Lie detection accuracy and beliefs about cues to deception in adult children of alcoholics

Abstract

When one or both parents misuse alcohol, it can lead to the development of particular and varied traits in their children. The present study tested whether adult children of alcoholics (ACoAs) who participated in therapy had better veracity assessment skills and more reliable beliefs about cues to deception than the control group of non-ACoAs. The results revealed that individuals who grew up in a family with alcohol misuse problems detected truth – but not lies – significantly better than the control group. The groups did not differ in accuracy of their beliefs about cues to deception. It is possible that the ACoAs’ higher truth detection accuracy is to some extent attributable to their participation in therapy, which increased their level of trust in others.

Introduction

Telling the truth is a socially desirable behavior but research has shown that lying is part and parcel of communication (e.g. Cantarero, Van Tilburg, & Szarota, 2018; DePaulo & Kashy, 1998; DePaulo, Kashy, Kirkendol, Wyer, & Epstein, 1996; but see Serota, Levine, & Boster, 2010). In a diary study, DePaulo et al. (1996) found that participants lied approximately 1.5 times per day, which meant that they tried to deceive people they interacted with 34% of the time.

Prolific liars

Studies on the frequency of lying have also expanded the ways to identify personality traits of individuals who are capable of lying more often than average. It was established that people who lied more often were also more outgoing but less socialized, sought to impress others and had a stronger tendency to manipulate others (Kashy & DePaulo, 1996). Furthermore, research has uncovered that attachment styles affect how often people lie and that participants with attachment-related anxiety lie more frequently to strangers and best friends, while participants with attachment avoidance are more likely to deceive romantic partners (Ennis, Vrij, & Chance, 2008). Moreover, some studies have suggested a relationship between the frequency of lying and psychopathy (see Vrij, 2008, for a review).

People with substance addictions represent another group that has been associated with the frequent use of deception as they tend to deny and hide their use or abuse of addictive substances (Rogers & Bender, 2018). Lying could be especially frequent in the early stages of their problematic drinking or substance abuse when their personal and work life is still relatively intact (Paredes, 1974). Additionally, they tell lies to justify the failure to carry out professional or personal duties caused by addiction or to enable them to acquire addictive substances (Rogers & Bender, 2018). Although it seems very difficult to precisely assess the frequency of lying in persons with alcohol misuse (see Midanik, 2009, for a discussion), it was revealed that students who admitted to risky drinking behaviors also admitted both to using deception more often in order to avoid taking exams and to actual cheating during exams (Blankenship & Whitley, 2000). Furthermore, alcohol and illicit substance abusers had a higher tendency to engage in self-deception that involved active denial, confabulation and selective amnesia (Martínez-González, Vilar López, Becoña Iglesias, & Verdejo-García, 2016). The high level of self-deception by active alcohol abusers and their impression management tendencies (i.e. intentional deception used to gain a social advantage; Paulhus, 1984) was related to their positive self-esteem and drinking-control beliefs. This was confirmed by Sirvent, Herrero, Moral, and Rodriguez (2019), who found significant differences in levels of manipulation (i.e. self-presentation with the intention of influencing others’ behavior) and mystification (i.e. inaccurate perceptions and distortion about one’s lifestyle) between alcohol and drug abusers and the general population. The level of self-deception in active alcohol abusers was significantly higher than in recovering alcoholics (Strom & Barone, 1993).

Lying about substance use and abuse is also one of the most frequently reported topics that therapy clients keep secret from their therapists (e.g. Baumann & Hill, 2016). They justify this dishonesty with shame, fear of being judged and concern about possible consequences of their problems (Farber, 2020).

Accurate detectors

This universality of lying implies that people have developed methods of defense against deception. However, studies on lie detection accuracy clearly demonstrate that veracity assessment based on observation of another’s behavior rarely exceeds the level of chance (Bond & DePaulo, 2006). The low accuracy of truth–lie judgments was explained by small differences in behavior between liars and truth-tellers (Hartwig & Bond, 2011) and by reliance on stereotypical cues to deception (e.g. Bogaard & Meijer, 2018).

However, other studies have identified certain professional groups in which deception detection abilities are greater than average (Ekman & O’Sullivan, 1991; Ekman, O’Sullivan, & Frank, 1999). Furthermore, studies that attempted to determine the characteristics of individuals with these outstanding abilities suggested that the experts were introverted, quiet and observation-oriented, and many of them had interests that complement these traits like bird watching or portrait painting (O’Sullivan & Ekman, 2004). Several of them also had untypical childhoods; for example, some had not learned English until grade school, others were children of working mothers (which was exceptional in their environment) or were children of alcoholics. However, these observations were not confirmed by the results of personality tests or other findings, and the research conducted by Ekman and colleagues (Ekman & O’Sullivan, 1991; Ekman et al., 1999; O’Sullivan & Ekman, 2004) stirred up methodological and statistical controversy (Bond, 2008; Bond & Uysal, 2007; Ekman, O’Sullivan, & Frank, 2008; O’Sullivan, 2007).

Nevertheless, the relationship between difficult childhood experiences and deception detection ability was confirmed in a study with participants who were physically abused as children (Bugental, Shennum, Frank, & Ekman, 2001). It is possible that negative life experiences or patterns of unreliability resulted in these children learning quicker than others that not all information given by other people is true. This coincides with various studies that have determined that high levels of suspicion increased deception detection accuracy (DePaulo, Jordan, Irvine, & Laser, 1982; Vrij, 2008). The influence of negative life experiences (including childhood experiences) on deception detection ability was also found by Draheim (2004): girls and boys housed in young offender institutions were able to detect deception more accurately than groups of various lie experts. Moreover, educators employed in these institutions confirmed that the majority of these children came from dysfunctional families (Draheim, 2004).

These results are also consistent with studies showing that individuals high in attachment anxiety are better than more secure persons in a poker game and in detecting deceitful statements (Ein-Dor & Perry, 2014), also when cooperating in that task with others (Ein-Dor, Perry-Paldi, Daniely, Zohar-Cohen, & Hirschberger, 2016). According to social defense theory (Ein-Dor, Mikulincer, Doron, & Shaver, 2010), these enhanced abilities are a consequence of the fact that people who are highly anxious about separation and abandonment are quicker and more accurate in detecting ambiguous signs of threat (Ein-Dor et al., 2016). This seems to be in accordance with the results revealing that women who were victims of sexual assault exhibited better immediate and delayed facial recognition skills than women without such a negative experience (Islam-Zwart, Heath, & Vik, 2005).

Furthermore, the influence of the social environment and life experiences on veracity assessment ability has been confirmed by a study of adult prison inmates (Hartwig, Granhag, Strömwall, & Andersson, 2004). The offenders were significantly better than students at lie detection but not in concealment of the truth, yet their overall veracity assessment accuracy was above the level of chance. This pattern of results stemmed from prisoners’ tendency to assess viewed statements as deceptive rather than truthful (known as lie bias), which was most likely due to their high levels of suspicion. Moreover, it was also revealed that offenders have more accurate knowledge about cues to deception than a control group or a judiciary staff (Strömwall & Granhag, 2003; Ulatowska, 2009; Vrij & Semin, 1996). More accurate lie detection, as well as less stereotypical beliefs about deception indicators, are probably due to the environment in which offenders lived that required increased vigilance since the consequences of being deceived or being caught in deception were very high (Hartwig et al., 2004). This also suggested a correlation between knowledge of cues to deception and detection accuracy, which was later directly confirmed (Bogaard & Meijer, 2018; Reinhard, Scharmach, & Müller, 2013; Ulatowska, 2011).

Children of alcoholics

Previous studies suggest that individuals with negative life experiences that date even back to childhood may have higher veracity assessment abilities. A specific group noted in this context are people whose parents misuse alcohol. According to the World Health Organization (2018), alcohol is the most commonly used addictive substance, and 5.1% of the world’s population aged 15 and older suffer from alcohol use disorders (2.6% of this population suffer from alcohol dependence and 2.5% from harmful use of alcohol). This means that children and adult children of alcoholics (ACoAs) account for a considerable proportion of society. They might have experienced unreliability and neglect in adolescence (e.g. Sher, 1991) and might have encountered parental deception on a regular basis (Rogers & Bender, 2018). Consequently, these people might have developed specific traits and behaviors.

While research shows that this group is more prone to alcohol and substance abuse in adulthood (e.g. Christoffersen & Soothill, 2003; Schuckit & Smith, 1996), findings on emotional or cognitive problems in ACoAs do not yield a homogeneous picture, which suggests that ACoAs differ from non-ACoAs to a lesser degree than the assumptions in the popular literature (Baker & Stephenson, 1995). The majority of studies show that ACoAs are more prone to depression and anxiety disorders (e.g. Omkarappa & Rentala, 2019; Thapa, Selya, & Jonk, 2017) than non-ACoAs and have fewer coping strategies. Also, ACoAs have lower self-esteem and are higher in dependency and neuroticism, as well as more frequently engage in manipulative behaviors (Baker & Stephenson, 1995; Rangarajan, 2008; Roebuck, Mattson, & Riley, 1999). However, other studies revealed no such patterns (for a review see Sher, 1997). These discrepancies could stem from the heterogeneity of this group caused by factors such as temperament, gender, the attitude of the non-addicted parent and the child’s social environment (Sher, 1991).

Studies also suggest that due to their experiences with an untrustworthy parent, ACoAs exhibit a lower level of trust in others (Bradley & Schneider, 1990; Haverfield & Theiss, 2014). They are also taught – explicitly or implicitly – that they should not share their internal or external experience with anyone from outside their family (Zakrzewska & Samochowiec, 2017). Whereas the lack of trust can make it more difficult to enter into close relationships with other people, for accurate deception detection suspicion is necessary to implement an appropriate level of analysis of cues to deception (Vrij, 2008). Furthermore, some studies suggest that parental alcohol misuse is related to the development of other adaptive traits in ACoA women as compared to a control group, such as flexibility, openness to change, outgoingness, openness to criticism, self-assurance and independence (Baker & Stephenson, 1995). In addition, when choosing a field of study in higher education, ACoAs are more likely to choose majors connected with helping others such as psychology, mental health, social work or nursing (Baker & Stephenson, 1995).

The present study

Previous studies have suggested that people with negative childhood experiences and people high in attachment anxiety have better veracity assessment ability. It is possible that frequent exposure to deception and having to hide certain information or one’s own emotional states from other people lead to the development of a heightened ability to detect deceit and to notice ambiguous cues, including indicators of deception. Although the influence of parental alcoholism on their children’s behavioral functions has been investigated for years, we are not aware of any studies testing deception detection ability and knowledge of cues to deception in ACoAs. As shown in prior studies, ACoAs are less trustful and more anxious and were probably the victims of lies told by their alcohol-abusing parents. Likewise, it appears that they had more opportunities to quickly verify their suspicions – for instance, when a parent denied drinking or promised not to drink again. According to the feedback hypothesis (e.g. Hartwig et al., 2004), this might have enabled them to develop more accurate beliefs about cues to deception and more accurate veracity assessment ability than non-ACoAs. Thus, this study is the first to verify whether lie-detection ability and beliefs about cues to deception of ACoAs are more accurate than those of individuals without such childhood experiences. These results will add to our understanding of social and individual factors influencing lie detection process.

Method

Participants

In total, 56 participants took part in the study. Half of them (N = 28) were adult children of at least one alcoholic parent (M_age = 34.71 years, SD = 8.05; 78.6% women). All participants from this group attended therapy for ACoAs and were tested after one of the meetings in the therapeutic center. They attended this kind of therapy for one year on average (M = 12.23 months, SD = 9.99) and self-assessed that their parents’ problems with alcohol started when they were approximately three and half years old (M = 3.57 years, SD = 4.14). The control group (N = 28) was recruited by social-media ads and was composed of participants who declared that their parents had no alcohol misuse problems and did not participate in any kind of psychotherapy (M_age = 36.04 years, SD = 11.27; 82.1% women) but matched ACoAs in terms of demographic characteristics. The participants in both groups did not differ in terms of age, t(54) = 0.50, p = .616, d = 0.13, gender, χ²(1, N = 56) = 0.11, p = .737, or level of education, χ²(2, N = 56) = 1.96, p = .375. All participants were volunteers and gave their informed consent.

Materials

The procedures for video footage preparation and detection were approved by the research ethics committee of The Maria Grzegorzewska University. The video footage used was recorded prior to the study during a special session in which 20 young people (M_age = 22.4 years, SD = 3.13; 55% women) were asked to describe the most stressful experience from their lives. The reasoning behind choosing this lie scenario was twofold. First, it was possible to obtain credible truthful and deceptive stories on this topic from an available sample of senders (i.e. undergraduate students). Second, the narratives on the emotional experiences (i.e. accidents, illnesses, family emergencies) could be common among substance abusers as they might use such stories in an attempt to cover their failures in professional and personal lives or to acquire addictive substances (Rogers & Bender, 2018).

The senders were asked to prepare one truthful and one deceptive description of this stressful experience. They knew that they were being filmed and that the recordings would be used for scientific purposes. As compensation, they received a gift card worth approximately $8. Before the session the senders were also told they would receive an addition $35 gift card if they were able to convince the experimenter of their veracity. The purpose of this additional reward was to increase the lie stake and thus the chance of occurrence of cues to deception. The order of truthful and deceptive statements was counterbalanced, and before each recording the senders were given 2 min to prepare their description.

A set of five truthful statements and five deceptive statements of different senders were randomly chosen from all the recordings and were then used in the current study. Each statement lasted about 60 s, and they were presented in random order.

A questionnaire on beliefs about cues to deception (Ulatowska, 2017) was also used. It contained a list of 33 verbal and nonverbal behaviors connected to deception that had been tested in previous studies (e.g. DePaulo et al., 2003). Participants were asked to decide whether the intensity of each cue was considerably higher when the truth was told, did not differ between liars and truth tellers, or was considerably higher when a lie was told.

Procedure

Participants from both groups were asked whether they were interested in taking part in a lie detection study and were apprised of the procedure. After giving their informed consent, they were presented with video footage of truthful and deceptive statements, and after each video they were asked to determine whether the sender was lying or telling the truth. Additionally, they were asked to rate whether they were confident in this decision using a 7-point scale (1 = definitely not to 7 = definitely). Subsequently, participants were given the questionnaire on beliefs about deception cues, and an explanation of how some of the cues were to be understood was provided. After recording their judgments participants were debriefed and dismissed. The entire test period lasted approximately 30 min. The experimenters were blind to the veracity of senders on the video footage.

Results

Experience with lying and lie detection

Participants from both groups were also asked to assess their experience with lying and lie detection using a 7-point scales (1 = I strongly disagree to 7 = I strongly agree). All questions are presented in Table 1. The analysis revealed significant differences in two questions. The control group assessed their own knowledge of cues to deception, t(54) = 2.28, p = .027, d = 0.62, and their own abilities to detect deception, t(54) = 4, p < .001, d = 1.08, significantly higher than the ACoA group.

Table 1.

Different personal experience variables as a function of group.

Personal experiences	Group
	ACOAs		Control
	M	SD	M	SD
I often encounter lying in everyday life	4.71	1.67	5.04	1.60
I am a good lie detector	3.71	1.44**	5.04	1
I rarely lie in everyday life	3.39	1.83	4.04	1.77
I am a good liar	3.32	1.89	3.57	1.48
I know cues to deception	4.32	1.36*	5.07	1.09

Note: ACoAs = adult children of alcoholics.

*p < .05. **p < .001.

Accuracy of veracity judgments

Each participant assessed both truthful and deceptive statements; thus, two kinds of mean accuracy rates were computed: the truth detection accuracy rate and the lie detection accuracy rate. To compare detection accuracy in both groups, a 2 (group: ACoAs vs. control) × 2 (veracity of statement: truthful vs. deceptive) mixed-model analysis of variance (ANOVA) was utilized. Group type was a between-subjects factor, and veracity was a within-subjects factor. There was a main effect of veracity, F(1, 54) = 4.81, p = .033, η² = .08. Participants were significantly more accurate in lie detection (M = .46, SD = .21) than in truth detection (M = .38, SD = .23). The main effect of group was not significant, F(1, 54) = 1.26, p = .267, η² = .02, indicating that ACoAs (M = .45, SD = .13) were no more accurate in veracity assessment than the control group (M = .40, SD = .19). An interaction between veracity and group was significant, F(1, 54) = 5.72, p = .020, η² = .10. A simple effect analysis revealed that the groups tested differed significantly only in truth detection (p = .024), with ACoAs being more accurate than the control group but not in lie detection (p = .527). Furthermore, only the control group (p = .002), but not the group of ACoAs (p = .888), was better in lie detection than truth detection. All means are presented in Figure 1.

Figure 1.

Veracity assessment accuracy in both groups. ACoAs = adult children of alcoholics.

One-sample t tests were used to test whether detection accuracy in both groups differed from the level of chance (.5). In the ACoA group only the overall accuracy rate (M = .45, SD = .13) differed significantly from the level of chance, t(27) = 2.15, p = .041, d = 0.41. In the control group the truth detection rate, t(27) = 4.32, p < .001, d = 0.83, and the overall rate (M = .40, SD = .19), t(27) = 2.81, p = .009, d = 0.54, were both significantly lower than the level of chance.

Furthermore, lie bias – that is, the tendency to believe that most messages one hears are deceptive – was compared in both groups. It was revealed that lie bias was more pronounced in the control group as 58.21% of their judgments were lie judgments (one-sample t test, p < .001) as compared to 49.64% of judgments in the ACoA group (one-sample t test, p = .904). The difference in lie bias between both groups was significant, t(54) = 2.39, p = .02, d = 0.64.

No difference between ACoAs (M = 4.91, SD = 0.64) and the control group (M = 5.12, SD = 0.69) was revealed in the confidence of veracity judgments, t(52) = 1.17, p = .249, d = 0.32.

Beliefs about cues to deception

In the next stage of analysis, the mean accuracy of beliefs about cues to deception was computed. A group comparison found that the difference in accuracy of beliefs between ACoAs (M = .39, SD = .09) and the control group (M = .37, SD = .09) was not significant, t(54) = 0.97, p = .338, d = 0.26. However, it was discovered that for both the ACoAs, t(27) = 3.76, p = .001, d = 0.72, and the control group, t(27) = 2.18, p = .038, d = 0.42, the mean accuracy of beliefs differed significantly from the level of chance (i.e. .33¹). Furthermore, a test of whether the accuracy of beliefs about cues to deception was related to veracity assessment accuracy was performed, and the correlation between both accuracy rates for both groups was not significant [ACoAs: r(26) = .18, p = .371; control group: r(26) = −.21, p = .276].

Discussion

The aim of this study was to compare veracity assessment accuracy and beliefs about cues to deception between ACoA and non-ACoA groups. However, the results did not fully confirm our predictions about differences in veracity assessment accuracy between these groups. Both groups had similarly low accuracy scores that were below the level of chance, which could stem from small differences in behavior between truthful and “lying senders”: (Hartwig & Bond, 2011) and from inaccurate knowledge of cues to deception (Bogaard & Meijer, 2018; Reinhard et al., 2013).

However, a separate analysis of deception detection and truth detection found that ACoAs were equally as effective in detecting liars and truth-tellers. They did not exhibit the lie bias observed in the control group. This is inconsistent with predictions based on previous studies suggesting that individuals who were more mistrustful and suspicious or who had more frequent contact with deception – as ACoAs were (Bradley & Schneider, 1990; Haverfield & Theiss, 2014; Rogers & Bender, 2018) – might exhibit a tendency to judge others as liars rather than truth-tellers (Hartwig et al., 2004). However, this is in accordance with the lack of differences between studied groups in self-perceptions concerning the frequency of contact with deception in everyday life: ACoAs did not differ from the control group in the assessment of the frequency of lying and being lied to. Future studies should test whether such groups would differ when assessing the frequency of being lied to by parents, both now and during childhood.

Furthermore, ACoAs were significantly more accurate at detecting the truth than controls. On one hand, this can be explained by the low accuracy rate in the control group, which was perhaps the result of low motivation during performance of this task (e.g. see Wu, Cai, Zhang, Liu & Jin, 2018; but see also Porter, McCabe, Woodworth, & Peace, 2007, for opposite results). On the other hand, this could be an effect of increased scrutiny resulting from an enhanced ability to notice ambiguous signs by people higher in attachment anxiety (Ein-Dor & Perry, 2014; Ein-Dor et al., 2016), which might be experienced by some of the ACoAs. However, it must be stressed that better lie detection, rather than better truth detection reported here, seems to be more consistent with social defense theory (Ein-Dor et al., 2010).

Finally, it is also possible that the difference in truth detection is an outcome of participation in therapy. This kind of therapy is often based on restoration of a sense of security and trust in oneself and in others, working on forgiveness, improvement of relationships with one’s social environment and self-development (Ackerman & Gondolf, 1991; Osterndorf, Enright, Holter, & Klatt, 2011). Therefore, it is possible that ACoAs who have taken part in group therapy of this kind were aware of the consequences of their family history of alcohol misuse and tried to put into practice the abilities acquired during therapy. Although the follow-up analysis of the correlation between veracity assessment accuracy and the duration of therapy in the ACoA group was not significant, r(26) = −.08, p = .695, the issue of possible therapy influence should be addressed in future studies.

This study did not confirm the predictions based on the results of earlier studies suggesting that individuals with negative childhood experiences detect deception more effectively. Apart from being attributed to the influence of therapy, these results can be explained in several other ways, but this is only speculative, and further studies should address these issues. First, some studies have shown that ACoAs were somewhat deficient in social intelligence as measured by theory of mind tasks (e.g. Baron-Cohen’s Eyes Task, (Baron-Cohen, Wheelwright, Hill, Raste, & Plumb, 2001)), which require one to decode relevant social information about other people based on observable information (i.e. photos of the eyes region; Hill et al., 2007). This skill contributes to a person’s ability to predict or explain others’ emotions and behavior (but see Kopera et al., 2014). Therefore, this kind of deficit could impair the ability to detect subtle cues to deception.

Furthermore, some of the previous studies on deception detection accuracy revealed that individuals with high social anxiety and low self-esteem were worse at veracity assessment (DePaulo & Tang, 1994). It is also suggested that ACoAs are exposed to a higher risk of anxiety and depressive disorders (e.g. Omkarappa & Rentala, 2019; Thapa et al., 2017) as well as low self-esteem (Rangarajan, 2008), which in the present study was indirectly expressed by their significantly lower self-assessment of lie-detection abilities and knowledge of cues to deception than in the control group. Thus, it is possible that ACoAs have difficulty focusing attention on cues to deception or misinterpret cues they have noticed. In turn, this leads to problems with discernment of liars from truth-tellers (but see Vrij & Baxter, 1999; Vrij, Harden, Terry, Edward, & Bull, 2001).

Finally, as mentioned above, ACoAs are not a homogeneous group, which means that their childhood experiences and the influence of these experiences on their functioning in adulthood can vary considerably. For this reason, when analyzing ACoAs’ various abilities, further studies should take into account not only the past experience of having a parent with an alcohol problem but other individual factors as well.

The present study also revealed a lack of differences between ACoAs and non-ACoAs in beliefs about cues to deception. This is not consistent with earlier studies showing that groups who lived in an environment in which deception was more common were more familiar with these cues (Strömwall & Granhag, 2003; Ulatowska, 2009; Vrij & Semin, 1996). Lastly, we found no relationship between deception detection ability and familiarity with cues to deception. Although this is inconsistent with some of the previous studies (Bogaard & Meijer, 2018; Reinhard et al., 2013; Ulatowska, 2011; but see Wright & Wheatcroft, 2017), Hartwig and Bond’s (2011) analysis suggests that it stems not so much from reliance on the wrong cues when detecting deception as from the absence of marked differences in behavior between liars and truth tellers.

Limitations and future directions

The present study was the first to test ACoAs’ accuracy in lie detection. Thus, it is necessary to replicate these results, preferably using different truth–lie statements as the low accuracy of detection could indicate that the differences in cues to deception displayed by senders might be too small to notice (Hartwig & Bond, 2011). Furthermore, apart from the internal heterogeneity of the group of ACoAs, a limiting factor of the present study that should be taken into account in the future is participation in therapy. Individuals who partake in ACoA therapy might initially differ from the non-clinical population of ACoAs (for a discussion, see Baker & Stephenson, 1995; Newcomb, Stollman, & Vargas, 1995; Tweed & Ryff, 1991). Moreover, participation in therapy might have affected the participants’ level of trust in other people. Therefore, future studies should include participants from a group of ACoAs who do not participate in therapy as well as participants taking part in other kinds of therapy that is unrelated to their being an ACoA. Lastly, although participants from the control group did not grow up in families with addiction problems, they could have experienced other negative events in childhood, which might have influenced their truth–lie decisions. Thus, future studies should control this possibility and, for example, utilize a negative life experiences scale. Despite the limitations, these preliminary findings add to the literature on individual and social factors related to veracity assessment and to our understanding of specific traits that can develop in ACoAs.

Ethical standards

Declaration of conflicts of interest

Joanna Ulatowska has declared no conflicts of interest

Iga Nowatkiewicz has declared no conflicts of interest

Sylwia Rajdaszka has declared no conflicts of interest

Ethical approval

All procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee 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