Skip to main content
NCBI home page
Psychiatry, Psychology, and Law logoLink to Psychiatry, Psychology, and Law
. 2022 Feb 20;30(2):177–191. doi: 10.1080/13218719.2021.2003269

Differences between Japanese and British participants in self-reported verbal strategies to appear convincing

Naoya Tabata a,✉, Aldert Vrij b
PMCID: PMC10026762  PMID: 36950192

Abstract

We compared the self-reported verbal strategies employed to appear convincing when lying and truth telling from 101 British (a low-context culture) and 149 Japanese (a high-context culture) participants. They completed a web-based survey and rated the degree to which they would use 16 verbal strategies when telling the truth and lying. British participants were more concerned with providing innocent reasons and avoiding/denying incriminating evidence when lying than when truth telling (no veracity effect emerged for Japanese participants). Japanese participants were less concerned with avoiding hesitations and lack of consistency when lying than when truth telling (no veracity effect emerged for British participants). The findings suggest that it is important to examine whether interview protocols developed to determine veracity in low-context cultures, such as the Strategic Use of Evidence and Cognitive Credibility Assessment, are equally effective in high-context cultures.

Key words: Bayes factors, deception, Grice’s cooperative principle, high-context culture, low-context culture, veracity, verbal strategies

It is generally believed that lie tellers leak cues to deception. Much research has been conducted to examine both verbal and nonverbal cues to deception (e.g. DePaulo et al., 2003; Vrij, 2008). At present, verbal cues are considered to be more promising than nonverbal cues (e.g. Vrij et al., 2019). For example, a comprehensive meta-analysis of cues to deception concluded that verbal cues are more revealing indicators of deception than nonverbal cues (DePaulo et al., 2003). Several experiments have further shown that verbal cues to deception can be elicited or enhanced when specific interview techniques are used (e.g. Granhag & Hartwig, 2015; Leal et al., 2015; Vrij et al., 2017).

Verbal cues to deception research have mostly been conducted in the so-called WEIRD cultural groups (Western, educated, industrialised, rich and democratic; Gerlach et al., 2019). Recently, there has been increased interest in possible cultural differences in verbal cues to deception (e.g. Taylor et al., 2015). Examining cultural differences in verbal cues to deception provides insight whether the findings obtained in WEIRD cultures can be generalised to other cultures.

There is evidence to suggest that there are likely to be some cultural differences in verbal cues to deception (Leal et al., 2018; Taylor et al., 2017; Vrij, Leal, et al., 2020). For example, Vrij, Leal, et al. (2020) compared verbal communication of British and Arab truth tellers and lie tellers in interviews. British participants provided more details and more complications than Arab participants. In both cultural groups, truth tellers provided more details than lie tellers, but the effect was most pronounced in British participants. Only in the British sample did truth tellers and lie tellers differ in reporting complications: truth tellers reported more complications than lie tellers. Complications are occurrences that make the story more complex (‘We did not see him at first, because he was waiting at a different entrance’).

Low-context versus high-context cultures

Different reasons for the existence of cultural differences in verbal communication when lying have been suggested (Vrij, Leal, et al., 2020). This includes differences in communication style defined as the way people communicate with others (Hall, 1976; Liu, 2016). Cultures can be categorised by whether people rely more heavily on language or on context in communication (Liu, 2016). Low-context cultures use a communication style that relies heavily on language. When communicating, those in low-context cultures provide more information to make the message understood than those in high-context cultures. In contrast, high-context cultures use a communication style that relies heavily on context. It is more essential for those in high-context cultures to read the atmosphere when communicating than for those in low-context cultures. This suggests that in conversations, speakers from low-context cultures rely more heavily on language to avoid obscurity and ambiguity and are thus more likely to provide more details than speakers from high-context cultures. In alignment with this, British participants (low-context culture) reported more details than Arab participants and Chinese participants (both high-context culture) in Leal et al. (2018).

A good example of cultural differences in communication style is the use of personal pronouns (e.g. I, we). British speakers (low-context culture) almost always use personal pronouns, whereas Japanese speakers (high-context culture) almost never use them. The Japanese are not an exception. So-called personal pronoun drop occurs in many languages (Y. Kashima et al., 2014). Kashima and Kashima (E. S. Kashima and Kashima, 1998, 2005; Y. Kashima and Kashima, 2003) found a correlation between personal pronoun drop and belonging to a collectivistic culture. There is an overlap between collectivistic cultures and high-context cultures (Liu, 2016).

Cultural differences in communication style could produce cultural differences in verbal cues to deceit. People infer deceptiveness when the actions of the communicator violate what the receiver expects as normative behaviour (Bond et al., 1992; Levine et al., 2000). To appear convincing to the people around them, lie tellers need to adjust the construction of the statements to the norms and expectations of the culture in which the communication is being made. If these norms and expectations culturally differ, lie tellers in different cultures are likely to use different strategies to appear convincing (e.g. Gudykunst, 1997). Therefore, culturally defined cues to deceit may exist, and lies may sound different in different cultures.

In the present study, we focused on differences between British and Japanese participants in the self-reported verbal strategies used when lying in order to appear convincing in conversations. In their meta-analysis, DePaulo et al. stressed the importance on measuring strategies amongst truth tellers and lie tellers because it could explain differences in responses between them (DePaulo et al., 2003). Because the verbal cues to deception depend on the strategy of what one intends to say, focusing on self-reported verbal strategies makes it possible to discover what is actually going on in a lie teller’s mind (Vrij et al., 2010) and on which verbal cues to focus on in future cross-cultural studies. Strategies research was absent at the time of the DePaulo meta-analysis but has emerged since (e.g. Colwell et al., 2006; Hartwig et al., 2007, 2010). Strategies research has indeed been successful in explaining verbal veracity differences. For example, it showed that truth tellers want to ‘tell it all’ whereas lie tellers prefer to ‘keep things simple’ (Hartwig et al., 2007). This is in alignment with truth tellers (vs. lie tellers) typically including more details (Amado et al., 2016) and more complications (Vrij, Palena, et al., 2021) in their statements.

Verbal strategies to appear convincing

We examined various verbal strategies to appear convincing in conversations derived from two perspectives. First, we chose verbal strategies based on Grice’s cooperative principle (Grice, 1975). It posits four maxims that participants in a conversation are expected to abide by: quantity, quality, relevance and manner. According to the maxim of quantity, people should express the amount of information that is expected. According to the maxim of quality, people should not express what they think is not true. According to the maxim of relevance, people should keep to the point. According to the maxim of manner, people should avoid obscurity and ambiguity. Statements that violate these maxims are likely to be regarded as deceptive (McCornack, 1992; McCornack et al., 1992).

Lie tellers who intend to demonstrate explicitly that their statements are convincing should obey Grice’s cooperative principle. This tendency should be more readily observed in people belonging to low-context cultures, for whom language dominates when they communicate with others. Indeed, a study with Japanese participants showed that scenarios that violate the Grice cooperative principle (which is developed in a low-context culture) did not necessarily make a deceptive impression on them (Murai, 1998). There is evidence that the Grice maxims are not universally applied (Vrij, Leal, et al., 2020). Research has shown that they are often disregarded by speakers belonging to high-context cultures, including Arabic (Al-Qaderi, 2015), Chinese (He, 2012) and Indonesian (Herawati, 2013) speakers.

In terms of Grice’s cooperative principle, we thus examined the following two hypotheses: British participants will endorse Grice’s cooperative principle more strongly than Japanese participants (Hypothesis 1). If British participants find adhering to Grice’s cooperative principle more important than Japanese participants, differences between truth telling and lying regarding endorsing the Grice cooperative principle will be smaller in British participants than in Japanese participants (Hypothesis 2).

Second, we chose additional verbal strategies based on several previous studies of strategies used by lie tellers and truth tellers (Hartwig et al., 2007, 2010; Hines et al., 2010; Strömwall et al., 2006; Strömwall & Willén, 2011). In these studies, carried out in Sweden and the United States, both low-context cultures, participants' self-reported descriptions of the strategies that they used in order to be convincing during mock interrogations were coded, and data-driven categories of strategies were established. We prepared a list of 12 verbal strategies: see Table 1, other verbal strategies. Our lack of knowledge about cultural differences in verbal strategies makes it impossible to make predictions for each of the 12 strategies. Instead, we formulated general hypotheses based on the concept that for British participants (low-context culture) language dominates more when they communicate with others than for Japanese participants (high-context culture). British participants will endorse these strategies more than Japanese participants (Hypothesis 3). If British participants find adhering to these strategies more important than Japanese participants, differences between truth telling and lying regarding these verbal strategies will be smaller in British participants than in Japanese participants (Hypothesis 4).

Table 1.

Verbal strategies.

Variable Item
Grice strategies  
Maxim of quantity I try to be informative, but not more than required.
Maxim of quality I try to give only information that is reliable and I remember.
Maxim of relevance I try to keep to the point.
Maxim of manner I try to avoid obscurity and ambiguity and, instead, I try to be brief and orderly.
Other verbal strategies  
Rich in detail I try to give as much detail as possible about what has happened.
No hesitation I try to appear decisive.
Close to truth I try to stay close to the truth.
Plausibility I try to give a statement that sounds plausible (that sounds as if it really could have happened).
Consistent story I try to stick with a story and don’t change elements within it.
Unrehearsed story I try to make the story sound spontaneous.
Provide innocent reason I try to provide an innocent reason for an activity
Deny/avoid incriminating details I try to avoid reporting incriminating details, but go into more detail about innocent elements of the story.
Complete detail I try to tell as much detail that I could recall.
Minimal detail I try to say as little as possible so that if the story needed to be repeated there would be less room for error.
Coherent and consistent I try to explain everything the same way even if asked the same question again.
Emotions I try to explain what I was feeling.
Open in a new tab

Method

Participants

A total of 101 British participants (34 men, 67 women) and 149 Japanese participants (49 men, 100 women) voluntarily took part in this study; their average age was 20.00 years (SD = 4.28) and 20.30 years (SD = 0.97), respectively. The British participants, who identified themselves as British citizens, were recruited at British universities. The Japanese participants, who identified themselves as Japanese citizens, were recruited at Japanese universities. There were no significant differences between the British and the Japanese participants in the distributions of gender, χ2(1) = 0.02, p = .898, V = .01, and age, t(106.99) = 0.70, p = .488, d = 0.11 (–0.15, 0.36) in Welch's t test.

Design

This study used a mixed design with ethnic group (British vs. Japanese) as a between-participants factor and veracity (truth vs. lie) as a within-participants factor. We present effect sizes d and Bayes factors (BFs). A d score of 0.20 represents a small effect, a d score of 0.50 a medium effect and a d score of 0.80 a large effect (Cohen, 1988). BFs between 1 and 3 suggest weak evidence, BFs between 3 and 10 suggest strong evidence, and BFs >10 suggest very strong evidence for the alternative hypothesis (Jeffreys, 1961). BFs between 1.00 and 0.33 suggest weak evidence, BFs between 0.33 and 0.10 suggest strong evidence, and BFs <0.10 suggest very strong evidence for the null hypothesis. We used the default Cauchy’s prior of .707 for the Bayesian t tests (Lakens, 2016).

Procedure

Participants were recruited via online advertisements, student portals and flyers posted on campuses. They were invited to complete a web-based survey using Qualtrics. Those who agreed to participate were required to tick a box indicating their consent to participate in the study. Next, they were asked about their gender and age. Then they read the following instructions: ‘Below are possible ways to present oneself in conversations. Please indicate for each statement the extent to which you endorse it when you tell the truth and when you tell a lie.’ Of the 16 strategies, four were about the four maxims of Grice’s cooperative principle (Grice, 1975): ‘quantity’, ‘quality’, ‘relevance’ and ‘manner’. The 12 remaining strategies were taken from the deception literature that described strategies that truth tellers and lie tellers typically use. Two strategies were derived from Strömwall et al. (2006): ‘rich in detail’ and ‘no hesitation’. Two strategies were derived from Strömwall et al. (2011): ‘close to truth’ and ‘plausibility’. Two strategies were derived from Hartwig et al. (2007): ‘consistent story’ and ‘unrehearsed story’. Two strategies were derived from Hartwig et al. (2010): ‘provide innocent reason’ and ‘deny/avoid incriminating details’, and four strategies were derived from Hines et al. (2010): ‘complete detail’, ‘minimal detail’, ‘coherent and consistent’ and ‘emotions’. The order of the questions (telling the truth vs. lying) was randomly counterbalanced across each strategy. Responses to all questions were made on 10-point scales ranging from 1 (not at all) to 10 (entirely). The questionnaire was originally developed in English. The authors of this paper discussed the meaning of each item and then asked a native Japanese speaker to translate the questionnaire into Japanese. The translation was checked by the Japanese author of this paper. Table 1 provides the definitions of the strategies we gave to the participants.

Results

Grice’s cooperative principle

A 2 (ethnic group) × 2 (veracity) multivariate analysis of variance (MANOVA) with Grice’s cooperative principle (maxims of quantity, quality, relevance and manner) as dependent variables resulted in significant multivariate effects for ethnic group (Hotelling’s T2 = 0.12), F(4, 245) = 7.12, p < .001, ηp2 = .10), veracity (Hotelling’s T2 = 0.19), F(4, 245) = 11.70, p < .001, ηp2 = .16, and Ethnic Group × Veracity (Hotelling’s T2 = 0.12), F(4, 245) = 7.12, p < .001, ηp2 = .10. Tables 2, 3 and 4 provide the univariate statistics.

Table 2.

Verbal strategies as a function of ethnic group.

  British
 Japanese
 F p d BF10
M SD 95% CI M SD 95% CI
Maxim of quantity 6.75 1.61 [6.43, 7.07] 6.05 1.78 [5.76, 6.34] 9.96 .002 0.41 (0.15, 0.66) 6.71
Maxim of quality 6.97 1.52 [6.67, 7.27] 6.11 1.81 [5.81, 6.40] 15.50 <.001 0.51 (0.25, 0.76) >100
Maxim of relevance 6.97 1.68 [6.64, 7.30] 6.11 1.91 [5.80, 6.42] 13.40 <.001 0.47 (0.21, 0.73) 56.59
Close to truth 7.75 1.79 [7.39, 8.10] 7.28 1.66 [7.01, 7.55] 4.49 .035 0.27 (0.02, 0.53) 0.94
Consistent story 7.63 1.67 [7.30, 7.96] 6.71 1.69 [6.44, 6.99] 18.02 <.001 0.55 (0.29, 0.80) >100
Provide innocent reason 6.55 1.86 [6.18, 6.92] 5.08 2.13 [4.73, 5.4] 31.71 <.001 0.73 (0.46, 0.98) >100
Minimal detail 6.21 1.55 [5.91, 6.52] 5.13 1.74 [4.84, 5.41] 25.55 <.001 0.65 (0.39, 0.91) >100
Open in a new tab

Note: Only variables that resulted in significant effects are reported. CI = confidence interval; BF = Bayes factor.

Table 3.

Verbal strategies as a function of veracity.

  Truth
 Lie
 F p d BF10
M SD 95% CI M SD 95% CI
Maxim of quantity 5.62 2.58 [5.30, 5.95] 7.04 2.14 [6.77, 7.31] 39.82 <.001 0.60 (0.42, 0.78) >100
Rich in detail 7.74 1.79 [7.52, 7.97] 5.08 2.20 [4.80, 5.35] 192.84 <.001 1.33 (1.14, 1.52) >100
No hesitation 7.29 2.09 [7.03, 7.55] 6.51 2.30 [6.22, 6.71] 16.30 <.001 0.35 (0.18, 0.53) >100
Close to truth 8.22 1.86 [7.98, 8.45] 6.72 2.45 [6.41, 7.03] 78.31 <.001 0.69 (0.51, 0.87) >100
Plausibility 7.56 2.24 [7.29, 7.84] 7.94 1.78 [7.72, 8.17] 7.99 .005 0.19 (0.01, 0.36) 2.76
Consistent story 7.64 2.10 [7.38, 7.90] 6.53 2.33 [6.24, 6.82] 30.33 <.001 0.50 (0.32, 0.68) >100
Unrehearsed story 5.17 2.45 [4.86, 5.47] 5.81 2.50 [5.50, 6.12] 8.63 .004 0.26 (0.08, 0.43) 9.94
Provide innocent reason 5.40 2.69 [5.06, 5.74] 5.94 2.62 [5.62, 6.27] 10.49 .001 0.20 (0.03, 0.38) 4.52
Deny/avoid incriminating details 6.82 2.21 [6.55, 7.10] 7.12 2.00 [6.87, 7.37] 4.77 .030 0.14 (–0.04, 0.32) 0.50
Complete detail 7.71 2.01 [7.46, 7.96] 5.91 2.52 [5.59, 6.22] 92.74 <.001 0.79 (0.61, 0.97) >100
Minimal detail 4.29 2.44 [3.99, 4.60] 6.84 2.20 [6.57, 7.11] 170.92 <.001 1.10 (0.91, 1.28) >100
Coherent and consistent 6.47 2.57 [6.15, 6.78] 6.84 2.38 [6.54, 7.14] 5.32 .022 0.15 (–0.03, 0.32) 1.07
Emotions 7.22 2.05 [6.97, 7.48] 5.44 2.32 [5.16, 5.73] 89.58 <.001 0.81 (0.63, 0.99) >100
Open in a new tab

Note: Only variables that resulted in significant effects are reported. CI = confidence interval; BF = Bayes factor.

Table 4.

Verbal strategies as a function of ethnic group and veracity.

  Truth
 Lie
 F p d BF10
M SD 95% CI M SD 95% CI
Maxim of quantity  
 British 6.50 2.27 [6.05, 6.94] 7.00 1.91 [6.62, 7.38] 3.58 .061 0.24 (0.06, 0.41) 0.87
 Japanese 5.03 2.62 [4.61, 5.46] 7.07 2.29 [6.70, 7.44] 53.74 <.001 0.83 (0.64, 1.01) >100
Maxim of manner  
 British 6.00 2.56 [5.49, 6.51] 6.44 2.39 [5.96, 6.91] 1.34 .250 0.18 (0.00, 0.35) 0.35
 Japanese 6.79 2.11 [6.45, 7.13] 5.76 2.47 [5.36, 6.16] 14.55 <.001 0.45 (0.27, 0.63) >100
Rich in detail  
 British 7.45 2.03 [7.05, 7.85] 5.60 2.13 [5.18, 6.02] 31.61 <.001 0.89 (0.70, 1.07) >100
 Japanese 7.95 1.58 [7.69, 8.20] 4.72 2.18 [4.37, 5.07] 253.58 <.001 1.70 (1.49, 1.90) >100
No hesitation  
 British 6.70 2.22 [6.27, 7.14] 6.60 2.21 [6.17, 7.04] 0.20 .656 0.05 (–0.13, 0.22) 0.17
 Japanese 7.68 1.91 [7.37, 7.99] 6.44 2.36 [6.06, 6.83] 29.53 <.001 0.58 (0.40, 0.76) >100
Consistent story  
 British 7.53 2.31 [7.08, 7.99] 7.73 1.96 [7.34, 8.12] 0.55 .459 0.09 (–0.08, 0.27) 0.20
 Japanese 7.71 1.96 [7.39, 8.03] 5.71 2.10 [5.35, 6.07] 100.24 <.001 0.98 (0.80, 1.17) >100
Provide innocent reason  
 British 5.97 2.45 [5.49, 6.45] 7.13 2.07 [6.72, 7.54] 20.37 <.001 0.51 (0.33, 0.69) >100
 Japanese 5.01 2.79 [4.56, 5.47] 5.14 2.65 [4.71, 5.57] 0.21 .645 0.05 (–0.13, 0.22) 0.14
Deny/avoid incriminating details  
 British 6.69 2.29 [6.24, 7.15] 7.38 1.95 [6.99, 7.76] 6.85 .010 0.32 (0.15, 0.50) 3.91
 Japanese 6.91 2.16 [6.56, 7.26] 6.95 2.03 [6.62, 7.27] 0.03 .870 0.02 (–0.16, 0.19) 0.13
Complete detail  
 British 7.50 2.22 [7.07, 7.94] 6.47 2.40 [5.99, 6.94] 13.35 <.001 0.45 (0.27, 0.62) 77.89
 Japanese 7.85 1.85 [7.55, 8.14] 5.53 2.54 [5.12, 5.94] 118.92 <.001 1.04 (0.86, 1.23) >100
Open in a new tab

Note: Only variables that resulted in significant effects are reported. CI = confidence interval; BF = Bayes factor.

At the univariate level, three ethnic group effects were significant. The univariate results are presented in Table 2. British participants endorsed the maxims of quantity, quality and relevance significantly more strongly than Japanese participants. The effects were strong for the maxim of quantity and very strong for the maxims of quality and relevance. There was strong evidence for the null hypothesis regarding the maxim of manner. This supports Hypothesis 1 for the maxim of quantity, quality and relevance, but the hypothesis was not supported for manner.

At a univariate level, one significant main effect for veracity emerged; see Table 3. Participants endorsed the maxim of quantity more strongly when lying than when telling the truth. The evidence for the maxim of quantity was very strong.

At a univariate level significant univariate Ethnic Group × Veracity effects emerged for the maxims of quantity, F(1, 248) = 14.44, p < .001, ηp2 = .06, and manner, F(1, 248) = 10.57, p = .001, ηp2 = .04. The univariate results are presented in Table 4. Comparisons between lying and truth telling revealed that the maxims of quantity and manner distinguished telling the truth from lying only in Japanese participants. Japanese participants endorsed the maxim of quantity more strongly, F(1, 248) = 53.74, p < .001, d = 0.83 (0.64, 1.01), BF10 > 100, and the maxim of manner less strongly, F(1, 248) = 14.55, p < .001, d = 0.45 (0.27, 0.63), BF10 > 100, when lying than when telling the truth. Both effects showed very strong evidence for the alternative hypothesis. In contrast, in British participants, there were no significant differences between telling the truth and lying for the maxims of quantity, F(1, 248) = 3.58, p = .061, d = 0.24 (0.06, 0.41), BF10 = 0.87, and manner, F(1, 248) = 1.34, p = .250, d = 0.18 (0.00, 0.35), BF10 = 0.35. This shows that the differences between truth telling and lying were smaller in British participants than in Japanese participants, supporting Hypothesis 2.

Verbal strategies for deception

A 2 (ethnic group) × 2 (veracity) MANOVA with 12 verbal strategies for deception as dependent variables resulted in significant multivariate effects for ethnic group (Hotelling’s T2 = 0.31), F(12, 237) = 6.21, p < .001, ηp2 = .24, veracity (Hotelling’s T2 = 1.79), F(12, 237) = 35.38, p < .001, ηp2 = .64, and Ethnic Group × Veracity (Hotelling’s T2 = 0.37), F(12, 237) = 7.27, p < .001, ηp2 = .27. Tables 2, 3 and 4 provide the univariate statistics.

At the univariate level, four ethnic group effects were significant. The univariate results are presented in Table 2. British participants making stronger endorsements of the strategies of ‘close to truth’, ‘consistent story’, ‘provide innocent reason’ and ‘minimal detail’ than Japanese participants. The effects were very strong for ‘consistent story’, ‘provide innocent reason’ and ‘minimal detail’. The results of these three strategies support Hypothesis 3, whereas the hypothesis was not supported for the other strategies.

At a univariate level, all 12 strategies showed significant main effects for veracity; see Table 3. Participants endorsed ‘plausibility’, ‘unrehearsed story’, ‘provide innocent reason’, ‘deny/avoid incriminating details’, ‘minimal detail’ and ‘coherent and consistent’ strategies more strongly when lying than when telling the truth. They also endorsed ‘rich in detail’, ‘no hesitation’, ‘close to truth’, ‘consistent story’, ‘complete detail’ and ‘emotion’ strategies more strongly when telling the truth than when lying. Except for the ‘deny/avoid incriminating details’ strategy, which showed strong evidence for the null hypothesis, the evidence for these effects varied from strong to very strong.

At a univariate level, significant univariate Ethnic Group × Veracity effects emerged for the following six strategies: ‘rich in detail’, F(1, 248) = 14.43, p < .001, ηp2 = .05, ‘no hesitation’, F(1, 248) = 11.84, p < .001, ηp2 = .05, ‘consistent story’, F(1, 248) = 45.13, p < .001, ηp2 = .15, ‘provide innocent reason’, F(1, 248) = 6.74, p = .010, ηp2 = .03, ‘deny/avoid incriminating details’, F(1, 248) = 3.92, p = .049, ηp2 = .02, and ‘complete detail’, F(1, 248) = 13.41, p < .001, ηp2 = .05. The univariate results are presented in Table 4. Comparisons between lying and telling the truth revealed that British participants endorsed the strategies ‘provide innocent reason’, F(1, 100) = 20.37, p < .001, d = 0.51 (0.33, 0.69), BF10 > 100, and ‘deny/avoid incriminating details’, F(1, 100) = 6.85, p = .010, d = 0.32 (0.15, 0.50), BF10 = 3.91, more strongly and ‘rich in detail’, F(1, 100) = 31.61, p < .001, d = 0.89 (0.70, 1.07), BF10 > 100, and ‘complete detail’, F(1, 100) = 13.35, p < .001, d = 0.45 (0.27, 0.62), BF10 = 77.89, less strongly when lying than when telling the truth. The Bayes factors showed strong evidence for ‘deny/avoid incriminating details’ and very strong evidence for the other three variables. No significant differences were found for truth telling versus lying for ‘no hesitation’, F(1, 100) = 0.20, p = .656, d = 0.05 (–0.13, 0.22), BF10 = 0.17, and ‘consistent story’, F(1, 100) = 0.55, p = .459, d = 0.09 (–0.08, 0.27), BF10 = 0.20. Japanese participants made weaker endorsements of the following four strategies when lying than when telling the truth: ‘rich in detail’, F(1, 148) = 253.58, p < .001, d = 1.70 (1.49, 1.90), BF10 > 100, ‘no hesitation’, F(1, 148) = 29.53, p < .001, d = 0.58 (0.40, 0.76), BF10 > 100, ‘consistent story’, F(1, 148) = 100.24, p < .001, d = 0.98 (0.80, 1.17), BF10 > 100, and ‘complete detail’, F(1, 148) = 118.92, p < .001, d = 1.04 (0.86, 1.23), BF10 > 100. All four variables revealed very strong evidence for the alternative hypothesis. Japanese participants showed no significant differences between lying and truth telling for ‘provide innocent reason’, F(1, 148) = 0.21, p = .645, d = 0.05 (–0.13, 0.22), BF10 = 0.14, and ‘deny/avoid incriminating details’, F(1, 148) = 0.03, p = .870, d = 0.02 (–0.16, 0.19), BF10 = 0.13. In other words, four significant results emerged in both cultures, which means that Hypothesis 4, which predicted more differences in Japanese than in British participants, was not supported.

Discussion

In the present study, we examined differences in self-reported verbal strategies when telling the truth versus lying in a low-context culture (British) and high-context culture (Japanese). In low-context cultures it is essential to provide more information to make the message understood, whereas high-context cultures rely heavily on context when communicating. Table 5 shows the schematic summary of the significant main effects and interaction effects. We examined 16 strategies; four were derived from Grice’s cooperative principle and 12 from the verbal deception strategies literature. All 16 verbal deception strategies yielded significant effects.

Table 5.

Schematic summary of the significant main effects and interaction effects.

  Ethnic group Veracity Ethnic group × Veracity
Maxim of quantity British > Japanese T < L British: — Japanese: T < L
Maxim of quality British > Japanese      
Maxim of relevance British > Japanese      
Maxim of manner     British: — Japanese: T > L
Rich in detail   T > L British: T > L Japanese: T > L
No hesitation   T > L British: — Japanese: T > L
Close to truth British > Japanese T > L    
Plausibility   T < L    
Consistent story British > Japanese T > L British: — Japanese: T > L
Unrehearsed story   T < L    
Provide innocent reason British > Japanese T < L British: T < L Japanese: —
Deny/avoid incriminating details   T < L British: T < L Japanese: —
Complete detail   T > L British: T > L Japanese: T > L
Minimal detail British > Japanese T < L    
Coherent and consistent   T < L    
Emotions   T > L    
Open in a new tab

Note: T = truth; L = lie; > denotes significantly more than; < denotes significantly less than; — denotes non-significant difference (Ethnic Group × Veracity column only); blank cells indicate non-significant differences.

British participants tended to endorse the Grice cooperative principle more than Japanese participants. This replicates previous findings that Grice’s maxims are often disregarded by speakers belonging to high-context cultures (Al-Qaderi, 2015; He, 2012; Herawati, 2013). For the other verbal strategies, we also found stronger endorsements amongst British than Japanese participants. This is not surprising because these verbal strategies were derived from studies in low-context cultures. It shows the importance of carrying out this type of research in high-context cultures to explore whether it will reveal verbal strategies that are not prevalent in low-context cultures.

Veracity main effects occurred for only one of the four Grice cooperative principle items but for all 12 verbal strategies items. The strongest veracity effects for the latter 12 items reflect that these were based on previous deception research whereas the Grice cooperative principle is not. Significant veracity effects suggest that verbal cues to deception are likely to occur, reflecting research in this domain (Amado et al., 2016; DePaulo et al., 2003). Interestingly, these veracity findings do not always predict the direction of verbal veracity cues. For example, although lie tellers seem to strive more than truth tellers to be plausible (Table 5), research has shown that truth tellers typically sound more plausible than lie tellers (Vrij, Deeb, et al., 2021). Striving to achieve plausibility does not actually mean achieving it. Telling a plausible story is difficult to achieve (Leal et al., 2015), because the concept plausibility is a cluster of multiple verbal cues that need to be satisfactorily incorporated in a statement. That is, it takes into account the contextual information (e.g. whether the statement is conventional or reasonable given the situation), the quantity of details (e.g. total details) and the quality of these details (complications and verifiable sources; Vrij, Deeb, et al., 2021). In addition, conflicting findings emerged for the variables ‘consistent’ (higher during truth telling) and ‘coherent and consistent’ (higher during lying). This could be the result of participants misunderstanding the meaning of consistency. Confusion about this term is common. For example, researchers and laypersons appear to be in disagreement how to conceptualise consistency (Hudson et al., 2020).

Perhaps most relevant are the interaction effects because they shed light on possible cross-cultural differences in verbal indicators of veracity. Japanese participants revealed differences related to the Grice cooperative principles whereas British participants did not. This suggests that Grice-related verbal veracity cues may be more likely to occur when these principles are examined in Japanese speakers than in British speakers. In particular, Japanese lie tellers may provide more information than expected (maxim of quantity) or their statements may sound vaguer and more ambiguous (maxim of manner) than Japanese truth tellers’ statements.

For the verbal strategies, four differences emerged in both Japanese and British participants. In both groups, participants endorsed the strategies that a statement needs to be rich in detail and complete more when truth telling than when lying. This suggests that ‘total details’ will be a veracity cue in both low-context and high-context cultures, which indeed has been found in deception research. In both types of culture, truth tellers have been found to report more details than lie tellers (Leal et al., 2018; Vrij, Leal, et al., 2020).

In British participants, the strategies of providing innocent reasons and denying and/or avoiding incriminating evidence were more endorsed when lying than when truth telling whereas no differences emerged for these strategies in Japanese participants. This suggests that British participants put more effort in covering up their lies than Japanese participants. It also suggests that British and Japanese participants may respond differently in a Strategic Use of Evidence interview which has as core that lie tellers more than truth tellers try to deny or avoid incriminating evidence (Granhag & Hartwig, 2015). This requires cross-cultural Strategic Use of Evidence research.

In Japanese participants, the strategies not to hesitate and to be consistent were more endorsed when truth telling than when lying, whereas no differences emerged in British participants regarding these two strategies. This thus suggests that more hesitations and lack of consistency may be more likely to be indicators of deceit in Japanese than in British participants. The interview protocol Cognitive Credibility Assessment exploits the concept of consistency. In Cognitive Credibility Assessment the same question is asked repeatedly in different formats. Truth tellers are more likely than lie tellers to add new information at each stage, in part because lie tellers strive to be consistent (Vrij, Mann, et al., 2020). Providing new information makes a statement more inconsistent. If Japanese participants are less concerned with consistency when they lie than British participants, then Cognitive Credibility Assessment may be more effective in British than in Japanese participants or, perhaps more generally, in low-context than in high-context cultures. This is worth examining.

The results of our study demonstrated substantial cultural differences in self-reported verbal strategies when lying. There is growing attention to cultural differences in verbal veracity cues (e.g. Leal et al., 2018; Taylor et al., 2015; Vrij, Leal, et al., 2020), but cross-cultural research examining cultural differences in self-reported verbal strategies to sound convincing has been absent to date. Such studies are important because they could give insight into which verbal cues are relevant to examine in different cultures (DePaulo et al., 2003). They could also assist researchers in developing interview protocols that exploit these differences in specific cultures. Different cultures may well require different interview protocols to be effective, and researchers and practitioners should be careful with applying findings from low-context cultures to high-context cultures.

Several limitations of this study should be noted. First is the way we measured self-reported verbal strategies. We captured these strategies in a general way through asking participants how they would present themselves in conversations. An alternative method would have been by introducing an experiential basis of guilt and innocence, for example by exposing participants to some sort of event, which they are then to convey either truthfully or deceptively. We decided against this approach because it would mean tapping into strategies used in that specific event that may not be necessarily generalisable to other events.

Second, we measured self-reported verbal strategies rather than verbal responses. We agree with DePaulo et al. (2003) that too much research focuses on verbal behaviour and that strategies underlying these responses are largely neglected. We argued that measuring verbal strategies is an important first step because it gives insight into what is going on in the mind of truth tellers and lie tellers. We only measured self-reported verbal strategies. In other words, whilst in most deception research the strategies are neglected, we neglected the verbal responses. We suggested a link between self-reported verbal strategies and verbal behaviour but this needs to be empirically examined.

Third, the study did not clarify whether cultural differences in verbal strategies may be affected by communication style, because communication style was not measured directly, and other factors may have affected the results. More empirical evidence that directly focuses on effects of communication style on verbal strategies is warranted before reaching definitive conclusions about the influence of communication style. For example, to explain why Dutch interviewees (low-context culture) reported more details than Ghanaian interviewees (high-context culture), Anakwah et al. (2020) used the collectivistic–individualistic cultures distinction (Hofstede, 1983). It is related to the high-context (collectivistic)–low-context (individualistic) culture distinction (Liu, 2016). Anakwah et al. (2020) argued that, in conversations, parents in individualistic cultures provide more feedback to their children than parents in collectivistic cultures and that these differences in linguistic elaborations are transferred to children and continue in adulthood (Vrij, Leal, et al., 2020).

Fourth, our findings were exclusively based on data from British and Japanese people. According to Liu (2016), all cultures can be categorised as high or low context in terms of communication style. Thus, the findings of this study should be replicated with other samples to establish generalisability.

In summary, we argued that cultural differences in communication styles exist and that these differences may result in cultural differences in verbal cues to deceit. To start examining this we examined the self-reported verbal strategies of British participants (low-context culture) and Japanese participants (high-context culture). Cultural differences in verbal strategies emerged. The next step is to examine whether these differences in self-reported verbal strategies are reflected in differences in verbal responses. We hope that our research stimulates researchers to carry out such cross-cultural deception research.

Ethical standards

Declaration of conflicts of interest

Naoya Tabata has declared no conflicts of interest

Aldert Vrij has declared no conflicts of interest

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the Ethical Committee for Policy Studies Association in Aichi Gakuin University and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study

References

  1. Al-Qaderi, I. (2015). Conversational implicature in Arabic: A pragmatic analysis of applying flouting the Maxims to the Yemeni dialect. International Journal of Linguistics, 7(6), 53–68. 10.5296/ijl.v7i6.8745 [DOI] [Google Scholar]
  2. Amado, B. G., Arce, R., Fariña, F., & Vilarino, M. (2016). Criteria-Based Content Analysis (CBCA) reality criteria in adults: A meta-analytic review. International Journal of Clinical and Health Psychology, 16(2), 201–210. 10.1016/j.ijchp.2016.01.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Anakwah, N., Horselenberg, R., Hope, L., Amankwah‐Poku, M., & Koppen, P. J. (2020). Cross-cultural differences in eyewitness memory reports. Applied Cognitive Psychology, 34(2), 504–515. 10.1002/acp.3637 [DOI] [Google Scholar]
  4. Bond, C. F., Omar, A., Pitre, U., Lashley, B. R., Skaggs, L. M., & Kirk, C. T. (1992). Fishy-looking liars: Deception judgment from expectancy violation. Journal of Personality and Social Psychology, 63(6), 969–977. 10.1037//0022-3514.63.6.969 [DOI] [PubMed] [Google Scholar]
  5. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Erlbaum. [Google Scholar]
  6. Colwell, K., Hiscock-Anisman, C., Memon, A., Woods, D., & Michlik, P. M. (2006). Strategies of impression management among deceivers and truth tellers: How liars attempt to convince. American Journal of Forensic Psychology, 24, 31–38. [Google Scholar]
  7. DePaulo, B. M., Lindsay, J. L., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H. (2003). Cues to deception. Psychological Bulletin, 129(1), 74–118. 10.1037/0033-2909.129.1.74 [DOI] [PubMed] [Google Scholar]
  8. Gerlach, P., Teodorescu, K., & Hertwig, R. (2019). The truth about lies: A meta-analysis on dishonest behavior. Psychological Bulletin, 145(1), 1–44. 10.1037/bul0000174 [DOI] [PubMed] [Google Scholar]
  9. Granhag, P. A., & Hartwig, M. (2015). The Strategic Use of Evidence (SUE) technique: A conceptual overview. In P. Granhag A., Vrij A., & Verschuere B. (Eds.), Deception detection: Current challenges and new approaches (pp. 231–251). Wiley. [Google Scholar]
  10. Grice, H. P. (1975). Logic and conversation. In Cole P., & Morgan J. L. (Eds.), Syntax and semantics: Speech acts (Vol. 3, pp. 41–58). Academic Press. [Google Scholar]
  11. Gudykunst, W. (1997). Cultural variability in communication. Communication Research, 24(4), 327–348. 10.1177/009365097024004001 [DOI] [Google Scholar]
  12. Hall, E. T. (1976). Beyond culture. Doubleday. [Google Scholar]
  13. Hartwig, M., Granhag, P. A., & Strömwall, L. A. (2007). Guilty and innocent suspects’ strategies during police interrogations. Psychology, Crime & Law, 13(2), 213–227. 10.1080/10683160600750264 [DOI] [Google Scholar]
  14. Hartwig, M., Granhag, P. A., Strömwall, L. A., & Doering, N. (2010). Impression and information management: On the strategic self-regulation of innocent and guilty suspects. The Open Criminology Journal, 3(2), 10–16. 10.2174/1874917801003020010 [DOI] [Google Scholar]
  15. He, Y. (2012). Cooperative principle in English and Chinese cultures. Theory and Practice in Language Studies, 2(1), 132–137. 10.4304/tpls.2.1.132-137 [DOI] [Google Scholar]
  16. Herawati, A. (2013). The cooperative principle: Is Grice’s theory suitable to Indonesian language culture? Lingua Cultura, 7(1), 43–48. 10.21512/lc.v7i1.417 [DOI] [Google Scholar]
  17. Hines, A., Colwell, K., Hiscock-Anisman, C., Garrett, E., Ansarra, R., & Montalvo, L. (2010). Impression management strategies of deceivers and honest reporters in an investigative interview. The European Journal of Psychology Applied to Legal Context, 2, 73–90. [Google Scholar]
  18. Hofstede, G. (1983). Dimensions of national cultures in fifty countries and three regions. In Deregowksi J. B., Dziurawiec S., & Annis R. C. (Eds.), Expiscations in cross-cultural psychology (pp. 335–355). Swets and Zeitlinger. [Google Scholar]
  19. Hudson, C., Vrij, A., Akehurst, L., Hope, L., & Satchell, L. (2020). Veracity is in the eye of the beholder: A lens model examination of consistency and deception. Applied Cognitive Psychology, 34(5), 996–1004. 10.1002/acp.3678 [DOI] [Google Scholar]
  20. Jeffreys, H. (1961). Theory of probability. UK Oxford University Press. [Google Scholar]
  21. Kashima, E. S., & Kashima, Y. (2005). Erratum to Kashima and Kashima (1998) and reiteration. Journal of Cross-Cultural Psychology, 36(3), 396–400. 10.1177/0022022104273659 [DOI] [Google Scholar]
  22. Kashima, E. S., & Kashima, Y. A. (1998). Culture and language: The case of cultural dimensions and personal pronoun use. Journal of Cross-Cultural Psychology, 29(3), 461–486. 10.1177/0022022198293005 [DOI] [Google Scholar]
  23. Kashima, Y., & Kashima, E. S. (2003). Individualism, GNP, climate, and pronoun drop: Is individualism determined by affluence and climate, or does language use play a role? Journal of Cross-Cultural Psychology, 34(1), 125–134. 10.1177/0022022102239159 [DOI] [Google Scholar]
  24. Kashima, Y., Kashima, E., & Kidd, E. (2014). Language and culture. In Holtgraves T. M. (Ed.), Oxford library of psychology. The Oxford handbook of language and social psychology (pp. 46–61). Oxford University Press. [Google Scholar]
  25. Lakens, D. (2016, January 14). Power analysis for default Bayesian t-tests [Blog post]. http://daniellakens.blogspot.com/2016/01/power-analysis-for-default-bayesian-t.html
  26. Leal, S., Vrij, A., Vernham, Z., Dalton, G., Jupe, L., Harvey, A., & Nahari, G. (2018). Cross-cultural verbal deception. Legal and Criminological Psychology, 23(2), 192–213. 10.1111/lcrp.12131 [DOI] [Google Scholar]
  27. Leal, S., Vrij, A., Warmelink, L., Vernham, Z., & Fisher, R. (2015). You cannot hide your telephone lies: Providing a model statement as an aid to detect deception in insurance telephone calls. Legal and Criminological Psychology, 20(1), 129–146. 10.1111/lcrp.12017 [DOI] [Google Scholar]
  28. Levine, T. R., Anders, L. N., Banas, J., Baum, K. L., Endo, K., Hu, A. D. S., & Wong, N. C. H. (2000). Norms, expectations, and deception: A norm violation model of veracity judgements. Communication Monographs, 67(2), 123–137. 10.1080/03637750009376500 [DOI] [Google Scholar]
  29. Liu, M. (2016). Verbal communication styles and culture. Oxford Research Encyclopedia of Communication. Oxford University Press. 10.1093/acrefore/9780190228613.013.162 [DOI] [Google Scholar]
  30. McCornack, S. A. (1992). Information manipulation theory. Communication Monographs, 59(1), 1–16. 10.1080/03637759209376245 [DOI] [Google Scholar]
  31. McCornack, S. A., Levine, T. R., Solowczuk, K. A., Torres, H. I., & Campbell, D. M. (1992). When the alteration of information is viewed as deception: An empirical test of information manipulation theory. Communication Monographs, 59(1), 17–29. 10.1080/03637759209376246 [DOI] [Google Scholar]
  32. Murai, J. (1998). Perceived deceptiveness of verbal message: An examination of Information Manipulation Theory. The Japanese Journal of Psychology, 69(5), 401–407. 10.4992/jjpsy.69.401 [DOI] [Google Scholar]
  33. Strömwall, L. A., & Willén, R. M. (2011). Inside criminal minds: Offenders’ strategies when lying. Journal of Investigative Psychology and Offender Profiling, 8(3), 271–281. 10.1002/jip.148 [DOI] [Google Scholar]
  34. Strömwall, L. A., Hartwig, M., & Granhag, P. A. (2006). To act truthfully: Nonverbal behaviour and strategies during a police interrogation. Psychology, Crime, and Law, 12(2), 207–219. 10.1080/10683160512331331328 [DOI] [Google Scholar]
  35. Taylor, P. J., Larner, S., Conchie, S. M., & Menacere, T. (2017). Culture moderates changes in linguistic self-presentation and detail provision when deceiving others. Royal Society Open Science, 4(6), 170128. 10.1098/rsos.170128 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Taylor, P. J., Larner, S., Conchie, S. M., & van der Zee, S. (2015). Cross-cultural deception detection. In Granhag P. A., Vrij A. & Verschuere B. (Eds.), Detecting deception: Current challenges and cognitive approaches (pp.175–201). Wiley. [Google Scholar]
  37. Vrij, A. (2008). Detecting lies and deceit: Pitfalls and opportunities (2nd ed.). Wiley. [Google Scholar]
  38. Vrij, A., Deeb, H., Leal, S., Granhag, P. A., & Fisher, R. P. (2021). Plausibility: A verbal cue to veracity worth examining? The European Journal of Psychology Applied to Legal Context, 13(2), 47–53. 10.5093/ejpalc2021a4 [DOI] [Google Scholar]
  39. Vrij, A., Fisher, R., & Blank, H. (2017). A cognitive approach to lie detection: A meta-analysis. Legal and Criminological Psychology, 22(1), 1–21. 10.1111/lcrp.12088 [DOI] [Google Scholar]
  40. Vrij, A., Granhag, P. A., & Mann, S. (2010). Good liars. The Journal of Psychiatry & Law, 38(1–2), 77–98. 10.1177/009318531003800105 [DOI] [Google Scholar]
  41. Vrij, A., Hartwig, M., & Granhag, G. A. (2019). Reading lies: Nonverbal communication and deception. Annual Review of Psychology, 70, 295–317. 10.1146/annurev-psych-010418-103135 [DOI] [PubMed] [Google Scholar]
  42. Vrij, A., Leal, S., Mann, S., Vernham, Z., Dalton, G., Serok-Jeppa, O., Rozmann, N., Nahari, G., & Fisher, R. P. (2020). Please tell me all you remember: A comparison between British’ and Arab’ interviewees’ free narrative performance and its implications for lie detection. Psychiatry, Psychology, and Law, Advance online publication. 10.1080/13218719.2020.1805812 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Vrij, A., Mann, S., Leal, S., & Fisher, R. P. (2020). Combining verbal veracity assessment techniques to distinguish truth tellers from lie tellers. The European Journal of Psychology Applied to Legal Context, 13(1), 9–19. 10.5093/ejpalc2021a2 [DOI] [Google Scholar]
  44. Vrij, A., Palena, N., Leal, S., & Caso, L. (2021). The relationship between complications, common knowledge details and self-handicapping strategies and veracity: A meta-analysis. The European Journal of Psychology Applied to Legal Context, 13(2), 55–77. 10.5093/ejpalc2021a7 [DOI] [Google Scholar]

Articles from Psychiatry, Psychology, and Law are provided here courtesy of Taylor & Francis

RESOURCES