ministrative Science Quarterly, 59(3): 409–441.

Kogut, B., & Zander, U. 1996.What firms do? coordination,identity, and learning. Organization Science, 7(5):502–518.

Kotha, S., & Srikanth, K. 2013. Managing a global part-nership model: Lessons from the Boeing 787 ‘Dream-liner’ Program. Global Strategy Journal, 3(1): 41–66.

Koumakhov, R. 2009. Conventions in Herbert Simon’stheory of bounded rationality. Journal of EconomicPsychology, 30(3): 293–306.

Kretschmer, T., & Puranam, P. 2008. Integration throughincentives within differentiated organizations. Orga-nization Science, 19(6): 860–875.

Kunda, G. 2009. Engineering culture: Control and com-mitment in a high-tech corporation. Philadelphia:Temple University Press.


Laamanen, T., Maula, M., Kajanto, M., & Kunnas, P. 2018.The role of cognitive load in effective strategic issuemanagement. Long Range Planning, 51(4): 625–639.

Lawrence, P. R., & Lorsch, J. W. 1967. Organization andenvironment:Managingdifferential and integration.Boston, MA: Harvard University Press.

Lee, E., & Puranam, P. 2016. The implementation impera-tive: Why one should implement even imperfectstrategies perfectly. Strategic Management Journal,37(8): 1529–1546.

Leonardi, P. M. 2014. Social media, knowledge sharing, andinnovation: Toward a theory of communication visi-bility. Information Systems Research, 25(4): 796–816.

Levinthal, D., & Posen, H. E. 2007. Myopia of selection: Doesorganizational adaptation limit the efficacyof populationselection?Administrative ScienceQuarterly, 52(4):586–620.

Levinthal, D., & Workiewicz, M. 2018. When two bosses arebetter than one: Nearly decomposable systems and orga-nizational adaptation.Organization Science, 29(2): 207–224.

Levinthal, D. A. 1997. Adaptation on rugged landscapes.Management Science, 43(7): 934–950.

Levinthal, D. A. 2011. A behavioral approach to strategy—What’s the alternative? Strategic Management Jour-nal, 32(13): 1517–1523.

Malone, T. W., Laubacher, R. J., & Johns, T. 2011. The ageof hyperspecialization. Harvard Business Review,89(7/8): 56–65.

March, J. 1978. Bounded rationality, ambiguity, and theengineering of choice. The Bell Journal of Economics,587–608.

March, J. G., & Simon, H. A. 1958. Organizations. JohnWiley & Sons. New York

March, J. G., & Simon, H. A. 1993. Introduction to thesecond edition. organizations (2nd ed.). Cambridge,MA: Blackwell Publishers.

Marschak, J., & Radner, R. 1972. Economic theory ofteams. New Haven, CT: Yale University Press.

Martignoni, D., Menon, A., & Siggelkow, N. 2016. Conse-quences of misspecified mental models: Contrastingeffects and the role of cognitive fit. Strategic Man-agement Journal, 37(13): 2545–2568.

Martin, J., &Eisenhardt,K. 2010.Rewiring:Cross-business-unit collaborations in multibusiness organizations.Academy of Management Journal, 53(2): 265–301.

McNamara, G., Moon, H., & Bromiley, P. 2002. Banking oncommitment: Intendedandunintendedconsequencesof an organization’s attempt to attenuate escalationof commitment. Academy of Management Journal,45(2): 443–452.

Mintzberg, H. (Ed.). 1979. The structuring of organiza-tions: A synthesis of the research. Englewood Cliffs,NJ: Prentice-Hall.

Newell, A., & Simon,H. 1956. The logic theorymachine--Acomplex information processing system. IRE Trans-actions on Information Theory, 2(3): 61–79.

Newell, A., & Simon, H. A. 1972.Humanproblem solving.Englewood Cliffs, NJ: Prentice-Hall.

Nickerson, J., Yen, C. J., & Mahoney, J. T. 2012. Exploringthe problem-finding and problem-solving approachfor designing organizations. Academy of Manage-ment Perspectives, 26(1): 52–72.

Nickerson, J. A., & Zenger, T. R. 2002. Being efficientlyfickle: A dynamic theory of organizational choice.Organization Science, 13(5): 547–566.

Nickerson, J. A., & Zenger, T. R. 2004. A knowledge-basedtheory of the firm—The problem-solving perspective.Organization Science, 15(6): 617–632.

Nisbett, R. E., & Ross, L. 1991. Person and the situation:Perspectivesof socialpsychology. NewYork:McGraw-Hill.

Obloj, T., & Zenger, T. 2017. Organization design, prox-imity, and productivity responses to upward socialcomparison. Organization Science, 28(1): 1–18.

Ocasio,W. 1997. Towards an attention‐based viewof the firm.Strategic Management Journal, 18(S1): 187–206.

Ocasio, W., & Joseph, J. 2005. An attention-based theory ofstrategy formulation: Linking micro-and macro-perspectives in strategy processes. In G. Zulanski, J.Porac, & Y. Doz (Eds.), Advances in Strategic Man-agement: Strategy Process: 39–61. Bingley, UK:Emerald Group Publishing.

Ocasio, W., & Joseph, J. 2008. Rise and fall-or trans-formation?: The evolution of strategic planning at thegeneral electric company, 1940–2006. Long RangePlanning, 41(3): 248–272.

Ocasio, W., Laamanen, T., & Vaara, E. 2017. Communica-tion and attention dynamics: An attention-based viewof strategic change. Strategic Management Journal,39(1): 155–167.

Ocasio, W., Loewenstein, J., & Nigam, A. 2015. Howstreams of communication reproduce and change in-stitutional logics: The role of categories. Academy ofManagement Review, 40(1): 28–48.

Perkmann, M., McKelvey, M., & Phillips, N. 2018. Pro-tecting scientists from Gordon Gekko: How organiza-tions use hybrid spaces to engage with multipleinstitutional logics.Organization Science, 30(2): 298–318.

Perrow, C. 1961. Analysis of goals in complex organiza-tions. American Sociological Review, 26: 335–341.


Piezunka, H., & Dahlander, L. 2015. Distant search, narrowattention: How crowding alters organizations’ filtering ofsuggestions in crowdsourcing.AcademyofManagementJournal, 58(3): 856–880.

Posen, H. E., Keil, T., Kim, S., & Meissner, F. D. 2018.Renewing research on problemistic search—A reviewand research agenda. Academy of Management An-nals, 12(1): 251.

Posen, H. E., & Levinthal, D. A. 2012. Chasing a movingtarget: Exploitation and exploration in dynamic en-vironments. Management Science, 58(3): 587–601.

Postrel, S. 2002. Islands of shared knowledge: Specializa-tion and mutual understanding in problem-solvingteams. Organization Science, 13(3): 303–320.

Puranam, P., Alexy, O., & Reitzig, M. 2014. What’s “new”

about new forms of organizing? Academy of Man-agement Review, 39(2): 162–180.

Puranam, P., & Hakonsson, D. D. 2015. Valve’s way. Jour-nal of Organization Design, 4(2): 2–4.

Puranam, P. 2012. A future for the science of organizationdesign. Journal of Organization Design, 1(1): 18–19.

Puranam, P. 2018. The microstructure of organizations.Oxford, UK: Oxford University Press.

Puranam, P., Raveendran, M., & Knudsen, T. 2012. Orga-nization design: The epistemic interdependence per-spective. Academy of Management Review, 37(3):419–440.

Puranam, P., Shrestha, Y., He, V., & von Krogh, G. 2018.Algorithmic induction through machine learning:Opportunities for management and organizationresearch. INSEAD Working Paper no. 2018/11/STR.

Puranam, P., Singh, H., & Chaudhuri, S. 2009. Integratingacquired capabilities: When structural integrationis (un) necessary.Organization Science, 20(2): 313–328.

Puranam, P., Stieglitz, N., Osman, M., & Pillutla, M. M.2015.Modelling bounded rationality in organizations:Progress and prospects. Academy of ManagementAnnals, 9(1): 337–392.

Puranam, P., & Swamy, M. 2016. How initial representa-tions shape coupled learningprocesses.OrganizationScience, 27(2): 323–335.

Radner, R. 1993. The organization of decentralized infor-mation processing. Econometrica: Journal of theEconometric Society, 61: 1109–1146.

Raffaelli, R., Glynn, M. A., & Tushman, M. 2019. Frameflexibility: The role of cognitive and emotional fram-ing in innovation adoption by incumbent firms.Strategic Management Journal, 40(7): 1013–1039.

Rao, H., & Giorgi, S. 2006. Code breaking: How entrepre-neurs exploit cultural logics to generate institutional

change. Research in Organizational Behavior, 27:269–304.

Raveendran, M., Puranam, P., &Warglien, M. 2015. Objectsalience in the division of labor: Experimental evi-dence. Management Science, 62(7): 2110–2128.

Reitzig, M., & Maciejovsky, B. 2015. Corporate hierarchyand vertical information flow inside the firm—Abehavioral view. Strategic Management Journal,36(13): 1979–1999.

Reitzig, M., & Sorenson, O. 2013. Biases in the selectionstage of bottom‐up strategy formulation. StrategicManagement Journal, 34(7): 782–799.

Ren, Y., & Argote, L. 2011. Transactive memory systems1985–2010: An integrative framework of key dimen-sions, antecedents, and consequences. Academy ofManagement Annals, 5(1): 189–229.

Rerup, C. 2009. Attentional triangulation: Learning fromunexpected rare crises. Organization Science, 20(5):876–893.

Retelny,D., Robaszkiewisz, S., To,A., Lasecki,W., Patel , J.,Rahmati, N., Doshi, T., Valentine, M., Bernstein, M.2014. Expert crowdsourcing with flash teams. Pro-ceedings of 27th ACM symposium on User interfacesoftware and technology: 75–85.

Rhee, L., Ocasio,W., & Kim, T. 2019. Performance feedback inhierarchical business groups: The cross-level effects ofcognitive accessibility on R&D search behavior. Organi-zation Science, 30(1): 51–69.

Rindova, V., Dalpiaz, E., & Ravasi, D. 2011. A culturalquest: A study of organizational use of new culturalresources in strategy formation. Organization Sci-ence, 22(2): 413–431.

Rindova, V. P., & Kotha, S. 2001. Continuous “morphing”:Competing through dynamic capabilities, form, andfunction. Academy of Management Journal, 44(6):1263–1280.

Rivkin, J. W., & Siggelkow, N. 2003. Balancing search andstability: Interdependencies among elements of organi-zational design.Management Science, 49(3): 290–311.

Rivkin, J. W., & Siggelkow, N. 2007. Patterned interactionsin complex systems: Implications for exploration.Management Science, 53(7): 1068–1085.

Rowley, T. J., Shipilov, A. V., & Greve, H. R. 2017. Boardreform versus profits: The impact of ratings on theadoption of governance practices. Strategic Man-agement Journal, 38(4): 815–833.

Sah, R. K., & Stiglitz, J. E. 1985. Human fallibility andeconomic organization. The American EconomicReview, 75(2): 292–297.

Sah, R. K., & Stiglitz, J. E. 1986. The architecture of eco-nomic systems: Hierarchies and polyarchies. TheAmerican Economic Review, 76(4): 716–727.


Sah, R. K., & Stiglitz, J. E. 1988. Committees, hierarchiesand polyarchies. The Economic Journal, 98(391):451–470.

Sah, R. K., & Stiglitz, J. E. 1991. The quality of managers incentralized versus decentralized organizations. TheQuarterly Journal of Economics, 106(1): 289–295.

Sanchez, R., &Mahoney, J. T. 1996.Modularity, flexibility,and knowledge management in product and organi-zation design. Strategic Management Journal,17(S2): 63–76.

Schein, E. 2004. Organizational culture and leadership.San Francisco, CA: Jossey.

Sengul, M., & Gimeno, J. 2013. Constrained delegation:Limiting subsidiaries’decision rights and resources infirms that compete across multiple industries. Ad-ministrative Science Quarterly, 58(3: 420–471.

Seo, M., & Creed, W. D. 2002. Institutional contradictions,praxis, and institutional change:Adialectical perspective.Academy ofManagement Review, 27(2): 222–247.

Seshadri, S., Shapira, Z., Tucci, C. L. 2019. Are flatter or-ganizations more innovative: Hierarchical depth andthe importance of R&D ideas. Working Paper, NYUStern School of Business, 2019.

Siggelkow, N. 2001. Change in the presence of fit: The rise,the fall, and the renaissance of liz claiborne.Academyof Management Journal, 44(4): 838–857.

Siggelkow, N. 2002. Misperceiving interactions amongcomplements and substitutes: Organizational conse-quences. Management Science, 48(7): 900–916.

Siggelkow, N. 2011. Firms as systems of interdependentchoices. Journal of Management Studies, 48(5):1126–1140.

Siggelkow,N., & Levinthal, D.A. 2003. Temporarily divideto conquer: Centralized, decentralized, and reinte-grated organizational approaches to exploration andadaptation. Organization Science, 14(6): 650–669.

Siggelkow, N., & Rivkin, J. W. 2005. Speed and search:Designing organizations for turbulence and complex-ity. Organization Science, 16(2): 101–122.

Siggelkow, N., & Rivkin, J. W. 2009. Hiding the evidence ofvalid theories: How coupled search processes obscureperformance differences among organizations. Ad-ministrative Science Quarterly, 54(4): 602–634.

Simon, H. A. 1947. Administrative behavior: A study ofdecision-making processes in administrative orga-nization. Chicago, IL: Macmillan.

Simon, H. A. 1952. Comments on the theory of organiza-tions. American Political Science Review, 46(4):1130–1139.

Simon, H. A. 1955. A behavioral model of rational choice.The Quarterly Journal of Economics, 69(1): 99–118.

Simon, H. A. 1956. Rational choice and the structure of theenvironment. Psychological Review, 63(2): 129.

Simon, H. A. 1962. New developments in the theory of thefirm. The American Economic Review, 52(2): 1–15.

Simon, H. A. 1964. On the concept of organizational goal.Administrative Science Quarterly, 9(1): 1–22.

Simon, H. A. 1991. Organizations and markets. Journal ofEconomic Perspectives, 5(2): 25–44.

Simon,H.A. 1995. Comment onkagel. InG.Wolters, & J. G.Lennox (Eds.), Concept, theories, and rationality inthe biological sciences: 359–66. Pittsburgh, PA: Uni-versity of Pittsburgh Press.

Simon, H. A. 1997. Models of bounded rationality: Em-pirically grounded economic reason. Cambridge,MA: MIT Press.

Soda, G., & Zaheer, A. 2012. A network perspective onorganizational architecture: Performance effects of theinterplay of formal and informal organization. Stra-tegic Management Journal, 33(6): 751–771.

Soderstrom, S. B., & Weber, K. 2019. Organizational struc-ture from interaction: Evidence from corporate sus-tainability efforts. Administrative Science Quarterly.In Press.

Sonenshein, S. 2010. We’re changing—or are we? Untan-gling the role of progressive, regressive, and stabilitynarratives during strategic change implementation.Academy of Management Journal, 53(3): 477–512.

Sorenson, O. 2003. Interdependence and adaptability:Organizational learning and the long–term effect ofintegration. Management Science, 49(4): 446–463.

Sosa, M. E., Eppinger, S. D., & Rowles, C. M. 2004. Themisalignment of product architecture and organiza-tional structure in complex product development.Management Science, 50(12): 1674–1689.

Srikanth, K., & Puranam, P. 2011. Integrating distributedwork: Comparing task design, communication, andtacit coordination mechanisms. Strategic Manage-ment Journal, 32(8): 849–875.

Srivastava, S. B., Goldberg, A., Manian, V. G., & Potts, C.2017. Enculturation trajectories: Language, culturaladaptation, and individual outcomes inorganizations.Management Science, 64(3): 1348–1364.

Stan, M., & Puranam, P. 2017. Organizational adaptation tointerdependence shifts: The role of integrator structures.Strategic Management Journal, 38(5): 1041–1061.

Sullivan, P. 2010. Empirical evidence on occupation andindustry specific human capital. Labour Economics,17(3): 567–580.

Swidler, A. 1986. Culture in action: Symbols and strate-gies. American Sociological Review, 51: 273–286.


Sytch, M., Wohlgezogen, F., & Zajac, E. 2018. Collabora-tive by Design? How matrix organizations see/do al-liances. Academy of Management Proceedings,2016(1): 13400.

Tarakci, M., Ates, N. Y., Floyd, S. W., Ahn, Y., &Wooldridge, B. 2018. Performance feedback and mid-dlemanagers’divergent strategic behavior: The roles ofsocial comparisons and organizational identification.Strategic Management Journal, 39(4): 1139–1162.

Thompson, J. D. 1967. Organizations in action: Socialscience bases of administrative theory. New York:McGraw-Hill.

Thornton, P. H., Ocasio, W., & Lounsbury, M. 2012. Theinstitutional logics perspective: A new approach toculture, structure, and process New York: OxfordUniversity Press.

Tuggle, C. S., Sirmon, D. G., Reutzel, C. R., & Bierman, L.2010. Commanding board of director attention: In-vestigating how organizational performance and CEOduality affect board members’ attention to monitoring.Strategic Management Journal, 31(9): 946–968.

Turner, K. L., & Makhija, M. V. 2012. The role of individ-uals in the information processing perspective. Stra-tegic Management Journal, 33(6): 661–680.

Tushman, M. L., & Nadler, D. A. 1978. Information processingas an integrating concept in organizational design. Acad-emy of Management Review, 3(3): 613–624.

Valentine, M., Retelny, D., To, A., Rahmati, N., Doshi, T.,Bernstein, M. 2017. Flash organizations: Crowd-sourcing complex work by structuring crowds as or-ganizations. CHI 2017: SIGCHI Conference on HumanFactors in Computing Systems.

Valentine, M. A. 2019. Algorithms and the Organiza-tional Chart. Working Paper, Stanford University.

Van denSteen, E. 2010. On the origin of shared beliefs (andcorporate culture). The Rand Journal of Economics,41(4), 617–648.

VanKnippenberg, D., Dahlander, L., Haas,M. R., &George,G. 2015. Information, attention, and decision-making,Academy of Management Journal, 58: 649–657.

Van Zandt, T. 1999. Decentralized information processingin the theory of organizations. Contemporary eco-nomic issues: 125–160. Springer.

Vissa, B., Greve, H. R., & Chen, W. 2010. Business groupaffiliation and firm search behavior in India: Re-sponsiveness and focus of attention. OrganizationScience, 21(3): 696–712.

Walsh, J. 1995. Managerial and organizational cognition:Notes from a trip down memory lane. OrganizationScience, 6(3): 280–321.

Weber, K. 2005. A toolkit for analyzing corporate culturaltoolkits. Poetics, 33(3–4): 227–252.

Weber, K., & Dacin, M. T. 2011. The cultural constructionof organizational life: Introduction to the special issue.Organization Science, 22(2): 287–298.

Weber, K., & Glynn, M. A. 2006. Making sense with in-stitutions: Context, thought and action in karl weick’stheory. Organization Studies, 27(11): 1639–1660.

Weick, K. E., & Sutcliffe, K. M. 2015. Managing the un-expected sustained performance in a complexworld. Hoboken: Wiley.

Weick, K. E. 1976. Educational organizations as looselycoupled systems.Administrative ScienceQuarterly,21: 1–19.

Williams, C., &Mitchell, W. 2004. Focusing firm evolution:The impact of information infrastructure onmarket entryby U.S. Telecommunications Companies, 1984–1998.Management Science, 50(11): 1561–1575.

Wu, Y. 2015. Organizational structure and product choicein knowledge-intensive firms.Management Science,61(8): 1830–1848.

Yayavaram, S., & Ahuja, G. 2008. Decomposability inknowledge structures and its impact on the usefulnessof inventions and knowledge-base malleability. Ad-ministrative Science Quarterly, 53(2): 333–362.

Zhang, M. C., & Greve, H. R. 2019. Dominant coalitionsdirecting acquisitions: Different decision makers,different decisions. Academy of Management Jour-nal, 62(1): 44–65.

Zhou, Y. M. 2013. Designing for complexity: Using di-visions and hierarchy to manage complex tasks. Or-ganization Science, 24(2): 339–355.

John Joseph ([email protected]) is an Associate Professor ofStrategy at the University of California, Irvine. He receivedhis PhD from Northwestern University. He studies orga-nization design and its impact on cognition and decisionmaking in complex organizations. His current researchexamines the antecedents of different organizationalstructures and the impact of structure on innovation,growth, and performance.

Vibha Gaba ([email protected]) is Associate Pro-fessor of Entrepreneurship and the INSEAD Fellow inMemoryof ErinAnderson at INSEAD.Her current researchfocuses on understanding the implications of multiplegoals and aspirations on adaptive change, performancefeedback, and decision making with particular emphasison the role of organizational structure. Currently, sheserves as an Associated Editor at Strategic ManagementJournal.


mailto:[email protected]

mailto:[email protected]

ORGANIZATIONAL STRUCTURE, INFORMATION …agenda for the decision-making implications of orga-nizational structure and information processing. Therefore, the goal of this study is to

Documents