Metacognition and the Development of

Intercultural Competence
H. Chad LANE
Institute for Creative Technologies
University of Southern California, USA

Abstract. We argue that metacognition is a critical component in the development of

intercultural competence by highlighting the importance of supporting a learners selfassessment, self-monitoring, predictive, planning, and reflection skills. We also survey
several modern immersive cultural learning environments and discuss the role intelligent
tutoring and experience management techniques can play to support these metacognitive
demands. Techniques for adapting the behaviors of virtual humans to promote cultural
learning are discussed, as well as explicit approaches to feedback. We conclude with
several suggestions for future research, including the use of existing intercultural
development metrics for evaluating learning in immersive environments and to conduct
more studies of the use of implicit and explicit feedback to guide learning and establish
optimal conditions for acquiring intercultural competence.

Keywords. intercultural competence, metacognition, intelligent tutoring systems, serious

games, immersive learning environments, experience manipulation

Introduction
Learning and adapting to a new culture is a significant challenge. In different cultural
contexts, interpersonal and communicative behaviors that seem natural may produce
unexpected results. For example, simple habits such as nodding and other forms of
backchannel feedback can lead to unintended agreements that may, in turn, negatively affect
trust, reputation, and so on. It is certainly important for someone who will be spending time
in a new cultural context to prepare and be prepared for what awaits them. This is the
problem cross-cultural education programs attempt to solve.
A common approach is to provide a learner with a long list of dos and donts specific
to the country or culture they will be experiencing. While straightforward and easy, this
approach relies heavily on rote learning and produces little or no deep understanding of
culture. It also ignores empirical evidence that to develop intercultural competence in a
general way, people need to move through identifiable stages of development [3,11,13,20].
Rushing to the point of behavior adjustment with limited or no understanding of the
underlying cultural reasons can be problematic. True intercultural competence requires (at
least) a heightened sense of self-awareness, an ability to self-assess, enhanced perceptive
abilities, and a proclivity to reflect on experience. In other words, intercultural development
requires metacognitive maturity. This paper is about this process and how immersive
learning environments and intelligent tutoring can be used to promote intercultural learning.

1. Metacognition in learning
Metacognition involves active control over cognitive processes during problem solving. For
example, when one is solving an algebra equation, cognition refers to the activities necessary
to solve it, such as identifying rules to apply, applying them, finding a solution, and so on.
Metacognition refers to a higher order of thinking that operates on these cognitive activities,
such as planning, analyzing, assessing, monitoring, and reflecting on problem solving
decisions and performance. Metacognition also enables more effective learning. A learner
who is able to accurately gauge his or her own understanding is better equipped monitor his
or her own progress. This typically involves self-questioning and is part of the larger notion
of metacognitive regulation [6].
Metacognitive skills can be taught. Numerous classroom studies have shown that
explicitly teaching metacognitive strategies in the context of a specific domain (e.g., physics)
can improve learning outcomes [5, p.19]. Strategies taught in these studies integrate
metacognitive activities with cognitive and seek to make the steps of analyzing, planning,
assessing, and reflection habitual in the learner. Studies have also shown that learning is
more effective when learners explain worked out solutions to themselves [7]. This
phenomenon, which better learners do spontaneously, is known as the self-explanation effect.
It can also be taught [8]. More recently, computer tutors focusing on teaching metacognitive
skills have shown positive effects on learning behaviors (e.g., [1]).

2. Developing intercultural competence

2.1. The Peace Corps model
Peace Corps volunteers are told to expect four levels of cultural awareness when they begin a
new assignment in a foreign country [10, p.199]:
1. Unconscious incompetence: minimal awareness of cultural difference or mistakes
made; a state of blissful ignorance.
2. Conscious incompetence: basic realization of cultural difference; minimal
understanding of underlying reasons or their significance.
3. Conscious competence: increased understanding of differences; deliberate
behavioral adjustments are made to reduce cultural errors and misunderstandings.
4. Unconscious competence: culturally appropriate behavior is more or less
automatic; ones instincts have been reconditioned."
Peace Corps educators have therefore recognized that a certain level of metacognitive
maturity is needed to become interculturally competent. For example, to move from
unconscious to conscious incompetence, one must realize that what seems normal may be
considered strange by people from another culture. It must be recognized that any peculiar
observed behavior is likely in reaction to the learner. For people with underdeveloped
metacognitive abilities, there is a real risk these connections will fail to be made, leading
them to conclude the strangeness they perceive is inevitable and beyond their understanding.
2.2. The Developmental Model of Intercultural Sensitivity
The Peace Corps model is useful as a teaching tool, but was not designed as an explanatory
scientific model. The Development Model of Intercultural Sensitivity (DMIS), created by
Bennett [3], is such a model. It is intended to explain how people construe cultural difference

Figure 1. The Developmental Mode of Intercultural Sensitivity [3].

and how this ability becomes more flexible with time. By construe, Bennett is referring to
Kellys [16] constructivist view that experience is a function of how one assigns meaning to
events that occur in their lives. It is not simply a matter of being present during some event
or set of events, but rather how those events are interpreted, encoded into memory, and later
remembered. An underlying assumption of the DMIS is that as ones ability to construe
cultural differences evolves, intercultural competence also increases. According to Bennett,
it is the construction of reality as increasingly capable of accommodating cultural difference
that constitutes development [3, p. 24].
What constitutes a cultural difference for someone? It depends on that persons cultural
worldview, which is defined as the set of distinctions the person draws from to construe
events in the world. A monocultural person one who has primarily experienced a single
culture in his or her life will be unable to construe perceived differences from outside that
cultural worldview. On the other hand, a person with a broader understanding is generally
able to understand, even assume, other cultural worldviews. Hammer and Bennett
summarize: The crux of the development of intercultural sensitivity is attaining the ability
to construe (and thus to experience) cultural difference in more complex ways [11, p.423].
The DMIS is a posits two broad worldview orientations: ethnocentrism and ethnorelativism.
Each consists of three stages and are described below. The model is summarized in figure 1.
Bennett defines ethnocentrism as an assumption that the worldview of ones own
culture is central to all reality [3, p.30]. Further, an ethnocentric person will implicitly
assume that all others share this same worldview. The first ethnocentric stage is denial of
difference in which the learner ignores or neglects differences. The next stage is defense
against difference which includes recognition of cultural difference, but with negative
evaluation. This stage is characterized by an us vs. them mindset and overt, negative
stereotyping. The last ethnocentric stage is minimization of difference and includes the first
signs of considering another cultural worldview. In this stage, the learner emphasizes
similarities between cultures and recognizes only superficial cultural differences. Comments
such as we are all the same are common at this stage. Guidance is especially important
because some learners believe minimization is the ultimate stage of growth. When reality
sets in that cultural differences are truly significant, there is a risk of withdrawal [3, p. 44].
The remaining three stages represent a shift to the ethnorelative orientation and are
characterized by a basic understanding that ones culture is but one out of many valid
worldviews. The first ethnorelative stage is acceptance of difference in which the learner
recognizes and appreciates cultural differences. Cultural difference evokes positive feelings
in the learner for the first time. The next stage, adaptation to difference, is akin to the Peace
Corps conscious competence stage in that the learner makes an asserted effort to take the
perspective of others. Because of this frame shifting ability, the learner can more easily
interact with people from other cultures. The final stage is integration of difference: the
learner has internalized multiple cultural worldviews and can easily assume different
perspectives. Integration is an advanced stage often requiring years of experience to achieve.

2.3. Metacognition and the DMIS

Metacognitive skills are critical for advancement through the DMIS stages. Given that the
model is based on how one construes cultural differences, it follows that a learner must
become aware of the construal process [16], how it works and how to assess their own
construal abilities. The following metacognitive skills (at least) are related to the DMIS:
Enhanced perceptive abilities are needed to consciously recognize cultural
differences without reacting to them immediately.
Self-assessment failures on ones own interactions can hinder progress through the
DMIS stages. This feeds into self-monitoring and tracking through the stages.
Cultural self-awareness, defined as an understanding of ones own culture, is an
important aspect of movement from ethnocentrism to ethnorelativism [2].
Self-regulated learning is part of intercultural developmental: Generally, people in
the later phase of adaptation know how to orchestrate their own learning" [3, p. 59].
Planning and goal-setting can support progression through DMIS stages, such as
seeking to reach a specific stage or to understand specific cultural differences.
Bennett addresses metacognition in his discussion of the final stage, integration of difference.
He explains how ones identity is not lost in an ethnorelative state:
Rather, the integrated person understands that his or her identity emerges from the act of defining
identity itself. This self-reflective loop shows identity to be one act of constructing reality, similar to
others that yield concepts and cultures. By being conscious of this dynamic process, people can
function in relationship to cultures while staying outside the constraints of any particular one. (p.60)

The risk is that some learners may feel like their culture and individuality is lost once they
reach this advanced stage. Bennetts point is that the learner must also accept this redefined
and more advanced understanding of self one that relies on metacognitive maturity. Of
course, reaching this is well out of the scope of any educational approach; but, it does make a
strong case for nurturing an intercultural learners metacognitive skills. The Peace Corps
approach is consistent with it [10], as are other training programs [14,18]. The rest of this
paper explores how immersive learning environments may provide additional support.

3. Intelligent techniques for guided cultural learning

3.1. Immersive cultural learning environments and virtual humans
Technologies such as virtual reality and photoreal 3D graphics are particularly important
when considering cross-cultural training. High fidelity simulations make it possible to create
realistic portrayals of the products of different cultures, such as architecture, dress, sounds,
art, and even smells. This can promote the learners sense of immersion and provide a
foundation for identifying objective cultural differences. Two such environments are shown
in figure 2. The first is the Tactical Iraqi Language and Culture Training System (TLCTS)
developed by Tactical Language Training, Inc. [15]. TLCTS teaches Arabic language and
cultural skills. In the mission environment (shown in the screenshot), a learner is free to
explore an Iraqi village, hear the sounds, speak to locals, and make gestures. The clothing,
buildings, and surroundings are realistic and thus can give a learner a sense of what it might
be like to walk around an Iraqi village. In this way, the system is already in a position to aid
in the learners identification of cultural differences.

Figure 2. Examples of immersive cultural learning environments: (left) Tactical Iraqi Language and Culture
Training System [15] and (right) the Adaptive Thinking and Leadership simulation [21]. Used with permission.

The screenshot on the right in figure 2 is from the Adaptive Thinking and Leadership
(ATL) simulation game [21]. ATL is a team-training system that uses human role players for
both sides in intercultural scenarios. In-game assessment is performed by peers and
instructors who observe play and after-action review (AAR) facilities are available to convey
the outcomes to trainees. Learners are often assigned to role play as people from different
cultures, with appropriate backstories and goals. Role-playing is a well-developed technique
in the crosscultural training literature [18] and consistent with the DMIS with respect to the
goal of understanding different cultural worldviews.
In multi-player environments, like the ATL system, inhabitants are human roleplayers.
This can be costly and sometimes challenging to control from an educators point of view.
Research in virtual humans provides an alternative or supplement to cultural team-training in
immersive learning environments. Virtual humans combine artificial intelligence (AI)
research in cultural and emotional modeling, speech processing, dialogue management,
natural language understanding, and gesturing, among others, to enable natural feeling
communication and interaction with computer-controlled characters that listen and respond
to the user. Virtual humans are driven by rich models of tasks, emotion, body language, and
communication [22]. The underlying representations readily support explanation, which can
be useful for learning [9]. In the case of intercultural education, it is therefore important to
endow virtual humans with models of culture and the ability to explain their actions and
reactions in terms of their cultural worldview. It is also important that their behavior be
controllable in order to establish conditions that best promote learning.
3.2. Experience manipulation and implicit feedback
Generally speaking, computer simulations simulate real world phenomena as accurately as
possible. There are circumstances when it is appropriate to consider goals other than fidelity
when deciding how a simulation should behave and what events should occur. For instance,
to enhance entertainment value, a popular basketball video game includes special modes that
allow players to jump well over ten feet high. In this case, the goal of entertaining the human
player trumps the goal of simulating basketball completely realistically. In the case of
learning, the same idea applies: if a certain event or situation will promote learning, then the
simulation should seek to make that event happen. We refer to this general technique as
experience manipulation and now discuss several ways it might be used to promote
metacognitive growth and cultural learning.

Figure 3. Culturally influenced expressions of anger, skepticism, and appreciation by virtual humans [9,22].

When a cultural error is committed, or when appropriate actions are taken, learners need
support in (at least):
recognizing that an error was committed (or that a good action was taken)
finding a causal link between the action taken and the observed reaction
understanding the reason(s) and culpable underlying cultural differences
learning how to avoid the same mistake in the future (or sustain good actions)
It is important to go beyond simply concluding to avoid the same behavior in the future since
this will contribute little in the learners progression through the DMIS stages. Also, the
stage a person is in impacts how cultural differences are interpreted. Someone in the denial
phase may not even be willing to accept the fact that a cultural error even occurred, for
example. Someone in the other two ethnocentric stages (defense and minimization) may be
aware an error occurred, but unwilling to take blame or perhaps place the onus on the virtual
human to be the one who should adapt. Based this understanding of cultural growth, the
reaction of a virtual human to a cultural error should be appropriate for that learner.
Feedback from the simulation itself, such as the oral and gestural reactions of virtual
humans, is called implicit feedback. To support recognition of cultural errors, there are
several strategies that can be used adjust implicit feedback to promote learning. One of the
simplest is to accentuate verbal responses of characters to draw more attention to anger or
negative feelings, in the case of an error. Similarly, implicit positive feedback can be
achieved by accentuating positive and laudatory responses to correct user actions. In some
cases, it may even be appropriate for the virtual human to deliver an impassioned minilecture regarding the cultural issues in question. The choice of words by the virtual human
can be designed to refer directly to actions taken by learner to support the pedagogical goal
of linking cause and effect in the learners mind. In addition, the virtual human might also
drop hints regarding the underlying cultural differences. Body language and gestures can
have a dramatic effect on the communicative power of utterances. Figure 3 shows several
virtual humans in different emotional states and displaying a variety of gestures. The timing
and emphasis of these gestures can be adjusted to meet pedagogical goals in a way similar to
the utterance content. Aside from body language, other features that might be adjusted in
virtual humans are facial expressions, speech rate, intonation, and tone, emotional state, and
personality traits.

3.3. Experience management and interactive narrative

The techniques described in the previous section all focused on emphasizing specific details
of interacting with virtual humans to support the learner in recognizing cultural difference
and improving their ability to self-assess in an interpersonal context. Of course, explanations
for why certain behaviors are culturally offensive can be very complex. They may involve
fundamental differences between worldviews, varying ethical standards, social structure,
historical and geographical factors, and so on. A deep understanding of culture includes
these advanced notions and may enable a learner to go beyond rote learning by providing the
knowledge needed to reconstruct appropriate surface behaviors later. Cultural simulations
should provide diverse cultural experiences that go beyond one-on-one interactions and
carefully manage how events are presented and experienced by the learner.
One such technology focusing on experience management is automated story directing
(ASD) [23]. The goal is similar to that of a traditional tutoring system: allow users to feel as
much freedom as possible, but keep them on certain paths that consist of certain experiences.
The path in an interactive storytelling system is a storyline consisting of plots, arcs, events,
and other narrative elements. Users may break a storyline by taking actions in the virtual
world, and so ASD systems use a variety of techniques, like reactive planning, to repair
storylines and re-plan when new events are deemed desirable. Often, the aim is to maximize
engagement. For cultural training, the additional aim is to create situations and conditions
that promote practice and learning.
Metacognition comes into play when we consider the learners role in the narrative. She
or he must be aware that the actions being taken are being observed by the AI agents in the
simulation and that choices being made have observable outcomes. Just as minute details of
interaction can be manipulated to highlight differences, so can story elements. For example,
if a learner makes a gender-related error early in a game, the ASD may decide to propagate
this knowledge through the social network to force the learner to enter future encounters with
this baggage. Here the goal is not only to teach gender specific cultural differences, but also
to encourage consideration of unintended cultural consequences of earlier actions. This
requires the metacognitive abilities to self-assess over an extended period, reaching back
further than just the most recent action, and the ability to predict (another metacognitive
skill). After dealing with negative consequences of actions, it is hoped that a learner will
become more likely to consider possible unintended outcomes of actions before taking them.
3.4. Guidance and feedback
Inherent risks exist in unguided environments, such as inefficient learning, the formation of
misconceptions, and development of incomplete or fragmented knowledge [17]. Experience
manipulation and implicit feedback can certainly mitigate these risks to a certain degree, but
to adequately address the needs of novice and intermediate learners, explicit feedback from a
human tutor, pedagogical agent, disembodied coach, or other form of intelligent tutor has the
potential to greatly enhance learning (e.g., [4]). Explicit guidance can come in different
forms in an immersive learning environment. A pedagogical agent who plays a role in the
underlying simulation is a popular approach. The TLCTS [15] and the mission rehearsal
exercise described in [22] both provide pedagogical agents in the form of a knowledgeable
Sergeant who maintains an understanding of the cognitive goals and can give hints on how
to succeed. Another form is that of a disembodied tutor that posts messages in a special area
of a GUI or via speech. No matter what the modality, explicit feedback provides more direct
and understandable guidance than implicit this is especially important for novices [17].

Immersive learning environments can be overwhelming at times. With respect to

cultural learning, explicit hinting and feedback can help learners in several ways:
1. confirm a learners interpretation of observed virtual humans behaviors
2. explain the cultural differences in play during specific interactions
3. explain the under the hood reasoning of a virtual human
4. hint about ideal actions to take or warn against certain risks
5. suggest the learner identify possible outcomes and desirable end states
Explicit tutorial feedback removes a level of interpretation for the learner. Rather than
guessing or inferring the cognitive and emotional states of virtual humans, a clear statement
by a tutor can act as a strong scaffold for learning in immersive cultural environments. There
are certainly cognitive aspects to the tactics listed above, but they also address the
metacognitive demands of intercultural development. Tactics 1-3 encourage self-assessment
by describing the impact of a learners actions on a virtual human. Because this is feedback
being delivered in a real-time environment, the issues of distraction and cognitive overload
need to be considered. Thus, it is ideal to keep in-game feedback short and precise, saving
the longer explanations for post-practice reflection. This is the approach taken in the ELECT
Bi-Lateral negotiation game [12]). Hinting (tactic 4) can be direct (and at the cognitive level),
but also can be used to encourage the learner to think about pros and cons of taking different
actions this is especially important in ill-defined domains where assessment is inherently
challenging [19]. The content of tactics 1-4 are precisely the things we want the learner
speculating on before acting in the environment. In other words, the ultimate aim is for the
learner to self-guide in these precise ways. Such cognitive activities by the learner would
constitute attempts at self-explanation. Tactic 5, identifying potential end states, is a purely
metacognitive tactic that is geared towards supporting goal formation and identifying what
right looks like. Encouraging the learner to think before acting, to engage in planning and
simulate hypothetical actions, and reflect after acting are at the core of metacognitive
maturity and a fundamental requirement for growth through the DMIS stages.
Tutorial sub-dialogues are rarely possible in real-time environments, and so there is time
only for very brief periods of reflection. However, once a practice session or exercise is
complete, there is time to carefully target metacognitive skill development. Immersive
learning environments should therefore provide supporting tools such as video playback.
Reflective tutoring is an appropriate supplement to guide the use of these tools and to fill in
the gaps from feedback that was delivered during practice. The reflective tutoring system
built for the virtual humans [9,12] walks the student through three questions: (1) What
happened? (2) Why did these events occur the way they did? and (3) How can good
performance be sustained and poor performance be improved? A promising approach here is
to leverage explanation facilities of virtual humans to uncover their thinking via explanation
[9] and discover what other actions may have produced better outcomes. An advanced tactic
is to restart the simulation to give the learner a second chance (a mulligan). Reflection at
this point may enhance self-assessment skills and intercultural growth.

4. Conclusions and suggestions for future research

This paper has argued that achieving intercultural competence requires strong metacognitive
abilities. Although cross-cultural training programs frequently adopt metacognitive
approaches to teaching intercultural competence, the connection is rarely made explicit. The
Peace Corps approach is an example of this that describes growth as a continuum from
unconscious incompetence (not knowing anything and being blissfully unaware of

differences) to unconscious competence (full awareness of differences and appropriate

behavior is second nature). By describing these stages, the learner put in a position to selfmonitor their advancement. This then requires application of other metacognitive skills, such
as self-assessment, self-explanation and self-regulation, to progress through the stages. The
Developmental Model of Intercultural Sensitivity (DMIS) is an empirically derived and
validated model of intercultural development based on notion of how cultural differences are
construed by a learner [3,11]. To develop intercultural competence, it is necessary to
construe cultural difference in progressively more complex ways, such as from different
cultural worldview perspectives. Growth here also requires mature metacognitive abilities
and it may be possible to promote these skills in modern immersive learning environments
through a combination of experience manipulation and explicit guidance. The DMIS
suggests cultural difference as the pivot point for intercultural growth, and so careful
direction of virtual humans and delivery of feedback that targets self-assessment, predictive,
and reflection skills has the potential to speed the growth to intercultural competence.
Early evaluations of TLCTS [15] and the ELECT BiLAT [12] serious games have
shown some promising results with respect to learning aspects of specific cultures. Most of
the computer simulations built for cultural education have not undergone rigorous
experimental evaluations for learning or for intercultural development. Hammer and
Bennetts Intercultural Development Inventory (IDI) [11] has been used to validate the
DMIS and may provide a suitable metric for determining the value-added, if any, that comes
from augmenting cultural training programs [18] with immersive learning environments
especially if the IDI can be administered repeatedly and rates of change can be tracked.
Other more general questions about feedback are suggested for further study. For example,
Does implicit pedagogical feedback break immersion?
If so, what is the cost (if any) to breaking immersion with respect to learning?
What is the interplay between implicit and explicit feedback?
What situations merit the use of explicit feedback?
The use of implicit feedback and experience manipulation is perhaps one of the most
important open questions to address. The instructor interface to the ATL serious game [21]
provides the ability to throw curve balls to teams during their missions, such as helicopter
fly-overs. These are intended to support the development of adaptive thinking skills under
stress. This is related to research in the ITS community on intelligent problem selection and
finding the appropriate level of challenge for a learner. These connections need further
exploration, as does the reasoning behind expert instructors decisions to throw curve balls:
What are the triggers? How do instructors decide which curve ball to throw? The answers
may not always involve metacognitive skills, but as this paper has attempted to argue,
manipulations of this sort in an intercultural context may be ideal to highlight cultural
differences to give learners practice in dealing with them. This may enhance the
development of intercultural competence, but empirical research is certainly needed to
confirm this hypothesis.

Acknowledgements
I thank Dr. Jonathan Kaplan and Dr. Jim Belanich of the U.S. Army Research Institute for
many interesting conversations on experience manipulation, implicit feedback, and difficulty
that seeded the ideas presented in this paper. The project described here has been sponsored
by the U.S. Army Research, Development, and Engineering Command (RDECOM).

Statements and opinions expressed do not necessarily reflect the position or the policy of the
United States Government, and no official endorsement should be inferred.

References
[1]
[2]
[3]
[4]
[5]
[6]

[7]
[8]
[9]
[10]
[11]
[12]

[13]

[14]

[15]

[16]
[17]

[18]

[19]

[20]

[21]
[22]
[23]

V. Aleven, B. McLaren, I. Roll, & K. Koedinger. Toward meta-cognitive tutoring: A model of help-seeking
with a cognitive tutor. Int. Journal of Artificial Intelligence in Education, 16:101130, 2006.
J. M. Bennett. Intercultural communication training in cultural self-awareness for study abroad. PhD thesis,
University of Minnesota, October 1985.
M. J. Bennett. Towards ethnorelativism: A developmental model of intercultural sensitivity. In R. M. Paige, Ed.,
Education for the Intercultural Experience, 21-71. Intercultural Press, Yarmouth, ME, 1993.
B. S. Bloom. The 2 sigma problem: The search for methods of group instruction as effective as one-to-one
tutoring. Educational Researcher, 13:416, 1984.
J. D. Bransford, A. L. Brown, and R. R. Cocking, editors. How People Learn: Brain, Mind, Experience, and
School. National Academy Press, Washington, DC, 2000.
A. L. Brown. Metacognition, executive control, self-regulation, and other more mysterious mechanisms. In F. E.
Weinert and R. H. Kluwe, editors, Metacognition, motivation, and understanding, pages 65-116. Lawrence
Erlbaum Associates, Hillsdale, New Jersey, 1987.
M. T. H. Chi, N. de Leeuw, M. H. Chiu, and C. LaVancher. Eliciting self-explanations improves understanding.
Cognitive Science, 18(3):439477, 1994.
M. T. H. Chi, S. Siler, H. Jeong, T. Yamauchi, and R. G. Hausmann. Learning from tutoring: A studentcentered versus a tutor-centered approach. Cognitive Science, 25:471533, 2001.
M. G. Core, D. Traum, H. C. Lane, W. Swartout, J. Gratch, M. vanLent & S. Marsella. Teaching negotiation
skills through practice and reflection with virtual humans. SIMULATION, 82(11):685-701, 2006.
Peace Corps. Culture matters: The peace corps cross-cultural workbook, 1999.
M. R. Hammer, M. J. Bennett, and R. Wiseman. Measuring intercultural sensitivity: The intercultural
development inventory. International Journal of Intercultural Relations, 27:421443, 2003.
R. W. Hill, J. Belanich, H. C. Lane, M. G. Core, M. Dixon, E. Forbell, J. Kim, and J. Hart. Pedagogically
Structured Game-based Training: Development of the ELECT BiLAT Simulation. In Proceedings of the 25th
Army Science Conference, 2006.
D. S. Hoopes. Intercultural communication concepts and the psychology of intercultural experience. In M. D.
Pusch, editor, Multicultural education: A cross cultural training approach, pages 1038. Intercultural Network,
Inc., La Grange Park, IL, 1979.
D. S. Hoopes and M. D. Pusch. Teaching strategies: The methods and techniques of crosscultural training. In M.
D. Pusch, editor, Multicultural education: A cross cultural training approach, pages 104204. Intercultural
Network, Inc., La Grange Park, IL, 1979.
W. L. Johnson, H. Vilhjalmsson, and S. Marsella. Serious games for language learning: How much game, how
much AI? In C.-K. Looi, G. McCalla, B. Bredeweg, and J. Breuker, editors, Proceedings of 12th International
Conference on Artificial Intelligence in Education, pages 306313. IOS Press, 2005.
G. A. Kelly. Theory of personality: The psychology of personal constructs. Norton, New York, 1963.
P. A. Kirschner, J. Sweller, and R. E. Clark. Why minimal guidance during instruction does not work: An
analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching.
Educational Psychologist, 41(2):7586, 2006.
M. Mendenhall, G. Stahl, I. Ehnert, G. Oddou, J. Osland, and T. Khlmann. Evaluation studies of cross-cultural
training programs: A review of the literature from 1988 to 2000. In D. Landis, J. M. Bennett, and M. J. Bennett,
eds., Handbook of intercultural training, pp. 129144. Sage, Inc., Thousand Oaks, CA, 2004.
A. Ogan, V. Aleven, and C. Jones. Culture in the classroom: Challenges for assessment in ill-defined domains.
In Proceedings of the Workshop on Intelligent Tutoring Systems for Ill-Defined Domains at Intelligent Tutoring
Systems. Jhongli, Taiwan, 2006.
R. M. Paige, M. Jacobs-Cassuto, Y. A. Yershova, and J. DeJaeghere. Assessing intercultural sensitivity: An
empirical analysis of the Hammer and Bennett Intercultural Development Inventory. International Journal of
Intercultural Relations, 27:467486, 2003.
E. M. Raybourn. Applying simulation experience design methods to creating serious game-based adaptive
training systems. Interacting with Computers, 19:206214, 2007.
W. Swartout, J. Gratch, R. Hill, E. Hovy, S. Marsella, and J. Rickel. Toward virtual humans. AI Magazine,
27(2):96108, July 2006.
R. M. Young. Towards an Architecture for Intelligent Control of Narrative in Interactive Virtual Worlds. In
Proceedings of the 2003 International Conference on Intelligent User Interfaces. Miami, FL, 2006.