 

 

2008

This is to certify that the
thesis entitled

NUNCH’I AND DECEPTION: THE CULTURAL DIFFERECNCE IN
DECEPTION BETWEEN KOREA AND AMERICA

 

presented by

Kitae Kim

has been accepted towards fulfillment
of the requirements for the

LIBRARY
Michigan State
University

 

 

 

M.A degree in Communication

 

 

 

74%”;

 

Major Professor’s Signature
0 "xi/'0 ‘l [*0 C 8‘

Date

MSU is an aﬂirmative—action. equal-opportunity employer

gn-n-ncn-.-------.-u----9----q--

--.--------—----------n-.—.—-.-.--¢---<-—-----.-I-a--—---o-_A

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

070915

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/ClRC/DaleDue.indd-p.1

 

NUNCH’I AND DECEPTION: THE CULTURAL DIFFERECNCE IN DECEPTION
BETWEEN KOREA AND AMERICA

By

Kitae Kim

A THESIS
Submitted to
Michigan State University

In partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS
Department of Communication

2008

ABSTRACT

NUNCH’I AND DECEPTION: THE CULTURAL DIFFERECNCE IN DECEPTION
BETWEEN KOREA AND AMERICA

By
Kitae Kim

Information Manipulation Theory (IMT) views deception as arising from covert
violations of one or more of Grice’s (1989) four maxims (quality, quantity, relevance,
and manner). This study predicted that although messages produced by nunch ’i usually
involve violations of the maxims, Koreans, who are accustomed to nunch ’1’
communication, would not judge the message as deceptive so long as the nunch ’1'
message is interpretable and appropriate to the situation. On the contrary, for the message
violating the maxims covertly, Koreans as well as Americans will judge the message as
deceptive, but there will be a difference in ratings on the deceptiveness of each type of
violation between Koreans and Americans due to nunch ’i as a Korean form of indirect
communication. The predictions were partially supported. Koreans viewed nunch ’1'
messages for quantity and manner violation as signiﬁcantly more deceptive than
Americans. However, the deceptiveness ratings between two nations, in general, were not

moderated by cultural difference in communication indirectness.

Copyright by
KITAE KIM

2008

To my loving one-year-old daughter, Slhan

iv

Acknowledgements

First and foremost I would like to thank my advisor, Dr. Tim Levine. Tim have
guided me every step of the way and without your patience, I do not think I could have
made it this point. Thank you so much for all of the time you spent reading and re-
reading draﬁs. I am so indebted to you for all you have done for me. I will be forever
grateful to you.

To my committee members, Hee-sun and Nicole, I am grateful to you both for
guidance and insight that you have given me. I have learned so much from two of you.
Thank you to everyone in the department who helped me along my path. A special thanks
to Dr. Mary J. Bresnahan. You are an inspiration to me. If you ever needed me, feel free
to send me an e-mail and I will gladly assist you.

I would like to thank my family for their love and support throughout this process.
To my parents, thank you so much for your encouragement and understanding as I
ﬁnished this degree. I love all and am so blessed to have you in my life. I pray for your
health all the time. To my beautiful wife, Jihee, ﬁ'om start to ﬁnish you have been there
for me and there are not words enough to describe how thankful I am to have you in my
life. Finally to my daughter, Slhan, you are just everything to me. I pray for your health

all the time.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................ viii
LIST OF FIGURES ................................................................................ ix
INTRODUCTION .................................................................................. 1
CHAPTER 1
LITERATURE REVIEW ........................................................................... 3
The Traditional Approach to Deception ....................................................... 3
Information Manipulation Theory .............................................................. 5
Deception and Culture .......................................................................... 11
Nunch’i in Korea, Deception and [MT ....................................................... 19
CHAPTER 2
HYPOTHESES AND RESEARCH QUESTIONS ............................................. 27
CHAPTER 3
METHOD ............................................................................................ 34
Participants ...................................................................................... 34
Research Design ................................................................................ 35
Procedures ....................................................................................... 35
Measurements ................................................................................... 36
Proposed Statistics Analysis ................................................................... 37
CHAPTER 4
RESULTS ........................................................................................... 38
Measurement Analysis ........................................................................ 38
Manipulation Check ........................................................................... 38
Primary Results ................................................................................. 39
Further Analysis for the Main Effect for Violation Type ................................. 40
Further Analysis for Interaction between Violation Type and Culture ................. 42
Further Analysis for 3-Way Interaction among Violation Type, Culture and
Condition ....................................................................................... 44
The Analysis for Dichotomous Deception Ratings ....................................... 45
The Analysis for Distribution of Koreans’ Ratings on Deceptiveness .................. 46
CHAPTER 5
DISCUSSION ....................................................................................... 48
APPENDICES ...................................................................................... 54
Appendix A: Tables ........................................................................... 54
Appendix B: Figures .......................................................................... 59

vi

Appendix C: Scenario .......................................................................... 76

Appendix D: Message Examples ............................................................. 79
Appendix D: Deceptiveness Scale ............................................................ 81
REFERENCES ...................................................................................... 85

vii

Table 1

Table 2

Table 3

Table 4

LIST OF TABLES

Reliability for Each Scale ........................................................ 55

The Deceptiveness Ratings of Each Violation Type for
Country and Condition .......................................................... 56

Cell Means for the Current Study, Yeung et al. (1999),
Jacobs et a1. (1996), Lapinski (1995) and McComack et a1. (1992) ........ 57

The Dichotomous Judgment upon Types of Violations ..................... 58

viii

Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

LIST OF FIGURES

Interaction between Violation Type and Culture .............................. 60
Interaction between Violation Type and Culture in Nunch’i Condition. . .61
Interaction between Violation Type and Culture in Covert Condition. . . ..62
Interaction between Violation Type and Condition for US ................. 63
Interaction between Violation Type and Condition for Korea .............. 64
Culture by Condition Interaction for Quantity Violation Messages ........ 65
Culture by Condition Interaction for Manner Violation Messages. . . . . ....66

Culture by Condition Interaction for Relevance Violation Messages. . ....67
Culture by Condition Interaction for Quality Violation Messages ......... 68
Culture by Condition Interaction for Baseline Messages .................... 69
Stern and LeafPlots of Deceptiveness Ratings by Manner Violation. . ....70
Stern and LeafPlots of Deceptiveness Ratings by Quantity Violation. . ..71

Stern and Leaf Plots of Deceptiveness Ratings by Covert Quantity
Violation ........................................................................... 72

Histogram of Deceptiveness Ratings by Covert Quantity Violation ....... 73

Stem and Leaf Plots of Deceptiveness Ratings by Nunch’i Quantity
Violation ........................................................................... 74

Histogram of Deceptiveness Ratings by Nunch’i Quantity Violation. . ....75

ix

Nunch ’i and Deception:
The Cultural Difference in Deception between Korea and America
INTRODUCTION

Historically, the research on deception has typically explored the issues related to
deception detection and identiﬁcation of nonverbal cues distinguishing truth from lies.
The type of deception that most deception research has dealt with was a barefaced lie, or
falsiﬁcation (McComack, 1992; 1997). Some deception research, however, has focused
on more subtle forms of deception (e.g., Bavelas, Black, Chovil, & Mullet, 1990; Turner,
Edgley, and Olmstwd, 1975).

However, there was no theoretical framework to explain these types of deception.
Information Manipulation Theory (IMT) proposed by McComack (1992) provides such a
framework, and in addition, explains why deception ﬁmctions to deceive. Deception is
deceptive because deceptive messages derive from covert violations of the conversational
maxims, and the listeners are misled by their belief that the speaker (i.e., the deceiver) is
still adhering to the Cooperative Principle (CP) and its maxims (McComack, 1992). The
four dimensions of violation, which are derived from Grice’s conversational maxims
(Grice, 1989), include the violation of quantity, the violation of quality, the violation of
relevance, and the violation of manner.

To date, there have been several tests of IMT (e.g., Jacobs, Dawson, and Brashers
1996; Levine, 1998; McComack, Levine, Solowczuk, Torres, and Campbell, 1992).
Some IMT studies have focused on cultural difference in deception (e.g., Lapinski, 1995;
Murai 1998; Yeung, Levine, and Nishiyama, 1999). Unfortunately, previous IMT cultural

deception studies did not pay attention to distinct and culture-speciﬁc concepts related to

communication style in the cultures examined. Furthermore, past cultural deception
research suggested that cultural difference in deception might not be found across all
situations, and what is counted as deception should vary across cultures. The previous
IMT studies interested in cultural deception, however, have not been sensitive to
important situation-speciﬁc contingencies.

As a remedy to these limitations in the past research, this study derives
hypotheses specifying cultural differences and similarities in what counts as deception
between Korea and America, focusing on the concept of nunch ’i in Korea. Nunch ’i is a
Korean form of indirect communication style to make or interpret implication (Park,
1997). When to use nunch ’i or what the nunch ’i means in a given situation is often
ritualized.

Messages involving mmch ’i usually violate quantity, relevance, manner, or some
combination of these maxims at the level of what is said, but nunch ’i messages still
adhere to the CP at the level of what is implied. Consequently, it is proposed that that if
the nunch ’i message recipients can sufﬁciently infer what the messages mean with
mmch 'i and the use of nunch ’i is appropriate to the situation, Koreans will not count the
nunch '1' messages as deception despite deviation from baseline honest message. On the
other hand, for messages covertly violating the maxims, Koreans as well as Americans
will judge the message as deceptive.

This paper ﬁrst introduces the traditional trends in deception research, and then
examines the previous research on IMT and cultural deception. After investigating the
cultural difference in communication style, focusing on nunch ’i in Korea, several

hypotheses are advanced.

CHAPTER 1: LITERRATURE REVIEW
The Traditional Approach to Deception

The majority of research on deception has centered on issues regarding how to
detect deception, and has focused on nonverbal cues to detect lies. It has been generally
accepted by researchers on deception is that humans’ ability to detect deception is fairly
limited. The consistent ﬁnding regarding people’s ability to distinguish truths ﬁ'om lies is
that detection accuracy tends to be signiﬁcantly, but only slightly, better than chance
levels. For example, the accuracy rate reported in past studies almost always ranges from
45% to 70% (e.g., Miller and Stiff, 1993; Vrij, 2000), and the most recent meta-analysis
indicated that the mean accuracy rate was about 54%, correctly classifying 47% of lies as
deceptive and 61% of truths as truthful (Bond and DePaulo, 2006)

One of the factors that is commonly believed to determine deception detection
accuracy is truth-bias, referring to the inclination to judge more messages as truth than
lies (McComack & Parks, 1986; Levine, Park, and McComack, 1999). People tend to be
more accurate at judging truth than lies, and accuracy is a linear function of the
percentage of the messages that are truths and lies (Levine, Kim, Park, & Hughes, 2006;
Levine et al., 1999; Park and Levine, 2001). Gilbert’s (1991) Theory of Mental
Representation provides a plausible explanation for why people are so truth-biased. The
theory suggests that in order for people to deal with the message, they should initially
comprehend the message as if it is true. If the message is false, such a initial belief has to
be ‘rebuked.’ That is, for human beings to process the message, it is the prerequisite that

mentally representing all message as true initially (Gilbert, 1991).

Another widely accepted determinant factor of the detection accuracy is people’s
incorrect beliefs about verbal or nonverbal cues to detect deception (Miller and Stiff,
1993; Stiff and Miller, 1986). There exists a pervasive assumption with regard to the
centrality of verbal and especially nonverbal behavior cues in deception detection, and
these beliefs have little correspondence with actual diagnostic validity (DePaulo et al.,
2003; Levine, Feeley, McComack, Hughes, and Harms, 2005).

More recently, deception research has started turning toward everyday lies (e. g.,
DePaulo, Kashy, Kirkendol, and Epstein, 1996; McComack, 1997; Park, Levine,
McComack, Morrison, and Ferrara, 2002). The previous research on everyday lies has
revealed that the lies in everyday life are mostly unplanned and insigniﬁcant “light” lies
(DePaulo et al., 1996). In addition, real-life lies are so successful that if detected at all,
they are most often uncovered long after the lies are told (Park, et al., 2002). In sum,
deceptive communication in everyday life is ubiquitous, casual, low stakes and successful
in nature (McComack, 1997).

Given the nature of the everyday lies, some scholars argue that behavioral
differences may be limited to high stakes lies (Frank and Feeley, 2003), and speculated
that behavioral differences between everyday truths and lies are minimal (Levine, F eeley,
McComack, Hughes, & Harms, 2005; McComack, 1997; Park et al., 2002). Consistent
with this rational, the most through meta-analysis of deception-related behaviors
indicates that the effects of source veracity on speciﬁc source behaviors are often small
and unreliable (DePaulo et al., 2003). In a similar vein, Levine et a1. (2005) tested
whether training in nonverbal behaviors associated with deception improved deception

detection accuracy by comparing the accuracy between no training and bogus training

control groups (Levine et al., 2005). The result suggested that valid training did not
produce large improvements over a bogus training group, and that bogus training could
produce signiﬁcant improvement over no-training group (Levine et al., 2005). These
results imply that the improvement in the detection accuracy may be independent of the
training content (i.e., nonverbal cues to deception). This accumulated dissatisfaction with
the traditional approach evoked the need to change the attention from behavioral cues
related to the deception detection to message design in terms of the content and
production of deceptive messages (Levine et al., 2005).
Information Manipulation Theory

Researchers who have viewed deception in terms of the content and production of
deceptive messages have conceptualized deceptive messages as resulting ﬁ'om the
manipulation of information in particular ways (McComack, 1992). One way to
manipulate information that all relevant literatures include is falsiﬁcation, referring to
providing information contradictory to true information or explicitly denying the validity
of the true information. In fact, the type of deception studied by experimental deception
researchers belongs exclusively to falsiﬁcation (McComack, 1997).

However, deception research using the falsiﬁcation-only deceptive messages has
failed to capture the nature of most of deception in everyday life, given the fact that the
vast majority of deceptive messages are subtle and complex (McComack, 1992; 1997).
Research on deceptive message design has found at least three different ways besides
falsiﬁcation that information can be manipulated to produce deceptive messages (e.g.,
Bavelas et al., 1990; Ekman, 1985; Turner et al., 1975). Ekman (1985) argued that

individuals can adjust the amount of information that is disclosed in order to produce

deceptive messages. Turner et al. (1975) found “diversionary responses,” acts that
produce deceptive messages by diverting the conversation away from sensitive
information that they desired to withhold. Bavelas et al. (1989) have demonstrated that
the manner in which information is presented, such as the clarity or ambiguity of the
information, can produce deceptive messages.

While each of four forms of information manipulation (i.e., distortion,
concealment, diversionary response, and equivocation) has been recognized as impacting
the deceptiveness of messages, prior to IMT there was no theory that incorporates each of
the forms into one workable framework. The Information Manipulation Theory (IMT)
proposed by McComack (1992) provides such a framework to describe these different
forms of information manipulation.

According to McComack (1992; 1997), deceptive message design is not so
different from the truthful message design, but in most cases, involves only subtle
deviation ﬁ'om the general principal for the truthful message design. IMT employs the
Cooperative Principle (CP) in Grice’s Theory of Conversational Implicature as the
general principal that forms the foundation for understanding and making sense of
another’s communication (McComack, 1992). Grice argued that in order for
communicators to understand what others really mean, people tacitly assume that
communicators follow the CP (Grice, 1989). According to Grice, participants within any
interaction are expected to, “Make their contribution such as is required, at the stage at
which it occurs, by the accepted purpose or direction of the talk exchange in which they

are engaged” (Grice, 1989, p26).

Further, Grice also speciﬁed “four categories under one or another of which will
fall certain more speciﬁc maxims, the following of which will, in general, yield results in
accordance with the CP” (Grice, 1989, p.26). The maxim of quality refers to the
truthfulness of the information given. Participants in conversation are expected not to
present information that they know is false, or that they are nor sure of (Grice, 1989). The
maxim of quantity refers to expectations regarding the quantity of information that a
message should contain. It is assumed that in a conversation, each contribution should be
as informative as is required given the situation (Grice, 1989). The maxim of relation
maintains that participants should make their contributions relevant to the preceding
disclosure (Grice, 1989): “By failing to response in a relevant fashion (given the
constraints established by the partner’s previous utterance), an individual can divert the
course of the conversation away from potential disclosure of “dangerous” information”
(McComack, 1992, pp] 1-12). As a ﬁnal maxim, “the maxim of manner relates not to
what is said, but rather to how what is said is to be said” (Grice, 1989, p.27, emphasis in
the original). This is based on the expectation that the participants will present the
information clearly, directly and without ambiguity. People, of course, do not always
follow these maxims. Grice’s point, however, is that others need to assume that the
maxims are being followed at least at some level, in order to make sense out of what is
said.

Grice’s four maxims were originated to describe truthful messages rather than
deceptive messages. However, the IMT employs the four maxims as a framework to
account for the various forms of deceptive messages noted by previous researchers (i.e.,

Bavelas et al., 1989; Ekman, 1985; Turner et al., 1975), and explains why deceptive

messages deceive. McComack (1992) indicates how IMT can account for the different
types of deception, described by the previous authors mentioned above. The violation of
quantity is to alter information in terms of the amount of information that is presented.
Concealment (Ekman, 1985) can ﬁt into this category. The violation of quality is to
completely distort or fabricate information. The falsiﬁcation or fabrication can be
conceptualized in this way. The violation of relevance is to respond in an irrelevant
fashion. Diversionary responses found by Turner et al. (1975) can be included in this
category. Finally, the violation of manner is to state information in an equivocal manner,
or avoid clear messages. Bavelas et al.’s (1989) equivocation ﬁts into this category.
Furthermore, IMT provides the explanation for why deceptive messages deceive.
Listeners are misled by their inference that the speaker (i.e. the deceiver) is adhering to
the CP and its maxims when in fact they are secretly violated (McComack, 1992). Even
when individuals observe the deviations ﬁom the maxims, they often still assume other
interactants’ adherence to the CP at some level, rather than considering that the other
interactant said deceptive. In this way, IMT is consistent with the prevalent observation
of truth-bias in the literature. Simply put, since the receiver assumes that the quantity,
quality, relevance and manner of the information will be appropriate, if the manipulation
of these elements pass unnoticed, then the receiver will be deceived. That is, deception
involves the exploitation of the very belief that forms the foundation for accurate
communication, which is the CP. In sum, according to the IMT, deceptive messages are
deceptive because deceivers covertly violate one or more maxims while exploiting

listeners’ fundamental belief that speakers are still adhering to the CP.

IMT thus views deception in terms of quantitative variation in message
characteristics along multi-dimensions (i.e., quantity, quality, relevance, and manner).
Unlike earlier approaches, IMT is not a categorical approach to classifying deceptive
messages, but a continuous approach. Extending on this idea, researchers have
increasingly moved from dichotomous deception judgment to continuous deception
judgment. IMT research (e.g., Jacobs, Dawson, and Brashers 1996; Lapinski, 1995;
Levine, 1998; McComack et al., 1992; Yeung et al., 1999) as well as other studies of
deception (e. g., Burgoon, Buller, Ebesu, & Rockwell, 1994; Burgoon, Buller, Guerrero,
Aﬁﬁ, & Feldman, 1996) have employed continuous ratings. The results of the studies
suggest that some types of deception are seen more deceptive than others. Violations of
quantity and relevance are typically more deceptive than violations of quantity and
manner (Jacobs et al., 1996; Lapinski, 1995; Levine, 1998; McComack et al., 1992;
Yeung et al., 1999).

However, Levine (2001) questions the interpretation of these ﬁndings. It is
possible that ﬁndings reﬂect quantitative variation in some message features (i.e., the
percentage of message content that is false). Alternatively, it might reﬂect individual
difference in the perception. Some people may see a message as deceptive while others
see it as honest, resulting in a mean near the scale midpoint. To address the issue, Levine
(2001) investigated the distributions of continuous honesty ratings. The results show that
the distribution in the manner violation condition was approximately at the scale
midpoint, which suggests that manner violations seem to reﬂect the moderate degree of

perceived deceptiveness. The distribution in the quantity violation was, however,

bimodal, which suggests that quantity violation may reﬂect the individual difference in
the rating of the same message (Levine, 2001).

Another factor that may affect the degree of perceived deceptiveness is the degree
of violation, i.e., the extent to which a message deviates from the baseline messages.
Yeung et al. (1999), for example, reasoned that the participants in Hong Kong did not see
quantity and manner violation as deceptive because the extent to which messages
provided for the participants violate maxim of quantity or manner were not sufficient for
the participants to judge the violations as deceptive. This reasoning was supported by
further analysis revealing that ratings of violations were correlated with rating of honesty.
In other words, if the participants had perceived the message as one or more violations of
each maxim, they should have seen the message as deceptive.

Jacobs and his colleagues (Jacobs et al. 1996a; 1996b) have challenged the
validity of the IMT and proposed a Primacy of Quality Model (PQM). While the PQM
accepts the claim that Grice’s maxims are central to deception, the PQM contends that
deception necessarily involves quality violation at the level of what is said or what is
implied (Jacobs et al. 1996a; 1996b). In other words, quality violation occurs at the level
of what is said and quantity, relevance, manner violation requires quality violation at the
level of what is implied. Simply put, unless quality violation occurs, deception does not
occur (Jacobs et al., 1996a; 1996b).

Levine (1998) tested the consistency of the IMT and the PQM with empirical
data. Speciﬁcally, Jacobs et al. (1996a) proposed a multidimensional measurement model
that scales related to quality violations and perceived honesty should load on the same

factor, which should be empirically distinct forms other violation types (i.e., quality

10

violation and perceived honesty, relevance, manner, quantity violation). In contrast, the
IMT proposed second-order unidimensional model that quality, quantity, relevance,
manner, and perceived honesty all should contribute a single more abstract, higher-order
construct, the CP (McComack, 1996). Levine (1998) tested it using Conﬁrrnatory Factor
Analysis (CF A) and showed that the data were consistent with second-order
unidimensional model.

To date, there have been several tests of IMT: McComack et al.’s (1992) original
study and replications by Jacobs et al. (1996), Lapinski (1995) and Yeung et al. (1999).
McComack et al.’s (1992) study provided subjects with a hypothetical situation and one
of ﬁve message forms. Participants were asked to rate the messages in terms of honesty.
The messages either violated one of Grice’s maxims (quality, quantity, relevance, and
manner) or were baseline messages designed to be honest. The study found that each type
of violation messages was viewed signiﬁcantly more deceptive than baseline honesty
message. Jacobs et al. (1996) and Lapinski (1995) obtained nearly identical ﬁndings to
the ﬁndings of McComack et al.’s (1992) original study. Yeung et al. (1999), however,
failed to ﬁnd between group differences in message quantity and manner violation and
the baseline message in data collected in Hong Kong .
Deception and Culture

Given that deception is pervasive in everyday communication (DePaulo et al.,
1996), it is reasonable to think that culture will inﬂuence deceptive communication
because the characteristics of everyday communication of individuals within a particular
culture reﬂect their own culture. There have been relatively few studies of deception

across cultures, but some works suggest that what is considered deceptive may vary from

11

culture to culture, whereas other research points to similarity. Some research on cultural
deception comes ﬁom the anthropological perspective. For example, Lewis and Saarni
(1993) deal with deception in non-industrialized societies such as the Waika Indians of
Orinoco, Venezuela, and suggest several deceptive behaviors that occur across these
cultures. These include deception centered on clandestine affairs, protecting one’s
possessions ﬂour a competitor, and feigning emotion one does not really feel. Warner
(1976), based on his experience with a shaman of the Huichol Indians of Mexico and on a
literature search, compared the self-deceptions practiced by Indian medicine men and
Western psychiatrists to maintain self-respect. Sincere, effective healers, regardless of
culture, are convinced of their healing powers and integrity, while deceiving patients with
placebo treatments. While these studies provide some insight into the deception in
different culture, they have failed to examine deception in a more systemic manner.

The research addressing issues of cultural difference in deception in a more
systemic way have focused on two assumptions at large. First, there should be cultural
difference in deception cues and second, the speciﬁc situation should play a role as a
moderator in cultural impact on deception. First, parallel to traditional deception
research, the attempts to identify culturally speciﬁc deception cues has been made in the
ﬁeld of cross-cultural deception research (e.g., Bond, 1990; Bond, 2006; Bond & Atoum,
2000; O’Hair, Cody, Wang, & Choa, 1990; Kam, 2004). For example, O’Hair et al.
(1990) investigated vocal stress in the truthﬁrl and deceptive messages of Chinese
immigrants. Chinese had higher levels of vocal stress when revealing negative emotions.
Kam (2004) examined the impact of cultural orientation on nonverbal behaviors related

to deception. When considering partner perceptions of deceivers' behavior, higher

12

degrees of independence were found to be associated with less positive affect under
deception. When outside-observers viewed the behaviors of deceivers, higher degrees of
independence were found to be associated with greater kinetic involvement and
pleasantness, less nervousness, and greater vocal pleasantrress and vocal relaxation under
truth. The study also reported that higher degrees of interdependence were associated
with lower detection accuracy and greater truth bias.

Bond and Atoum (2000) investigated whether or not culture-speciﬁc learning as
to deception cues was required for the deception detection, employing Americans,
Jordanians and Indians as participants. The result showed that the difference in the
accuracy between within a culture and across cultures were not signiﬁcant.

Bond and his colleagues (2004) carried out the exhaustive study on stereotypical
deception cues in 75 different countries. The most salient result was that a dominant pan-
cultural stereotype was that liars avert gaze. According to the ﬁndings of meta-analysis
on deception cues by DePaulo et al. (2003) gaze aversion is not related to honesty, and
thus does not have diagnostic utility. Thus, people may rely on a similar unreliable cue to
detect deception. In sum, no culturally-unique deception one has been reported so far.

Second, some research on cultural deception has focused on speciﬁc condition
under which culture have impact on deceptiveness. This research assumed that under
speciﬁc circumstance, the motivation to deceive would vary depending on cultures.
Although it did not take a speciﬁc situation into consideration, Lalwani, Shavitt, and
Johnson (2006) examined cultural difference in deception in terms of cultural difference
in self-presentation style and image management. The study expected that European

Americans compared to Asian Americans would like to see oneself in a positive light and

13

to give inﬂated assessment of one’s skills and abilities while Asian Americans would be
more likely to misrepresent their self-reported actions to appear more normatively
appropriate. The data, unfortunately, did not support the idea.

Kim, Kam, Singles, and Aune’s (1999) Cultural Model of Deceptive
Communication Motivation proposes that cultural orientation such as self-construals
should predict the conditions under which people are likely to deceive others. Kim et al.’s
predictions specify locus of beneﬁt (i.e., self-beneﬁt, or other-beneﬁt) as the crucial
moderator. The results of empirical research that tests the Model of Deceptive
Communication Motivation are ambiguous. On the one hand, Anne and Waters (1994)
found that the collectivistic American Samoan participants indicated they would be more
likely to deceive another on an issue related to family or other in-group concerns. U.S.
Americans, in contrast, were motivated to deceive when they felt an issue was private.
Kam (2004) also reported that degree of independence was the single best predictor of
one's motivation to protect the self, whereas degree of interdependence was the best
predictor of one's motivation to protect the other. On the other hand, Lapinski (1995)
reported that no signiﬁcant interaction between self-construal and beneﬁt were observed.
Finally, Levine et al., (2002) concluded that self-construal was related to self-reported
deceptiveness, but not to the actual messages generated.

In a similar vein, Chunsheng’s (2002) study showed that the Chinese participants
reported that higher likelihood in telling lies under the situation under which the lies were
not harmful. Seiter, Bruschke and Chunsheng (2002) explored the impact of culture on
the acceptability of deception, employing Chinese and American participants. The study

shows that culture does not exert an overpowering inﬂuence, but it does manifest itself in

14

particular situations: for example, the Chinese participants view deception almost three
times more acceptable to deceive a spouse for malice than do the American participants,
while the Chinese participants than the American participants view deception less
acceptable in case that they deceive teachers to manage the impression of oneself. Lee,
Carmen, Xu, Fu, & Boaed (1997) also examined the effect of culture in a speciﬁc
situation. They investigated difference of Chinese and US. children’s evaluation of lying
and truth telling, comparing pro-social situations and anti-social situations. The Chinese
children rated the lie telling positively when a person tells a lie to conceal his or her pro-
social conduct, while Canadian children rated it negatively.

In conclusion, the previous cultural deception studies call for a more nuanced role
of culture in deception for ﬁrture studies. In other words, in order for cultural deception
scholars to ﬁnd cultural difference in deception, they should focus on speciﬁc situations
under which perception of deception will vary across cultures.

Some cultural deception research has employed the IMT as a research framework.
Lapinski (1995) conducted the survey with culturally diverse sample in Hawaii,
investigating the inﬂuence of cultural disposition such as self-construal on ratings of
message honesty. Additionally, the study took the speciﬁc context in which the deception
occurs, such as how the inﬂuence self-construals vary depending on locus of beneﬁt (i.e.,
self and other beneﬁt). In general, the results of this study replicated previous IMT
ﬁndings. The results also indicated that those high in interdependent, presumably
dominant in collectivistic cultures, tend to see relevance violations as more honest. The

study failed to ﬁnd cultural variance on honesty ratings in terms of locus of beneﬁt, but in

15

general, violations of quantity and manner were seen as more deceptive when for
another’s beneﬁt regardless of self-construals.

The replication of IMT has been attempted employing non-American participants.
Murai (1998) tested IMT in Japan. Generally, the results were consistent with IMT.
However, there were no difference between messages violating multiple maxims and
messages involving a single violation in terms of ratings of message deceptiveness.
Yeung et al. (1999) also attempted to replicate IMT in Hong Kong. Contrary to IMT
predictions, violations of quantity and manner were not rated as deceptive. However,
consistent with IMT, messages seen as violation were rated as deceptive.

While the previous IMT research on cultural deception is successfully replicated
across culture, the IMT cultural deception research has failed to ﬁnd reliable cultural
difference in deception. Such a failure, however, does not necessarily mean that there is
no cultural difference in deception or that IMT is not a workable framework to ﬁnd
cultural differences in deception. As mentioned earlier, there may not be universal
difference in deception from culture to culture across all situations and all message forms,
but there should be a certain situation to magnify cultural impact on the deceptiveness.
Therefore, theoretical reﬁnement is needed. Specially, the cheating situation in romantic
relationships, which has been used frequently in previous IMT studies, might not be the
adequate situation to ﬁnd the cultural difference in deception. Given honesty is generally
accepted the most important value in romantic relationships (C01, 2001) and the
discovery of deception is usually consequential (Jang, Smith, and Levine, 2001;
McComack and Levine, 1990), the concern for the relationship by participants could

overwhelm the concern for culture, resulting in little or no impact of culture on deception.

l6

As a result, Lapinski (1995) used the situation the students deceive their ﬁiends for
themselves and for their friends, and assumed that the situation should exert the cultural
difference in deceptiveness via self-construal as a potential moderator. The data, however,
did not support that assumption. Self-construal did not interact with beneﬁt to affect
honesty ratings (Lapinski, 1995). Similar to Lapinski’s (1995) study, Yeung et al. (1999)
and Murai (2000) investigated cultural differences in deception based on traditional
cross—cultural framework (i.e., individualism/collectivism or independent/ interdependent
construal). However, it is doubtful that such a broad perspective can grasp the
complicated phenomenon like deception and more subtle variation in cultural difference.

Although it was not cultural deception research, Levine (2001) argued that degree
of conﬁdence or uncertainty in judging deception, degrees of moral condemnation, and/or
degrees of lie importance could be confounding ratings of deceptiveness of messages
violating each type of violation, and, as a result, could inﬂuence the degree of
deceptiveness of messages. Testing this reasoning, Levine (2003) investigated the relative
impact of violation type and lie severity (importance) on judgment of message
deceptiveness. The ﬁndings indicate that quantity violations are rated as more deceptive
than quantity violations when lie severity is low, but this difference diminishes as lie
severity increases (Levine, 2003). Thus, differences between violation types are
moderated by the situation.

In terms of cultural difference in deception, the confounding factors that Levine
(2003) indicated may moderate a certain situation and cultural variance on deceptiveness
of each type of violation. Thus, under a particular situation, if degrees of moral

condemnation vary between cultures, then the deceptiveness of each type of violation

l7

will be inﬂuenced by the different level of perceived moral condemnation. For example,
Lee et al. (1997) explained the result of the study such that Chinese children rated
deception positively under pro-social situation because they accepted the behavior (i.e.
deception) as morally acceptable or more appropriate one.

In addition, Iizuka, Patterson, and Matchen (2002) investigated the conﬁdence on
judgment of deception as a part of the study comparing Japanese and Americans. They
found that American participants were much more conﬁdent in their judgment of
deception than the Japanese participants. Thus, it would also be possible to ﬁnd cultural
difference in deception by focusing on the cultural difference in a level of confidence
with the judgment under a certain situation.

Thus, the traditional cross-cultural ﬁamework may not be the best considering all
the issues, such as culture-speciﬁc situations, and culture by situation by message form
interactions on ratings on deceptiveness. Furthermore, given that previous cultural
research points out that cultural difference in deception may not be found across all
situations (e.g., Chunsheng, 2002; Lee et al., 1997; Seiter et al., 2002;), it is critical to
ﬁnd the aspect of situation that interacts with culture to impact the degree of
communication indirectness.

Consequently, it may be the useful ﬁrst step to focus on the culturally speciﬁc
concept related to communication style unique to a speciﬁc culture. Because this study is
based on IMT, a culturally unique way to manipulate the message along Grice’s
dimensions would be of particular interest, and this study thus investigates such a unique

way to manipulate it should be moderated under a situation in which the culturally

18

speciﬁc concept plays the most active role. Consequently, this study uses the nunch ’i in
Korea as such a useful framework to study cultural deception IMT research.
Nunch ’i in Korea, Deception and IMT

There is little qualitative research regarding nunch ’i and no quantitative research
on nunch ’i. Martin (1967) deﬁned nunch ’i as the attempt to read one’s mind and probe
one’s motives. From communication point of view, Park (1997) deﬁned nunch ’i as the
communicative acts that got message recipients to infer what the message senders meant
through the indirect or situational cues (verbal or nonverbal cues) in the message. In short,
nunch ’i is a communication style in Korea to make or interpret implication.

Nunch ’i is a type of topic-associate communication style. Au (1993) identiﬁes two
different kinds of communication style between European Americans and Asian
Americans: topic-centered and topic-associate style. European Americans prefer the ﬁrst
while Asian Americans are inclined toward the latter (Au, 1993). In a topic-centered
style, speakers focus on one issue at a time; arrange facts and ideas in logical, linear
order; and make explicit relationship between facts and ideas. Quality is determined by
clarity of descriptive details, or sufficient elaboration and how well explanations focus on
the essential features of the issue being analyzed. On the other hand, in a topic-associate
style, more than one issue is addressed at once. Related explanations unfold in
overlapping, intercepting loops. Relationships among segments within the statement are
assumed or inferred rather than explicitly established (Au, 1993).

Choi and Choi (1991) identiﬁed two nunch ’i processes in commrmication: nunch ’i
execution and nunch ’i ﬁguring out. Nunch ’i execution involves that message senders

deliver what they intent to mean to the message receivers without directly referring to

19

what they intent to mean. On the other hand, nunch ’i ﬁguring out involves that receivers
catch what is implied in the message through situational cues provided by senders. In
order for nunch ’i interaction to be successful, both execution and ﬁguring-out moves are
required (Choi & Choi, 1991).

For example, consider the situation in which someone from your ofﬁce drops by, but
you do not have time to talk with him or her because of a pressing deadline. A nunch ’i
emitted verbal message might be, “What time is it now?” the visitor should then realize
that the co-workers know very well what time it is and is simply providing an indirect
means of communicating the fact that the visitor should leave so that the co-worker can
continue with the work at hand. The visitor would then announce the intent of leaving,
and one appropriate response would be, “Oh, it’s already 4’ o’clock, I would better leave
now. I have an appointment at 4: l 5” (Martin, 1967)

The existing theoretical approach to indirect communication style can shed light on
the mechanism of nunch ’i communication. Holtgraves (1997) analyzes conversational
indirectness based on cultural dimensions such as individualism and collectivism and
argued that collectivistic culture prefers indirect communication, while individualistic
culture prefers direct communication. For example, an average Korean prefers to choose
“hint” while an average American “requests” as a favor asking strategy. Similarly,
applying conversation constraints theory to cross-cultural context, Kim and Wilson
(1994) found that members of individualistic cultures perceive direct requests as the most
effective strategy for accomplishing their goals, whereas members of collectivistic
cultures perceive direct request the least effective. According to Holtgraves (1997),

conversational indirectness has two dimensions. The ﬁrst dimension is a production,

20

which refers to an individual’s tendency to speak either indirectly or directly. The second
dimension is an interpretation dimension, which refers to the extent to which individuals
look for indirect meaning in the remarks of others. Relating two dimensions to nunch ’i,
nunch ’i execution corresponds to production dimensions of conversational indirectness
while nunch ’i ﬁguring out corresponds to interpretation dimensions of it. Hara and Kim
(2004) conducted a survey with Korean and American participants, testing effects of
cultural orientation on conversational indirectness. The results showed that the
communication style of Korean participants was positively associated with both
interpretation and production dimensions of conversational indirectness. Given nunch ’i is
a Korean form of indirect communication style, based on the results of Hara and Kim’s
(2004) study, it can be said Koreans are very accustomed to production and interpretation
of nunch ’i messages.

Nunch ’i execution and ﬁguring out reﬂect a prevalent communication style in Korea,
and the extent to which a person is socially competent largely depends on whether or not
he or she has nunch ’i (Park, 1997). In other words, nunch ’i is considered one of the most
critical social skills in Korea. As a result, substantial degree and ﬁequency of
conversational indirectness are ritualized and tolerated. One reason that nunch ’i is so
pervasive in Korea may be that the concept of nunch ’i is closely related to the concept of
face (Choi and Choi, 1991; Park, 1994). Ting-Toomey’s (1988) Face Negotiation Theory
suggests that individuals in collectivistic cultures place more value on the positive-face
needs of themselves and others than those in individualistic cultures. In order to maintain
face, collectivists prefer indirect communication style, such as nunch ’i for Koreans

(Ting-Toomey, 1988). Indeed, it is often speculated that collectivists’ presumed

21

preference for indirectness in communication due to face concerns (Kim, 1994; Kim and
Wilson, 1994; Hara and Kim, 2004; Gudykunst, Matsumoto, Ting-Toomey, Nishida,
Kim, and Heyman, 1996; Gudykunst & Ting—Toomey, 1988; Okabe, 1983). In short,
because of their need to maintain harmonious relationships, collectivists have come to
value more indirect, face-saving communication style. Accordingly, the overreaching
goal of the collectivists is to correctly infer the other’s mind and to know what the other
is thinking or feeling.

Nunch ’i also functions within a situation, and the success of nunch ’i execution and
ﬁguring out rely mainly on a shared understanding between the speaker and the listener
regarding the use of nunch ’i in the particular situation. Given that nunch ’i function counts
on situation, nunch ’i can be considered Korean form of High-Context (HC)
Communication. Hall (1976) proposes that “HC communication or message is one which
most of the information is either in the physical context or internalized in the person
while very little is in the coded, explicit, transmitted part of the message” (Hall, 1976,
p.79). In other words, using HC communication involves using and interpreting messages
that are not explicit, minimizing the content of the verbal message, and being sensitive to
others. Much drinking on cultural differences in communication style holds that HC
communication is predominant in collectivistic cultures (Gudykrmst et al, 1996; Kim,
2004; Kim & Wilson, 1994).

In a sense, nunch ’i is not only ability to read one’s mind, but also the ability to read
situation. To read situation in a manner consistent with another, one should be
accustomed to not only the culturally ritualized manner, but also the situation as it relates

to the speciﬁc person with whom one is interacting. Thus, nunch ’i often involves both

22

ritualized situational cues and person-speciﬁc situational cues. The personalized
situational cues activate the common relational and interaction knowledge that the
interactants share. Theory of Communication Responsibility that Anne (1998) articulated
bears important implication for the nunch ’r' mechanism. Aune contends that, “Responding
to the demand of a communicative situation can be thought of as a communicator’s
responsibility and that communicator’s burden of communicative responsibility is not
uniform across communicative situations” (Aune, Levine, Park, Asada, & Banas, 2005, p
360). Finally, he argues that our judgments of each other’s communicative responsibility
will inﬂuence the extent to which we engage in communicative use of implicature (e.g.,
nunch ’i execution) and inference-making (e.g., nunch ’i ﬁguring out) (Aune et al. 2005).

Speaking of nunch ’i, how to make messages using nunch ’i, how to interpret the
messages produced by nunch ’i, and extent to which nunch ’i is activated depend on
particular situations and common ground. For the nunch ’i communication using
ritualized situational cues, the cues pertaining to the situation is limited to some extent
and rather ﬁxed, so that people can execute or ﬁgure out nunch ’i without thinking about
it elaborately. On the contrary, for the nunch ’i communication using the personalized
situational cues, because the way that the interactants encode or decode nunch ’i is
personalized based on unique common ground that the interactants share and extent to
which nunch ’i is activated is very deep, sometimes only the interactants will notice what
the nunch ’i implies.

While there are both ritualized and personalized uses of nunch ’i, the ritualized use of
nunch’i will stand out when we compare Koreans with Americans in terms of

communication indirectness. This is because although nunch ’i communication is needed

23

in a very broad context among Koreans, the situation under which nunch ’i is most active
is the one where hierarchical social structure is involved. In hierarchical relationships in
Korea, there are number of ritualized nunch ’i communication situations.

One reason that a number of nunch ’i communications is ritualized in hierarchical
relationships in Korea can be explained in terms of face concern. Ting-Toomey and
Kurogi (1998) pointed out that power distance should be taken into consideration in
explaining face negotiation. Power distance, from the cultural value analysis level, refers
to the way a culture deals with status difference and social hierarchies. Culture differs in
the extent to which they view status inequalities, such that people in large power distance
cultures tend to accept unequal power distributions whereas people in small power
distance tend to value evenly distributed power. Typically, America scores on the low
side of power distance, while Korea scores on the high side (I-Iofstede, 2001). For large
power distance cultures, such as Korea, vertical-based facework interactions in
accordance with hierarchical roles are predominant (T ing-Toomey et al. 2001). In
addition, Ide (1998) investigated cultural difference in apology types, reporting that in
collectivistic cultures such as Japan, social distance and the power relationships between
the speaker and the listener constraint the speaker’s choice of speech types. Given the
fact that the different interpretations and enactment of ‘power” issues are tied closely to
everyday facework practice, and power distance value is often signaled through the
linguistic habits of a cultural community (e.g., nunch ’1'), it is natural that a number of
linguistic practice (i.e., nunch ’i) has been developed and ritualized in accordance with

hierarchical roles among Koreans.

24

Another reason that a number of nunch ’i communications is ritualized in
hierarchical relationships in Korea can be explained in terms of politeness. Politeness
Theory invokes face management concerns to explain why we use indirectness and how
others can understand the indirect communicative behavior by considering face needs
(Brown & Levinson, 1987). A bulk of research indicated that there are cultural
differences in face needs, such that members of collectivistic cultures tend to express a
greater degree of other-face or mutual-face concerns than members of individualistic
cultures (see Oetzel and Ting-Toomey, 2003). Furthermore, the recent study argued that
there might be cultural differences in the concept of politeness itself (Koyama, 2001;
Koyama and Jacobs, 2001). For instance, Koyama and Jacobs (2001) argued that there
would be divergences in perceived politeness within a particular cultural orientation, such
that Japanese, Chinese, and Korean tend to strongly associate politeness with formality,
while Americans tend to associate the politeness with the concept of ﬁiendliness.
Similarly, the importance of status in communication style is also found among Japanese
as high as Koreans in power distance. Nixon and Bull (2006) compared the way to
express emotion between Japanese and British, and found that Japanese expressed status
cues more overtly than British, while British expressed intimacy cues more overtly
(Nixon and Bull, 2006). Taken together, in Korea, it is likely that facework related to
politeness will be salient between people different in status rather than people equal in
status. Consequently, nunch ’i will be salient in hierarchical relationships because nunch ’r'
as a form of indirect communication style reﬂects such a speciﬁc facework orientation.

The fact that the face concern or concept of politeness is closely related to status

difference and formality provide explanation for the result of the recent study focusing on

25

cross-cultural comparison of communication indirectness. Sanchez-Burks, Lee, Choi,
Nisbett, Zhao, and K00 (2003) examined the communication style in the working
situation and the social situation comparing Americans with Koreans, Chinese, and
Japanese. The results indicated that Koreans were more indirect than Americans in
working situations while Americans were just as indirect as Koreans in social situations
(Sanchez-Burks et al., 2003). It is reasonable to think that status difference and formality
is more emphasized in working situation than in social situation, thus it is not surprising,
as Sanchez-Burks et al. (2003) showed, that communication indirectness such as nunch ’1'
stands out in working situation.

In addition, given that politeness is one of the main reasons that people are to be
deceptive as past research maintained, there should be cultural difference in deception
when people with different cultural background are deceptive, and under the situation in
which status difference and formality is involved, the nunch ’i should impact on the

deceptiveness of the message depending on the type of deceptive messages.

26

CHAPTER 2: HYPOTHESES AND RESEARCH QUESTIONS

The concept of nunch ’i is closely related to the concept of implicature in Grice’s
theory of implicature. In Grice’s theory, the implicature means that which can be inferred
ﬁ'om what is said, and communicators should assume the CP in order to produce the
implicature (Grice, 1989). As the previous example, although the statement, “What time
is it now?” sounds irrelevant at the face value, what is actually meant by the statement is
“I am busy now, so could you come by later. ” That is, at the level of what is implied (i.e.,
implicature), the statement is not only relevant but it is also clear if what is implied is
understood. Likewise, even if the visitor does not have an appointment at 4:15, the
statement is not false statement because the other communicator are able to infer the
actually meaning, or what is implied by the statement, “ Since this is not a good time, I
will be back later.”

Given the fact that nunch ’i is pervasive in Korea and nunch ’i usually includes
messages violating the maxim of quantity, manner, and relevance, it is likely that the
range of quantity, manner, and relevance violations that Koreans consider culturally
deceptive would be narrower than those with Americans. In other words, it is likely that
the level of tolerance on violations of quantity, relevance, and manner will be higher
among Koreans than Americans. Thus, unless the degree of violation is beyond the level
of tolerance Koreans have, they may not judge the nunch ’i messages as deceptive
because the violation do not function to, nor are they meant to deceive. However,
Koreans, same as Americans, will judge messages with signiﬁcant violation of each
maxim should judge the message as deceptive regardless of deception type if the

violation is such that an accurate understanding is unlikely to be achievable through

27

nunch ’i. In fact, Yetmg et al.’s (1999) replication study of the IMT in Hong Kong implies
that what counts as a violation differs signiﬁcantly depending on a level of the preference
for the conversational indirectness. In other words, Koreans count the violation within
range of their normal indirect communication style as not covert violation, but overt
violation, which is not a deception (McComack, 1992). Accordingly, if the manipulation
of quantity, manner and relevance violation is beyond some threshold and Koreans
perceive them as beyond the range of what can be considered overt violation caused by
communication indirectness, such violation of quantity, relevance and manner should
differ from the baseline message.

Therefore, this study ﬁrst investigates the main effect for type of covert violations
between Koreans and Americans. For quality violations, several studies (e.g., Jacobs et
al., 1996; Lapinski, 1995; Levine, 1998, 2001, McComack et al., 1992b; Yeung et al.,
1999) have found that quality violations are rated more deceptive than any other
violations. In fact, quality violations or falsiﬁcation is the most direct form of deception.
Levine’s (2001) study that examined the distributions of ratings of each violation found
that quality violations formed a negatively skewed uni-modal distribution with the most
extreme deception rating possible as the mode. Given the nature of nunch ’i involving
indirect communication style with subtle situational cues, nunch ’i will not play a strong
role in deceptiveness ratings of messages covertly violating the quality maxim.
Consequently, it is expected that there should be no difference between Korea and US. in
the ratings on deceptiveness for the quality violation messages. Therefore, this study
predicts that:

H]: There will be a main effect for violation type.

28

Hla: The deceptiveness ratings of quality violation messages will be higher
than those of other message forms such as the baseline honest message and the messages
violating quantity, relevance, or manner.

Hlb: The deceptiveness ratings of quality violation messages will not be
moderated by culture.

However, nunch ’i may play a role in the perceived deceptiveness of quantity,
relevance, and manner violation messages. As mentioned in the previous chapter, nunch ’i
is a Korean indirect communication style. These cultural differences in communication
style might cause cultural variation on ratings of deceptiveness in three of the information
manipulation dimensions. For example, for the quantity violation, the concept of nunch ’i
is associated with the value that Koreans view communicators being reserved as
competent. Extreme form of being reserved involves being silent. For example, a Korean
proverb has it that silence is better than thousands of words, or silence talks truth. In
nunch ’i communication, “silence is a communicative act rather than mere void in
communicational space” (Lebra, 1987, p343), rather than interpreted as violating the
quantity maxim. In terms of nunch ’i as HC communication style, Okabe (1983) argues
that HC communication such as nunch ’i requires transmitting messages through
understatement and hesitation (the opposite of Grice’s quantity maxim).

Furthermore, Au (1993) argues that the message of people in a culture that
prefers topic-associate style such as nunch ’i should be more ambiguous in manner, too
short or too long in the quantity of the messages, or more irrelevant to the topic. In
addition, some research reports that collectivists that prefer indirect communication style

are less concerned with clarity in conversation (Kim, 1994) and view clarity as necessary

29

for effective communication (Kim & Wilson, 1994) less than do individualists. Given
nunch ’i communication, which is usually irrelevant to the topic or ambiguous, it is
expected that Koreans are more likely tolerant to violation of relevance and manner. Thus,
the following hypotheses are provided:

H2: There will be a signiﬁcant interaction between violation type and culture.

H2a: The deceptiveness ratings of quantity violation messages will be lower
among Koreans than Americans.

H2b: The deceptiveness ratings of relevance violation messages will be lower
among Koreans than Americans.

H2c: The deceptiveness ratings of quantity violation messages will be lower
among Koreans than Americans.

Choi and Choi (1991) point out that nunch ’i is a kind of ‘mind reading.’ This ability
of ‘reading one’s mind’ is not only associated with producing and interpreting indirect
statements by nunch ’i, but also related to deception. McComack et al. (1996) noted that it
was necessary to produce and interpret deception that interactants should have the ability
to represent beliefs and desires about beliefs and desires of the other interactant. In other
words, the interactants should have the ability of representing one’s mental state. For
example, Peskin (1992) showed that children, who lack such ability, ‘reading one’s
mind,’ could not produce or even understand deception. In other words, the production
and the interpretation of deception and the production and the interpretation of nunch ’i
commonly need the ability to read one’s mind in mind. Thus, If an individual who lacks
the cultural knowledge enough to read the ritualized cues pertain to the situation with

nunch ’i, he/or she will judge the messages violating each maxim at the face value

30

deceptive. Or if an individual who lacks the common ground with the other individual
enough to read personalized cues with nunch ’i, he/or she will also judge the messages
deceptive. In contrast, If an individual who has ability to ﬁgure out what is implied with
nunch ’i, he/or she will judge the message deceptive despite the violation of each maxim.

Here, it should be noted that The Social Norm Violation Model proposed by Levine,
Anders, Banas, Baum, Endo, Hu, and Norman (2000) suggested that each culture had
their own range of culturally appropriate and desirable verbal and nonverbal behaviors
and that regardless of whether the behaviors are expected or not, the behaviors beyond
the range would be perceived deceptive. Relating the model to nunch ’i, while messages
produced by nunch ’i usually involves shorter, more irrelevant, and more ambiguous
messages at the level of what is said than fully stated truthful messages, under the
particular situation, if Koreans judge the nunch ’i messages as more culturally appropriate
to the situation than fully stated relevant and clear messages, they should be reluctant to
judge the nunch ’i messages as deception despite violation of each maxim.

Recall that The IMT deﬁnes the deception as covertly exploiting the belief that the
other person adhere to the CP (McComack, 1992a). However, while nunch ’i usually
violate quantity, relevance, or manner maxim at the level of what is said, nunch ’1'
messages stick to the CP at the level of what is implied. Thus, if the nunch ’i message
recipients can sufficiently infer what the messages mean by nunch ’i, the recipients who
are accustomed to nunch ’i ﬁguring out will not count the nunch ’i messages as deceptions
despite the deviation of baseline message. In other words, if nunch ’i message provides
the participants with sufficient cues to ﬁgure out what it really means, the message may

not be a covert violation, but a overt violation, which is a ‘ﬂout’ according to Grice

31

(1989). Therefore, under the nunch ’i message condition (the overt violation message), the
following hypothesis are provided

H3: The interaction between violation types and culture will be further
qualiﬁed, such that

H3a: Nunch ’i messages will be seen less deceptive by Koreans than Americans.

H3b: Koreans will see nunch ’1' messages less deceptive than covert violation
messages.

Gudykunst et al. (1996) point out that Grice’s maxims, in particular quantity,
relevance and manner maxim, are not characteristic of HC communication such as
nunch ’i, and argue that Gricean maxims on which the IMT is based may not hold true
cross-culturally. However, what is considered a violation may differ between Korea and
America (Yeung et al., 1999). Consequently, as long as the extent to which the message
violates each maxim is beyond the range that Koreans normally consider it appropriate,
Koreans will judge the messages deceptive. That is, for covert violation, regardless of
culture, the message should be judged as deception, as IMT predicts. Thus, it is predicted
that: under covert violation condition,

H4: Koreans will see covert violation messages nearly as deceptive as
Americans.

In addition, this study explores the distributions of these ratings. Levine (2001)

showed that quantity violation had bimodal distribution reﬂecting individual difference in
the judgment of deceptiveness, while the manner violation was approximately at the scale

midpoint, reﬂecting the moderate degree of perceived deceptiveness. It is, however,

32

unclear whether Levine’s (2001) interpretation will be applicable to other culture than
US. Thus, this study proposes two research questions:

RQl: Will the distribution of Koreans’ ratings on deceptiveness of manner
violation show moderate degree of deceptiveness?

RQ2: Will the distribution of Koreans’ ratings on deceptiveness of quantity

violation show bimodal?

33

CHAPTER 3: METHOD
Participants

The previous IMT studies on cultural deception have a problem with the sample as a
cross-cultural study. For example, while Lapinski (1995) conducted an experiment with
ethnically and culturally diverse sample in Hawaii, there might be cultural gaps between
Hawaiian Americans and mainland Americans and/or Hawaiian Asians and Koreans,
Japanese, and Chinese. In addition, the Hong Kong participants that Yeung et al. (1999)
recruited seems fairly westernized, given that Yeung et al. (1999) reported all participants
had no problem to understand the scenarios written in English. Thus, for this study,
participants were considered Americans if they were born in the United States and used
English as their ﬁrst language. Participants were considered as Koreans if (a) they held
Korean citizenship and were born in Korea, (b) their ﬁrst language was Korean, and (c)
they had lived in the United States for less than 5 years.

Consequently, this study recruited 256 undergraduate students in Korea and 199
American undergraduate students. In recruiting the participants, for Korean participants,
one professor from the College of Social Science at Seoul National University helped
recruit male participants while one professor from the Department of Business
Management at Sook-Myung Women’s University helped recruit female participants.
The US. participants were recruited from Michigan State University. The professor in
the Department of Communication helped recruit them. All participants gained course

research credits in exchange for their participation. The participants completed survey.

34

Research Design

This experiment used a 2 X 2 X 5 mixed groups factorial design with two levels of
ethnicity (Korean/American), two levels of message example (covert violation message /
nunch ’i message: overt violation message), and ﬁve levels of maxim violation. The
between subject factors were ethnicity and message example while the within subject
factor was types of maxim violation. Each type of messages involved a baseline truthful
message (no violation) a false message (quality violation), a message omitting crucial
information (quantity violation), an evasive message (relevance violation) or an
equivocal message (manner violation). The ratings on message deceptiveness served as
the dependent measure.
Procedures

Participants were provided with a questionnaire. The questionnaire for the Korean
participants was written in Korean. Discrepancies between the American and the Korean
version of the questionnaire were reviewed by the author and bilingual translators, and
then were resolved.

The ﬁrst page of the questionnaire contained a hypothetical situation that author
created. Participants were instructed to read the situation carefully and to imagine that
they were in the situation. The participants took a role of an outside observer of the
situation. The scenario involved the situation in which the student (Min-Young for
Korean participants or Mike for American participants) made an excuse to the professor
(Prof. Parker for American participants or Prof. Kim for Korean participants). This
situation was adopted because nunch ’i might play a critical role in making request or

excuse between those high in social status and low in social status. Min-Young or Mike

35

provided a fully disclosive message and a message violating one of the maxims described
in IMT. The scenario and messages are presented in detail in Appendix A.

After reading the scenarios, the participants were asked to rate the messages on a
4-item deceptiveness scale. Participants were not told that the focus of the study was on
deception, nor were they given any further instructions on how to respond. Manipulation
check items for each violation type were also completed. All items were taken from
McComack et al. (1992) and translated into Korean by one bilingual translator. After
completing the survey, the participants were debriefed.

Measurements

Perceptions of message deceptiveness were measured using the four-item
semantic differential scale with a seven- point response (7 = most deceptive) format
developed by McComack et al., (1992). McComack et al., (1992) provided evidence for
the reliability, validity, and dimensionality of the scales measuring respondents’
perceptions of messages.

On the other hand, Levine et al. (2003) suggested that the future research should
use both dichotomous and continuous measurement on the deceptiveness of the messages,
pointing out that scaling judgments of deception with continuous measures could
confound conceptually and operationally distinct dimensions. Thus, this study also used
dichotomous measures.

The items for manipulation were also included in the questionnaires. The survey

questions were presented in Appendix B.

36

Proposed Statistics Analysis
To test the proposed hypnoses, this study used the mixed model of Analysis of

Variance (ANOVA) by the SPSS version 14.0 to analyze the data.

37

CHAPTER 4: RESULTS
Measurement Analysis

Item retention decision was made by assessing item contribution to scale
reliability. As a result, all items were retained and all scales exhibited acceptable
reliabilities. The reliabilities of all scales except for the scale for quantity violation of the
baseline message (a. = .80), were near above .90. The alpha for each scale is reported in
Table 1.

Manipulation Check

To insure the messages were perceived as violations of the intended maxims,
participants’ ratings of those messages in which information was manipulated were
compared with ratings of truthful/baseline messages. To this aim, paired samples t-tests
were conducted. The tests showed that regardless of cultures and conditions, all of the
manipulations were successful.

For US. participants in covert violation condition, manipulations of quality were
seen as differing signiﬁcantly in terms of distortion from baseline messages, ((96) =
23.35, p < .0001, r2 = .85. Manipulations of quantity and baseline messages were seen as
signiﬁcantly different in terms of informativeness, t(95) = 11.45, p < .0001, r2 = .58.
Messages which involved manipulations of relevance and baseline messages were
signiﬁcantly different in terms of relevance, t(96) = 13.98, p < .0001, r2 = .67, and those
messages which involved manipulations of manner and baseline messages differed
signiﬁcantly in terms of clarity, t(96) = 11.28, p < .0001, r2 = .57. Thus, in the US,

covert violation conditions, all manipulations were successful.

38

For US, nunch’i condition, manipulations of quality, t(101) = 32.19, p < .0001,
r2 = .9], quantity, 1(100) = 21.90,p < .0001, r2 = .83, relevance, t(101)=13.21,p
< .0001, r2 = .63, and manner, t(101) = 18.52,p < .0001, r2 = .77, were successful.

For Korean covert violation condition, manipulations of quality, t(130) = 27.71 , p
< .0001, r2 = .86, quantity, t(130) = 15.32,p < .0001, r2 = .64, relevance, t(130) = 12.79,
p < .0001, r2 = .56, and manner, t(130) = 10.50, p < .0001, r2 = .46, were successful.

For Korean nunch’i condition, manipulations of quality, t(122) = 24.23, p < .0001,
r2 = .83, quantity, t(123) = 23.74,p < .0001, r2 = .82, relevance, t(121) = 13.38,p
< .0001, r2 = .60, and manner, ((124) = 16.65, p < .0001, r2 = .69, were successful.
Primary Results

To examine the main effects and interaction effects for violation type, culture and
condition, a 2(country: US. and Korea) X 2(condition: covert violation and nunch’i) x 5
(violation type: baseline, quality, quantity, relevance, and manner violation messages)
mixed model analysis of variance (AN OVA) was conducted with violation type as a
repeated factor, and country and condition as independent groups variables, and
participants’ ratings of message deceptiveness as the dependent variable.

H1 predicted that there would be a main effect for violation type. The results of
the AN OVA indicated a large and signiﬁcant main effect for violation type on
respondents’ ratings of message deceptiveness, F (4, 1804) = 689.51, p < .0001, 772 = .51.
Thus, the data were consistent with H1. The cell means and the standard deviations for
ratings of each violation type and each condition are presented in Table 2, and cell means
for the current study is compared to cell means for Yeung et al. (1999), Jacobs et al.

(1996), Lapinski (1995) and McComack et al. (1992) in Table 3.

39

Both the interaction between violation type and culture, F (4, 1804) = 42.48, p
< .0001 , 172 = .02, and the interaction between violation type and condition, F (4, 1804) =
15.15,p < .0001, 772 = .01 were signiﬁcant, but much smaller than the main effect. In
addition, the three-way interaction among violation type, country, and condition was also
signiﬁcant, F(4, 1804) = 10.91,p < .0001, 172 = .01.
Further Analysis for the Main Eﬂect for Violation Type

The Hla predicted that the deceptiveness ratings of quality violation would be
higher than those of other message forms such as the baseline message and the message
violating quantity, relevance, and manner. To test this, paired sample t-tests were
conducted for each culture and condition. Support was found across cultures and
conditions.

For Americans in covert condition, deceptiveness for quality violation (M = 6.45 ,
SD = 1.14 ) was signiﬁcantly higher than deceptiveness for baseline (M = 1.51, SD
= .99 ), t(96) = 32.12,p < .0001, r2 = .91, quantity (M= 5.15 , SD = 1.79 ), t(96) = 6.67,
p < .0001, r2 = .32, relevance (M=5.20 , SD = 1.57 ), t(96) = 7.12,p < .0001, r2 = .35,
and manner (M= 5.57 , SD = 1.39 ), t(96) = 3.77,p < .0001, r2 = .13. Thus, Americans
viewed the covert quality violation message as more deceptive than any other types of
covert violation messages.

For Americans in nunch’i condition, deceptiveness for quality violation (M =
6.47 , SD = 1.02) was signiﬁcantly higher than deceptiveness for baseline (M = 1.51 , SD
= .95 ), t(101) = 32.9,p < .0001, r2 = .82, quantity (M= 5.01 , SD = 1.64 ), t(101) = 7.41,
p < .0001, r2 = .36, relevance (M= 4.87 , SD = 1.59 ), t(101) = 9.38,p < .0001, r2 = .48,

and manner (M= 5.44 , SD = 1.60 ), ((101) = 6.32,p < .0001, r2 = .29. Thus, Americans

40

viewed the nunch’i quality violation message as more deceptive than any other types of
nunch’i violation messages.

For Koreans in covert condition, deceptiveness for quality violation (M = 6.26 ,
SD = 1.03 ) was signiﬁcantly higher than deceptiveness for baseline (M = 2.56 , SD =
1.29 ), t(130) = 26.36,p < .0001, r2 = .84, quantity (M= 5.44 , SD = 1.76 ), t(l30) =
4.92,p < .0001, r2 = .16, relevance (M= 5.85 , SD = 1.18 ), t(130) = 3.06,p < .01,
r2 = .07, and manner (M= 5.96, SD = 1.23 ), t(130) = 2.27,p < .05. r2 = .04. Thus,
Koreans viewed the covert quality violation message as more deceptive than any other
types of covert violation messages.

For Koreans in nunch’i condition, deceptiveness for quality violation (M = 6.33 ,
SD = .87 ) was signiﬁcantly higher than deceptiveness for baseline (M = 2.57 , SD =
1.27 ), t(124) = 27.27,p < .0001, r2 = .86, quantity (M= 4.07 , SD = 1.65 ), t(124) =
15.06,p < .0001, r2 = .65, relevance (M= 5.68 , SD = 1.41 ), t(124) = 4.80,p < .0001,
r2 = .16, and manner (M= 4.91 , SD =1.79), t(124) = 8.37,p < .0001, r2 = .36. Thus,
Koreans viewed the nunch’i quality violation message as more deceptive than any other
types of nunch’i violation messages.

The Hlb predicted that the deceptiveness ratings of quality violation messages
will not be moderated by culture. To test for H l b, l (deceptiveness ratings for quality
violation) X 2 (culture) AN OVA was conducted. The results were consistent with Hlb in
that difference in terms of deceptiveness for quality violation messages between Koreans
(M = 6.29, SD = .95) and Americans (M = 6.46, SD = 1.08) was not signiﬁcant, F (1,
453) = 3.15, p > .05. 172 = .006. That is, Koreans viewed quality violation messages

nearly as deceptive as Americans (See Figure 1).

41

Further Analysis for Interaction between Violation Type and Culture
The H2 predicted that there would be a signiﬁcant interaction between violation
type and culture. As the previous 2X2><5 repeated measures ANOVA indicated, the

interaction between violation type and culture was signiﬁcant, F (4, 1804) = 42.48, p
< .0001, 772 = .02. However, further analysis showed that the interaction was not in the

direction that this study predicted. The results are presented in Figure 1. It was predicted
that the deceptiveness ratings of quantity violation messages (HZa), relevance violation
messages (I-12b), and manner violation messages (H2c) will be lower among Koreans
than Americans. Data did not support H23, H2b, and H2c. The difference in terms of
deceptiveness for quantity violation messages between Koreans (M = 4.77, SD = 1.84)
and Americans (M =5.08, SD = 1.71), was not signiﬁcant, t(453) = 1.85, p > .05.
Similarly, the difference in terms of deceptiveness for manner violation messages
between Koreans (M = 5.45, SD = 1.61) and Americans (M = 5.55, SD = 1.64), was not
signiﬁcant, t(453) = .68, p > .05. Consequently, Koreans viewed quantity and manner
violation messages as similar in deceptiveness as Americans.

Interestingly, the difference in terms of deceptiveness for relevance violation
messages between Koreans (M = 5.77, SD = 1.29) and Americans (M = 5.03, SD = 1.59)
was signiﬁcant, ((453) = -5.43, p < .0001, r2 = .06, but in the opposite direction. That is,
contrary to the prediction, Koreans than Americans viewed relevance violation messages
signiﬁcantly more deceptive. In addition, Koreans (M = 2.56, SD = 1.28) than Americans
(M = 1.51, SD = .97) viewed baseline messages as signiﬁcantly more deceptive, ((453) =
9.66, p < .0001, r2 = .17. In conclusion, data did not support the second hypothesis of this

study.

42

This study further examined violation type by culture interaction for each
condition. H3a expected that Koreans than Americans will see nunch’i messages as less

deceptive. A 2 x 5 repeated measures ANOVA found support for H3a, F (4, 900) = 23.32,
p < .0001, n2 = .03 That is, violation type by culture interaction was signiﬁcant when

nunch’i messages were evaluated. The results are in Figure 2. Independent sample t-test
was conducted to investigate which type of nunch’i messages Koreans and Americans
viewed differently in terms of deceptiveness. Koreans (M = 4.07, SD =1.65) viewed
nunch’i quantity violation messages as less deceptive than Americans (M = 5.01, SD
=1.64), t(225) = 4.32, p < .0001, r2 = .08. Furthermore, Koreans (M = 4.91, SD = 1.79)
viewed nunch’i manner violation messages as less deceptive than Americans (M = 5.44,
SD = 1.60), t(225) = 2.32, p < .05, r2 = .02. On the other hand, Koreans (M = 5.68, SD =
1.41) viewed nunch’i relevance violation messages as signiﬁcantly more deceptive than
Americans (M = 4.87, SD = 1.59), t(225) = - 4.06, p < .0001, r2 = .07. This last ﬁnding
was inconsistent with the predictions.

This study also investigated violation type by culture interaction in covert
violation condition, and expected that the interaction in covert violation condition should
not be signiﬁcant (H4). On the contrary to the prediction, it was signiﬁcant, F(4, 904) =
12.03, p < .0001, 172 = .01, but effect size was small. Independent sample t-test was

conducted to investigate the interaction between each type of violation messages and
culture in covert violation condition. The results are presented in Figure 3. Although
Koreans did not view covert quantity violation messages, t(226) = - 1.22, p > .05, and
covert manner violation, t(226) = -1.49, p > .05, as less or more deceptive than

Americans, Koreans(M = 5.85 , SD = 1.18 ) viewed covert relevance violation messages

43

as significantly more deceptive than Americans(M = 5.20 , SD = 1.57 ), t(225) = -3.55, p
< .01, r2 = .05.
Further Analysis for 3- Way Interaction among Violation Type, Culture and Condition
The previous 2XZ><5 repeated measures AN OVA showed that the interaction
among violation type, culture, and condition was signiﬁcant. That is, the interaction
between type of violation and condition was moderated by culture. To investigate the role
of culture as a moderator in the violation type by condition interaction, 2 (condition) x5
(type of violation) repeated measures AN OVA for each culture was conducted. H3b
predicted that nunch’i messages relative to covert violation messages should be viewed as
less deceptive by Koreans than Americans. The results indicated that Koreans did see
nunch’i messages as less deceptive than covert violation, F(4, 1016) = 16.84, p < .0001,

772 = .02. Thus, data were consistent with H3b. On the other hand, Americans did not

see nunch’i messages as less deceptive than covert violation messages, F (4, 788) = 1.14,
p > .ns.

Further analysis was conducted to investigate which kind of nunch’i messages
Koreans or Americans viewed less deceptive than covert violation messages. The results
are presented in Figure 4 and 5. Independent samples t-test revealed that Koreans viewed
nunch’i quantity violation messages(M = 4.07 , SD = 1.65 ) as signiﬁcantly less deceptive
than covert quantity violation messages (M = 5.44 , SD = 1.76 ), t(254) = 6.44, p < .0001,
r2 = .14, bur not for Americans, t(197) = .57, p > .05 (see Figure 6) Furthermore,
Koreans viewed nunch’i manner violation messages(M = 4.91 , SD = 1.61 ) as

signiﬁcantly less deceptive than covert manner violation messages (M = 5.96 , SD =

1.23 ), 1054) = 5.48, p < .0001, r2= .11, bur not for Americans, t(197) = -.32, p> .05

(see Figure 7). However, nunch’i relevance violation messages was seen nearly as
deceptive as covert relevance violation messages by Koreans, t(254) = 1.03, p >.05 as
well as Americans, t(197) = 1.48, p >.05 (see Figure 8). In addition, the deceptiveness of
baseline messages and quality violation messages were not different depending on the
condition for both Koreans and Americans (see Figure 9 and 10).
The Analysis for Dichotomous Deception Ratings

Chi- square was conducted to investigate the association between dichotomous
judgments on types of violation and condition for Koreans and Americans. The number
of judgment of honest on each type of violation and the percentages are presented in

Table 4. The results showed that for Americans, there were no signiﬁcant associations
between condition and types of violation: for baseline message, 12 (1, n = 199) = 1.90, p
>.05, ¢ = .10 for quality violation message, 12 (1, n = 199) = 1.18, p >.05, ¢ = .08 for
quantity violation message, 12 (1, n = 199) = 0.04, p >.05, ¢ = .01, for relevance
violation message, 12 (1, n = 199) = 1.46, p >.05, ¢ = .09, and for manner violation
message, 12 (1, n = 199) = .52, p >.05. ¢ = .05. Thus, consistent with the analysis of
continuous ratings, analysis of dichotomous judgment revealed that Americans did not

view the nunch’i violation messages less deceptive than covert violation messages.

However, chi-square showed that for Koreans, there were signiﬁcant associations
between conditions and some of types of violations: for quantity violation, 12 (1, n =
256) = 26.33, p <.0001, ¢ = .32, for manner violation, 12 (1, n = 256) = 18.36, p <.0001,
¢ = .27 and for relevance violation, 12 (1, n = 256) = 6.86, p <.01, ¢ = .16. Interestingly,

when we examined continuous ratings, Koreans did not see nunch’i relevance violation

45

messages as less deceptive than covert relevance violation messages. However, when we
examined dichotomous judgment, the results showed that Koreans judged nunch’i
relevance violation messages as honest signiﬁcantly more often than covert relevance

violation messages. On the other hand, for Koreans there were no signiﬁcant associations

between condition and baseline messages, 12 (1, n = 256) = 0.20, p >05, (6 = .01 and for
quality violation messages, 12 (1, n = 256) = 1.1 1, p >.05, ¢ = .07.
The Analysis for Distribution of Koreans ’ Ratings on Deceptiveness

This study investigated whether the distribution of Koreans’ ratings on
deceptiveness of manner violation would show a moderate degree of deceptiveness
(RQl). The Figure 11 shows negatively skewed distribution, (Skewness = - .98, SD
= .15) and Kolrnogorov-Smirnov test (KS- test) revealed that the distribution was
deviated from normality signiﬁcantly, D(256) = .17, p < .001 Thus, the distribution of
Koreans’ ratings on deceptiveness of manner violation did not show moderate degree of
deceptiveness.

This study also investigated whether the distribution of Koreans’ ratings on
deceptiveness of quantity violation would show bimodal distribution (RQ2). The Figure
12 shows negatively skewed distribution, (Skewness = - .37, SD = .15). Subsequently,
the analysis for distribution of Koreans’ ratings on deceptiveness of quantity violation in
each condition was followed. As Figure 13 illustrates, for covert quantity violation, the
distribution of Koreans’ ratings on deceptiveness showed skewed distribution (Skewness
= - 1.16, SD = .21), and deviated from normality signiﬁcantly, (see Figure 14). However,
for nunch’i quantity violation, the distribution of Koreans’ ratings on deceptiveness

showed moderate degree of deceptiveness, (Skewness = - .25, SD = .22, see Figure 15),

46

and similar to normal distribution (See Figure 16). In conclusion, while the distribution of
Koreans’ ratings on deceptiveness of quantity violation did not show bimodal, Koreans

viewed nunch’i quantity violation messages as moderately deceptive or honest.

47

CHAPTER 5: DISCUSSION

First, this study investigated a main effect of violation types on participants’
ratings on message deceptiveness. Consistent with the result of the previous IMT
research (e.g., Yeung et al., 1999, Jacobs et al., 1996, Lapinski, 1995, and McComack et
al., 1992), the participants rated the deceptiveness of each message differently depending
on the type of violation (i.e., baseline, quality, quantity, relevance, and manner violation).
Alternatively stated, Americans and Koreans viewed some type of violation messages as
more deceptive, but other types of messages as less deceptive.

Speciﬁcally, as expected, quality violation messages were viewed as more
deceptive than any other type of violation messages. Furthermore, the deceptiveness
ratings on quality violation were not moderated by culture; that is, Koreans did not see
quality violation messages as more or less deceptive than Americans. These results are
consistent with the previous IMT cross-cultural studies, such that violation of the maxim
of quality is universally seen as deceptive by either collectivistic or individualistic
cultures (e.g., Yeung et al., 1999, Lapinski, 1995). Given that violation of quality has
nothing to do with communication indirectness, it is not surprising that cultures did not
affect the deceptiveness ratings on quality violation messages.

Although it is not hypothesized formally in this study, it is important to note that
for quality violation messages, the interaction effect between condition and country was
not signiﬁcant. In other words, not only cultures but also even condition (i.e. nunch ’i
messages / covert violation messages) did not moderate the relationship between quality
violations and ratings on deceptiveness. This result implies that participants may consider

the violation of quality as a ‘prototype’ of deception. That is, so long as Koreans or

48

Americans see that messages as explicitly denying the validity of the true information or
contradictory to true information (McComack, 1992), they may unconditionally judge the
message as a lie.

The most contradictory results in this study were found in the interaction between
violation type and culture. The previous cross-cultural studies on communication style
have found that communication style of those in collectivistic cultures such as Koreans
were more tolerant to incomplete, irrelevant, and ambiguous, compared to that of those in
individualistic cultures such as Americans (Hara and Kim, 2004; Kim, 1994; Kim and
Wilson, 1994). As a result, it was expected that tolerance level of such a violation
involving quantity, relevance, and manner would be higher among Koreans than
Americans. However, the result indicated that Koreans viewed quantity and manner
violation messages as nearly deceptive as Americans, and furthermore viewed relevance
violation messages more deceptive than Americans.

Those ﬁndings suggest that the cultural difference in communication indirectness
between collectivism and individualism may not affect individuals’ judgment on
deceptive messages. Alternatively stated, even though indirect communication style is
more prevalent and tolerated among Koreans than among Americans, it may be true only
if communication partners cooperate. If Koreans consider the partner is not cooperative,
or covertly providing incomplete, irrelevant, and ambiguous information, they may not be
more tolerant to those messages.

This interpretation is more qualiﬁed by the ﬁndings that Koreans judged nunch ’i
messages signiﬁcantly less deceptive than covert violation messages. On the contrary,

Americans did not see nunch ’i messages less deceptive than covert violation because

49

while Koreans perceived nunch ’1' messages more ‘overt’ violations rather than ‘covert’
violation, Americans might view nunch ’i messages as anther examples of ‘covert’
violation. This result suggests that when individuals judge the deceptiveness of messages,
they may use more profound standard, such as Grice’s CP.

While it is not surprising that Koreans judged nunch ’i quantity and manner
violations as less deceptive than Americans, they did not see nunch ’i relevance violations
as less deceptive than Americans. Furthermore, Koreans viewed relevance violation
messages signiﬁcantly more deceptive than Americans regardless of condition. Recall
that the violation of relevance maxim in IMT means, in fact, “diversionary responses,”
acts. Koreans might view diversionary responses to professor’s inquiry as inappropriate
acts, no matter what the intention of the act was. Obviously, this would be also the case
for Americans. However, Korean students might view the acts more inappropriate than
American students.

The Social Norm Violation Model proposed by Levine et al. (2000) provides
plausible explanation for this ﬁnding. The theory suggests that each culture has their own
range of culturally appropriate and desirable verbal and nonverbal behaviors and that
regardless of whether the behaviors are expected or not, the behaviors beyond the range
would be perceived as deceptive. For this study, Koreans might see the relevance
violation presented in this study as beyond the range of culturally appropriate and
desirable behaviors.

This study also included the dichotomous measurement, which ask the
participants if they think the message that they received is honesty or deceptive in a

dichotomous way. The results from dichotomous measurement were consistent with

50

those from continuous measurement, except for the result for the interaction effect
between culture and condition for relevance violation. For the dichotomous measurement,
inconsistent with continuous measurement, Koreans viewed nunch ’i relevance messages
as honesty signiﬁcantly more often than covert violation messages. Levine et al. (2003)
pointed out that scaling judgments of deception with continuous measures could
confound conceptually and operationally distinct dimensions. For this study, continuous
measurement might confound the level of appropriateness associated with social norm
regarding status difference in Korea, and this confounding might stand out for relevance
violation. It would be reasonable to think that the confounding factor might affect
deceptive ratings across all types of violation. However, for other types of violations, the
confounding might not have signiﬁcant impact on the ratings given that no deceptiveness
ratings on quantity and manner violation were universally lower and higher depending on
cultures. However, it is recommended that future research should include dichotomous
measurement besides continuous measurement. In addition, as Levine et al. (2003)
pointed out, when the research use continuous measurement, possible confounding
factors (e.g., conﬁdence in judgment or morality condemnation) should be taken into
consideration. This study did not measure the level of appropriateness of each message,
which should be one of limitation of this study.

Another limitation comes ﬁom the use of mixed experimental design, such that
participants evaluated each type of violation messages repeatedly. It is possible that
sensitization effect could inﬂuence the results of the study, and after participants were
exposed to one of the messages involving the violation of one of the maxims, their

fatigue or lack of naivete’ might make them unsuited for consecutive evaluation of

51

another message. Furthermore, this study employed strategic ordering instead of full-
crossed ordering, such that the students always evaluated quality violation messages prior
to other types of violation messages. As a result, the judgment on quality violation
messages might affect the judgment on other types of messages. This sort of ordering
effect may be plausible given the fact that the deceptive ratings on quantity, relevance,
and manner violation for this study were signiﬁcantly higher than for previous IMT
studies, and then the differences in deceptiveness among different types of messages for
this study were narrower than for previous IMT studies (see Table 3). That is, students
who viewed quality violation messages as substantially deceptive (i.e., choosing 7 out of
1-7 scales) might evaluated subsequent other types of violation messages as substansially
deceptive like the evaluation on quality violation messages. However, this ordering effect
may not weaken some signiﬁcant ﬁndings in this study because if such an ordering effect
took place, the effect might be disadvantageous to the ﬁndings. Nonetheless, the
recommendation is that the future research should employ independent design rather than
mixed group design that this study used.

Although the signiﬁcantly higher deceptiveness ratings may be caused by the
strategic ordering, the degree of lie potential of the situation in this study (i.e., the extent
to which the situation allow people to choose to ell the truth or lie) may be signiﬁcantly
lower than the situations in previous IMT studies. In fact, McComack et al.(1992)’s
original study evaluated lie potential of the scenario and used the situation with moderate
level of lie potential. Thus, it is recommended that future research evaluate lie potential

of the scenario and use the situation with moderate level of lie potential.

52

In conclusion, despite the limitations, some contribution was made for the deception
and cross-cultural research. First, the previous IMT study was successfully replicated
using Korean subjects. Second, in general cultural difference in communication
indirectness did not affect deceptive ratings. However, when the covertrress of the
violations was at issue (i.e., covert /nunch ’i violation messages), Koreans ﬁgured out
what is implied in nunch ’1' messages, viewing the violations as overt rather than covert,

and then judged the messages as less deceptive than Americans.

53

APPENDIX A

Tables

54

Table 1

Reliability for Each Scale

 

 

 

Types of Scale
Quantity Quality Relevance Manner Deceptiveness
Baseline .80 .93 .89 .95 .89
Quality .87 .90 .89 .95 .88
Quantity .92 .94 .91 .95 .93
Relevance .92 .93 .93 .96 .92
Manner .92 .92 .92 .96 .94

 

55

Table 2

The Deceptiveness Ratings of Each Violation Type for Country and Condition

 

BASE_dec

QUALl_dec

QUANTI_dec

REL_dec

MAN_dec

 

Country Condition Mean SD N
US Covert l .5 1 .99 97
Nunch'i l .51 .95 102

KOREA Covert 2.56 ‘1 .29 13 l
Nunch'i 2.57 1.27 125

US Covert 6.45 1.14 97
Nunch'i 6.47 l .02 102

KOREA Covert 6.26 1.03 13 1
Nunch'i 6.33 .87 125

US Covert 5.15 1.79 97
Nunch'i 5.01 1.64 102
KOREA Covert 5 .44 l .76 13 l
Nunch'i 4.07 1.65 125

US Covert 5 .20 1.57 97
Nunch'i 4.87 1.59 102

KOREA Covert 5.85 1.18 131
Nunch'i 5 .68 1.41 125

US Covert 5 .67 1.69 97
Nunch'i 5 .44 l .60 102

KOREA Covert 5 .96 1 .23 13 l
Nunch'i 4.91 1.79 125

 

56

Table 3

Cell Means for the Current Study, Yeung et al. (I 999), Jacobs et al. (1996), Lapinski

( I 995 ) and McComack et al. (1992).

 

Violation Type

 

Baseline Quality Quantity Relevance Manner

 

Current US 6.49 1.54 2.92 2.97 2.45
Study Korea 5.44 1.71 3.23 2.24 2.55
Hong Kong 4.94 3.34 4.90 3.50 4.61
Arizona 5.41 1.82 4.51 3.25 3.92
Hawaii 5.46 2.13 3.85 3.43 3.67
Michigan 5.47 1.74 4.20 2.93 3.43

 

Note: All scores were averaged on a scale of l to 7, where higher scores reﬂect higher

honesty ratings.

57

Table 4

The Dichotomous Judgment upon Types of Violations

 

 

Country Condition N %

BASE_dec U.S Covert 90 92.8
Nunch'i 99 97. l

KOREA Covert l 18 90. 1

Nunch'i 1 12 89.6

QUALl_dec U.S Covert 2 2. l
Nunch'i 5 4.9

KOREA Covert l .8

Nunch'i 3 2.4
QUANTI_dec U.S Covert l 8 18.6
Nunch'i 20 19.6
KOREA Covert 26 19.8
Nunch'i 63 50.4

REL_dec U.S Covert 1 7 l 7.5
Nunch'i 25 24.5

KOREA Covert l 2 9.2
Nunch'i 26 20.8
MAN_dec U.S Covert l 7 l 7.5
Nunch'i 22 21 .6

KOREA Covert l 2 9.2
Nunch'i 38 30.4

 

58

APPENDIX B

Figures

59

 

7 r
’ 6.46
l 6.29
l
6 1 5.77
i 5.45 5'55 I Korea
5.08 503 in Us W 1
5 4.77
§ .
l
4 L l.
QUALIﬁdec QUANTI_dec REL_dec MAN_dec
Figure 1

Interaction between Violation Type and Culture

60

  

65 [ 63
5.5 a: 5.44
5-01 l Korea

I

4.5
4.0

l

3.5 L, #1 l. . i L ;

 

 

 

QUALLdec QUANTLdec REL_dec MAN_dec
Figure2

Interaction between Violation Type and Culture in Nunch’i Condition

61

5.85 5.96
l .67
5.5 l 5.44 5 2
o 15

I Koreal
ll] US.

4.5 -

3.5 ~4— *4

 

 

 

QUALI_dec QUANTI_dec REL_dec MAN_dec
Figure 3

Interaction between Violation Type and Culture in Covert Condition

62

6.47

 

 

 

 

 

 

 

 

6.45
6 ..
5 .-
4 L L l
QUALI_dec QUANTI_dec REL_dec MAN_dec
—O-— Covert - «I» - Nunch'i
Figure 4

Interaction between Violation Type and Condition for US.

63

 

 

 

 

6 ' —-05.96
5 ' 3 ”£4.91
4 . i407 . .

QUALI_dec QUANTI_dec REL_dec MAN_dec

 

 

—0—Covert - I- - Nunch'i

 

 

Figure 5

Interaction between Violation Type and Condition for Korea

64

4.5 '

 

3.5

Figure 6

5.44

5.15... _____
' - - ~ I501

 

4.07

Covert N unch'i

 

 

+Korea
- «I- -U.S.

 

Culture by Condition Interaction for Quantity Violation Messages

65

 

5.5 -

 

i i 5' 5-44 +Korea
- .- -U.S.

 

 

 

 

4.91

 

4. 5 ' '
Covert Nunch'i

 

Figure 7

Culture by Condition Interaction for Manner Violation Messages

66

 

 

 

—O-Korea
- I- -U.S.

 

 

5.68
5.5 -
5.2 I .
~‘:I487
4.5
Covert Nunch'i
Figure 8

Culture by Condition Interaction for Relevance Violation Messages

67

 

6.45 . .............. I 6.47

6.26

 

 

 

—O—Korea
- I- -U.S.

 

 

Covert Nunch'i

Figure 9

Culture by Condition Interaction for Quality Violation Messages

68

 

 

 

 

 

 

+Korea
- l -U.S.

 

 

2.560 #257
2 .-
151 I - -' ----------- I 1.51
‘l r
Covert Nunch'i
Figure 10

Culture by Condition Interaction for Baseline Messages

69

 

. 5

. 0000

. 557

. 0002

. 557

. 00000022

. 57777

. 0000000000000222
. 5555577777

. 00000000000002222
. 5555577

. 0000000000000000000000000000000000000000

\JO\Cthtm-h~h-Uabohabou—r—

Figure 11

Stern and Leaf Plots of Deceptiveness Ratings by Manner Violation.

70

. 00002

. 577

. 000022

. 5555777

. 0000002222

. 5577

. 000000000002

. 555557777

. 000000222

. 5555577

. 000000000000222
. 5557777

. 00000000000000000000000000

\IQONMLh-h-bWUJNNi—sr—

Figure 12

Stern and Leaf Plots of Deceptiveness Ratings by Quantity Violation

71

. 2

. 57

. 00002

. 557

. 0000

. 55

. 000002

. 5555577

. 000000000022

. 555557777

. 0000000000000000022222
. 55555557777777

. 000000000000000000000000000000000000000

\IO\O\UrUr-§-§UJUJNNr——'

[ﬁgunel3

Stern and Leaf Plots of Deceptiveness Ratings by Covert Quantity Violation

72

 

oo-l

 

 

 

 

 

 

 

 

QUANTI_dec

Figure [4

Histogram of Deceptiveness Ratings by Covert Quantity Violation

73

1. 00002

. 57777

. 000002222

. 5555555777777

. 00000000022222222
. 55577777

. 0000000000000000002
. 555555777777

. 00022222

. 555557

. 00000002

. 7

. 00000000000000

\IQQUIUI-h-bUJUJNN"

Figure 15

Stern and Leaf Plots of Deceptiveness Ratings by Nunch’i Quantity Violation

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20-1

15"-1

 

7.

6.

.1

 

3.

2.

I.

QUANTI doc

Figure 16 011

Histogram of Deceptiveness Ratings by Nunch’i Quantity Violation

75

APPENDIX C

Scenario

76

Instructions: Please read the hypothetical situation below carefully

Mike has known Prof. Daniel Parker for nearly two years, and he has already
taken two classes taught by Prof. Parker. Because he received excellent grades in those
two classes, he has been building a good relationship with the professor.

On Monday, Prof. Parker emailed Mike and asked him if he could help the
professor at an upcoming conference in Chicago this weekend. Mike was very glad
because he had wanted to make a good impression so that he could get a strong
recommendation letter from Prof. Parker later and it would be a good networking
opportunity. So, Mike replied right away and he was happy to help. The Prof. Parker told
Mike that he would send Mike a conﬁrmation email on Friday because his help might not
be necessary.

On Wednesday, one of Mike’s friends told him that they were going to Niagara
Falls this weekends, and the girl that Mike has liked secretly would join the trip. Despite
pre-appointrnent with Prof. Parker, he promised to join the trip to Niagara Falls, and just
hoped the professor would not email him. On Friday, Mike checked his mail nearly every
hour in the morning and afternoon. Unfortunately, the professor sent him the email
Friday evening. Because Mike wanted to go to the trip with the girl he liked anyway, he
just ignored the mail and pretended as if he had not checked the mail. He went to Niagara
Falls, and had a great time.

On the following Monday, Mike went to the student lounge and ran into Prof.

Parker. Prof. Parker asked Mike, “Did you get my email on Friday?”

77

The following messages are possible replies to the professor ’s question. Please
evaluate each of Mike ’s answers below.

Message: Mike says, (Message Example inserted here).

78

APPENDIX D

Message Examples

79

Baseline (Diselosive) message

“I got the email you sent last Friday. But, I ignored it. I am really sorry for that.
To tell you the truth, I went to Niagara Falls with ﬁiends, including the girl I like. The
reason I ignored your mail was that I wanted to go to the trip with ﬁiends and at the same
time did not want to disappoint you.”
Quality Violation
Example 1(Covert violation message): “No, I didn’t. I couldn’t check my email on
Friday because last Friday, my grandmother was sick that I had to take her to the
emergency room. I was there all day and I had no chance to check my email. Sorry.”
Example 2 (Nunch ’i message): “Oops! I forgot to check. I wasn’t feeling well.”
Quantity Violation
Example] (Covert violation message): “I checked my email Friday afternoon. There
wasn’t a message.”
Example 2 (Nunch ’i message): “ .......... (Awkward silence)”
Relevance Violation
Example 1(Covert violation message): “Oh, I am glad I ran into you. I have been
meaning to set up an advising appointment with you.”
Example 2 (Nunch ’i message): “Oh, Prof. Parker, I’ve heard you have been so busy
these days. Is there anything else I can help you?”
Manner Violation
Example 1(Covert violation message): “I didn’t see your message until it was too late.”

Example 2 (Nunch ’1' message): “Well, I usually check it, though”

80

APPENDD( E

Deceptiveness Scale

81

Now, we would like to rate Mike ’3 message on the several scales. Circle a number ﬁ'om

I to 7 that represents your opinions about the message. Please answer each and every

scale.

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29
30

31

What Mike said was:

Uninforrnative 1

Incomplete
Nondisclosive
Concealing
Distorted
Altered
Fabricated
False
Irrelevant
Inappropriate
Nonapplicable
Irnpertinent
Ambiguous
Indeﬁnite
Vague
Obscure
Dishonest
Deceitful
Deceptive

Misleading

1

82

Informative
Complete
Disclosive
Revealing
Accurate
Authentic
Genuine
True
Relevant
Appropriate
Applicable
Pertinent
Clear
Deﬁnite
Precise
Straightforward
Honest

Truthful
Not Deceptive

Not Misleading

Was Mike’s reply honest, or was Mike’s reply deceptive. Please check one:

What Mike said was: Honest Deceptive

83

REFERENCES

84

REFERENCES

Au, K. H. (1993) Cultural dimension of discourse. Childhood Education, 67, 270-284.

Aune, R. K. (1998) A theory of attribution of responsibility for creating
understanding. Annual meeting of the International communication Association,
Jerusalem, Israel.

Aune, R, K., & Waters, L. L. (1994). Cultural difference in deception: Motivation to
deceive in Samoans and North Americans. International Journal of Intercultural
Relations, 18, 159-172.

Aune, R. K., Levine, T. R., Park, H. S., Asada, K. J. K., & Banas, J. A. (2005). Tests of a
theory of communicative responsibility. Journal of Language and Social
Psychology, 24, 358-381.

Bavelas, J. 3., Black, A., Chovil, N., & Mullet, J. (1990). Truths, Lies, and
Equivocations: The effects of conﬂicting goals on disclosure. Journal of
Language and Social Psychology, 9, 135-161 .

Bradac, J ., Fridman, E., & Giles, H. (1986). A social approach to propositional

communication: Speakers lie to hearers. In G. McGregor (Ed.) Language for
hearers (pp. 127-151. Oxford: Pergamon Press.)

Brown, R, & Levinson, S. (1987). Politeness: Some universals in language usage.
Cambridge, UK: Cambridge University Press.

Bond, C. F., & The Global Deception Research Team (2004). A world of deception.
Unpublished manuscript. Texas Christian University.

Bond, C. F., & Atoum, A. O. (2000). International Deception. Personality and Social
Psychology, 26, 385-395.

Bond, C. F ., Omar, A., Mahmoud, A., & Bonser, R. N. (1990). Cross-cultural nonverbal
deception cues. Journal of Nonverbal Behavior, 14, 189-204.

Bond, C. F ., DePaulo, B. M. (2006). Accuracy of deception judgments. Personality and
Social Psychology Review, 10, 214-234.

Buller, D. B., & Burgoon, J. K. (1996) Interpersonal deception theory. Communication
Theory, 6, 203-242

Choi, S. H., & Choi, S. C. (1991). Nunch ’i: An indigenous form of Koreans’ politeness
communication. J. Irn (Ed.) Korean Communication (pp. 127-151. Young: Seoul).

85

Chunsheng, B. (2002). Cultural impact on perceived acceptability of deception:
Unpublished Ph. D. thesis. Tucson, AZ: University of Arizona.

Cole, T. (2001). Lying to the one you love: The use of deception in romantic
relationships. Journal of Social and Personal Relationships, 18, 107-129.

DePaulo, B. M., & Bell, K. L. (1996). Truth and Investment: Lies are told to those who
care. Journal of Personality and Social Psychology, 71, 703 -71 6.

DePaulo, B. M., Kashy, D. A., Kirkendol, M. M., & Epstein, J. A. (1996). Lying in
everyday life. Journal of Personality and Social Psychology, 70, 979-995.

DePaulo, B. M., Lindsay, J. J ., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper,
H. (2003). Cues to deception. Psychological Bulletin, 129, 74-118.

Ekman, P. (1985).Telling lies: Clues to deceit in the marketplace, politics, and marriage
. New York: Norton.

Gilbert, D. T. (1991). How mental systems believe. American Psychologist, 46, 107-119.

Grice, HP. (1989). Logic and conversation (Chapter 1): Studies in the way of words.
Harvard.

Gudykunst, W. B., Matsumoto, Y, Ting-Toomey, S., Nishida, T., Kim, K., & Heyman, S.
(1996), Human Communication Research, 22, 510-543.

Gudykunst, W. B., & Ting-Toomey, S. (with Chua, E.). (1988). Culture and
interpersonal communication. Newbury Park, CA: Sage.

Hall, E. T. (1976). Beyond culture. New York: Doubleday

Hara, K., & Kim, M. S. (2004). The effect of self-construals on conversational
indirectness. International Journal of Intercultural relations, 28, 1-12

Hofstede, G. (2001 ). Culture ’s consequences: Comparing values, behaviors, institutions,
and organizations across cultures. (2"d ed.). Thousand Oaks, CA: Sage.

Holtgraves. T. (1997) Styles of language use: Individual and cultural variety in
conversational indirectness. Journal of Personality and Social Psychology, 73,
624-63 7.

Ide, R. (1998). ‘Sorry for your kindness’: Japanese interactional ritual in public discourse.
Journal of Pragmatics, 29, 509-529.

86

Iizuka, Y., Patterson, M. L., & Matchen, J. C. (2002). Accuracy and conﬁdence on the
interpersonal perception task: a Japanese and American comparison. Journal of
Nonverbal Behavior, 26, 1 5 1-165 .

Jacobs, 8., Dawson, E. J ., & Brashers, D. (1996a). Information manipulation theory: A
replication and assessment. Communication Monographs, 63, 70-82.

Jacobs, 8., Brashers, D., & Dawson, E. J. (1996b). Truth and deception. Communication
Monographs, 63, 70-82.

Jang, S. A., Smith, S. W., & Levine, T. R. (2002). To stay or to leave? The role of
attachment styles in communication patterns and potential termination of

romantic relationships following discovery of deception. Communication
Monographs, 69, 236-252.

Kam, K. Y. (2004). A cultural model of nonverbal deceptive communication: the
independent and interdependent self-construals as predictors of deceptive
communication motivation and nonverbal behaviors under deception.
Unpublished doctorial dissertation, Tucson, AZ: University of Arizona

Kim, M. S. (1994). Cross-cultural comparisons of the perceived importance of
conversational constraints. Human Communication Research, 21, 128-151.

Kim, M. S., & Wilson, S. R. (1994) A cross-cultural comparison of implicit theories of
requesting. Communication Monographs, 61, 210-235.

Kim, M. S., Kam, K, Y., Singelis, T., & Aune, R. K. (1999). A cultural model of
deceptive communication motivation. Human communication, 2, 55-74.

Koyama, T., & Jacobs, C. S. (2001). Universals and diversences in perceived politeness.
Paper presented at the annual meeting of the International communication
association, Washington, DC.

Koyama, T. (2001). Universals in perceived politeness: comparison of native and non-
native speakers of English. Unpublished master’s thesis, University of Arizona,
Tucson, AZ.

Lalwani, A. K., Shavitt. S., & Johnson, T. What is the relationship between cultural
orientation and socially desirable responding? (2006). Journal of Personality and
Social Psychology, 90, 165-178.

Lapinski, M. K. (1995). Deception and the self: A cultural examination of Information

Manipulation Theory. Unpublished master’s thesis. Honolulu, HI: University of
Hawaii.

87

Lee, K., Cameron, C. A., Xu, F ., Fu, G., & Board, J. (1997). Chinese and Canadian
children’s evaluation of lying and truth telling: Seminaries and differences in the
context in the pro- and antisocial behaviors. Child Development, 68, 924-934.

Lebra, T. S. (1976). Japanese patterns of behavior. Honolulu: University of Hawaii Press.
Lewis, M., & Saarni, C. (1993). Lying and deception in everyday life: New York:
The Guilford Press.

Levine, T. R. (1998). Modeling the Psychometric Properties of Information Manipulation
Ratings. Communication Research Reports, 15, 218-225.

Levine, T. R., Lapinski, M. K., Banas, J., Wong, N. C. H., Hu, A. D. S., Endo, K., Baum,
K. L., & Anders, L. N. (2002). Self-construal, self and other beneﬁt, and the

generation of deceptive messages, Journal of Intercultural Communication
Research, 31, 20-47

Levine, T. R (2001). Dichotomous and continuous views of deception: A reexamination
of deception ratings in information manipulation theory. Communication
Research Reports, 1 8, 230-240.

Levine, T. R., Park, H. S., & McComack, S. A. (1999). Accuracy in detecting truths and
lies: Documenting the “veracity eﬂ‘ects”. Communication Monographs, 66, 125-
14-4.

Levine, T. R, Park. H. S., & Asada, K. J. (2005). The lying chicken and the ﬁdgeting
egg: Eye contact, deception, and causal order. Paper presented at the annual
meeting of the International Communication Association, 2005, New York, NY,

Levine, T. R., Feeley, T. H., McComack, S. A., Hughes, M. & Harms, C. M. (2005).
Testing the effects of nonverbal behavior training on accuracy in deception
detection with the inclusion of a bogus training control group. Western Journal of
Communication, 69, 203-217.

Levine, T. R., Kim, R. K., Park. H. S., & Hughes, M. (2006). Deception detection
accuracy is a predictable linear function of message veracity base-rate: A formal
test of Park and Levine’s Probability Model. Communication Monographs, 73,
243-260.

Levine, T. R, Asada, K. J ., & Linsay, L. L. M. (2003). The relative impact of violation
type and lie severity on judgments of message deceitfulness, Communication
Research Reports, 20, 208-218.

Levine, T. R., Anders, L. N., Banas, J ., Baum, K. L., Endo, K., Hu, A. D. S., & Norman,

C. H.. W. (2000). Norms, expectations and deception: A norm violation model of
veracity judgments. Communication Monographs, 6 7, 123-137.

88

Martin, E. M., Lee, Y. H., & Chang, S. U. (1967). A Korean, English Dictionary. New
Haven, CT: Yale University Press

McComack, S. A. (1992). Information manipulation theory. Communication
Monographs, 59, l- 16.

McComack, S. A., Levine, T. R., Torres, H. 1., Solowczuk, K. A., & Campbell, D. M.
(1992). When the alteration of information is viewed as deception: An empirical
test of information manipulation theory. Communication Monographs, 59, 17-29.

McComack, S. A., Levine, T. R., Morrison, K., & Lapinski, M. K. (1996). Speaking of
information manipulation: A critical rejoinder. Communication Monographs, 63,
83-91.

McComack, S. A. (1997). The generation of deceptive messages: Laying the groundwork
for a viable theory of interpersonal deception. In J. O. Greene (Ed.). Messages
Production. Mahwah, NJ: LEA.

McComack, S. A., & Parks, M. R. (1986) Deception detection and relationship
development: the other side of trust. Communication Monographs, 5 7, 219-230.

McComack, S. A., & Levine, T. R. (1990). When lies are uncovered: Emotional and
relational outcomes of discovered deception. Communication Monographs, 5 7,
1 1 9-1 3 8 .

Miller, G.R., & Stiff, J .B. (1993). Deceptive communication. Newbury Park, CA: Sage,
Chapt. 4.

Murai, J. (1998). Perceived deceptiveness of verbal messages: An examination of
Information Manipulation Theory. The Japanese Journal of Psychology, 69, 401-
407.

Nixon, Y. & Bull, P. (2006). Cultural communication styles and accuracy in cross-
cultural perception: A British and Japanese study. Journal of Intercultural
Communication, 12, 178-210.

Oetzel, J. G., & Ting-Toomey, S. (2003). Face concerns in interpersonal conﬂict: A
cross-cultural empirical test of the face-negotiation theory, Communication
Research, 30, 599-624

O’hair, D., Cody, M.. J., Wang, X. T., & Chao, E. Y. (1990). Vocal stress and deceptions
detection among Chinese. Communication Quarterly 38, 158-169.

Okabe, R. (1983). Cultural assumptions of East and West: Japan and the United States. In

W. Gudykunst (Ed.), Intercultural communication theory: Current perspectives
(pp. 21-44). Beverly Hills, CA: Sage.

89

Park, H. S., Levine, T. R., McComack, S. A., Morrison, K., & Ferrerra, M. (2002). How
people really detect lies. Communication Monographs, 69, 144-157.

Park, H. S., & Levine, T. R. (2001). A probability model of accuracy in deception
detection experiments, Communication Monographs, 68, 201 -21 0.

Park, K. S. (1997) Han-KuK-in-oi Communication: nunch ’i, che-myoun and Ki-pun,
(Communication of Koreans: nunch ’i, face, and mood), The paper presented in
103rd Korean Psychology Conference, 1997, Seoul. Korea.

Peskin, J. (1992). Ruse and representation: On children's ability to conceal information.
Developmental Psychology, 28, 84-89.

Sanchez-Burks, J., Lee, R, Choi, 1., Nisbett, R, Zhao, s., & Koo. J. (2003). Conversing
across cultures: East-West communication style in work and non-work contexts.
Journal of Personality and Social Psychology, 85, 363 -3 72.

Stiff, J. B., & Miller, G. R0986) “Come to think of it. . .”: Interrogative probes,
deceptive communication, and deception detection. Human Communication
Research, 12, 339-357.

Seiter, J. S., Bruschke, J ., and Chunsheng, B. (2002). The Acceptability of deception as a
function of perceivers' culture, deceiver's intention, and deceiver-deceived
relationship. Western Journal of Communication, 66, 158-180.

Ting-Toomey, S. (1988), Intercultural conﬂict styles: A face negotiation theory. In Y. Y.
Kim & W. B. Gudykunst (Eds), Theories in Intercultural communication (pp.
213-238). Newbury Park, CA: Sage.

Ting-Toomey, S., Oetzel, J ., & Y-J, Kim, (2001). Self-construal types and conﬂict
management styles. Communication Reports, 14, 87-104.

Turner, R. E., Edgley, C., & Olmstead, G. (1975). Information control in conversations:
Honesty is not always the best policy. Kansas Journal of Speech, 1 I, 69-89.

Yeung, L. N. T., Levine, T. R., & Nishiyama, K. (1999) Information manipulation theory
and perceptions of deception in Hong Kong. Communication Reports, 12, 1-11.

Virj, A. (2000). Detecting lies and deceit. NY: Wiley. Chapter 1.

90

..............

   

innillimiginm‘glji[mini