The long-term impact of early adverse experience on adaptive functioning: a pilot study integrating measures of mental status, nonverbal communication, and heart rate variability

ABSTRACT

Background: Childhood maltreatment (CM) can disrupt the development of behavioural and physiological systems, increasing the risk of physical and psychological adverse outcomes across the lifespan. CM may cause interpersonal dysfunctions that impair social communication and lead to dysfunctional activation of the autonomic nervous system. The present exploratory study analyzed the long-term impact of CM from an integrated perspective through the simultaneous assessment of psychological symptoms, social and behavioural communication, and physiological regulation.

Methods: Participants were 55 healthy university students (9 males and 46 females; mean age ± SD = 25.26 ± 2.83 years), who filled out a battery of questionnaires to assess the presence of CM (Childhood Trauma Questionnaire) and psychopathological symptoms (Symptom Check-List-90 Item Revised). Participants were then subjected to a videotaped interview for the assessment of non-verbal behaviour (Ethological Coding System for Interviews) and measurement of tonic heart rate variability (HRV), a measure of physiological adaptability to the environment. We performed Pearson's correlation analysis to evaluate the associations between non-verbal behaviour, HRV, and CM variables. Multiple regression analysis was used to evaluate the independent associations between CM variables on HRV and nonverbal behaviour.

Results: We found an association between more severe CM, increased symptoms-related distress (ps < .001), less submissive behaviour (ps < .018), and decreased tonic HRV (ps < .028). As a result of multiple regression analysis, participants with a history of emotional abuse (R² = .18, p = .002) and neglect (R² = .10, p = .03) were more likely to display decreased submissive behaviour during the dyadic interview. Moreover, early experience of emotional (R² = .21, p = .005) and sexual abuse (R² = .14, p = .04) was associated with decreased tonic HRV.

Conclusion: Our preliminary findings show the utility of analyzing the long-term effects of adverse early experiences at different levels of ‘adaptive functioning’ (the capabilities needed to respond effectively to environmental demands).

HIGHLIGHTS

• Substantial traumatic experiences during childhood, such as emotional abuse, emotional neglect, and sexual abuse were associated with lower heart rate variability levels in a sample of young adults.

• Young adults with a history of emotional abuse and neglect were more likely to display decreased submissive behaviour (ethological behaviour) during the dyadic interview.

• Severe childhood maltreatment was associated with increased symptoms-related distress, less submissive behaviour (ethological behaviour), and decreased tonic heart rate variability in young adults.

Antecedentes: El maltrato infantil (MI) puede interrumpir el desarrollo de los sistemas fisiológicos y de comportamiento, lo que aumenta el riesgo de consecuencias adversas físicas y psicológicas a lo largo de la vida. A nivel conductual, el MI puede causar disfunciones interpersonales que perjudican la comunicación social y conducen a una activación disfuncional del sistema nervioso autónomo. El objetivo del presente estudio exploratorio es analizar el impacto a largo plazo del MI desde una perspectiva integrada a través de la evaluación simultánea de los síntomas psicológicos, la comunicación social y conductual, y la regulación fisiológica.

Métodos: Participaron 55 estudiantes universitarios sanos (9 hombres y 46 mujeres; edad media ± DE = 25,26 ± 2,83 años), que completaron una batería de cuestionarios para evaluar la presencia de MI (Cuestionario de Trauma Infantil – versión abreviada) y síntomas psicopatológicos (Lista de Verificación de Síntomas de 90 ítems Revisado). Luego, los participantes fueron sometidos a una entrevista grabada en video para la evaluación del comportamiento no verbal (Sistema de codificación etológica para entrevistas) y la medición de la variabilidad de la frecuencia cardíaca tónica (HRV, por sus siglas en inglés), una medida de adaptabilidad fisiológica al entorno. Realizamos el análisis de correlación de Pearson para evaluar las asociaciones entre las variables de comportamiento no verbal, HRV y MI. Luego, se utilizó un análisis de regresión múltiple para evaluar las asociaciones independientes entre las variables de MI en HRV y el comportamiento no verbal.

Resultados: Encontramos una asociación entre MI más severa, mayor malestar relacionado con los síntomas (ps < .001), menos comportamiento sumiso (ps < .018) y disminución de la HRV tónica (ps < .028). Como resultado del análisis de regresión múltiple, los participantes con antecedentes de abuso emocional (R² = 0,18, p = 0,002) y negligencia (R² = 0,10, p = 0,03) tenían más probabilidades de mostrar un comportamiento de sumisión disminuido durante la entrevista diádica. Además, la experiencia temprana de abuso emocional (R² = 0,21, p = 0,005) y sexual (R² = 0,14, p = 0,04) se asoció con una disminución de la HRV tónica.

Conclusión: Nuestros hallazgos preliminares muestran la utilidad de analizar los efectos a largo plazo de las primeras experiencias adversas en diferentes niveles de "funcionamiento adaptativo" (las capacidades necesarias para responder de manera efectiva a las demandas ambientales). Se necesitan estudios futuros para replicar la validez del presente estudio piloto centrándose en poblaciones clínicas con tasas de prevalencia más altas de maltrato infantil.

背景：童年期虐待 (CM) 会破坏行为和生理系统的发育，增加一生中身体和心理不良后果的风险。 在行为层面，CM 可能会导致人际功能障碍，从而损害社会沟通并导致自主神经系统功能失调。 本探索性研究旨在通过同时评估心理症状、社会和行为交流以及生理调节，从综合角度分析 CM 的长期影响。

方法：参与者是 55名健康的大学生（9 名男性和 46 名女性；平均年龄 ± SD = 25.26 ± 2.83 岁），他们填写了一组问卷以评估 CM（简版童年期创伤问卷）和精神病症状的存在 （修订版90 条目症状检查表）。 然后对参与者进行录像访谈，以评估非语言行为（采访行为学编码系统）和强直心率变异性 (HRV) 测量，这是一种对环境的生理适应性的测量。 我们进行了皮尔逊相关性分析来评估非语言行为、HRV 和 CM 变量之间的关联。 然后，使用多元回归分析评估 CM 变量与 HRV 和非语言行为之间的独立关联。

结果：我们发现更严重的 CM、增加的症状相关痛苦(ps < .001)、降低的顺从行为 (ps < .018) 和减小的强直性 HRV (ps < .028) 之间存在关联。 作为多元回归分析的结果，有情感虐待史 (R² = .18, p = .002) 和忽视史 (R² = .10, p = .03) 的参与者更有可能在二元访谈期间表现出更少的顺从行为。 此外，早期的情绪经历 (R² = .21, p = .005) 和性虐待 (R² = .14, p = .04) 与强直 HRV 降低有关。

结论：我们的初步研究结果表现出在不同水平‘适应性功能’（有效响应环境需求所需的能力）下分析不良早期经历的长期影响的实用性。 未来的研究需要通过关注儿童虐待流行率较高的临床人群来重复本试点研究的有效性。

KEYWORDS:

• Early-life stress
• nonverbal communication
• heart rate variability

PALABRAS CLAVE:

• Estrés en la vida temprana
• Comunicación no verbal
• Variabilidad del ritmo cardíaco

关键词:

• 早年生活压力
• 非口头交流
• 心率变异性

1. Introduction

Childhood maltreatment (CM) is any intentional act of commission (abuse) or omission (neglect) from a caregiver or a trusted adult, that results in potential/threatened harm, or actual physical or emotional harm, sexual abuse, or exploitation to a child (Child Abuse Prevention and Treatment Act, CAPTA – 2010). Despite it is well established that CM is a social issue that affects children worldwide and is associated with health and socioeconomic burden (CISMAI and Terres des Hommes, 2015; Ferrara et al., 2015; Hughes et al., 2017), it still persists to be an underreported phenomenon (CISMAI and Terres des Hommes, 2015).

Childhood and adolescence are pivotal developmental stages for the maturation of biological and behavioural systems, and CM can disrupt a child's development, increasing the risk of physical and mental health issues across the lifespan (Duncan et al., 2015; Guo et al., 2021; Halpern et al., 2018; Herrenkohl & Jung, 2016; Hovdestad et al., 2020; Humphreys et al., 2020; Kaufman & Torbey, 2019; Nakayama et al., 2020; Romano et al., 2015; Shields et al., 2016; Soares et al., 2020; Thompson et al., 2017; Ting et al., 2022).

There is evidence that CM may contribute to interpersonal dysfunctions (Felitti et al., 1998; Klein et al., 2015; Maier et al., 2020; Struck et al., 2021) and attachment disorders (Doyle & Cicchetti, 2017), as reflected by reduced prosocial behaviour (Puetz et al., 2021; Yu et al., 2020), altered interpersonal distance (Maier et al., 2020), ambivalent relationships (Waldron et al., 2018) and angry reactivity (Kim & Cicchetti, 2010).

Ethological psychiatry has been offering an original approach to the study of social communication and its disorders. Building on the description of human nonverbal communication (Eibl-Eibesfeldt, 1989; Jakobs et al., 1999; Schmidt & Cohn, 2001), clinical studies during interviews have applied the ethological method (Ethological Coding System for Interviews, ECSI; Troisi, 1999) to investigate social dysfunctions in psychiatric disorders, diagnostic classification, and prediction of the treatment outcome (Geerts et al., 1996; McGuire & Polsky, 1979; Schelde, 1998; Schelde et al., 1988; Troisi, 1999; Troisi et al., 1989; Troisi et al., 1991; Troisi et al., 1998; Troisi et al., 2007). In light of the well-established literature describing that individuals subjected to child maltreatment have interpersonal and social communication problems (for a systematic review see: Hyter, 2021; Stacks et al., 2011; Stivanin et al., 2016), it is possible to hypothesize the presence of an alteration in non-verbal behaviour belonging to the area of social communication (e.g. submission [i.e. pacification signals], displacement activities, etc.), which is optimally measured by the ECSI paradigm (Troisi, 1999).

Among CM exposure consequences, at a physiological level, the chronic activation of the endocrine and the autonomic nervous systems (ANS) results in an increase in metabolic demand, with detrimental effects on health (Kibler et al., 2014; Orr & Roth, 2000; Porges, 2001; Ridout Samuel et al., 2015; Santangelo et al., 2020; Schneider & Schwerdtfeger, 2020; Villada et al., 2014). According to the Neurovisceral Integration Model (NVM – Thayer & Lane, 2009) and the Polyvagal Theory (Porges, 2001), chronic stressful events may prompt hyperarousal, mediated by the chronic activation of the sympathetic branch of the ANS (Ader et al., 1995). Consistently, the prevailing model on the development of individual differences in stress responsiveness – the Biological Sensitivity to Context Theory (Boyce & Ellis, 2005; Ellis et al., 2005), postulates that children exposed to highly hostile environments would develop a hypersensitivity, characterized by chronic sympathetic activation and parasympathetic inhibition when exposed to such environments. However, the opposite pattern of hypoarousal has also been reported particularly in association with neglect (e.g. Shirtcliff et al., 2021).

A reliable economic index of ANS modulation is heart rate variability (HRV), which reflects parasympathetic control of the heart via the vagus nerve and has been associated with self-regulation as well as with adaptive and flexible responses to meet various situational demands (Reynard et al., 2011). It is now well established that by measuring tonic (resting) cardiac vagal tone via HRV analysis, we can indirectly estimate the activity of prefrontal brain regions, due to the bidirectional communication between the heart and the brain. Pharmacological studies as well as two independent meta-analyses of existing studies combining neuroimaging techniques with HRV assessment support the notion that higher resting HRV reflects better inhibitory capacity by prefrontal brain regions (Beissner et al., 2013; Thayer et al., 2012). In contrast, low resting-state HRV has been related to hypoactive prefrontal regulation, leading to hyperactive subcortical structures and the release of the default physiological defensive responses (e.g. Thayer & Lane, 2009). Hence, HRV may serve as an indicator of the degree to which the prefrontal cortex provides flexible control over the periphery (ultimately reflecting ‘adaptive functioning). For this reason, resting (tonic) HRV has been reported to be significantly reduced in a plethora of psychopathological conditions, up to the point that it is now considered a transdiagnostic biomarker of psychopathology (Beauchaine & Thayer, 2015).

In sum, there is a large body of evidence showing that the long-lasting consequences of CM affect a wide range of functional systems, including emotional regulation, resilience to stressful events, physiological homeostasis, and interpersonal communication.

This pilot study aimed to assess not only the presence of psychological but also behavioural and physiological dysfunctions in young adults in association with exposure to CM. The identification of parallel alterations in these three systems would produce an important diagnostic advance in the treatment of the long-term CM consequences, in terms of the identification of biomarkers of risk/resilience. Given the exploratory nature of the current investigation, we have no specific hypotheses on the direction and specificity of the associations.

2. Materials and methods

2.1. Participants

Participants included a group of 55 university students (9 men and 46 women; mean age ± SD = 25.26 ± 2.83 years). Inclusion criteria were being of age between 18 and 28 years and being native Italian speakers. Exclusion criteria were the presence of major disorders of the central nervous system (e.g. epilepsy, dementia, Parkinson's disease), self-reported history of major psychiatric disorders, and any condition affecting the ability to participate, including denial of informed consent. The experimental procedure took place in the Department of Dynamic, Clinical and Health Psychology (‘Sapienza’, University of Rome) during the 2019–2021 period. The study was approved by the Ethical Committee of the Department of Dynamic and Clinical Psychology, Sapienza, University of Rome (Prot. no. 0000453 and Prot. no. 0000112).

2.2. Procedure

Participants were informed of the following restrictions: no caffeine, alcohol, nicotine, or strenuous exercise for 2 h prior to the appointment. First, each participant gave written informed consent prior to participation. Then, the participants were asked to fill out a battery of questionnaires in order to assess the presence of childhood maltreatment and psychopathological risk. The sample was divided into subgroups based on the presence of Emotional Abuse and Emotional Neglect (at least one of these two subscales scoring moderate at the CTQ-SF). After that, participants were subjected to a videotaped interview for the assessment of non-verbal behaviour. The interview lasted 10 min and focused on neutral (i.e. non-emotionally charged) topics concerning participants’ involvement in their academic courses as well as on life and lifestyle matters. The choice of basing the interview on neutral issues responded to the aim of focusing on the participants’ baseline levels of affective activation. On a different day, inter-beat intervals (IBIs) at rest were measured for 5 min.

2.3. Clinical assessment

2.3.1. Childhood trauma questionnaire – short form

The Childhood Trauma Questionnaire-Short Form (CTQ-SF; Bernstein David et al., 2003) is a self-report test consisting of 28 items investigating the traumatic events experienced within the first 15 years of life. Specifically, the questionnaire is made of five subscales assessing Emotional Abuse; Physical Abuse; Sexual Abuse; Emotional Neglect; Physical Neglect; and an overall score that reflects the severity of the suffered maltreatment. The Emotional Abuse subscale evaluates verbal aggression, humiliating and threatening behaviours that undermine a child's self-esteem and psychological well-being; the Physical Abuse subscale evaluates the voluntary aggressive behaviours from an adult toward the child, that cause harm or injury; the Sexual Abuse subscale concerns the experience of sexualizing behaviours that involve the child in an unconscious way and/or forced way; the Emotional Neglect subscale is related to the caregiver's failure to satisfy basic psychological needs of the child, while the Physical Neglect subscale concerns the failure of caregivers to meet the basic physical needs of the baby. Each item of the CTQ-SF is evaluated using a 5-point Likert scale, from ‘never true’ to ‘very often true’. For each dimension, the score can vary from 5 (absence of traumatic events) to 25 (history of severe maltreatment). All scales provide a single cut-off score, which allows the identification of different levels of severity belonging to each childhood maltreatment dimension (absence of CM, low CM, moderate CM, severe CM). The CTQ-SF has excellent psychometric qualities, characterized by good reliability and validity. Bernstein and colleagues (2003) report good internal coherence (Cronbach's alpha between .66 and .96), and a predictive validity ranging from .79 to .86. Cronbach's alpha in the present study was .80. In the present study, the Italian CTQ – SF was administered, translated, and validated by Sacchi and colleagues (2018).

2.4. Symptom check-list-90 item revised (SCL-90-R)

SCL-90-R is a 90-item self-report, which evaluates psychopathological symptoms and psychological distress in adults in both the general and the clinical population (Derogatis, 1994). The SCL-90-R is rated on a 4-point Likert scale from 0 (not at all) to 4 (extremely), and asks participants to report if, during the last seven days, they have suffered from symptoms of somatization (e.g. headaches), obsessive-compulsivity (e.g. having to check and double-check what you do), interpersonal sensitivity (e.g. feeling that people are unfriendly or dislike you), depression (e.g. feeling blue), anxiety (e.g. feeling fearful), hostility (e.g. having urges to beat, injure, or harm someone), phobic anxiety (e.g. feeling afraid to go out of your house alone), paranoid ideation (e.g. the idea that you should be punished for your sins), and psychoticism (e.g. having thoughts that are not your own). Aside from these nine primary scales, the questionnaire provides a global severity index (GSI), which is used to determine the severity of psychological distress. For the purpose of the present study, the Italian-validated version of the SCL-90-R was employed (Prunas et al., 2012). Cronbach's alpha in the present study was .96.

2.5. Assessment of nonverbal behavior

The Ethological Coding System for Interviews (ECSI, Troisi, 1999) is an ethogram, which was developed in Italy, and precisely designed for measuring nonverbal behaviours during interviews. It includes 37 different behavioural patterns, including body posture, whole-body movement, and gesture as well as different head movements and specific facial expressions. The various behaviours are then arranged into different behavioural categories, including.

• Eye contact characterizes a fundamental aspect of social interaction. The amount of eye contact, in fact, indicates attention and involvement;

• Affiliation concerns a set of behaviours that favour and facilitate interaction;

• Submission is a set of pacification signals, which help prevent aggressive responses in the interlocutor;

• Flight contemplates a set of behaviours used to ‘cut out’ negatively perceived social stimuli;

• Assertion, which indicates low levels of aggression and disagreement;

• Gestures, which are used to emphasize or illustrate what is being communicated verbally;

• Displacement activities include those behaviours usually oriented to the body, such as playing with hair, awkward movements, or manipulating objects. These behaviours indicate motivational conflict;

• Relaxation is about behaviours that indicate low levels of emotional arousal.

The interviews were videotaped with a camera positioned behind the interviewer. The shot was adjusted so the subject's face and trunk were visible. The interviewers were female students of the same age as the interviewees, who were trained by the developer of the ECSI to conduct a neutral interview using standardized questions, for a total duration of about 10 min, which was focused on ‘comfortable topics.’ Then, a trained observer examined the video tracks for each subject in silent mode, following the taxonomy of the different aforementioned behaviours. The presence or absence of each of the behavioural elements was coded with the zero-sampling method, through which the recording of the session was divided into successive intervals of 15 s each. The instant of time at the end of each sampling interval, called the ‘sample point’, was identified by an acoustic signal. At each sample point, the observer recorded whether the behaviour had occurred or not. The score of each behaviour for each participant was expressed in terms of duration and frequency (‘quantity’) of behaviour during the interview. The decoding of nonverbal behaviour was made by a trained observer who scored at least 0.80 for each behaviour on the reliability assessment.

2.6. Physiological measurements

HR and HRV-related measures were assessed by the Firstbeat Bodyguard 2 (Firstbeat Technologies Ltd.; Jyväskylä, Finland), which allows short – and long-term recordings of the beat-to-beat range, with an accuracy of 1 ms (sampling frequency 1000 Hz). The device is not intrusive and can be directly applied on the skin using two electrodes, and then it automatically starts recording (https://www.firstbeat.com). Data were analyzed using the Kubios software HRV standard (Tarvainen et al., 2014) which allows for deriving HRV-related measures both in the time – and in the frequency-domain (Lipponen & Tarvainen, 2019). This software uses an advanced detrending method based on smoothness priors formulation in which the filtering effect is attenuated at the beginning and the end of the data, thus avoiding the distortion of data endpoints. Moreover, the frequency response of the method is adjusted with a single smoothing parameter, selected in such a way that the spectral components of interest are not significantly affected by the detrending.

The root means square of successive differences between normal heartbeats (rMSSD; in the time domain) and high-frequency HRV (HF-HRV; in the frequency domain: 0.15–0.4 Hz) were assessed as they appear to reflect the functionality of the parasympathetic nervous system via the vagus nerve (Task Force, 1996). Since the two indices were highly intercorrelated (r = .927; p < .000), in the following analyses, we focused exclusively on the rMSSD, since it is less susceptible to the effects of respiratory movements (Penttilä et al., 2001).

2.7. Statistics

The data were checked for normal distribution with the Shapiro–Wilk test. rMSSD was log-transformed to approximate better a normal distribution (ln(rMSSD)). From now on, we will use the term HRV to refer to ln(rMSSD). Given the normality of the distributions of the remaining variables, in order to evaluate the association between all non-verbal behaviour (ECSI), psychopathological risk (SCL-90-R), and CM (CTQ-SF), Pearson's correlation analysis was first used. Then, to evaluate the role of CM variables as predictors of alterations in non-verbal behaviour and in vagal tone, multiple regression analysis was used. In order to control non-specific potential predictors, we have included scores on the GSI (SCL-90-R) as a covariate in the model. The GSI is designed to measure overall psychological distress and it is a symptomatological severity index; we have opted for including it as a covariate to test for specific associations between CM and physiological and behavioural outcomes, ruling out the role of overall distress. To check for normality in the context of the regression analysis, we have used the normal probability plot of the residuals (P–P plot). Moreover, homoscedasticity of residuals was checked by using the homoscedasticity plot. The Statistical analyses were performed by SPSS Version software 25 for Windows (SPSS, Inc., Chicago, IL). For each of the examined variables, outliers were checked on the basis of the deviation from the mean of two standard deviations.

3. Results

3.1. Association between childhood maltreatment, nonverbal behaviour, HRV, and symptoms-related distress

Data for the bivariate correlational analysis were available for the entire sample (N = 55). No outliers were in the sample. Overall, emotional abuse (r = .574, p < .0001), physical abuse (r = .453, p = .001), and physical neglect (r = .496, p < .0001) were correlated with symptoms-related distress assessed by GSI (SCL-90-R, Table 1). As shown in the Table, emotional abuse (r = −.451, p = .002), sexual abuse (r = −.332, p = .028) and emotional neglect (r = −.504, p < .0001) were inversely correlated with HRV. Emotional abuse (r = −.400, p = .002) and emotional neglect (r = −.318, p = .018) were inversely correlated with Submission, physical abuse was positively correlated with Flight (r = .286, p = .034), and physical neglect with Displacement Activities (r = .268, p = .048). HRV positively correlated with Submission only (r = .353, p = .019). No correlation between HRV and other measured behaviours was observed. Overall, individuals with a history of childhood maltreatment were likely to display more severe symptoms-related distress, less submissive behaviours, and decreased HRV. Subsequent analyses focused only on Submission, as it was the only behaviour that had a relationship with both CM and HRV.

Table 1. The association (Pearson correlations) between CM, symptoms-related distress, HRV, and submissive behaviours.

	Submission	Flight	Displacement Activities	HRV	Global Severity Index (SCL-90-R)
Emotional Abuse	−.400**	.213	.120	−.451**	.574**
Physical Abuse	−.210	−.286*	−.054	−0.200	.453**
Sexual Abuse	−.150	.122	−.042	−.332*	0.180
Physical Neglect	−.214	.200	.268*	−0.153	.496**
Emotional Neglect	−.318*	.144	−.046	−.504**	0.263
HRV	.353*	.156	0.124	1	−0.213

Note. HRV = Heart Rate Variability;*p < 0.05, **p < 0.01.

3.2. Emotional trauma was associated with poor submissive behaviours

To further investigate the predictive role of Emotional Abuse and Emotional Neglect on Submission, we conducted two multiple regression analyses, inserting non-specific potential predictors into the model, such as the GSI (SCL-90-R). Table 2 shows that Emotional Abuse (B = −.493, R² = .177, p = .002) and Emotional Neglect (B = −.308, R² = .103, p = .028) were negatively associated with Submission (Supplementary Table 1). Notably, a history of Emotional trauma corresponded to decreased Submissive behaviours.

Table 2. Multiple Regression models having Submissive behaviour as outcome and emotional abuse and neglect as predictors, controlling for Global Severity Index.

	Beta	t	R-square	Sig.
	Submission
Emotional Abuse	−0.493	−3.208	0.177	0.002
Global Severity Index (SCL-90-R)	0.162	1.055		0.297
Emotional Neglect	−0.308	−2.262	0.103	0.028
Global Severity Index (SCL-90-R)	−0.040	−0.293		0.771

Note. SCL-90-R = Symptom Check-List-90 Item Revised.

3.3. Childhood maltreatment was associated with decreased levels of HRV

To investigate the predictive role of Emotional Abuse, Emotional Neglect, and Sexual Abuse on HRV, we conducted multiple regression analyses, controlling for scores on the GSI (SCL-90-R). From the results (Table 3 and Supplementary Table 2), we could see how Emotional Abuse (B = −.524, R² = .211, p = .005), Emotional Neglect (B = −.486, R² = .257, p = .001), and Sexual Abuse (B = −.307, R² = .137, p = .043) were negatively associated with HRV levels. Namely, the presence of substantial child traumatic experiences corresponded to lower HRV levels.

Table 3. Multiple Regression models having HRV as outcome and emotional, sexual abuse, and neglect as predictors, controlling for Global Severity Index.

	Beta	t	R-square	Sig.
	HRV
Emotional Abuse	−0.524	−2.939	0.211	0.005
Global Severity Index (SCL-90-R)	0.117	0.657		0.515
Emotional Neglect	−0.486	−3.419	0.257	0.001
Global Severity Index (SCL-90-R)	−0.056	−0.392		0.697
Sexual Abuse	−0.307	−2.088	0.137	0.043
Global Severity Index (SCL-90-R)	−0.165	−1.122		0.268

Note. SCL-90-R = Symptom Check-List-90 Item Revised.

4. Discussion

It is well established that exposure to stressful life events, especially during critical stages of development, leads to the alteration of normal developmental trajectories, by affecting the physiological and behavioural systems (Joca et al., 2007; Ray et al., 2017). Despite the poor and contradictory results regarding the effects of childhood maltreatment on resting-state vagal function and on non-verbal behaviours related to the domain of social communication, this pilot study presents some standpoints that can be considered as ‘food for thought’, and re-open the debate on these topics.

As a first result, we can see how substantial traumatic experiences during childhood, such as Emotional Abuse, Emotional Neglect, and Sexual Abuse were associated with lower HRV levels. These findings align with studies reporting that childhood maltreatment may induce fundamental changes in the stress response system, by disrupting its functioning (Marques-Feixa et al., 2021; McCrory et al., 2011). In fact, it has been widely discussed how childhood maltreatment has a broad-spectrum impact on individuals’ biology, as it affects brain areas involved in the response to stressful stimuli (Bremner, 2006), such as the amygdala, hippocampus, and prefrontal cortex (Jedd et al., 2015), involved also in autonomic regulation (Dale et al., 2022). Despite the large use of HRV in psychopathology studies, that globally reported decreased HRV levels in a wide spectrum of psychiatric conditions, such as mood disorders, anxiety, and PTSD (Alvares et al., 2016; Chalmers et al., 2014; Kemp et al., 2010; Nagpal et al., 2013; Pole, 2007; Schneider & Schwerdtfeger, 2020), only few studies have investigated the role of childhood maltreatment on resting vagal tone (MacMillan Harriet et al., 2009; McLaughlin et al., 2014; McLaughlin et al., 2015; Miskovic et al., 2009; Winzeler et al., 2017), with contradictory results. Interestingly, a recent meta-analysis by Sigrist and colleagues (2021) failed to report an association between CM severity and resting state HRV. However, the authors identified that this relationship was affected by the presence of some mediators, such as participants’ psychopathological state and age (Sigrist et al., 2021). The present pilot study differs from the other studies in the homogeneity of the sample, almost exclusively composed of women, and the homogeneity of age (all young adults, see Participants). It is then plausible that such associations appear only for females and/or only for young individuals.

Our study further showed an association between HRV and submissive behaviour, which was in turn associated with Emotional Trauma (Emotional Abuse and Emotional Neglect). As hypothesized by the Polyvagal Theory (Porges, 2007), the Neurovisceral Integration Model (Thayer & Lane, 2009), and the Pause and Plan Model (Segerstrom et al., 2012), the ANS is central in supporting self-regulatory processes in emotional expression in a social context, and this influences the quality of interpersonal functioning and behaviours (Snyder et al., 2006). This may be a putative pathway by which childhood maltreatment can alter social functioning.

Non-verbal behaviours related to social communication are part of the social skills, i.e. all those skills employed to modulate and favour interpersonal relationships. In particular, submission is a set of movements that serve to send pacification messages to prevent aggressive responses from the interlocutor (Burgoon et al., 1998). It is interesting to note how on a phylogenetic level, signs of pacification appeared in non-human primates to favour the transition from relationships based on social ranking (competitive) to forms of relationship based on cooperation (De Waal, 1989). Overall, the presence of submissive behaviours is crucial for the establishment of complex forms of relationships, based on intersubjectivity, shared emotions, and shared objectives (Avry et al., 2020; Keltner & Haidt, 1999). However, in order to implement this set of behaviours, the individual needs to have a positive self-representation that underlies a representation of the other as trustworthy. The ability to readily recognize a trustworthy person from one who is not allowed the activation of behavioural systems aimed at affiliation and cooperation. Notably, individuals with lower levels of resting HRV display difficulties in the recognition of trustworthy faces, even when associated with affectively relevant information, highlighting the importance of vagal modulation on social engagement and on the flexible adjustment to the social environment (Mattarozzi et al., 2019).

Overall, these results show that early adverse experiences impact several levels of an individual's ‘adaptive functioning’ (the skills needed to respond effectively to the demands of the environment). A relevant aspect of ‘adaptive functioning’ is the ability to communicate with others, to express our thoughts to others and understand what others want to communicate to us, and to be engaged in prosocial activities and positive relationships with friends. Previous research has documented that CM is inversely associated with adaptive functioning (Meng et al., 2018), so it may be worthwhile to study the impact of CM on clinically relevant social, educational, and emotional outcomes in the future.

It is well-known that childhood maltreatment exposure (Felitti et al., 1998; Finzi-Dottan & Karu, 2006; Rose & Abramson, 1992; Zweig-Frank & Paris, 1991) is associated with the formation of Internal Working Models (IWM) characterized by a lack of trust toward others, often experienced as unreliable and refusing (Bowlby, 1969/1982). Bernstein (2002) has further proposed a model that explains the onset of personality disorders in individuals with childhood maltreatment, in which, the frustration of basic needs leads to the formation of Early Maladaptive Schemas (EMS; Young et al., 2003), which influence self-perception and interpersonal relationships. In light of the above, alterations in IWM/EMS may account for the low levels of Submission in individuals exposed to Emotional Trauma. As an indirect index of cooperation aptitude, Submission can be linked to interpersonal functioning, a domain strongly undermined by Emotional Trauma (Felitti et al., 1998). In fact, Emotional Abuse and Emotional Neglect have detrimental effects on the child’s adaptation abilities to social context, as Emotional Trauma often occurs in conjunction with other types of childhood maltreatment, magnifying their influence (Chapman et al., 2004; Ferguson & Dacey, 1997; Kinard, 1999; Mattar, 2018; Zamir, 2021).

Overall, the present pilot study has several limitations. Participants were healthy volunteers, and this may explain the small size of the effects. It is likely that studies of clinical populations would yield larger effects of childhood maltreatment on the examined dependent variables. For example, a recent meta-analysis has suggested that the association between childhood adversity and vagal regulation becomes statistically significant for individuals with a diagnosis of psychiatric disorder (Wesarg et al., 2022). The small sample employed, together with a majority of females in the sample, reduces the generalizability of the results. Moreover, although an attempt was made to keep the content of the interview as neutral as possible, it cannot be excluded that it elicited anxiety in individuals characterized by anxious traits/personality.

In the future, it will be interesting to expand the sample and balance the number of males and females. Although the present study emphasizes the relationship between childhood maltreatment and reduced HRV levels as well as submissive behaviours, it is likely that other variables may explain these associations. On the relationship between childhood maltreatment and Submission, on the other hand, IWM / EMS, as factors that develop in conjunction with intersubjectivity and interpersonal motivational systems that act as an inner drive in individuals (Estévez et al., 2017; Hawkins & Haskett, 2014; May et al., 2022).

Acknowledgements

The authors thank Gloria Paganelli and Flaminia Brogna for the help provided during the experiments.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Facoltà di Medicina e Psicologia, Sapienza Università di Roma [grant no. RG11916B50652E41].