Impact of Adult Attachment Anxiety on Deception Judgments: Examining the Moderating Effect of Motivation

Abstract

Several psychologists have paid attention to individual differences in deception detection, but only a few studies have found significant results. The present study aimed to explore the relationship between attachment anxiety and deception judgment when there are no obvious cues to distinguish lies from truth, and to examine the moderating effect of motives. Participants were instructed to judge each of 10 audios on whether they were true or false. Subsequently, the attachment anxiety of participants was assessed using the Experiences in Close Relationships questionnaire. Results revealed that, compared with people who had low attachment anxiety, those with high attachment anxiety tend to have higher truth biases in the low-motive condition and lower accuracy in the high-motive condition.

Keywords

Attachment anxiety deception detection accuracy truth bias motivation

Introduction

Many lies are told for maintaining or facilitating interpersonal relationships in daily life. Therefore, factors that influence interpersonal interactions or relationships may also influence lie-related behaviors. One of those factors is adult attachment anxiety. The purpose of the present study was to examine the relationship between attachment anxiety and deception judgment.

Attachment is a behavior system that helps people to promote safety and feel secure through childhood relationships with an attachment caregiver (Bowlby, 1982). This propensity of human beings to make strong affection bonds to significant others has important effects not only during childhood but also for throughout the life span (Bartholomew & Horowitz, 1991).

Previous studies suggested that there are two distinct kinds of insecurity dimensions, attachment avoidance and attachment anxiety (Brennan, Clark, & Shaver, 1998). Attachment avoidance is characterized by discomfort with interpersonal close relationships and a propensity toward self-reliance, and individuals who have high attachment avoidance are unwilling to self-disclose and reluctant to establish relationships (Gudjonsson, Sigurdsson, Lydsdottir, & Olafsdottir, 2008). In contrast, attachment anxiety refers to a preoccupation with relationships, and people with high attachment anxiety are likely to focus on pleasing others and avoid conflict, and they experience fear of rejection and abandonment (Griffin & Bartholomew, 1994).

Although many studies have found that insecure individuals with high attachment avoidance or attachment anxiety are more prone to telling lies than secure individuals are (Gillath, Sesko, Shaver, & Chun, 2010), researchers have found that attachment anxiety predicts poor subjective ability to detect lies (Elaad et al., 2012). Considering that individuals frequently estimate their own abilities, emotions, and behaviors incorrectly, it is necessary to examine the relationship between people’s attachment orientation and their objective ability to detect deception. Recently, Ein-Dor and Perry (2014) conducted three studies to examine this relationship and found that attachment anxiety predicted the accurate detection of deceitful statements but not that of truthful statements. However, all of their studies had some limitations.

First, they intentionally provided some cues that could directly corroborate or contradict the target characters’ statements in Study 1. Therefore, the positive correlation of lie detection accuracy and attachment anxiety could only demonstrate that participants with higher attachment anxiety are more sensitive and accurate than participants with lower attachment anxiety are in detecting these obvious lie-related cues (but the same was not true for subtle lie-related cues). Second, although the authors used other materials without adding obvious lie-related cues in Study 2, the participant samples were not enough for conducting a regression analysis in both the Study 2 (58 participants) and Study 3 (35 participants). Third, all three of their studies included both verbal and nonverbal information (visual-audio videos in Study 1 and 2, and face-to-face information in Study 3), so it is still unknown how attachment anxiety influenced deception judgments in the audio-only or visual-only condition.

Many studies have found that individuals who have high attachment anxiety are more likely to please others (Bartholomew & Horowitz, 1991), have a higher compliance score (Gudjonsson et al., 2008), and have a higher fear of rejection and abandonment (Mikulincer & Shaver, 2012). Taken together, these results actually suggest that high attachment anxiety individuals may be less likely to do things that may damage their relationships. Directly pointing out lies, especially when make a wrong judgment, may threaten or damage the relationship with a liar. Consequently, people with high attachment anxiety will be less motivated to judge others as liars in public. Therefore, when there are no obvious cues to corroborate (or contradict) truth tellers’ (or liars’) statements, individuals with high attachment anxiety may adopt a conservative strategy by which they are more likely to judge statements as true. This tendency will lead individuals to have a higher truth bias in deception detection.

However, because this higher truth bias is caused by the motive to please others, if participants are provided a strong incentive to make accurate judgments, the motive to please others will be inhibited. In fact, previous studies have found that the deception judgment was influenced by whether participants’ accuracy was associated with their monetary rewards (Hancock, Woodworth, & Goorha, 2010; Wu, Cai, & Jin, 2015). In addition, studies found that when participants had a higher motivation to make accurate judgments, their truth bias increased (Reinhard & Sporer, 2008; Wu et al., 2015). Therefore, the effect of motivation may obstruct the effect of attachment anxiety, and it is possible that the negative correlation of attachment anxiety and truth bias will disappear when participants can have a higher monetary reward for higher accuracy.

Besides the truth bias, accuracy is also a very important aspect of deception judgment. Although Ein-Dor and Perry (2014) found that participants with higher attachment anxiety obtained higher accuracy rates in detecting lies, it is unclear if the same results will be obtained when there are no obvious lie-related cues. In real lie-related interactions, people will express some fake social cues to make themselves trustworthy when they are telling a lie, and these cues can corroborate what they have said. If individuals are sensitive to such fake social cues, they may be more likely to be duped in deception judgment situation. In fact, researchers have found that although people with high emotional intelligence exhibited a higher ability to perceive emotions, they would be more easily duped by deceivers (Baker, ten Brinke, & Porter, 2013). In addition, because both nonverbal and verbal information were presented in the studies conducted by Ein-Dor and Perry (2014), it is difficult to anticipate what will happen in an audio-only condition.

Therefore, audio-only materials and a larger sample were used in the present study to examine whether it is possible to replicate the results of previous studies (Ein-Dor & Perry, 2014) that revealed that individuals with high attachment anxiety are better at detecting deception. We also manipulated participants’ motivation which was operationally defined as whether participants’ rewards were associated with their accuracy of deception judgment (Wu et al., 2015). For participants in the high-motive condition, the higher the accuracy was, the higher were the monetary rewards that could be achieved. In contrast, in the low-motive condition, participants would receive a fixed amount regardless of their accuracy of deception judgment. Specifically, we hypothesized that higher attachment anxiety scores will predict higher percentage of messages judged as true. In considering the possible interaction effect of attachment anxiety and motivation on truth bias, we also anticipated an interaction effect of attachment anxiety and motivation on detection accuracy.

Method

Participants

One hundred and seventy-three students (99 females; M = 20.87 years, SD = 2.01 years) were recruited from a university in China through an online advertisement. The advertisement and an experimenter’s cell phone number were published on a bulletin board system, so that students who wanted to participate could call the experimenter to plan for a specific date and time. Students from all programs except for psychology were allowed to participate in our study. Participants received monetary remuneration for their involvement. This study employed a single-factor between-subjects design with motivation (low vs. high) as the independent variable.

Stimulus materials

Audio materials

Ten audio clips were used. In each of these audio clips, a person narrated a traveling experience. Half of these persons (three females and two males) narrated a real experience of traveling to a city, while the other half (three females and two males) narrated a fabricated experience of traveling to a city. There was no significant difference in the duration of the narration between truth tellers and liars (M_truth = 120.60 seconds, SD = 14.47 seconds; M_lie = 112.60 seconds, SD = 21.66 seconds), t(8) = 0.687, p = .512.

Adult attachment questionnaire

Adult attachment was assessed using the Experiences in Close Relationships Questionnaire (ECR; Brennan et al., 1998) modified by Li and Kato (2006). The ECR is a 36-item self-report questionnaire that assesses two dimensions of individuals’ attachment, attachment anxiety (18 items) and attachment avoidance (18 items). Participants rated their extent of agreement or disagreement with each statement on a 7-point scale (1=strongly disagree, 7 = strongly agree). Scores were averaged into a single score for each dimension, with higher scores suggesting a higher level of anxiety or avoidance, respectively. This scale has adequate reliability and validity (Li & Kato, 2006). In the present study, the Cronbach’s alpha was .87 for attachment anxiety and .83 for attachment avoidance.

Procedure

Participants were informed that they would listen to some audio clips in which an individual talked about a travel experience. Their task was to judge whether the individual in each audio clip was lying or telling the truth. Participants were randomly assigned to the low- or high-motive conditions. In the low-motive condition, participants were informed that they would receive 20 yuan as a reward, regardless of their detection accuracy. In the high-motive condition, participants were informed that they would receive 5 yuan for each accurate judgment and lose 5 yuan for each wrong judgment. In fact, participants in the high-motive condition would receive at least 20 yuan for their participation.

Participants were required to make a binary judgment (truth or lie) for each audio clip. All participants needed to judge all 10 audio clips, and the order of presentation of the clips was random. Motive was measured using the following two items that were rated on a 7-point scale (r = .55, p < .001): “To what extent is it important for you to make accurate judgments in the task” (1 = very unimportant and 7 = very important) and “To what extent is it interesting for you to find out who the liars are in the task” (1 = very uninteresting and 7 = very interesting).

Subsequently, participants completed another independent study and then their adult attachment was measured using the ECR questionnaire.

Results

Manipulation check

An independent-sample t test was conducted to examine the manipulation. Results revealed that the manipulation was valid, and that participants’ motivations were significantly higher in the high-motive condition (M = 5.05, SD = 1.13) as compared to those in the low-motive condition (M = 4.32, SD = 1.36), t(171) = 3.833, p < .001, d = 0.584.

Descriptive statistics

Table 1 shows the intercorrelations, means, and standard deviations for attachment anxiety, attachment avoidance, truth bias, and total accuracy.

Table 1.

Summary of intercorrelations, means, and standard deviations for all dimensions of adult attachment, truth bias, and total accuracy.

	1	2	3	4	M	SD
1. Anxiety	1				3.80	0.95
2. Avoidance	.04	1			3.19	0.83
3. Truth bias	.05	−.01	1		59.42%	16.56%
4. Total accuracy	−.05	−.04	−.07	1	48.15%	13.94%

Truth bias

Overall, participants judged 59.42% (SD = 16.56%) of the audio clips as true, which was significantly higher than a 50% change occurrence, t(172) = 7.486, p < .001, d = 0.569. Further, a hierarchical multiple regression was conducted with percentage of truth judgments as the dependent variable. Attachment anxiety, attachment avoidance, motive (we coded low motive as −1 and high motive as + 1) were entered in the first block, while the product of motive and anxiety and that of motive and avoidance were entered in the second block. Results showed a significant main effect of motive, b = 0.046, t = 3.750, p < .001, and a significant interaction effect of motive and anxiety, b = −0.029, t = −2.209, p = .029. No other effects were significant (see Table 2).

Table 2.

Results of the hierarchical multiple regression analyses with truth bias, accuracy of truth detection, and accuracy of lie detection as dependent variables.

		Truth bias			Detection accuracy
		b	S.E.	β	b	S.E.	β
Step 1	Anxiety	0.014	0.013	0.081	−0.004	0.011	−0.027
	Avoidance	−0.003	0.015	−0.016	−0.007	0.013	−0.044
	Motive	0.047***	0.012	0.283	0.025*	0.011	0.182
	Truth bias				−0.096	0.066	−0.114
		R²= 0.082***			R²= 0.038
Step 2	Anxiety	0.010	0.013	0.056	−0.008	0.011	−0.054
	Avoidance	−0.002	0.015	−0.012	−0.009	0.013	−0.053
	Motive	0.046***	0.012	0.276	0.026*	0.011	0.184
	Truth bias				−0.124	0.066	−0.147
	Anxiety × Motive	−0.029*	0.013	−0.163	−0.025*	0.011	−0.166
	Avoidance × Motive	0.006	0.015	0.028	0.019	0.013	0.113
		ΔR² = 0.027			ΔR²= 0.045*

Notes: ***p < .001, * p < .05.

A simple slope analysis was conducted to examine the interaction effect of anxiety and motive (see Figure 1). Results supported our hypotheses, and only in the low-motive condition, anxiety significantly positively predicted the truth bias, b = 0.039, t = 2.352, p = .021. However, in the high-motive condition, the relationship of anxiety and truth bias was not significant, b = −0.019, t = −0.951, p = .344.

Figure 1.

The moderating effect of motive on the relationship between anxiety and truth bias.

Detection accuracy

The mean of participants’ total accuracy was 48.15% (SD = 13.94%), which was not significantly different from a 50% chance occurrence, t(172) = −1.746, p = .083. Further, participants’ accuracy of detecting truthful statements (M = 57.57%, SD = 20.94%) was higher than the chance level, t(172) = 4.758, p < .01, d = 0.51, whereas that of detecting deceitful statements (M = 38.73%, SD = 22.32%) was lower than the chance level, t(172) = −6.641, p < .01, d = 0.72.

A hierarchical multiple regression was conducted with total detection accuracy as the dependent variable. Because the truth bias was positively correlated with truth detection accuracy (r = 0.747, p < .01) and negatively correlated with lie detection accuracy (r = −0.782, p < .01) in the present study, we also controlled the truth bias in the regression analysis. As shown in Table 2, a significant interaction effect of attachment anxiety and motive was found.

In order to examine these moderating effects of motive, two simple slope analyses were conducted (see Figure 2). Results showed that, when controlling for the truth bias, attachment anxiety significantly negatively predicted the detection accuracy (b = −0.034, t = −2.149, p = .034) in the high-motive condition, but these effects were not significant in the low-motive condition (b = 0.014, t = 0.896, p = .373).

Figure 2.

The moderate effect of motive on the relationship of anxiety and total accuracy.

Discussion

As one of the two dimensions of adult attachment orientation, attachment anxiety has been found to have a strong impact on individuals’ recognition of behaviors in interpersonal interactions (Ahrens, Ciechanowski, & Katon, 2012; Mikulincer & Shaver, 2012). The present study expanded these results by linking attachment anxiety to actual deception judgment.

Previous studies suggested that insecure persons exhibited more dishonest behaviors (Gillath et al., 2010) and they had a lower sense of trust than secure persons did (Mikulincer, 1998). However, we found that participants with high attachment anxiety had higher truth biases than participants with low attachment anxiety did, at least in the low-motive condition. These results do not indicate that people with high attachment anxiety trust others, but they were more likely to indicate that such people are reluctant to express their distrust when there is no obvious evidence, because expressing distrust is considered to threaten or damage interpersonal relationships. This is consistent with previous findings that people with high attachment anxiety were more likely to do things against their real will to maintain and develop relationships (Gudjonsson et al., 2008). Therefore, if high anxiety participants are provided a strong incentive or a secure controllable environment to detect lies, they will make judgments according to their real will. Actually, in the high-motive condition, a nonsignificant trend for a negative correlation between attachment anxiety and truth bias was found.

Considering that participants with high attachment anxiety had a motive to please others, it is understandable that there was no significant correlation between attachment anxiety and accuracy in the low-motive condition because the high truth bias could not increase accuracy when the base rate comprised 50% truthful messages and 50% lies (Kim & Levine, 2011; Levine, Clare, Green, Serota, & Park, 2014; Levine, Kim, Sun Park, & Hughes, 2006). Specifically, if a participant judges all messages as true, she/he will obtain a total accuracy of 50%. The participants had a strong incentive to make accurate judgments and show their real ability only in the high-motive condition. In such a situation, we observed a negative correlation between attachment anxiety and accuracy, such that participants with higher attachment anxiety exhibited lower detection accuracy. This negative correlation was opposite to the results of studies conducted by Ein-Dor and Perry (2014) who found a positive correlation between attachment anxiety and detection accuracy. At least two reasons could explain these inconsistent results. First, Study 1 in Ein-Dor and Perry’s study (2014) included obvious cues that could corroborate (or contradict) truthful (or deceitful) statements in their, but such cues were much subtler in the present study. Absence of obvious and real cues may lead high-motivated individuals with high attachment anxiety to notice and misinterpret wrong cues (Baker et al., 2013). Second, it is possible that individuals with high attachment anxiety are only sensitive to visual cues but not verbal cues, especially because audio materials without nonverbal information were used in the present study. Previous studies did find that different people would use distinct cues in different situations (Reinhard, 2010; Reinhard, Scharmach, & Sporer, 2012; Reinhard & Sporer, 2008), but more studies are needed to examine these two reasons.

Our results also elaborated on the social defense theory (SDT), which suggests that both secure and insecure individuals have both adaptive advantages and disadvantages (Ein-Dor, Mikulincer, Doron, & Shaver, 2010). According to the SDT, attachment anxiety is positively associated with hypervigilant strategies through which people are more likely to monitor threat-related cues and exaggerate threat appraisal. However, the preset findings suggested that attachment anxiety is only associated with detecting obvious cues and not subtle cues. In other words, individuals with higher attachment anxiety are more likely to consider obvious cues as threat signals, but they do not have higher abilities to detect subtle cues. These results showed that the adaptive advantages of attachment anxiety proposed by the SDT are more likely to be a reactive tendency rather than an actual ability. Of course, this explanation needs to be substantiated by future research.

Although many studies have explored individual differences in deception detection, only a few studies have reported significant results (Baker et al., 2013; Reinhard, 2010). Coincidently, these significant results were more or less related to motive. Similarly, in the present study, we observed a remarkable moderation effect of motive. These consistent results suggest that motive may be a key point for exploring individual differences in deception judgment.

Footnotes

Article Notes

References

Ahrens

K. R.

Ciechanowski

Katon

(2012) Associations between adult attachment style and health risk behaviors in an adult female primary care population. Journal of Psychosomatic Research 72(5): 364–370. doi:10.1016/j.jpsychores.2012.02.002.

Baker

ten Brinke

Porter

(2013) Will get fooled again: Emotionally intelligent people are easily duped by high-stakes deceivers. Legal and Criminological Psychology 18(2): 300–313. doi:10.1111/j.2044-8333.2012.02054.x.

Bartholomew

Horowitz

L. M.

(1991) Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology 61(2): 226–244. doi:10.1037/0022-3514.61.2.226.

Bowlby

(1982) Attachment and loss: Retrospect and prospect. American Journal of Orthopsychiatry 52(4): 664–678. doi:10.1111/j.1939-0025.1982.tb01456.x.

Brennan

K. A.

Clark

C. L.

Shaver

P. R.

(1998) Self-report measurement of adult attachment: An integrative overview. In: Simpson

J. A.

Rholes

W. S.

(eds) Attachment theory and close relationships, New York, NY: Guilford, pp. 46–76.

Ein-Dor

Mikulincer

Doron

Shaver

P. R.

(2010) The attachment paradox: How can so many of us (the insecure ones) have no adaptive advantages? Perspectives on Psychological Science 5(2): 123–141. doi:10.1177/1745691610362349.

Ein-Dor

Perry

(2014) Full house of fears: Evidence that people high in attachment anxiety are more accurate in detecting deceit. Journal of Personality 82(2): 83–92. doi:10.1111/jopy.12035.

Elaad

Lavy

Cohenca

Berholz

Thee

Ben-Gigi

(2012) Lies, truths, and attachment orientations in late adolescence. Personality and Individual Differences 52(6): 670–673. doi:10.1016/j.paid.2011.12.018.

Gillath

Sesko

A. K.

Shaver

P. R.

Chun

D. S.

(2010) Attachment, authenticity, and honesty: Dispositional and experimentally induced security can reduce self- and other-deception. Journal of Personality and Social Psychology 98(5): 841–855. doi:10.1037/a0019206.

10.

Griffin

D. W.

Bartholomew

(1994) Models of the self and other: Fundamental dimensions underlying measures of adult attachment. Journal of Personality and Social Psychology 67(3): 430–445. doi:10.1037/0022-3514.61.2.226.

11.

Gudjonsson

G. H.

Sigurdsson

J. F.

Lydsdottir

L. B.

Olafsdottir

(2008) The relationship between adult romantic attachment and compliance. Personality and Individual Differences 45(4): 276–280. doi:10.1016/j.paid.2008.04.010.

12.

Hancock

J. T.

Woodworth

M. T.

Goorha

(2010) See no evil: The effect of communication medium and motivation on deception detection. Group Decision and Negotiation 19: 327–343. doi:10.1007/s10726-009-9169-7.

13.

Kim

R. K.

Levine

T. R.

(2011) The effect of suspicion on deception detection accuracy: Optimal level or opposing effects? Communication Reports 24(2): 51–62. doi:10.1080/08934215.2011.615272.

14.

Levine

T. R.

Clare

D. D.

Green

Serota

K. B.

Park

H. S.

(2014) The effects of Truth–Lie base rate on interactive deception detection accuracy. Human Communication Research 40(3): 350–372. doi:10.1111/hcre.12027.

15.

Levine

T. R.

Kim

R. K.

Sun Park

Hughes

(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(3): 243–260. doi:10.1080/03637750600873736.

16.

Kato

(2006) Measuring adult attachment: Chinese adaptation of the ECR scale. Acta Psychologica Sinica 38(3): 399–406.

17.

Mikulincer

(1998) Attachment working models and the sense of trust: An exploration of interaction goals and affect regulation. Journal of Personality and Social Psychology 74(5): 1209–1224. doi:10.1037/0022-3514.74.5.1209.

18.

Mikulincer

Shaver

P. R.

(2012) Adult attachment orientations and relationship processes. Journal of Family Theory & Review 4(4): 259–274. doi:10.1111/j.1756-2589.2012.00142.x.

19.

Reinhard

M.-A.

(2010) Need for cognition and the process of lie detection. Journal of Experimental Social Psychology 46(6): 961–971. doi:10.1016/j.jesp.2010.06.002.

20.

Reinhard

M.-A.

Scharmach

Sporer

S. L.

(2012) Situational familiarity, efficacy expectations, and the process of credibility attribution. Basic and Applied Social Psychology 34(2): 107–127. doi:10.1080/01973533.2012.655992.

21.

Reinhard

M.-A.

Sporer

S. L.

(2008) Verbal and nonverbal behaviour as a basis for credibility attribution: The impact of task involvement and cognitive capacity. Journal of Experimental Social Psychology 44(3): 477–488. doi:10.1016/j.jesp.2007.07.012.

22.

Cai

Jin

(2015) Motivation enhances the ability to detect truth from deception in audio-only messages. Journal of Investigative Psychology and Offender Profiling 12(2): 119–126. doi:10.1002/jip.1411.