|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2010. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 97
3
Verbal Conceptual and Cultural Models
I
n this chapter we discuss models that are not instantiated in formal
algorithms or software, but in words. Verbal conceptual models are
presented first, followed by verbal cultural models.1 These models are
important for their attempts to apply theoretical constructs to the behavior
of individuals and groups. The models and the terms and constructs they
encompass may provide a foundation for some of the more applied and
formal models discussed later. These models have been developed in the dis-
ciplines of social psychology, sociology, anthropology, and organizational
behavior studies.
vERBAL CONCEPTuAL MODELS
What Are verbal Conceptual Models?
Verbal conceptual models characterize entities, variables, or events/
processes/mechanisms and the relations among them in words, not in equa-
tions or other mathematical or operational formulations. Although they
may use mathematical terms—for example, Kurt Lewin’s statement that all
behavior is a “function” of the person and the situation (1951)—the nature
1 Note that, in general, both conceptual models and cultural models can be articulated in
formal logical, mathematical, algorithmic, or computational forms. Our focus in this chapter
is on verbal representations of conceptual and cultural models, as an initial stepping stone to-
ward computational implementations in individual, organizational, and societal (IOS) models
and simulations.
OCR for page 98
BEHAVIORAL MODELING AND SIMULATION
and form of relations described in a verbal model are commonly under-
specified compared with formal models. Verbal conceptual models include
very general classifications or broad characterizations that provide the
foundation for a new discipline, such as the brain-as-computer metaphor on
which modern cognitive science was founded, and mid-level frameworks,
such as “images” of organizations (Morgan, 1997) as machines, or organ-
isms, or brains. Typologies or taxonomies, such as a taxonomy of emotional
states (Borgatti, 1994) or a typology of small groups (Arrow, McGrath, and
Berdahl, 2000), are another form of verbal conceptual model. Most numer-
ous of all are the small-scale models that characterize relations among vari-
ables or processes relevant for understanding a specific phenomenon. The
“progression of withdrawal” and “compensatory behaviors withdrawal”
models, for example, are alternate models of job withdrawal (quitting and
absenteeism) (Hanisch, 2000); another example is a two-variable model of
how social norms emerge in a newly formed group, based on whether or
not new members’ characterizations of the situation and the “scripts” they
retrieve to guide behavior match (Bettenhausen and Murnighan, 1985).
The use of such terms as “theory,” “framework,” “model,” and
“paradigm” in psychology and the social sciences is as informal as the
models themselves.2 One person’s conceptual model is another person’s
theory or framework. In this chapter, we use the term “conceptual model”
(and, for brevity, sometimes just “model”) as a way to group theories,
frameworks, and paradigms into rough classification systems based on
common features in structure (for example, dual process models, dynamic
models, threshold models) or relevant domain (group development models,
organizational withdrawal models, visual attention models). What verbal
conceptual models have in common is that they tend to be “highly infor-
mal constructions, use the natural language system, are rich in metaphor,
and use lavishly nuanced statements” (Davis, 2000, p. 218). If rendered as
diagrams instead of in straight prose, they tend to be represented via two-
by-two tables, labeled boxes with arrows drawn between them, or perhaps
a flowchart for a process model.
In psychology and the social sciences, theorizing about a problem typi-
cally begins with verbal conceptual models, which then may be elaborated
and adjusted over time as relevant empirical data accumulate. Formal
mathematical models, computational models, statistical models, etc. rely on
verbal conceptual models to specify variables and relations among them,
although a host of extra assumptions and plausible estimates are typically
needed to translate a verbal theory into a workable implementation. Hence
2 This is also true in the behavior modeling and simulation community, which is why we
attempted in Chapter 1 to identify and differentiate four levels of representation: theory, ar-
chitecture (here, framework), model, and simulation (here, paradigm).
OCR for page 99
VERBAL CONCEPTUAL AND CULTURAL MODELS
a computational model of emotional response relies on a conceptual tax-
onomy of emotional states (Ekman and Davidson, 1995), and a process
simulation model of jury decision making, such as DISCUSS (Stasser, 1988),
relies on the guiding metaphor of “interacting minds” engaged in “col-
lective information processing” (which represents the mind-as-computer
metaphor generalized to groups-as-networked-computers).
State of the Art for verbal Conceptual Models
Sophisticated verbal conceptual models (whose authors often call
them theories) are typically more specific about the nature of relations
among variables or about the nature of processes described than are ad hoc
models or global metaphor models. They may also be more sophisticated in
incorporating contingencies, dynamics, and multiple levels of analysis. For
example, in the study of leadership effectiveness, a very simple model, the
leadership grid (Blake and Mouton, 1982), proposes that leadership effec-
tiveness is explained by two dimensions—concern for people and concern
for production—and the more a leader has of both, the better. This model
focuses entirely on the leader (single level), entertains no contingencies,
assumes linear additive components, and has no dynamic elements. A more
sophisticated model, situational leadership theory (Hersey and Blanchard,
1988), proposes that the optimal mix of task-oriented and relation-oriented
behavior by leaders depends on the level of maturity and corresponding
skill level of the subordinate, which is expected to change over time. In a
heterogeneous group of members at different levels, effective leadership will
require that the leader tailor her style to individual members and adjust that
style as each member progresses through four successive levels of maturity
and autonomy. This model incorporates three levels (individual, dyad, and
group), contingencies (different levels of member development), and change
over time.
Computational models are often used to model complex processes
that unfold over time, so verbal conceptual models that include attention
to dynamics are particularly useful as a resource for the implementation of
more formal models. Verbal conceptual models of groups and organizations
can be arrayed along a continuum of increasing complexity using the four
different levels of complexity in time research of Ofori-Dankwa and Julian
(2001).3 First-level models focus on mean differences in, for example, how
much time a process takes and assume stationarity (sometimes implicitly
rather than explicitly). Second-level models add change as a possibility, so
that the rate of a process may speed up or slow down across time. Third-
3 The following summary is adapted from Arrow, Henry, Poole, Wheelan, and Moreland
(2005, pp. 313–368).
OCR for page 100
00 BEHAVIORAL MODELING AND SIMULATION
level models incorporate more than one hierarchical level of a system. For
example, the rate of change at the member, small group, and organizational
levels may be expected to differ systematically. Fourth-level models allow
for multiple simultaneous and potentially nonstationary processes at dif-
ferent levels.
Zaheer, Albert, and Zaheer (1999) introduced the concept of “time-
scale completeness” for a process model. In essence, they define what a
process model needs to specify to provide sufficient guidance in designing
a research program. Although their focus was on empirical data collection,
the same desiderata apply for implementing a verbal model in a compu-
tational form. A theory is time-scale complete if it specifies time scale for
all of its variables, relationships, and boundary conditions. For example, it
needs to specify the time needed for a complete instance of the phenomenon
to occur, the nature and rate of change in variables, and the duration and
sequence of any subphases in the process. Otherwise researchers cannot
make theory-driven choices of observation, recording, and aggregation
intervals, and the criteria for evidence either in support of or contrary to
model predictions remain unclear.
Finally, state-of-the-art conceptual models allow for conceptual “dock-
ing” with other models by clarifying how the terms used relate to other,
closely related (or synonymous) terms in the literature, and note where
other models might “plug in” (for example, a structural model might
refer to possible plug-in models that address processes or mechanisms not
included in but relevant to the structural model) and clearly specifying
boundary conditions.
Relevance to Modeling Requirements
One way to demonstrate the relevance of verbal models is to give an
example of how a well-developed verbal conceptual model could be used
for rapid cultural awareness training. The conceptual model is the cross-
cultural framework of Fiske (1991, 2000), which proposes that human
beings in all cultures coordinate their social interactions using a mix of fun-
damental relational models: communal sharing, authority ranking, equal-
ity matching, and market pricing. The four models are organized sets of
associated concepts and rules that serve as a generative grammar for think-
ing about and coordinating relationships.4 When following the communal
sharing model, people emphasize the common identity of group members
and focus on what is good for the group as a whole. The preferred model
of decision making is consensus and people pool resources and draw on the
pool without keeping track of individual contributions and withdrawals.
4 The following summary is adapted from Arrow and Burns (2004, pp. 176–178).
OCR for page 101
0
VERBAL CONCEPTUAL AND CULTURAL MODELS
Prototypical contexts and domains in which this model is used are family
and food. Families commonly share food resources freely, and people who
are defined as the “in group” in a particular context (such as invited guests
at a party) are expected to help themselves to whatever food and drink they
want. Violations of the rules occur when out-group members attempt to
access in-group resources (for example, someone crashes a party).
In relationships organized by the authority ranking model, people
structure their interactions according to status, position, and dominance
hierarchy. Military organizations commonly use this model, and personnel
wear insignias of rank to signal status. How people behave is strongly
governed by whether they have the higher or lower rank of the two people
in a given interaction. In distributing resources, high-status members get
more, and low-status members get less. Rank also comes with obligations:
superiors are expected to provide for or take care of inferiors. Violations
occur when lower status members are insubordinate, treating a higher
ranked person as an equal, for example, or when higher status members
abuse their rank and power and betray their obligations to lower status
followers or dependents.
When a relationship is governed by the equality matching model, people
reciprocate favors after some delay and maintain a balance between giving
and receiving. This model is commonly applied among people who consider
themselves to be of equal status, such as friends, classmates, or colleagues.
People in equality matching relationships often respond to favors by saying
“I owe you one” or “I’ll pay next time.” Note the difference from authority
ranking, in which a lower status person responds to favors with gratitude
and loyalty, rather than reciprocating in kind. If the relationship is using the
equality matching model, however, the failure to reciprocate a favor (or to
express one’s understanding of this obligation) would be a violation.
In market pricing relationships, people seek the best deal for themselves
and expect that others will do the same. This model commonly governs
trade and other social exchanges among strangers or acquaintances and
is guided by the equity principle of proportionality—so that price, for
example, should be proportional to value. Self-interested or selfish behavior
is not a violation (it is expected), but cheating or stealing (which violates
the equity rule) is.
Particular cultural implementations of these models organize social
exchange, distribution, contribution, decision making, social influence,
moral judgment, aggression, and conflict (Fiske, 1991). There is no prac-
tical limit, for example, to the types of objects or services that might be
deemed appropriate or inappropriate to reciprocate the gift of a chicken or
a radio. This sort of idiosyncratic and culturally specific content can be pro-
vided only by an informant who is very familiar with a culture. However,
more important to practical application in a field situation is simply detect-
OCR for page 102
0 BEHAVIORAL MODELING AND SIMULATION
ing whether the context in which a chicken is received is one that calls for
gratitude and acknowledgment of an in-group bond (communal sharing),
expression of deference and humility (authority ranking), reciprocation
with a favor of roughly equal value (equality matching), or direct payment
(market pricing). Mistakes involving specific cultural content (reciprocating
with an odd sort of gift) may invoke humor or surprise; violation of rela-
tional models (paying someone for a gift, failing to reciprocate) are more
likely to give offense and damage the relationship.
Major Limitations
The strengths of verbal models are also their weakness. Natural lan-
guage is a flexible and nuanced instrument in which one can express highly
sophisticated ideas, including multiple overlapping metaphors and embed-
ded narratives. However, because natural language is an encompassing
sea in which we all swim, shadings of meaning and idiosyncratic clouds
of associations allow four people to encounter the “same” verbal model
and understand it in four different ways. Some of the ambiguities and gaps
that are common in verbal models may become evident when designing an
experiment, and they are highlighted most sharply when one attempts to
extract a set of formal relations from a natural language model.
In psychology and the social sciences, the grand metaphors of con-
ceptual models often govern the whole direction of a field, but meta-
phors always direct attention to some features and lead to the neglect of
others. Once a broad conceptual framework such as this becomes perva-
sive, scholars tend to forget that a metaphor is involved. For example, the
information-processing metaphor for the brain, and the researchers who
focused on it, probably contributed to a pervasive neglect of research on
emotional and social processes for the first several decades of cognitive sci-
ence.5 Computers don’t have emotions and are not social beings. So if the
mind is not simply like a computer in some ways (simile with boundary
conditions), but is a biological computer (unreflective metaphor), the ways
in which minds are decidedly not like computers get overlooked, even by
researchers engaged in intensively social activities about which they have
strong feelings. The same curious social blindness is evident in the early
era of organizational research dominated by the organization-as-machine
metaphor. The related notion that workers are cogs in the machine led
researchers to study the impact of physical conditions, such as lighting, on
worker productivity while completely ignoring the possible impact of one
human being on another (Mayo, 1960).
5Asmall pocket of researchers clearly forged ahead in this important area; see, for example,
Ortony et al. (1988).
OCR for page 103
0
VERBAL CONCEPTUAL AND CULTURAL MODELS
verification and validation Issues
Verbal conceptual models are sometimes specific enough that they can
be tested and plausibly falsified, using empirical field studies or controlled
experiments. For example, in studies of subjects from Bengali, Chinese,
Korean, Vai (Liberia and Sierra Leone), and U.S. cultures (Fiske, 1992),
Fiske and colleagues have used social cognition experiments to demonstrate
that people organize acquaintances in memory according to the dominant
model that organizes the relationship and that for many subjects this clas-
sification accounts for more variance in recall and substitution errors than
such personal attributes as gender, race, and age.
In contrast to such well-developed conceptual frameworks, broad meta-
phors (brains as information-processing devices, organizations as cultures)
are not really subject to verification or falsification. Whether or not they
are used in a particular domain is likely to depend largely on face validity
and established precedent. In evaluating the usefulness of a verbal model of
this nature, the yardstick is often not how well supported the model is, but
how much interesting research it inspires. Even when a verbal model seems,
in principle, to be subject to falsification, the underspecification of relations
and processes often means that a rather broad array of different outcomes
can be presented as “consistent with” the theory. As Harris (1976) noted
in “The Uncertain Connection Between Verbal Theories and Research
Hypotheses in Social Psychology,” theoretical terms often are not defined,
boundary conditions are unspecified, and, under various plausible interpre-
tations of assumptions or conditions, several well-known theories include
internal contradictions and inconsistencies (as cited in Davis, 2000).
Future Research and Development Requirements
Verbal conceptual models can be highly influential and generative and
do not require intensive funding or technology to develop, yet the develop-
ment of such models is often overlooked as a funding priority. The scarce
resource in improving verbal theory is intellectual time and energy. Moti-
vation may also be an issue when grant funding is available primarily for
doing (conducting experiments, writing code, designing games, collecting
reams of data) and not for thinking. This can encourage the proliferation
of low-level, poorly specified, ad hoc conceptual models that get spawned
in discussion sections of journal articles to explain the results of a single
set of studies and, if they survive, are later herded together in introduction
sections of subsequent articles without actually being systematically inte-
grated into more comprehensive integrated models. That work is generally
left for the writers of literature reviews who are trying to make sense of a
mountain of facts and ideas and find a deeper order.
OCR for page 104
04 BEHAVIORAL MODELING AND SIMULATION
Stronger theory is needed for domains that social scientists still don’t
know how to think about and those in which numerous weak conceptions
have not been integrated. Verbal conceptual models are essential building
blocks for theory building. Bringing people together for conferences and
funding edited books and special issues that explore themes and issues in
depth are useful. Measurable advances in theory should also be specified as
a valuable deliverable for grants. Think tanks could be funded for scholars
to come together and work intensively for an extended period (three to
six months) on theory development and integration for issues and areas in
which it is increasingly clear not only that there are not enough data, but
also that it is difficult to know how to conceptualize the problem. Of course
this sort of conversation is going on in labs and institutes around the coun-
try, but the focus on generating data (at least in psychology) seems to eclipse
or marginalize the systematic development and integration of theory that
goes beyond the highly specific area in which people tend to do research.
CuLTuRAL MODELINg
What Is Cultural Modeling?
The term “cultural modeling” encompasses two broadly different areas
of research. One area is concerned with modeling growth and distribu-
tion of cultural phenomena, such as the evolution of norms or the diffu-
sion of beliefs. Research in this tradition typically treats culture (or, more
accurately, some characteristic of culture) as an outcome and concentrates
on the factors shaping those outcomes. This kind of cultural modeling is
distinguished from other kinds of modeling surveyed in this volume only
by the domain of study—namely, an element of culture. It does not imply
a particular modeling technique. For example, the evolution of norms may
be studied using a variety of methods, including multivariate statistics,
agent-based models, system dynamics models, event history models, and so
on. This kind of cultural modeling is discussed in several chapters in this
volume and is not discussed further here.
The other kind of cultural modeling, which is discussed here, is con-
cerned with describing (and often formally representing) a group’s culture.
Work in this tradition typically does not concern itself with how the culture
came to be but rather with how it is distributed in the population and, in
the best cases, what the consequences of having that culture might be.
Finally, it is appropriate to note that perhaps the most fundamental
verbal cultural models are those that are implicit in a region’s or society’s
language and history. It is abundantly clear—from Laurence of Arabia’s
exploits to today’s attempts to “democratize” Iraq—that deep and broad
knowledge of the local history and language are still fundamental for the
OCR for page 105
05
VERBAL CONCEPTUAL AND CULTURAL MODELS
kind of high-level understanding of societal dynamics that is the main focus
of this report and of today’s military. This committee acknowledges the
importance of both language and history as the foundational knowledge
base for any cultural model development, and as perhaps the starting point
for identifying “implicit” models embedded in the language and history—
models that can be built on in successive formalization efforts.
What Is Culture?
Culture can be defined in a number of different ways. Indeed, over
200 scholarly definitions have been documented (Kroeber and Kluckhohn,
1952). Researchers have defined culture in normative, historical, biological,
cognitive, functional, structural, categorical, and symbolic terms. Defini-
tions typically make use of some combination of the following elements:
beliefs, behaviors, values, customs, artifacts, organizational orientations,
preferences, experiences, attitudes, meanings, hierarchies, religions, percep-
tions, conceptions, material objects, possessions, symbols, motives, tradi-
tions, strategies, ideals, rules, habits, reasoning, identities, conventions,
customs, and institutions, among others. While definitions of cultures differ
on which of these elements constitute culture, most view a group’s culture
as an essential factor in problem solving, coping, and adapting to envi-
ronmental changes. In addition, they generally agree that culture is some-
thing possessed by groups (such as societies, organizations, occupations,
teams) and that it is learned, transmitted, and shared (albeit imperfectly
and unevenly). At the same time, scholars regard culture as being held in
individual minds and do not consider it an oxymoron to talk about an
individual’s culture.
State of the Art of Culture Models
There are four basic types of descriptive culture models popular today:
cultural inventory models, dominant trait models, semantic models, and
cultural domain models.
Cultural Inventory Models
Cultural inventory models are a way of describing cultures by list-
ing which of a list of traits they do or do not possess. Thus cultures are
conceived of as distinctive bundles of features that can be represented as a
string of 1s and 0s indicating the presence or absence of a trait. A number
of anthropologists have undertaken a compilation of cultural traits across
human societies. The best and most relevant example of such a compilation
across cultures is the Standard Cross-Cultural Survey developed by George
OCR for page 106
0 BEHAVIORAL MODELING AND SIMULATION
Murdock and others. The database consists of 186 societies and 22 cultural
categories involving almost 1,000 standard coded variables derived from
ethnographic sources (Murdock and Morrow, 1970). Essentially, a team of
researchers has combed through ethnographies written by anthropologists
and coded the cultures described using a universal codebook. The ability
to compare features across societies is critical for both developing models
and testing theories concerning patterns of and associations among cultural
traits, categories, and features. Table 3-1 provides examples of some of the
22 cultural categories and associated variables and their codes.
For military purposes (McFate, 2005), many of these traits may be
irrelevant, while others would need to be gathered, such as information on
cultural gestures (e.g., meaning of certain hand gestures), cultural greeting
etiquettes (e.g., rules for properly entering a village), cultural norms sur-
rounding conflict (e.g., cultural notions of courage, honor, and revenge),
etc. Such a database would considerably improve the ability to interact in
a satisfactory manner with natives and to accurately predict their reactions
to stimuli.
The key difficulty with cultural inventories is obtaining the necessary
data. Data need to be collected on an ongoing basis to ensure the quality
and timeliness of information. Also, it is important to recognize cultural
boundaries and subcultures. For example, a nation like China may form a
single political unit but may contain many different cultures. Furthermore,
collecting new cultural information can be particularly difficult during
periods of conflict, which means that the data need to be collected on an
ongoing basis regardless of whether it has immediate utility.
Another approach is the cultural classification system developed by
Karabaich, which is intended to cover the possible group types that might
be encountered in a military, business, or political context (Karabaich,
2004). These group types are summarized in Table 3-2.
TABLE 3-1 Examples of Cultural Categories and Coded Variables from
the Standard Cross-Cultural Survey
Examples of Cultural
Categories Examples of Labels for Variables Within Categories
Subsistence economy Marital residence
and supportive practices Matrilocal or uxorilocal—with wife’s kin
Avunculocal—with husband’s mother’s brother’s kin
Patrilocal or virilocal—with husband’s kin
Ambilocal—with either wife’s or husband’s kin
Neolocal—separate from kin
OCR for page 107
0
VERBAL CONCEPTUAL AND CULTURAL MODELS
TABLE 3-1 Continued
Examples of Cultural
Categories Examples of Labels for Variables Within Categories
Political organization Political power—most important source
Direct subsistence production
Warfare wealth
Tribute or taxes
Slaves
Contributions of free citizens
Large landholdings
Political office
Foreign commerce
Capitalistic enterprises
Priestly services
Cultural complexity Fixity of residence
Nomadic
Seminomadic
Semisedentary
Sedentary, impermanent
Sedentary
Sexual attitude and Frequency of premarital sex—male
practice Universal
Moderate
Occasional
Uncommon
Relative status of Mythical founders of the culture
women All male
Both sexes, but the role of men more important
Both sexes, and the role of both sexes fairly equal
Both sexes, but female role more important, or solely female
Cultural theories of Theories of soul loss
illness Absence of such a cause
Minor or relatively unimportant cause
An important auxiliary cause
Predominant cause recognized by the society
Female power and male Female economic control of products of own labor
dominance Absent
Present
Political decision Conflict between communities of the same society
making and conflict Endemic: high physical violence, feuding, and/or raiding occur
regularly
Moderately high, often involving physical violence
Moderate: disputes may occur regularly but tendency to
manage them in a more or less peaceful manner
Mild or rare
Nature of warfare Value of war: violence/war against nonmembers of the group
Enjoyed and considered to have high value
Considered to be a necessary evil
Consistently avoided, denounced, not engaged in
SOURCE: Adapted from Murdock and Morrow (1970).
OCR for page 111
VERBAL CONCEPTUAL AND CULTURAL MODELS
TABLE 3-4 Findings Regarding Cultural Differences in Human Inference:
Inductive Reasoning (ability to generalize from limited data) (Hudlicka,
2004)
Category of
Inference Findings
Covariation Ji, Peng, and Nisbett (2000)
judgment Chinese versus Americans
(identifying Simple stimuli presented on computer screen
correlations Chinese more confident about judgments
between cues) Chinese more correct in judgments
Chinese showed no primacy effect
Americans showing strong primacy effect
“East Asian cognition has been held to be relatively holistic; that is,
attention is paid to the field as a whole. Western cognition, in contrast,
has been held to be object focused and control oriented. In this study
East Asians (mostly Chinese) and Americans were compared on
detection of covariation and field dependence. The results showed
the following: (a) Chinese participants reported stronger association
between events, were more responsive to differences in covariation,
and were more confident about their covariation judgments; (b)
these cultural differences disappeared when participants believed
they had some control over the covariation judgment task; (c)
American participants made fewer mistakes on the Rod-and-Frame
Test, indicating that they were less field dependent; (d) American
performance and confidence, but not that of Asians, increased when
participants were given manual control of the test”
Causal Miller (1984)
attribution Americans versus Hindu Indians
(identifying Fundamental attribution error evident in Americans
causal Hindu Indians attribute behavior to social roles, obligations, physical
relations environment
between cues) Attributed to different beliefs regarding causality (content difference)
Social Morris, Nisbett, and Peng (1995)
Americans versus Chinese
Fundamental attribution (mass murderers, computer animations of fish)
Americans attributed behavior to individual dispositions
Chinese attributed behavior to environment
Lee, Hallahan, and Herzog (1996)
Americans and Hong Kong Chinese
Sportswriters’ descriptions of events
American writers focus on individuals
Hong Kong writers focus on situational factors
Nisbett (2003); Jones and Harris (1967)
Americans and Koreans
Judgment of another person’s attitude
Americans assume due to disposition
Koreans assume due to contextual influences
continued
OCR for page 112
BEHAVIORAL MODELING AND SIMULATION
TABLE 3-4 Continued
Category of
Inference Findings
Causal Asian folk physics is relational, emphasizing fields and force over distance
attribution Western folk physics focuses on nature of object itself, rather than its
Physical relation to the environment
Peng et al. (2001, p. 252)
Peng and Knowles (2003)
Chinese versus Americans
Force-over-distance explanations (aerodynamic, hydrodynamic, magnetic)
Americans referred more to nature of object
Chinese referred more to the field
Person Chiu, Hong, and Dweck (1997)
perception Hong Kong Chinese versus Americans
Judgment of self as fixed versus changing
Americans assume fixed, enduring traits
Chinese assume changing self
Peng et al. (2001)
Chinese versus Americans
Type of information used in person perception judgments
Americans focused on evidence provided by target
Chinese focused on evidence provided about the target by others
Inference of Americans prefer “what you see is what you get” norm of authenticity
mental states Asians would consider this impolite
Knowles, Morris, Chiu, and Hong (2001)
Chinese versus Americans
Judgment of mental states (thoughts, feelings, desires)
Americans: focus on what “they say”
Chinese: focus on what “they don’t say”
Categorization General findings:
• Some categories are more stable across cultures than others. Examples of
stable categories are: basic emotions, colors, basic shapes
• Westerners tend to categorize objects by color at an early age and by
function later. Africans tend to use color throughout their life. (This
finding may be related to formal education more than culture.)
• More cultural influence for goal-based categories than for environment-
based categories
• More salient categories for a given culture are more highly differentiated
(culture directs attention)
• Culture determines types of features used in defining categories
• Asians may be less attuned to categories in their inferences and
category learning
• Some evidence that Asians tend to use relational features as basis for
categorization
• Differences in “chronic accessibility” (less for Koreans than for Americans)
• Possible differences in category acquisition (exemplar based versus rule
based)
• Self-descriptions (Americans in terms of fixed traits; Asians in terms of
roles; more “socially diffused”)
SOURCE: Hudlicka (2004, Table 3.2.2-1 from Psychometrix Technical Report 0412, pp. 28–30).
OCR for page 113
VERBAL CONCEPTUAL AND CULTURAL MODELS
TABLE 3-5 Findings Regarding Cultural Differences in Human Inference:
Deductive Reasoning
Category of
Inference Findings
Syllogisms Luria (1931, Russia); Cole (1996, Africa)
Subjects did not engage syllogistic problems at the theoretical level
(i.e., if asked to deduce something based on a presented syllogism,
they would frequently think out of the box and suggest that the
experimenter go find out for himself; why would x be true, etc.)
Real-world (culturally relevant) grounding of topic makes a large
difference in success on task
Dialectical Asians: changing nature of reality and enduring presence of
reasoning contradictions versus Western: linear epistemology built on notions
of truth, identity, and noncontradiction
Resolving contradiction: Chinese seek compromise; Americans seek
exclusionary (either-or) truth and resolution (Peng and Nisbett, 1999)
Assumption in Eastern dialectical epistemology:
• Principle of change—everything is always in flux (thus x may not be
identical with itself because it may change over time)
• Principle of contradiction—opposing qualities coexist
• Principle of holism—everything is linked to everything else and
isolating phenomena may lead to misleading conclusions
• Folk wisdom: greater frequency and preference for dialectical
(apparently contradictory proverbs) among Chinese than Americans
Social Americans tended to blame one side versus Chinese tended to see fault
contradictions/ in both
conflicts
SOURCE: Hudlicka (2004, Table 3.2.2-2 from Psychometrix Technical Report 0412, p. 31).
behavior prediction, the most significant differences are those in emotion
elicitation; that is, in the specific situations and stimuli triggering particular
emotions. Variations were found both in the nature of the emotions elicited
and the intensity of those emotions. Some of these findings are summarized
in Table 3-6.
Semantic Models
Semantic models are not researcher-based models but rather the models
that ordinary people use to understand their worlds. The models are often
tacit, in the sense that individuals are not aware they have them. Anthro-
pologists discover the models by interviewing people and listening to their
accounts of daily life. They typically consist of chains of prototypical events
OCR for page 114
4 BEHAVIORAL MODELING AND SIMULATION
TABLE 3-6 Differences in Emotion Elicitors Across Cultures: Summary
of Findings
Situation Emotion Elicited
Birth of new family member More intense joy for Europeans/Americans than Japanese
Body-centered basic pleasures More intense joy for Europeans/Americans than Japanese
Achievement More intense joy for Europeans/Americans than Japanese;
more fear for Americans
Death of loved one More frequent triggers of sadness for Europeans/Americans
than Japanese
Physical separation from a More frequent triggers of sadness for Europeans/Americans
loved one than Japanese
World news More frequent triggers of sadness for Europeans/Americans
than Japanese
Strangers More frequent trigger of anger for Japanese than for
Europeans/Americans; more fear for Americans
Novel situations More fear for Japanese
Negative developments in More sadness for Japanese than Europeans/Americans
relationships
SOURCE: Hudlicka (2004, Table 3.2.2-3 from Psychometrix Technical Report 0412, p. 33).
that constitute plans of action. D’Andrade defined these sorts of models
as “a cognitive schema that is intersubjectively shared by a social group”
(D’Andrade, 1989, p. 809). Semantic models are qualitative or conceptual
rather than computational models.
As an example of a semantic model, Naomi Quinn (1987) has analyzed
hundreds of hours of interviews to discover concepts underlying American
marriage and to show how these concepts are tied together. She began
by looking at patterns of speech and at the repetition of key words and
phrases, paying particular attention to informants’ use of metaphors and
the commonalities in their reasoning about marriage. For example, one
of her informants said that “marriage is a manufactured product.” This
metaphor paints marriage as something that has properties like strength
and staying power and as something that requires work to produce. Some
marriages are “put together well,” while others “fall apart” like so many
cars or toys or washing machines (Quinn, 1987, p. 174).
OCR for page 115
5
VERBAL CONCEPTUAL AND CULTURAL MODELS
The objective is to look for metaphors in rhetoric and deduce the
schemas, or underlying principles, that might produce patterns in those
metaphors. Quinn found that people talk about their surprise at the breakup
of a marriage by saying that they thought the couple’s marriage was “like
the Rock of Gibraltar” or that they thought the marriage had been “nailed
in cement.” People use these metaphors because they assume that their
listeners know that cement and the Rock of Gibraltar are things that last
forever (i.e., they are intersubjectively shared).
Quinn reasons that if schemas or scripts are what make it possible for
people to fill in around the bare bones of a metaphor, then the metaphors
must be surface phenomena and cannot themselves be the basis for shared
understanding. Quinn found that the hundreds of metaphors in her corpus
of texts fit into just eight linked classes that she calls lastingness, shared-
ness, compatibility, mutual benefit, difficulty, effort, success (or failure), and
risk of failure. For example, Quinn’s informants often compared marriages
(their own and those of others) to manufactured and durable products (“it
was put together pretty good”) and to journeys (“we made it up as we went
along; it was a sort of do-it-yourself project”). Quinn sees these metaphors,
as well as references to marriage as “a lifetime proposition,” as exemplars
of the overall expectation of lastingness in marriage.
Other examples of the search for cultural schemas in texts include a
study of the reasoning that Americans apply to interpersonal problems
(Holland, 1985), a study of ordinary Americans’ theories of home heat
control (Kempton, 1987), and a study of what chemical plant workers and
their neighbors think about the free enterprise system (Strauss, 1997).
Cultural Domain Analysis
Cultural domain analysis refers to perspectives on and methods for
analyzing culture drawn from cognitive anthropology (Borgatti and Everett,
1992). A cultural domain is a collection of items that in some sense go
together or are all examples of a kind of x (e.g., animals, plants). Such
domains are often linguistic categories (e.g., semantic domains or concepts)
in that there is a simple name for the set of items, like fruit or vegetables.
What makes these domains cultural is that they are consensual. There is
general agreement on the part of cultural actors regarding membership of
most items in the domain. However, like all human things, the boundaries
of a domain can be porous or fuzzy. There are items that are clearly in the
domain, and items that are clearly outside, and many items that are in-
between. The general objective of this type of analysis and modeling is to
understand the cultural domain, which means to know what items belong
in it and how these items are perceived to relate to one another (i.e., the
extent to which they are similar or different). The data are collected and
OCR for page 116
BEHAVIORAL MODELING AND SIMULATION
analyzed in a systematic manner using data collection techniques, such as
pile sorts, sentence completion tasks, and triads tests (similar methods are
referred to as repertory grid analysis in psychology; see Johnson and Weller,
2002), and analytical methods, such as hierarchical clustering and multi-
dimensional scaling, to identify the conceptual organization and shared
dimensions among concepts. Analysis of domain items can also include
their attributes (e.g., diseases and their symptoms). A good example of a
general principle stemming from this form of analysis comes from Stefflre
(1972) in his proposition that people will behave similarly toward things
they perceive as being similar.
The importance of this approach lies in the ability to quickly assess
the nature of cultural beliefs and conceptions, albeit for a rather narrowly
delineated set of cultural items. However, such an understanding can facili-
tate the ability to alter or change cultural beliefs and ultimately human
behavior. These types of methods have been used in consumer research
for both product development and marketing (Stefflre, 1972). Johnson,
Griffith, and Murray (1987) and Murray, Griffith, and Johnson (1987),
for example, have used this approach in changing people’s beliefs about
underutilized fish species, leading to increased consumption of fish that
were traditionally considered “trash” fish.
Another branch of cultural domain analysis is the cultural consensus
model (CCM) of Romney, Weller, and Batchelder (1986). The model origi-
nated as a theoretical exploration of the formal conditions under which
similarity of beliefs would imply cultural knowledge. It was shown that,
in the context of a true/false questionnaire asking respondents to react to
propositions of fact, the degree of knowledge of each respondent could be
inferred when three conditions held. First, that a single culturally correct
right answer exists that is valid for all respondents in the sample. Second,
that conditional on the underlying cultural answer key, the responses of
subjects are independent (i.e., when they did not know the answer to a
question, their responses were uncorrelated). Third, that the questionnaire
contained questions about only one domain of knowledge (that is, a single
competence level for each person sufficed to characterize their probability
of answering any question correctly). When these three conditions held,
the model was capable of deriving both the culturally correct answer key
and the cultural competence of each respondent. The model allows for a
test of the degree to which cultural knowledge is shared, who has more
or less of this cultural knowledge, and how it varies among a group of
people in terms of, for example, gender, levels of human capital, and social
class. It also allows for the construction of the culturally correct answers
by working backward via Bayesian statistical techniques from the pat-
terns of agreement concerning a series of related cultural propositions or
statements.
OCR for page 117
VERBAL CONCEPTUAL AND CULTURAL MODELS
This approach has a number of advantages in terms of understanding
and modeling culture, particularly with respect to modeling aspects of intra-
cultural variation.6 CCM has been used in a variety of contexts, but it has
been applied practically to solving policy and management issues, model-
ing indigenous ecological knowledge, and understanding people’s cultural
beliefs concerning various aspects of health and illness. It has recently been
used to measure cultural consonance (i.e., the correspondence between
cultural beliefs and actual behavior) that has been shown to correlate with
health outcomes (e.g., low consonance is related to high blood pressure; see
Dressler and Bindon [2000]).
The CCM approach can be used to empirically determine shared beliefs
and knowledge that can be used in models incorporating cultural variables.
In addition, the approach can also be used to more finely tune an under-
standing of cultural beliefs and their variation that may be patterned in
terms of different social attributes (e.g., gender, age). Thus, cultural knowl-
edge (the correct cultural response) or individual cultural competency can
be treated as either a dependent or an independent variable in a model at
various levels of analysis.
Relevance to Modeling Requirements and Major Limitations
For the purposes of this study, a key limitation of all the models reviewed
in this section is that they were not built for military purposes. The variables
and dimensions they have focused on (such as power distance) have not been
shown to be relevant for any given military situation. More generally, dif-
ferent aspects of culture are relevant for different situations, and as a result
a new model must be built for each substantially different military purpose
and for each group of people (who have distinct cultures).
Another limitation of these models is that they do not explicitly link
culture and behavior and therefore do not provide direct guidance on how
to intervene in a group in order to change the culture. A partial exception
is cultural domain analysis, which posits that people behave similarly to
6 Understanding intercultural variations has benefited significantly from the approach taken
by Heinrich et al. (2004), in which an economic “game” (such as the Ultimatum Game, Güth,
Schmittberger, and Schwarze, 1982) is introduced across a number of different societies, and
the resulting behaviors correlated (or “normalized”) with respect to the interaction pattern
norms found in each society. As the authors note: “We draw two lessons from the experimental
results: first, there is no society in which experimental behavior is even roughly consistent with
the canonical model of purely self-interested actors; second, there is much more variation
between groups than has been previously reported, and this variation correlates with differ-
ences in patterns of interaction found in everyday life” (p. 5). Clearly there are implications
for war games, understanding cultural biases with respect to aggression across cultures, and
anticipating adversary tactics for a range of Department of Defense IOS modelers.
OCR for page 118
BEHAVIORAL MODELING AND SIMULATION
similar stimuli. As a result, it is possible to predict that people’s behavior
toward a new course of action will be similar to their reactions toward
other courses of action that are similar.
Another difficulty with predicting behavior is that the behavior of inter-
est to predict may often be that of individuals. However, some models, such
as the semantic models, unless they were based on a single individual, are not
intended to apply to any single individual. Other models, such as the trait-
based models of Hofstede, are based on individuals but then aggregated to
the group level. Cultures are then described by the traits of the majority.
Data, verification, and validation Issues
Cultural inventory models rely on ethnographic observation and are
therefore both time-consuming to develop and highly subjective. Having
multiple independent observers helps ameliorate the subjectivity problem
but is expensive.
Dominant trait models, such as the Hofstede dimensional models,
can involve two sets of data. The first set of data is used to derive the
dimensions. These can be validated by a number of different statistical
methods, such as factor analysis. Once these are fixed, another set of data
is obtained to score each new culture on the dimensions. These data have
to be obtained from willing natives of the culture, and the data have to be
updated over time because cultures change.
Future Research and Development Needs
In a certain sense, cultural models are critical for all the computational
models discussed in this volume, because the cultural models provide the
principles to be embedded in those models. For example, an agent-based
model of crowd behavior needs to know the cultural rules for behavior that
will govern the agents’ interactions.
The biggest limitation of cultural models at present is that existing
models were not designed with military purposes in mind. As a result, a key
research need is to develop models applicable to military needs. This would
include semantic models of how natives think about land, nation, war, for-
eigners, and so on, as well as cultural inventory models that include relevant
variables. Note that different models are needed for different cultures.
The semantic models are particularly powerful for military applica-
tions. However, they are currently not formal models, meaning that they
are expressed verbally and not in ways that are immediately amenable to
computational analysis. As a result, another key research direction is to
develop formal ways of expressing semantic models that are simple enough
to be used by field researchers and subject matter experts.
OCR for page 119
VERBAL CONCEPTUAL AND CULTURAL MODELS
REFERENCES
Arrow, H., and Burns, K.L. (2004). Self-organizing culture: How norms emerge in small
groups. In M. Schaller and C. Crandall (Eds.), The psychological foundations of culture
(pp. 171–199). Mahwah, NJ: Lawrence Erlbaum Associates.
Arrow, H., Henry, K.B., Poole, M.S., Wheelan, S.A., and Moreland, R.L. (2005). Traces,
trajectories, and timing: The temporal perspective on groups. In M.S. Poole and A.B.
Hollingshead (Eds.), Theories of small groups: Interdisciplinary perspectives (pp. 313–
367). Thousand Oaks, CA: Sage.
Arrow, H., McGrath, J.E., and Berdahl, J.L. (2000). Small groups as complex systems: Forma-
tion, coordination, development, and adaptation. Thousand Oaks, CA: Sage.
Benedict, R. (1934). Patterns of culture. New York: Houghton Mifflin.
Bettenhausen, K., and Murnighan, J.K. (1985). The emergence of norms in competitive
decision-making groups. Administrative Science Quarterly, 0, 350–375.
Blake, R.R., and Mouton, J.S. (1982). How to choose a leadership style. Training and Devel-
opment Journal, (2), 38–48.
Borgatti, S.P. (1994). Cultural domain analysis. Journal of Quantitative Anthropology, 4,
261–278.
Borgatti, S.P., and Everett, M.G. (1992). Notions of position in social network analysis.
Sociological Methodology, , 1–35.
Chiu, C.Y., Hong, Y.Y., and Dweck, C.S. (1997). Lay dispositionism and implicit theories of
personality. Journal of Personality and Social Psychology, , 19–30.
Cole, M. (1996). Cultural psychology: A once and future discipline. Cambridge, MA: Harvard
University Press.
D’Andrade, R.G. (1989). Cultural cognition. In M.I. Posner (Ed.), Foundations of cognitive
science (pp. 795–830). Cambridge, MA: MIT Press.
Davis, J.H. (2000). Simulations on the cheap: The Latané approach. In D.R. Ilgen and C.L.
Hulin (Eds.), Computational modeling of behavior in organizations: The third scientific
discipline (pp. 217–220). Washington, DC: American Psychological Association.
Dressler, W.W., and Bindon, J.R. (2000). The health consequences of cultural consonance:
Cultural dimensions of lifestyle, social support, and arterial blood pressure in an African
American community. American Anthropologist, 0(2), 244–260.
Ekman, P., and Davidson, R.J. (1995). The nature of emotion: Fundamental questions.
Oxford: Oxford University Press.
Fiske, A.P. (1991). Structures of social life: The four elementary forms of human relations.
New York: Free Press.
Fiske, A.P. (1992). The four elementary forms of sociality: Framework for a unified theory of
social relations. Psychological Review, (4), 689–723.
Fiske, A.P. (2000). Complementarity theory: Why human social capacities evolved to require
cultural complements. Personality and Social Psychology Review, 4(1), 76–94.
George, A. (1979). Cognitive beliefs about decision-making behavior. In L. Falkowski (Ed.),
Psychological models in international politics. Boulder, CO: Westview Press (Praeger).
George, A., and George, J. (1998). Presidential personality and performance. Boulder, CO:
Westview Press.
Güth, W., Schmittberger, R., and Schwarze, B. (1982). An experimental analysis of ultimatum
bargaining. Journal of Economic Behavior and Organization, (4), 367–388.
Hanisch, K.A. (2000). The Impact of organizational interventions on behaviors: An examina-
tion of different models of withdrawal. In D.R. Ilgen and C. L. Hulin (Eds.), Computa-
tional modeling of behavior in organizations: The third scientific discipline (pp. 33–60).
Washington, DC: American Psychological Association.
OCR for page 120
0 BEHAVIORAL MODELING AND SIMULATION
Harris, R.J. (1976). The uncertain connection between verbal theories and research hypotheses
in social psychology. Journal of Experimental Social Psychology, (2), 210–219.
Heinrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., and Gintis, H. (Eds.). (2004). Founda-
tions of human sociality: Economic experiments and ethnographic evidence from fifteen
small-scale societies. New York: Oxford University Press.
Hersey, P., and Blanchard, K.H. (1988). Management of organizational behavior: Utilizing
human resources. Englewood Cliffs, NJ: Prentice-Hall.
Holland, D. (1985). From situation to impression: How Americans get to know themselves
and one another. In J.W.D. Dougherty (Ed.), Directions in cognitive anthropology (pp.
389–412). Urbana: University of Illinois Press.
Hsu, F.L.K. (1972). Psychological anthropology. Cambridge, MA: Schenkman.
Hudlicka, E. (2004). Technical Report 04. Using cultural and group affiliations to infer indi-
vidual characteristics and predict behavior. Blacksburg, VA: Psychometrix Associates.
Hudlicka, E., Karabaich, B., Pfautz, J., Jones, K., and Zacharias, G. (2004). Predicting group
behavior from profiles and stereotypes. In Proceedings of Behavior Representation in
Modeling and Simulation (BRIMS) ’04 Conference, May 17–24, Arlington, VA.
Ji, L.-J., Peng, K., and Nisbett, R.E. (2000). Culture, control, and perception of relationships
in the environment Journal of Personality and Social Psychology, (5), 943–955.
Johnson, J.C., and Weller, S. (2002). Elicitation techniques in interviewing. In J. Gubrium
and J. Holstein (Eds.), Handbook of interview research (pp. 491–514). Thousand Oaks:
Sage.
Johnson, J.C., Griffith, D.C., and Murray, J.D. (1987). Encouraging the use of underutilized
marine fishes by southeastern U.S. anglers. Part I: The research. Marine Fisheries Review,
4(2), 122–137.
Jones, E.E., and Harris, V.A. (1967). The attribution of attitudes. Journal of Experimental
Social Psychology, , 1–24.
Karabaich, B. (2004). Towards a working taxonomy of groups. In Proceedings of Behavior
Representation in Modeling and Simulation (BRIMS) ’04 Conference, May 17–24,
Arlington, VA.
Kempton, W. (1987). Two theories of home heat control. In D. Holland and N. Quinn (Eds.),
Cultural models in language and thought (pp. 222–243). New York: Cambridge Uni-
versity Press.
Klein, H.A., and McHugh, A.P. (2005). National differences in teamwork. In W.B. Rouse and
K.R. Boff (Eds.), Organizational simulation (pp. 229–252). Hoboken, NJ: Wiley.
Klein, H.A., Pongonis, A., and Klein, G. (2000). Cultural barrier to multinational C2 deci-
sion making. In Proceedings of 000 Command and Control Research and Technology
Symposium, Monterey, CA. Available: http://stinet.dtic.mil/cgi-bin/GetTRDoc?AD=ADA
461631&Location=U2&doc=GetTRDoc.pdf [accessed Feb. 2008].
Knowles, E., Morris, M.W., Chiu, C., and Hong, Y. (2001). Culture and the process of person
perception: Evidence for automaticity among East Asians in correcting for situational
influences on behavior. Personality and Social Psychology Bulletin, , 1344–1356.
Kroeber, A.L., and Kluckhohn, C. (1952). Culture: A critical review of concepts and defini-
tions. Cambridge, MA: Peabody Museum.
Lee, F., Hallahan, M., and Herzog, T. (1996). Explaining real life events: How culture and
domain shape attributions. Personality and Social Psychology Bulletin, , 732–741.
LeVine, R.A. (1982). Culture, behavior, and personality: An introduction to the comparative
study of psychosocial adaptation. New York: Aldine.
Lewin, K. (1951). Field theory in social science: Selected theoretical papers. New York:
Harper.
Luria, A.R. (1931). Psychological expedition to central Asia. Science, 4, 383–384.
OCR for page 121
VERBAL CONCEPTUAL AND CULTURAL MODELS
Mayo, E. (1960). The human problems of an industrial civilization. New York: Viking
Press.
McFate, M. (2005). The military utility of understanding adversary culture. Joint Force
Quarterly, , 42–48.
Miller, J.G. (1984). Culture and the development of everyday social explanation. Journal of
Personality and Social Psychology, 4, 961–978.
Morgan, G. (1997). Images of organization, second edition. Thousand Oaks, CA: Sage.
Morris, M.W., Nisbett, R.E., and Peng, K. (1995). Causality across domains and cultures.
In D. Sperber and A.J. Pemack (Eds.), Causal cognition. New York: Oxford University
Press.
Murdock, G.P., and Morrow, D.O. (1970). Subsistence economy and supportive practices:
Cross-cultural codes I. Ethnology, (3), 302–330.
Murray, J.D., Griffith, D.C., and Johnson, J.C. (1987). Encouraging the use of underutilized
marine fishes by southeastern U.S. anglers. Part II: Educational objectives and strategy.
Marine Fisheries Review, 4(2), 138–142.
Nisbett, R.E. (2003). The geography of thought. New York: Free Press.
Ofori-Dankwa, J., and Julian, S.D. (2001). Complexifying organizational theory: Illustrations
using time research. The Academy of Management Review, (3), 415–430.
Ortony, A., Clore, G., and Collins, A. (1988). The cognitive structure of emotions. New York:
Cambridge University Press.
Peng, K., and Knowles, E. (2003). Culture, ethnicity and the attribution of physical causality.
Personality and Social Psychology Bulletin, (10), 1272–1284.
Peng, K., and Nisbett, R.E. (1999). Culture, dialecticism, and reasoning about contradiction.
American Psychologist, 54, 741–754.
Peng, K., Ames, D.R., and Knowles, E.D. (2001). Culture and human inference: Perspectives
from three traditions. In D. Matsumoto (Ed.), Handbook of culture and psychology (pp.
243–263). New York: Oxford University Press.
Quinn, N. (1987). Convergent evidence for a cultural model of marriage. In D. Holland and
N. Quinn (Eds.), Cultural models in language and thought (pp. 173–195). New York:
Cambridge University Press.
Romney, A.K., Weller, S.C., and Batchelder, W.H. (1986). Culture as consensus: A theory of
culture and informant accuracy. American Anthropologist, (2), 313–338.
Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution
process. In L. Berkowitz (Ed.), Advances in experimental social psychology (vol. 10, pp.
173–220). New York: Academic Press.
Schwartz, S.H. (1999). A theory of cultural values and some implications for work. Applied
Psychology: An International Review, 4(1), 23–47.
Stasser, G. (1988). Computer simulation as a research tool: The DISCUSS model of group
decision making. Journal of Experimental Social Psychology, 4(5), 393–422.
Stefflre, V.J. (1972). Some applications of multidimensional scaling to social science problems.
In R.N. Shepard, A.K. Romney, and S.B. Nerlove (Eds.), Multidimensional scaling:
Theory and applications in the behavioral sciences. New York: Seminar Press.
Strauss, C. (1997). Research on cultural discontinuities. In C. Strauss and N. Quinn (Eds.), A
cognitive theory of cultural meaning (pp. 210–251). New York: Cambridge University
Press.
Zaheer, S., Albert, S., and Zaheer, A. (1999). Time scales and organizational theory. The
Academy of Management Review, 4(4), 725–741.
