The Transactional Stress Model (Lazarus and Folkman, 1984, 1987) conceptualizes stress as an interaction between a person and their environment, in which the person experiences challenges that exceed their resources or threaten their well-being. This imbalance is accompanied by a hypervigilant psychophysiological state of arousal leading to stress-related symptoms of emotional exhaustion, which can lead to lasting physical and mental health impairments (Klusmann and Waschke, 2018). Teachers are an occupational group that experiences particularly high levels of stress and pronounced stress-related symptoms of emotional and physical exhaustion (e.g., Hasselhorn and Nübling, 2004; Lohmann-Haislah, 2012; Nübling et al., 2012). This burden is associated with limitations in subjective well-being (Meng and D’Arcy, 2015) and has been linked to negative teaching behaviors in the classroom (e.g., Klusmann et al., 2006, 2016). For example, Shen et al. (2015) found that teachers’ stress related emotional exhaustion at the beginning of the semester was a negative predictor of students’ motivation at the end of the semester.

Rothland (2013) proposed that contextual factors such as having a split workplace (school and home), unregulated working hours, coercion-based cooperation with students, or low opportunities for promotion can lead to high levels of stress experience among teachers. Research further suggests that classroom disruptions and student discipline problems are additional factors that predict higher levels of stress in teachers (e.g., Chang, 2009; Tsouloupas et al., 2010). However, Hillert et al. (2013) suggested that these challenges do not necessarily lead to high stress experiences for all teachers on an individual level. In line with this, Klusmann et al. (2008) found that contextual factors such as school size, the size of taught courses, the school district, the socioeconomic status of students, and their cognitive abilities are only slightly associated with stress-related symptoms of exhaustion among teachers. Instead, research suggests that negative personal characteristics, such as maladaptive patterns of self-regulatory abilities (e.g., Schaarschmidt et al., 1999; Schwarzer et al., 2022a) are more strongly associated with higher stress experiences. For instance, Lehr et al. (2008) found that teachers with negative patterns of coping strategies characterized by the tendency to resign report strong impairments in well-being, extensive emotional exhaustion, and higher depressive symptoms with overall large effect sizes. Notably, these results were independent of the average number of teachers’ working hours (part-time; full-time). These findings are of great importance: Apparently, high levels of stress-related exhaustion are not caused by the amount of time teachers spend in the classroom, but are instead linked to personal characteristics that are risk factors among prospective teachers for experiencing higher levels of stress even before they enter the profession.

As a consequence to these findings, Klusmann and Waschke (2018) have suggested that empirical studies need to pay greater attention to teachers’ personal characteristics and evaluate their impact on the development of global stress experiences at an early stage of teacher education. The university training phase of teachers reflects a promising window of prevention, as contextual and school-characteristic features seem to play a less important role in the development of stress than personal factors (Hillert et al., 2013). Therefore, it is important to focus on these personal features in the curriculum within the close cooperation between student teachers and lecturers at university prior to their entering teaching service in the field. This approach could help identifying individual differences and the underlying psychological mechanisms in early stages of teacher education and training, in order to prevent a more negative professional development of teacher candidates later on.

Attachment theory, although developed 60 years ago, continues to be considered an influential developmental theory in the behavioral and social sciences (Thompson et al., 2022). This theory, first proposed by John Bowlby (1944) suggests that humans have an innate drive to form close emotional relationships with attachment figures in order to create and maintain physical and psychological proximity (Bowlby, 1988). According to Bowlby (1969, 1988), when a child is exposed to threatening stimuli, the child’s attachment system is activated, leading to attachment behavior such as crying, calling, or seeking out the attachment figure. This behavior then triggers caring behavior from the attachment figure (Bowlby, 1969, 1988) creating a co-regulatory system that helps the child cope with stressful experiences (Bowlby, 1988).

The long-term effects of attachment relationships on the development of cognitive-affective representations, referred to as Internal Working Models [IWMs] have been studied extensively in recent decades (Bretherton, 1990; Bretherton and Munholland, 2018). IWMs are thought to be relatively stable over time (e.g., Fraley and Shaver, 2000; Mikulincer and Shaver, 2016), and shape behavior in challenging situations of allostatic load (Bowlby, 1988; Bretherton and Munholland, 2018).

Secure attachment representations, developed by the child on the basis of sensitive caring behavior by the attachment figure (Ainsworth et al., 1978), are considered an important factor positively influencing psychosocial development and mental health (Groh et al., 2014). In such cases, the attachment figure serves as a “safe haven” by providing the child with reliable and co-regulatory support in stressful or threatening situations (Bowlby, 1988). This leads to the development of attachment representations in which the child is represented as worthy of protection and self-efficacy, and which shape beliefs that even stressful situations are modifiable (Bretherton and Munholland, 2018). On the other hand, insecure attachment representations—caused by caregiving behavior that was dismissive or barely predictable—reflect a risk factor that has been consistently linked to a whole series of negative outcomes such as mental health impairments (Zhang et al., 2022), internalizing symptoms (Groh et al., 2012), externalizing symptoms (Fearon et al., 2010), or impairments in emotional (Groh et al., 2017) and social competencies (Groh et al., 2014).

Ainsworth et al.’s (1978) extension of Bowlby’s theory of attachment distinguished two forms of attachment insecurity: insecure-ambivalent and insecure-avoidant (Ainsworth et al., 1978). Insecure-ambivalent attachment is characterized by high levels of anxiety due to highly inconsistent and difficult to predict experiences with primary caregivers during stressful situations (Bretherton and Munholland, 2018). This leads to aversive experiences being highly stress-inducing, with little coping capacity available, and resulting in strong feelings of helplessness (Dykas and Cassidy, 2011). Insecure-avoidant attachment is argued to be rooted in experiences with primary caregivers who have been consistently rejecting and unavailable in stressful situations (Bretherton and Munholland, 2018). In these cases, attachment-related needs and whishes triggered in distressing situations are suppressed in a defensive manner as a way of self-protection despite significant psychophysiological arousal (Mikulincer et al., 2003; Dykas and Cassidy, 2011).

Mentalizing, or the ability to perceive and consider one’s own and others’ behavior as based on intentional mental states, is closely linked to attachment theory (Fonagy et al., 1991; Fonagy and Allison, 2014). It is conceptualized as a developmental achievement associated with the increasing awareness of the importance of mental states in organizing behavior (Fonagy et al., 2002; Luyten et al., 2020). Secure attachment relationships are thought to provide an adaptive learning environment in which children can develop effective mentalizing skills as a result of having their feelings reflected back sensitively by their attachment figures (Fonagy et al., 2002). Conversely, insecure attachment relationships can compromise the development of children’s mentalizing capacities due to the poorly attuned interaction experiences between the children and the attachment figures (Luyten et al., 2017).

Mentalizing has become an increasingly important aspect of mental health (Katznelson, 2014). Research has shown that mentalizing skills are impaired in patients with depression (Fischer-Kern et al., 2022), borderline personality disorder (Németh et al., 2018), pathological anxiety (Chevalier et al., 2023), or antisocial personality disorder (Newbury-Helps et al., 2017) compared to healthy controls. Further, psychotherapeutic interventions have been found to promote mentalizing (Levy et al., 2006; Fischer-Kern et al., 2015; Storebø et al., 2020), which in turn is associated with a decrease in symptom severity (Rossouw and Fonagy, 2012; De Meulemeester et al., 2018).

Recently, the focus on mentalizing as a clinically important factor in psychopathology has been extended to an understanding of mentalizing as a coping strategy buffering the negative impact of aversive experiences (Fonagy et al., 2017; Luyten et al., 2020). It has been argued that mentalizing protects against the effects of stressful experiences and enables a more resilient adaptation to, and processing of, stress (Fonagy et al., 2017; Holmes, 2017). This hypothesis assumes that mentalizing enables individuals to manage, process, and give meaning to their own and others’ behaviors, which helps them to maintain a robust comprehension of social interactions (Fonagy et al., 1994, 2002). Hence, mentalizing as an intrapsychic coping strategy is believed to reduce experiences of distrust, confusion and loss of control when confronted with emotionally challenging situations, and to foster a stable sense of self-coherence (Fonagy et al., 2019, 2021). Consistent with these assumptions, empirical data suggest that mentalizing might serve as a mediating state of mind reducing the impact of adversity in clinical (e.g. Bizzi et al., 2020; Huang et al., 2020) and non-clinical samples (e.g. Brugnera et al., 2021).

The present study aimed to examine the associations between attachment insecurity and global stress experience in prospective teachers. Furthermore, it was tested whether this relationship was mediated by the individual’s mentalizing capacity.

Our results confirm Hypothesis 1, i.e., the expectation of a positive association between attachment-related anxiety and current stress experience in preservice teachers. Increasing attachment-related anxiety was linked to higher levels of stress experience which is consistent with the attachment theory’s framework. Specifically, due to early inconsistent caregiving experiences and a failure to co-regulate in stressful situations, individuals with insecure-anxious attachment styles develop internal working models in which they are represented as having low competence and self-efficacy in dealing with challenging situations (Ainsworth et al., 1978; Bretherton and Munholland, 2018). From this point of view, anxious attachment styles can be conceptualized as a hypervigilant state of mind due to the hardly predictable reactions of the attachment figure which systematically and negatively distorts the perception of mental states and capabilities – even many years later in adulthood (Ehrenthal et al., 2018; Tironi et al., 2021). As a result, anxious individuals have only few options available to cope with challenging, demanding, or threatening experiences, which rapidly evoke stress and elicit strong feelings of helplessness and loss of control (Dykas and Cassidy, 2011). This study was able to replicate this link from other studies (Fearon et al., 2010; Groh et al., 2012, 2017; Bordoagni et al., 2021; Brugnera et al., 2021) among prospective teachers. These findings are consistent with results suggesting that attachment-related anxiety represents a risk factor associated with psychosocial distress, mental health problems, and adverse development (Zhang et al., 2022).

Hypothesis 2 predicted a direct association between attachment-related avoidance and current stress experience in preservice teachers, but it was rejected in light of the result as there was no significant relationship between the two variables. This finding contradicted attachment theory’s notion that individuals with insecure-avoidant attachment representations experience high levels of stress more rapidly (Bowlby, 1988; Tironi et al., 2021) and the findings of Brugnera et al. (2021), who empirically confirm such a relationship among psychotherapists. However, due to the self-report instruments used in the current study, it is likely that in particular avoidant subjects responded in a way that could be interpreted as trivializing or suppressing potentially threatening self-states for the purpose of self-protection. This defense process could have contributed to the fact that threatening self-states or self-states such as current stress experiences were negated, thus influencing the response behavior. However, the result may also be explained by certain characteristics of the avoidant attachment pattern. It is possible that prospective teacher candidates with higher levels of attachment-related avoidance are more capable of distancing themselves emotionally from stress experiences, which is consistent with the attachment theoretical framework (Ainsworth et al., 1978). Children with avoidant attachment representations have learned to respond to high levels of stress by suppressing their emotions as a form of self-protection and coping mechanism (Dykas and Cassidy, 2011). The presented results suggest that many preservice teachers continue applying this coping mechanism when they experience high levels of stress in their professional development.

Hypothesis 3, which predicted a mediation of the direct association of both forms of attachment insecurity and global stress experience through mentalizing, can be partially confirmed. In the studied sample, a partial mediation effect was found for the association between attachment-related anxiety and stress experience, but not between attachment-related avoidance and stress experiences. Consequently, the effects documented by Brugnera et al. (2021) among psychotherapists can only be partially replicated among prospective teachers. The results suggest that attachment-related anxiety, predicted by own experiences of inconsistent and unreliable caregiving behavior by the attachment figure in childhood, promotes limitations in mentalizing capacity, which in turn is linked to an increase in general stress experiences. This finding is in line with the mentalizing framework (Fonagy et al., 2002), which postulates that the quality of early attachment experiences is a determinant factor in the development of effective mentalizing skills (Luyten et al., 2017). Furthermore, according to Fonagy et al. (2017), effective mentalizing enables the inner-psychic processing of aversive experiences, so that a coherent self-experience is maintained even under demanding circumstances.

In contrast, the lack of a significant indirect path from avoidant attachment representations to global stress experience through impairments in mentalizing in our sample suggests that early experiences of rejection in attachment relationships do not necessarily lead to compromised mentalizing. This contradicts the mentalization theory’s postulate that secure attachment representations in particular are linked to effective mentalizing (Fonagy et al., 1991, 2002). However, the mentalizing concept has increasingly focused on the extended social environment, emphasizing that the development of effective mentalizing is not only determined by early attachment experiences, but also by other social relationships (Fonagy and Allison, 2014), for example with peers, teachers, or spouses (Fonagy et al., 2017; Luyten et al., 2020). Therefore, this novel finding needs replication in future studies. However, individuals with avoidant attachment representations may tend to have a positively biased assessment of their own abilities for the purpose of self-protection, which is consistent with results from other studies showing less self-concept clarity in avoidantly attached subjects (e.g., Emery et al., 2018). Following attachment theory (Ainsworth et al., 1978), a positively biased self-perception in avoidant individuals is an adaptation to a non-responsive environment. It allows subjects to avoid any thoughts about oneself that are too difficult to acknowledge or recognize, because of the high degree of mental stress such a recognition would bring about.

Data will be made available per request by the first author, N-HS, schwarzer@ph-heidelberg.de.

All subjects provided written informed consent to participate in the study. The research program was approved by the Ethics Committee of the Ludwigsburg University of Education (III-Sopead_NiSc-0009).

N-HS: conception and design, execution, interpretation of the data and writing of the original draft. N-HS and LD: analysis. TN, PF, SG, TB, and LD: review and editing. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.