IN SEARCH OF A MENTAL-MODEL-LIKE CONCEPT
FOR GROUP LEVEL MODELING
Hyunjung Kim
Department of Public Administration and Policy
Nelson A. Rockefeller College of Public Affairs and Policy
University at Albany
Milne 110, 135 Western Ave., Albany, NY 12222, U.S.A.
1-518-426-9640
hk8459@ albany.edu
Abstract
In system dynamics, we care about mental model, because it is what we attempt to
describe, understand, and improve using computer simulation models. But if we are developing
our simulation models from a group of individuals, can we still call what we are trying to study a
mental model? The main difference between individual and group level phenomenon is that in a
group setting, interaction and communication among individuals shape what is represented in the
maps and models. This paper attempts to clarify the object of our group level modeling by
surveying existing literature looking for concepts similar to “mental model” at the group level.
The survey leads to an insight that what we attempt to represent with our maps and models vary
widely by our selection of modeling methods, and how we define our object will draw a different
set of theoretical literature.
Key words
Mental model, collective mind, group model building
Relevance of Mental Model C oncept in Group Setting
In system dynamics, we create causal loop diagrams and computer simulation models to
describe, understand, and improve mental models. The field’s popular use of the term “mental
model” led to a need to define the term more explicitly. The definition suggested by Richardson,
Andersen, et al. (1994) integrated rich theories of human judgment processes involving the
means/ends model, Brunswikian lens model, and identification of designer and operator logic.
Their study describes how one perceives a system, selects and interprets information, plans
action, and modifies goals and models. They argued that a mental model that underlies this
process is “not directly accessible or observable, and efforts to elicit mental models distort what
is elicited.” (p. 182) Doyle and Ford (1998; 1999) attempted to define mental model by
emphasizing various attributes of a mental model identified in system dynamics and cognitive
psychology literature. With an input from Lane (1999), Doyle and Ford defined mental model of
a dynamic system as “a relatively enduring and accessible, but limited, internal conceptual
representation of an external system whose structure is analogous to the perceived structure of
the system.” (p. 414, 1999)
Two key assumptions are shared by the aforementioned studies and many other studies
on mental models (for example, Rouse and Morris 1986). The assumptions are that mental
models are metaphysical entities residing in individual minds and that we need a method or a
tool to access mental models. The method employed by system dynamicists are causal loop
diagramming and simulation modeling. However, what is represented in our maps and models
does not always come from an individual.
Much of our modeling work is carried out in a group environment. Important decisions in
organizations usually made by a group of experts (Steiner and Miner 1977; Walsh, Henderson et
al. 1988). Recognizing the importance of group decision making, system dynamics modeling and
systems thinking interventions have been actively carried out in group settings, and techniques
for group model building have been developed and elaborated (Richardson and Andersen 1995;
Vennix 1996; Andersen and Richardson 1997; Rouwette, Vennix et al. 2002; Luna-Reyes,
Martinez-Moyano et al. 2006).
A modeling at the group level raises a few questions: if we are developing causal loop
diagrams or simulation models from a group of individuals, what are we representing with our
maps and models? What is it that we are trying to describe, understand, and improve using our
methods? Should we still call it a mental model? The main difference between individual and
group level phenomenon is that in group settings, interaction and communication among
individuals play an important role in determining what is captured through representation
methods used by the modeler. In other words, in a group level modeling, what the modeler
attempts to portray using maps and models may be quite different from what we call a mental
model.
This paper attempts to clarify the issue by surveying existing literature looking for
concepts similar to “mental model” at the group level. The definitions of mental model laid out
by Richardson, Andersen, et al. (1994), Doyle and Ford (1998; 1999), and Lane (1999) all have
strong ties to cognitive psychology. This study will expand its scope of literature review from
cognitive psychology to organizational behavior and sociology. The survey leads to an insight
that the group-level concepts defined in the literature have different assumptions about the group
decision making process, and the paper introduces a framework to identify such differences. The
framework is then applied to the system dynamics context to discuss how different modeling
methods implicitly make different assumptions about what is represented in the maps and models
at the group level, and how those assumptions connect the different modeling methods to a
different set of theoretical literature. The goal of this paper is to promote a discussion on the
nature of our maps and models, and to introduce relevant literature to the system dynamics
community that will enrich our theoretical understanding of what is generated from our modeling
efforts.
Boundary of the Study: Decomposing Maps and Models
This paper will begin by setting a boundary for its theoretical discussion. We use maps
and models to access what we loosely call “mental models.” If this process of accessing a mental
model is decomposed, there is an elicitation of information about the mental model and a
representation of the elicited product. The process of generating maps and models is described in
Figure 1. The object of interest, or what we often refer to as a mental model, is portrayed in the
maps and models, but it is filtered—and likely it is distorted—by the elicitation and
representation methods used as well as the modeler’s mental models and priors. In other words,
the product of our modeling has only a partial relationship to what we intend to study, because it
is influenced by (1) the true object of interest, or what we have been calling mental model, (2)
the elicitation and representation methods being used, and (3) the modeler’s mental model.
“Maps and ~
i») Models =
Figure 1. What is Represented in a Map or a Model
The first part of paper will discuss how the existing literature defines the object of
interest of our maps and models in a group setting. If the group level phenomenon is different
from that of the individual level and if the term mental model does not appropriately capture the
object of interest at the group level, we need a group level alternative for the mental model
concept.
The second part of the paper will discuss how different lens used by the modeler
influence how the modeler defines his or her object of interest. A modeler’s lens is a combined
product of his or her prior or and a selection of research methods. Different lens will look at
different aspects of the object, and thus influence what is represented in the maps and models.
Who Studies Mental-Model-L ike C oncepts in Group Setting?
There are studies from various disciplines that have been interested in human cognition
and interactions as a processor for information and a generator for new decision or action.
Broadly stated, these researchers have been exploring the mediator that connects stimulus and
response, the decision processor that lies between decision input and decision output, and the
model that is used to perceive and understand environment. The processor has been often
referred to as a black box, because one cannot directly observe the mental process that
individuals or groups go though in converting information into decisions. The black box has
been labeled in many different ways, including mental model, schema, mind, knowledge
structure, and interpretation system. For a group or organizational level phenomenon, the
concept has been referred to with a qualifier that stands for collectivity, such as shared, team, or
organizational. An example would be a shared mental model or an organizational mind. Group
or organizational level studies gained much attention as one saw limitations in the application of
individual level analysis to understanding of a group or organizational phenomenon. Levine et al.
(1993) noticed that while cognition is frequently collaborative in reality, most experiments and
academic studies focus on individual cognition.
Researchers interested in the mental-model-like concept for group environments come
from disciplines such as cognitive psychology, organizational behavior, sociology, artificial
intelligence, and decision sciences.” Their research topics include- but are not limited to-
perception, interpretation, attention, memory, knowledge representation and learning, problem
solving, and social cognition (Huff 1990). While they all share a common interest in mental-
model-like concepts at the group or organizational level, their research motivation is quite
different.
Simon's (1947/1997) bounded rationality argument has been one of the major
motivations for research focused on information processing. It led to a recognition that different
individuals or organizations may come to different decisions in a given environment. Growing
interest in what drives these differences motivated researchers to focus on selection and
interpretation of information in organizations (Lant 2002). In psychology, a group of researchers
(for example, Cannon-Bowers and Salas 2001) studied shared mental model in an effort to
improve team performance. They assumed that shared knowledge among team members leads to
better team performance, and their research has been devoted to identifying and measuring
shared knowledge. The recent growth in the knowledge industry was another reason for the
stream of research on information processing and knowledge management in organizations
(Gibson 2001). And finally, there are those motivated by decision support. System dynamics
researchers and practitioners belong in this category. System dynamicists attempt to understand
how decisions are made in order to improve future decision performances. They believe that
mental models are inherently limited, and that by using various decision support tools, one can
substantially improve the decision making capacity (Forrester 1961; Sterman 1994).
A Plethora of Descriptors for the Black Box
Despite the common interest in defining and analyzing mental-model-like concepts at the
group level, there has been a little effort toward cross-fertilization among these disciplines.
' There may be an overlap in the listed disciplines as some of these disciplines are highly interdisciplinary in nature.
For example, there are psychologists and sociologists who pursue organizational behavior or decision sciences.
Klimoski and Mohammed (1994) noted that while there is a rich opportunity for interdisciplinary
cooperation, researchers from different fields rarely cite one another.
One barrier for not seeing inter-disciplinary communication as much as one would hope
for is the use of different terms to describe the black box. Researchers are interested what lies
between somewhat observable information inputs and decision outputs, but they use different
names for this black box. There are several factors contributing to the plethora of terms used to
describe the black box. One reason is a lack of communication between researchers. Although it
may seem like a circular argument, use of different terms in different disciplines discourages
interdisciplinary communication which then widens the language gap. But a more important
reason is that researchers using different languages are not looking at the same side of the black
box. In other words, what they are discussing is in fact not one concept, but it is more likely
different aspects of one concept or different concepts with similar attributes.
In order to discover a group-level mental-model-like concept that is grounded in the
established literature, it is necessary to explore various terms and definitions used in the
literature to describe the black box. For analytical convenience, the black box will be called “the
processor” in this study. The processor includes all aspects of cognition, information and
judgment processing in a group. The processor generates one or more decisions or actions in a
group. The term processor is comprehensive and general enough to be used as a higher level
concept for various mental-model-like concepts used in the literature. However, the term is used
only for an analytical purpose in this study, and it is not proposed as a name for mental-model-
like concept at the group level.
Lists formulated by Schneider and Angelmar (1993), Klimoski and Mohammed (1994),
and Walsh (1995) give a good overview of some of the terms used to describe the processor in
the literature. Schneider and Angelmar found over 65 terms from literature under what they call
“research on organizational cognition” (p.350). Klimoski and Mohammed listed over 30 terms
under “variations on a theme: application of the team mental model concept (p. 408-409).”
Walsh listed over 75 terms under “cognition in organization: the language of management
theory” (p.284-285). Even considering overlaps among these lists, there are over 200 variations
in the terms used. The number increases if we consider those not listed in the aforementioned
three studies but identified by other studies.
Table 1 shows some of the words used by researchers to formulate a descriptor for the
processor. Although what to include in the list would vary by how the boundary of the processor
is defined at the group level, there is a tendency to formulate a term by combining a qualifier
describing group-ness of the phenomenon with one or two words describing the processor. In a
number of cases, a qualifier that describes some specific attributes of the processor, such as
‘situated’ (Elsbach, Barr et al. 2005) or ‘transactive’ (Wegner 1986), is used instead of the ones
that show group-ness. If one reviews the previous research on the processor, he or she is likely to
come across many of the combinations of the listed words: shared mental model (Levesque,
Wilson et al. 2001; Mohammed and Dumville 2001; Salas and Cannon-Bowers 2001), team
mental model (Klimoski and Mohammed 1994; Mohammed and Dumville 2001), social
cognition (Schneider and Angelmar 1993), collective cognition (Gibson 2001), collective mind
(Weick and Roberts 1993), group mind (Wegner 1986; Sandelands and Stablein 1987), etc. How
they define and use these terms varies even more. There are cases where the same terms mean
significantly different concepts and cases where different terms are used interchangeably to mean
one concept.
Qualifiers for Group-ness: Name for the Processor:
Table 1. | Shared Knowledge Structure
Examples of Words | Team Cognitive Structure/System/Framework
Used to Formulate cor Cognition
‘ collective emory
a Descriptor Group Mental Model
for the Processor Organizational Schema
Negotiated Culture
we Minds
Representation
Special Qualifiers: Interpretation System
Managerial Belief Structure/System
Situated Social Order
Transactive Perceptual Filters
Having diverse terms and definitions has some benefits. It demonstrates the fact that
researchers have explored various attributes of the processor and exhibited that what they are
discussing is in fact not one concept, but different concepts with possible overlaps among them.
However, a problem is that there has been little effort to identify how these concepts differ from
one another. When a cross-referencing occurs, do the authors mean the same thing?
A failure to clearly identify differences among the terms and definitions resulted in
several problems. First, cases of circular definition can be found in the literature. Circular
definition exists when a term used to describe the processor is defined with another term for the
processor and the descriptor term requires a further clarification of definition. An example would
be defining a team mental model as a collective knowledge structure. Then, what is a collective
knowledge structure? Second, there are cases where the same words have different meanings.
For example, some use ‘shared’ to mean aggregated. Others use the same word to mean
commonly-held. It can also mean communicated in a group. Which one is it? A more serious
problem exists with the level of analysis. Schneider and Angelmar (1993) pointed out that
studies based on assumptions and methods from cognitive psychology tend to focus on the
individual level of analysis while those based on sociology and anthropology focus on group or
organization as the level analysis. They argued that these differences led to problems when
interdisciplinary integration results in use of the same term to describe cognitive phenomenon at
different levels of analysis. Finally, there has been confusion about what the content of the study
is: is it about a static product such as knowledge or memory or is it about process such as
interaction and communication in a group? It is an important distinction, because selection of
measurement and analysis methods depends on the form of the content being studied.
The most extensive work on classification of the processor concepts can be found in
Schneider and Angelmar (1993) and Walsh (1995). The former categorized previous studies on
the processor by the level of analysis (individual-group-organization) and the form of cognition
(structure-process-style). The latter used the level of analysis (individual-group-organization-
industry) and the content of knowledge structure (representation-development-use). These two
studies provided a useful framework for organizing various processor concepts and addressed the
problem of confusion in level of analysis and in the attributes of the processor being studied.
However, there are two issues that require further consideration:
First, some studies on the processor cannot be clearly categorized by level of analysis,
especially when studies are interested in the group or organization level processors. For example,
if cognition occurs at the individual level, but sharing of knowledge occurs at the group level,
what should be the level of analysis? The answer is closely related to the content of the study. Is
the focus of the study cognition at the individual level or sharing at the group level? What if one
phenomenon cannot be studied without looking at the other? In other words, because the
phenomenon at different levels interact and influence one another, it is not always possible to
identify one study with one level of analysis. Second, we cannot directly use the form or the
content categories proposed by the aforementioned studies, because the categories were
introduced to organize the previous research on the processor rather than the definitions of the
processor. For example, Walsh’s (1995) three-tier model categorized the previous studies on
knowledge structures into those that focused on representation, development, or usage aspects of
knowledge structures. Since this paper is interested in the organization of different processor
concepts rather than research objectives of the previous studies, it was necessary to develop a
framework that can be used to identify different contents of the processor. To address these
issues, the author developed a framework that has been modified from the aforementioned works.
A Model to Capture C onceptual Differences in the Terms
The framework proposed in this study is different from the previous works in that it is
based on the idea of a continuum (See Figure 2). It recognizes that it is not always easy to
identify different terms and definitions with clear-cut categories.
Figure 2. Framework of Location- Form Continuum
Form of the Processor
Product <——————_» Process
Individual
Location
of the
Processor
Collectivity
One dimension of the continuum is the location of the processor, or where the black box
is located. At one extreme, the entire processor could be located at the level of an individual. In
other words, thinking is done at the individual level only, and even in a group setting, the locus
of study is on individual mind process. At the other extreme, the entire processor could be
assumed to be located at the collective level. The collective level refers to the group,
organization, or even industry, and it is the collectivity that receives information, processes
information, and generates action. A similar criteria is briefly mentioned by Lant (2002). He
explained that the “locus of organizational cognition” (p. 355) is dependent on the perspectives
of organizational cognition studies, and he identified two key perspectives. In the information
processing perspective, organizational cognition takes place in individual minds, while from the
enactment perspective, it takes place at the collective level.
The location of the processor is an alternative framework to the level of analysis. It is
different from level of analysis in several ways. First, it explicates researchers’ assumptions
about the locus of thinking. Second, it differentiates where thinking takes place (i.e. location of
the processor) from where thinking is observed (i.e. level of analysis). Third, by using a
continuum, it acknowledges that definitions for the processors cannot be always clearly
categorized into one of the two locations.
The second dimension of the proposed continuum is the form of the processor. At one
extreme, researchers are interested in a static product form such as memory and knowledge. At
the other extreme, researchers are interested in process that changes in the course of group
dynamics. Examples would be communication and interaction. Again, there are studies that
define the processor in terms of both products and processes. Therefore, the continuum idea will
be useful to deal with such a problem.
With this framework, it is now possible to organize different terms and definitions for the
processor in the continuum.
Terms and Definitions and their Positions on the Continuum
Cognitive psychologists have been inclined to use the term mental model or cognition.
Although mental model became a wide spread term and is used in general to mean thinking,
many researchers using the term, including Doyle and Ford (1998), reference back to Johnson-
Laird (1983)’s book Mental Models. When used in a group setting, a mental model is often
referred to as a shared mental model or a team mental model. Cannon-Bowers and Salas have
been working on the development of a shared mental model concept since early 1990 (Cannon-
Bowers, Salas et al. 1993). They define a shared mental model as “knowledge held by a team
member that is either compatible, complementary, and/or overlapping with teammates” (Salas
and Cannon-Bowers 2001: p.87). Similarly, Klimoski and Mohammed (1994) define a team
mental model as “what is being shared and operating among team members as a collectivity” (p.
414). In the same study, they use the term shared cognition interchangeably with team mental
model, and emphasized that a team mental model or shared cognition is one’s knowledge about
his or her team members’ knowledge and is different from unarticulated subconscious
knowledge. In other words, their definition of team mental model is narrower than that of
Cannon-Bowers and Salas, because to qualify as a team mental model, the knowledge not only
has to be shared among team members, but in addition the fact that it is being shared must be
known to all members. Related to these definitions, Smith-Jentsch, Campbell et al. (2001) used
the term mental model of teamwork, to mean “individual’s understanding of the components of
teamwork that are critical for effective team performance” (p.180). In all these aforementioned
definitions, location of the processor is within individuals, and the processor is in product form,
ie. knowledge. In sum, these studies are interested in the part of individual knowledge that is
shared among the team members. Therefore, the position of their definitions in the continuum
framework is at the individual-product ends (See Figure 3).
In contrast to the above definitions, shared mental model defined by Levesque, Wilson et
al. (2001) moves slightly towards the collectivity-process ends of the continuum. Their interest is
in overlapping cognition which they associate with enhancement in team performance. Since
overlapping can only occur though coordination and communication, they look at interaction
leading to convergence of mental models over time. In other words, while mental models are
primarily located in individuals, the process of sharing modifies the individuals’ mental models.
When the location of the processor is at the individual end of the continuum, the same
concept may be used to describe a phenomenon at both the individual level and the group level.
One example would be the concept of a schema. Many definitions of schema can be found in the
literature (Fiske and Taylor 1984; Walsh 1995; Elsbach, Barr et al. 2005), but in general,
schemas are defined as cognitive templates or simplified representation of knowledge that are
used to identify elements of a situation and relationships between these elements (Elsbach, Barr
et al. 2005). Because schemas are individually held, are relatively stable, and are seldom
observed, schema discussions rarely differentiate between the individual and
group/organizational level processes. A concept similar to schema is knowledge structure.
According to Walsh (1995)’s definition, “a knowledge structure is a mental template that
individuals impose on an information environment to give it form and meaning” (p. 281). If
positioned in the location-form framework, schema and knowledge structure would be located at
the individual-product ends of the continuum.
There is a group of researchers who use the term cognition at the collective level.
Nicolini (1999) identifies organizational cognition as “social process of cognition and thinking
at organizational level” and differentiates it from “cognitive process at individual level in
organizational setting” (p.834). He makes it clear that organizational cognition is a social process
and the locus of the processor is at the collectivity. However, Tegarden and Sheetz (2003) use
the same term to mean “shared understanding that managers have in common with each other” (p.
114). Compared to Nicolini’s perspective, Tegarden and Sheetz’s organizational cognition
closely resembles the shared mental model defined by Cannon-Bowers and Salas (1993). There
are others who prefer to use the terms collective cognition. Gibson (2001) defines collective
cognition as the four phases of accumulation, interaction, examination, and accommodation that
take place at the group level. Langfield-Smith (1992) also emphasized that collective cognition is
an outcome of social process or collective encounter, and stressed its transitory nature. With an
exception for Tegarden and Sheetz’s, the definitions of organizational or collective cognition can
be positioned near the collectivity-process ends of the continuum.
Weick and Roberts (1993) called the processor collective mind or group mind. They
defined a collective as individuals in a group who inter-relate their actions with care, and mind as
an activity and not as an entity. They explained that collective mind is located in the process of
“heedful interrelating” (p.361). Therefore, their definition can be interpreted as individually
located processors which operate during an interrelating process among the individuals. This
definition thus falls on the individual-process side of the continuum. A similar term used is
organizational mind. Sandelands and Stablein (1987) defined organizational mind as an
ideational process that is carried out by organizational behavior. They contrast their view with
others who regard mind as “a substance or static pattern of relationships” (p.138), and
emphasized that complex interaction of ideas is at the heart of the organizational mind.
According to Sandelands and Stablein, organization is mind, and the criteria for it to qualify as
mind should be different from that for the human mind. Their definition of organizational mind
is clearly on the collectivity-process ends of the continuum. As a theory of group mind, Wegner
(1986) proposed the idea of transactive memory. He defined it as “a set of individual memory
systems in combination with the communication that takes place between individuals” (p.186). It
is cognitive interdependence among team members: a way that individuals use other people’s
memory as an external memory. In organizations, transactive memory is a “shared system for
encoding, storing, and retrieving information” (Wegner, Erber et al. 1991: p. 923) and Wegner
explains that it is neither memory of individuals nor process of interaction among these
individuals, but it is combination of these two. Therefore, the position of transactive memory
would be somewhere in the middle of the product-process and the individual-collectivity
continuum.
Shared belief system and collective belief system are used interchangeably with many of
the terms listed previously. Usually they are regarded as a higher level construct that embraces
attributes of the processor discussed by different researchers, and when used in the literature, the
definition for shared belief system or collective belief system is rarely provided. One special case
is the study by Langfield-Smith (1992). She differentiated the collective belief system from
collective cognition and explained that collective action or decision can exist without a collective
belief system as long as collective cognition takes place. In other words, Langfield-Smith
regarded collective belief system as shared value among a group of people. In contrast to his
definition of collective cognition, the definition of collective belief system places the term
towards the individual-product ends of the continuum.
In discussing the collective belief system, Walsh and Fahey (1986) and Walsh,
Henderson et al. (1988) focused more on the political process within a group. They posited that
power differences among individuals influence whose knowledge structures are better
represented in the group’s collective knowledge structure. They called the aggregated model a
negotiated belief structure. The position of the negotiated belief structure in the continuum
framework is somewhat less clear. Although the negotiated belief structure is a knowledge
structure, i.e. a product, what generates the product is a political process of negotiation and
influence. The location of the processor is also a combination of individual and collectivity,
because while negotiation takes place in collectivity, influence of powerful individual in creating
negotiated belief structures is substantial.
10
Figure 3. Positions of the Terms Used in the Literature to Describe the Processor
Collective Mind (Weick and Roberts 1993)
Perceptual Filters (Starbuck and Milliken 1988)
Organizational Interpretative System
(Daft and Weick 1984)
Organizational Cognition (Nicolini 1999)
Collective Cognition (Gibson 2001)
Collective Cognition (Langfield-Smith 1992)
Form of the Processor
Product
Individual
4 Shared Mental Model (Cannon-Bowers, Salas et al. 1993)
‘Team Mental Model (Klimoski and Mohammed 1994)
Mental Model of Teamwork (Smith-Jentsch, Campbell et al. 2001)
Schema (Elsbach, Barr et al. 2005)
Knowledge Structure (Walsh 1995)
Collective Cognition (Tegarden and Sheetz 2003)
Collective Belief System (Langfield-Smith 1992)
Shared Mental Model (Levesque, Wilson et al. 2001)
Location
‘Transactive Memory (Wegner 1986)
of the Negotiated Belief Structure (Walsh
Processor and Fahey 1986)
Organizational Memory (Levitt and March 1988) Organizational Mind (Sandelands and Stablein 1987)
v
Collectivity
11
There is a group of researchers who emphasize interpretation and meaning-creation
aspects of the processor. Daft and Weick (1984) called the processor organizational
interpretation systems. They defined it as “the process of translating events and developing
shared understanding and conceptual schemes among key managers” (p.286). Daft and Weick
explained that organizations differ in their attitudes towards interpretation and these attitudes are
determined by managers’ assumptions about the environment and by the organization’s
intrusiveness into the environment. A similar term used is organizational sensemaking (Weick
1995). Starbuck and Milliken (1988) combined the process of noticing and sensemaking and
called it perceptual filters. Noticing is a process of looking for information that matters, and
Starbuck and Milliken explain that individual’s previous experiences influence what information
he or she selects to look at. Sensemaking is creating meaning from the selected information, and
this process is also influenced by one’s previous experience. Perceptual filters operate at the
individual level, but the perceptual filters of executives or managers play a critical role at the
organization level. Daft and Weick’s organizational interpretation system and Starbuck and
Milliken’s perceptual filters are both process-oriented concept. However, while the former
assumes interpretation at both the individual and collective level, the latter is about individual
interpretation. Therefore, their positions the continuum differ in terms of the location of the
processor.
Levitt and March (1988) used the term organizational memory to describe the
organizational learning process. It is defined as routines, rules, and procedures in organizations
where organizational members’ experience and knowledge are accumulated and maintained.
This organizational memory is an artifact of organizational behaviors and is separate from
members of the organization. This leaves organizational memory at the product-collective ends
of the continuum.
Modeling Methods and the Processor
The survey of terminology and definitions for the processor reveals that while researchers
appear to discuss similar concepts, there are subtle differences among the concepts discussed.
When placed on the continuum framework, the differences become explicit and clear. The
literature review and the positioning exercise in the previous section lead to the conclusion that
the mental-model-like concept at the group level may not be one concept, but are more likely
different concepts along the continuum. Then how do the conceptual subtleties described in the
previous section allow system dynamicists to clarify what is represented in our maps and
models?
Maps and models are created, because we are interested in describing, understanding, and
improving the processor. However, the maps and models are not “the processor” per se, because
they are created from observable behaviors of individuals or a group which only partially reflects
the processor. What is represented in our maps and models is a combined product of the
processor of individuals or a group subject to study, the modeler’s own mental model, the
modeler’ s preferred methods of eliciting and representing the processor.
In this section, five ideal types of group-level modeling modes will be identified. Each
modeling modes makes different assumptions about the processor's location and form. In other
words, depending on the modeling modes used, the modeler will approach different aspects of
the processor.
12
The variation along the continuum is mainly determined by when and how a modeler or
facilitator intervenes with the subject or client group. The modeling process can be divided into
the elicitation, representation, and reflection/analysis phases. In the elicitation phase, the raw
data about the processor is collected from the subject of the study. If a system dynamics modeler
attempts to understand a mental model of an individual, the modeler would interact with the
subject or the client in order to obtain data that tells something about the person’s perception of
the system. A typical method for elicitation includes interview, survey, content analysis, and
observation. In the representation phase, the elicited part of the processor is represented using
causal loop diagrams or formal simulation models. Then in the reflection phase, the represented
maps or models are analyzed for insights.” Group level modeling adds another layer to the
process. There is a phase where individual level processors are integrated into a group level
processor. And when and how this integration takes place shapes what end up being elicited from
the subject or the client group and represented in the maps and models.
The following is a list of five ideal modes of group level modeling. Each modeling modes
are characterized by (1) a different sequence of the four phases of elicitation, integration,
representation, and reflection, and by (2) main player in each phase.
Mode 1: Interview with Individuals in a Group
A modeler can create a map or a model from the data collected from interviews of
individuals in the subject or the client group. In this case, the elicitation phase comes prior to the
integration phase. In order to understand the group level phenomenon, the modeler integrates the
data collected at the individual level. When data is collected at the individual level and combined
later by the modeler, there is no group interaction taking place among the individuals during the
data elicitation or integration period. In both phases, it is the modeler who takes the leading role.
In this modeling mode, the modeler is studying the processors located at the individual level and
the form of the processor is static knowledge or memory related to a problem (See Figure 4).
Mode 2: Content Analysis of Data Collected from Individuals in a Group
A modeler can create a map or a model from the data collected by individuals in the
subject or the client group. As in the modeling mode 1, the elicitation phase comes prior to the
integration phase, but the elicitation phase is now controlled by the subject or the client group. In
order to understand the group level phenomenon, the modeler integrates the individual level data
provided by the clients. Again, there is no group interaction taking place among the individuals
during the elicitation or integration period. The position of the processor being studied is similar
to the modeling mode 1: the individual-product end of the continuum. Integration is done either
by modeler identifying overlapping structures from individual mental models or by combining
all the structures mentioned by every individual. If the former is the case, what the modeler is
trying to capture is similar to that of a shared mental model or team mental model defined by
Cannon-Bowers and Salas (1993) and Klimosky and Mohammed (1994). If the latter is the case,
? What is represented in maps and models are determined by elicitation, integration, and representation phase.
However, if a modeling is a reiterative process, reflection phase also influence what is represented in the maps and
models.
13
the term aggregate, congregate, or composite used by Bougon (1992) may replace the term
shared.
Mode 3: Content Analysis of Data Collected from a Group Process
When individuals in a group interact and communicate, integration takes place prior to
elicitation. The degree of modeler’s participation in the integration phase determines the nature
of the elicited product. In the modeling mode 3, a modeler does not observe interactions among
individuals during the integration process. Rather, the integrated output from the group process is
compiled by the subject group and it is later presented to the modeler. In other words, the
modeler does not have influence over the elicitation or integration process, and it is the subject or
the client group who takes the leading role in these modeling phases. In this case, the processor
studied by the modeler is at the product-collective end of the continuum. It is a product, because
the modeler can only study an artifact of an organizational process. The processor is located at
the level of collectivity, because the modeler is forced to see the data as it is collected from the
group without having access to individual mental models.
Mode 4: Non-participant Observation of a Group Process
In some cases, a modeler has access to the integration process and is able to observe
group interactions and is thus able to collects data during the observation. The degree of modeler
participation in the integration phase differentiates the modeling mode 4 from 5. When a modeler
observes the group process and collects data during the observation, it is a non-participant
observation method. The processor studied is located somewhere in between the individual and
the collectivity. It is so, because the interaction of individuals within the collectivity is where the
processing takes place. In terms of the form of the processor, a modeler as a passive observer
will focus more on the process than the product, because instead of exploring what one may
think, the modeler is paying attention to conversation taking place in the meeting. What is being
said in the group process is often shaped by one’s motivation to influence others’ and a
substantial part of this is inseparable from the political and social process.
Mode 5: Modeler-leading Group Process
Finally, there is a modeler-leading group process, as in group model building (Richardson
and Andersen 1995; Vennix 1996; Andersen and Richardson 1997). In this mode, the modeler
plays an active role in the integration process. An active modeler becomes a leading facilitator
and induces the subject or client group to discuss parts of their mental models that are more
relevant or important to the problem at hand. The modeler also explores where the mental
models conflicts, and by promoting group communication, the modeler allows the group to come
to an agreement. As in the non-participant observation, the processor studied is located
somewhere in between the individual and collectivity. However, an active modeler may have
access to a product form of the processor as well as a process form, because the modeler can
probe knowledge and memory from individuals in the group.
14
The five modeling modes and their modeling sequences are summarized in Table 2. The
parentheses indicate the main player of each phase, which is either modeler or the subject/client
group. Figure 4 summarizes the location and form of the processor studied by each ideal modes
of modeling.
Table 2. Five Ideal Modeling Modes
Modeling Sequence
1. Interview with individuals in a Elicitation Integration Representation | Reflection
group (modeler) (modeler) (modeler) (both)
2. Content analysis of data collected Elicitation Integration Representation | Reflection
from individuals in a group (subject) (modeler) (modeler) (both)
3. Content analysis of data collected Integration Elicitation Representation | Reflection
from a group process (subject) (Subject) (modeler) (both)
4. Non-participant observation of a Integration Elicitation Representation | Reflection
group process (subject) (modeler) (modeler) (both)
Integration Elicitation Representation | Reflection
5. Modeler-leading group process (both) (modeler) (modeler) (both)
Implications for the System Dynamic Group Level Modeling
The five ideal types of modeling mode implicitly make different assumptions about the
processor. In other words, depending on our selection of modeling modes, we will look at
different aspects of a black box.
In real modeling practices, we rarely resort to one type of modeling mode. Rather we use
combination of different methods. A modeler using multiple methods will move along the
location-form continuum exploring different aspects of the processor. Depending on the location
and form of the processor, a set of theoretical literature relevant to the processor will be different.
For example, if a modeler is interviewing individuals in a group, the processor is relevant to a set
of literature on the individual-product end of the continuum. Schema (Elsbach, Barr et al. 2005),
knowledge structure (Walsh 1995), and collective cognition (Tegarden and Sheetz 2003) would
be a few examples. On the other hand, if the modeler is conducting a modeler-leading group
process, he or she might want to look at a different set of literature such as Walsh and Fahey
(1986)’s work on negotiated belief structure.
Although an organization of the processor using the location-form continuum provide a
useful guide to relate the processor assumed in our maps and models to the theoretical literature,
a co-location of a map-building method and a theoretical work does not mean that the two are
assuming the exactly the same processor. The co-location is a loose categorization, since the
location and the form are not the only criteria defining a processor. A similarity in the location
and the form does not necessarily guarantee the two processor share the exactly the same
characteristics.
15
Figure 4. Positions of the Processor Represented in SD Maps and Models
Form of the Processor
Product
Individual
4
Mode 1: Interview with individuals
inagroup
Mode 2: Content analysis of data
collected from individuals in a
Location
of the Mode 5: Modeler-leading group
Processor process
Mode 3: Content analysis of data
collected from a group process
¥
Collectivity
Mode 4: Non-participant
observation of a group process
16
Conclusions and Limitations
This paper attempted to find an alternative concept to “mental model” in a group
environment. The concept was called “the processor” for the convenience of the analytical
discussion. In order to ground the alternative term in the literature, this study surveyed
terms and definitions that have been used to describe the processor at group or
organizational level.
The main finding from the survey was that different terms used to describe the
processor have differences in terms of their assumptions about location and form of the
processor. This finding led to an insight that what system dynamics modelers try to
understand, represent, and improve in a group intervention processes is not a single concept
that parallels mental model at individual level. Depending on the data collection and
integration methods, the concept can vary in terms of the location and form of the processor.
In other words, we cannot have one single term that describes the mental-model like
concept at the group level, because sometimes we are studying an interaction process
within a group and at other times we are looking for knowledge held by individuals in a
group.
This study concludes without providing “the” alternative term that can be used like
“mental model” in group environment. However, this study highlights that the substance of
our study may differ despite we are all engaged in system dynamics group intervention, and
it is important to explicitly recognize what is represented with our maps and models.
The location-form continuum framework proposed in this study has some
limitations. Because it uses the concept of a continuum, determining relative positions of
different terms in the space requires the author's subjective interpretation of the definitions
given in the literature. While some definitions have clear statement of location and forms of
the processor, there are other definitions which are conceptually vague on these factors. If
that was the case, more in-depth study of the literature was carried out for clarification.
The scope of literatures reviewed in this paper may pose another problem. There are
terms and definitions relevant to the processor not included in this study. However, the
main goal of this paper was to show the diversity of terms used to describe the processor,
and to highlight the subtle differences among these terminologies, and to propose a way to
organize them.
Finally, it must be noted that there is a difference between what we are interested to
study and what has been actually elicited and represented using our maps and models. We
may say we are trying to understand, describe, and improve mental model. If we are
studying a group level phenomenon, we may say we are interested in eliciting shared
mental model, or negotiated belief structure, or whatever the descriptor we prefer to use.
However, we must stay away from referring to our maps and models as mental models or
the similar concept at the group level, because maps and models are only representation
tools used by modelers.
17
References
Andersen, D. F. and G. P. Richardson (1997). "Scripts for Group Model Building." System
Dynamics Review 13(2): 107-129.
Bougon, M. G. (1992). "Congregate Cognitive Maps: A Unified Dynamic Theory of
Organization and Strategy." Journal of Management Studies 29(3): 369-389.
Cannon-Bowers, J. A. and E. Salas (2001). "Reflections on Shared Cognition." Journal_of
Organizational Behavior 22(2): 195-202.
Cannon-Bowers, J. A., E. Salas, et al. (1993). Shared Mental Models in Expert Team Decision
Making. Individaul and Group Decision Making. N. J. Castellan. Hillsdale, NJ, Lawrence
Erlbaum Associates: 221-246.
Daft, R. L. and K. E. Weick (1984). "Toward a Model of Organizations as Interpretation
Systems." Academy of Management Review 9(2): 284-295.
Doyle, J. K. and D. N. Ford (1998). "Mental Models Concepts for System Dynamics Research."
System Dynamics Review 14(1): 3-29.
Doyle, J. K. and D. N. Ford (1999). "Mental Model Concepts Revisited: Some Clarifications and
a Reply to Lane." System Dynamics Review 15(4): 411-415.
Elsbach, K. D., P. S. Barr, et al. (2005). “Identifying Situated Cognition in Organizations."
Organization Science 16(4): 422-433.
Fiske, S. T. and S. E. Taylor (1984). Social Cognition. Reading, MA, Addison-Wesley
Publishing Company.
Forrester, J. W. (1961). Industrial Dynamics. Cambridge, MA, MIT Press.
Gibson, C. B. (2001). "From Knowledge Accumulation to Accommodation: Cycles of Collective
Cognition in Work Groups." Journal of Organizational Behavior 22(2): 121-134.
Huff, A. S. (1990). Mapping Strategic Thought. Mapping Strategic Thought. A. S. Huff. West
Sussex, England, John Wiley and Sons, Inc.: 11-49.
Johnson-Laird, P. N. (1983). Mental Models: Towards a Cognitive Science of Language,
Inference, and Consciousness. Cambridge, Cambridge University Press.
Klimoski, R. and S. Mohammed (1994). "Team Mental Model: Construct or Metaphor?" Journal
of Management 20(2): 403-437.
Lane, D. C. (1999). "Friendly Amendment: A Commentary on Doyle and Ford's Proposed Re-
definition of ‘Mental Model'." System Dynamics Review 15(2): 185-194.
Langfield-Smith, K. (1992). "Exploring the Need for a Shared Cognitive Map." Joumal of
Management Studies 29(3): 349-368.
Lant, T. K. (2002). Organizational Cognition and Interpretation. The Blackwell Companion to
Organization. J. A.C. Baum. Oxford, Blackwell Publishers.
18
Levesque, L. L., J. M. Wilson, et al. (2001). "Cognitive Divergence and Shared Mental Models
in Software Development Project Teams." Journal of Organizational Behavior 22(2):
135-144.
Levine, J. M., L. B. Resnick, et al. (1993). "Social Foundations of Cognition." Annual Review of
Psychology 44: 585-612.
Levitt, B. and J. March (1988). "Organizational Learning." Annual Review of Sociology 14:
319-340.
Luna-Reyes, L. F., I. J. Martinez-Moyano, et al. (2006). "Anatomy of a Group Model-Building
Intervention: Building Dynamic Theory from Case Study Research." System Dynamics
Review 22(4): 291-320.
Mohammed, S. and B. C. Dumville (2001). "Team Mental Models in a Team Knowledge
Framework: Expanding Theory and Measurement across Disciplinary Boundaries."
Journal of Organizational Behavior 22(2): 89-106.
Nicolini, D. (1999). "Comparing Methods for Mapping Organizational Cognition." Organization
Studies 20(5): 833-860.
Richardson, G. P. and D. F. Andersen (1995). "Teamwork in Group Model Building." System
Dynamics Review 11(2): 113-137.
Richardson, G. P., D. F. Andersen, et al. (1994). Foundations of Mental Model Research. The
12th Intemational System Dynamics Conference, Stirling, Scotland, System Dynamics
Society.
Rouse, W. B. and N. M. Morris (1986). "On Looking Into the Black Box: Prospects and Limits
in the Search for Mental Models." Psychological Bulletin 100(3): 349-363.
Rouwette, E.n.A.J.A., J. A. M. Vennix, et al. (2002). "Group Model Building Effectiveness :
A Review of Assessment Studies." System Dynamics Review 18(1): 5-45.
Salas, E. and J. A. Cannon-Bowers (2001). "Special Issue Preface." Journal of Organizational
Behavior 22(SI): 87-88.
Sandelands, L. E. and R. E. Stablein (1987). The Concept of Organization Mind. Research in
Sociology of Organizations. S. Bacharach and N. DiTomaso. Greenwich, CT, JAI Press.
5: 135-162.
Schneider, S. C. and R. Angelmar (1993). "Cognition in Organizational Analysis: Who's
Minding the Store?" Organization Studies 14(3): 347-374.
Simon, H. A. (1947/1997). Administrative Behavior: A Study of Decision-Making Processes in
Administrative Organizations. New Y ork, The Free Press.
Smith-Jentsch, K. A., G. E. Campbell, et al. (2001). "Measuring Teamwork Mental Models to
Support Training Needs Assessment, Development, and Evaluation: Two Empirical
Studies." Journal of Organizational Behavior 22(2): 179-194.
Starbuck, W. H. and F. J. Milliken (1988). Executives' Perceptual Filters: How They Notice and
How They Make Sense. The Executive Effect: Concepts and Methods for Studying Top
Managers. D. C. Hambrick. Greenwich, CT, JAI Press Inc.: 35-65.
19
Steiner, G. A. and J. B. Miner (1977). Management Policy and Strategy. New Y ork, MacMillan
Co.
Sterman, J. D. (1994). “Learning in and about Complex Systems." System Dynamics Review
10(2/3): 291-330.
Tegarden, D. P. and S. D. Sheetz (2003). "Group Cognitive Mapping: A Methodology and
System for Capturing and Evaluating Managerial and Organizational Cognition." Omega
31(2): 113-125.
Vennix, J. A. M. (1996). Group Model Building : Facilitating Team Learning Using System
Dynamics. Chichester, Wiley.
Walsh, J. P. (1995). "Managerial and Organizational Cognition: Notes from a Trip Down
Memory Lane." Organization Science 6(3): 280-321.
Walsh, J. P. and L. Fahey (1986). "The Role of Negotiated Belief structures in Stategy Making."
oumal of Management 12(3): 325-338.
Walsh, J. P., C. M. Henderson, et al. (1988). "Negotiated Belief Structures and Decision
Performance: An Empirical Investigation." Organizational Behavior and Humand Decisio
Processes 42(2): 194-216.
Wegner, D. M. (1986). Transactive Memoery: A Contemporary Analysis of the Group Mind.
Theories of Group Behavior. B. Mullen and G. R. Goethals. New Y ork, Springer-Verlag:
185-208.
Wegner, D. M., R. Erber, et al. (1991). "Transactive Memory in Close Relationships." Journal of
Personality and Social Psychology 61(6): 923-929.
Weick, K. E. (1995). Sensemaking in Organizations. Thousand Oaks: CA, Sage Publications.
Weick, K. E. and K. H. Roberts (1993). "Collective Mind in Organizations: Heedful Interrelating
on Flight Decks." Administrative Science Quarterly 38(3): 357-381.
20
