Conference Agenda

Introduction: The capacity to understand the intentions, thoughts and emotions of others is essential for effective social interactions. Also known as social information processing abilities more generally, such capacities include socio-cognitive skills like theory of mind (ToM) alongside socio-affective skills like empathy and compassion. Our study investigated the relationship between the latter three social information processing abilities and individual differences in adult attachment.

Methods: We employed the semi-behavioural paradigm EmpaToM that is based upon video-recorded narratives. These narratives vary in their negative versus neutral story content as well as in their requirement of ToM understand it. After each video, participants are asked to provide a rating of empathy and compassion and to respond to a ToM question. Attachment was assessed using the Adult Attachment Interview (AAI), coding participants' current ability to reflect on and discuss attachment-related information from their past, with a sample size of 85 participants (50.60% female, average age = 25.87±4.50 years). A self-report attachment questionnaire, the German version of the Experiences in Close Relationships revised (ECR-R), was also used in a larger sample of 158 participants.

Results: Individuals classified as autonomous/securely attached in the AAI exhibited higher ToM abilities. Additionally, there was a negative correlation between self-reported attachment avoidance from the ECR-R and compassion ratings.

Discussion: Our findings provide further evidence that different measures of attachment based on interviews versus self-reports do not necessarily align but may capture different aspects of social information processing. We conclude that individuals classified as autonomous/securely attached and individuals with low avoidant attachment tendencies show differential social skills that enhance their understanding of others' mental states and foster compassionate responses to others' distress.

Stress is a pervasive phenomenon with well-documented adverse effects on physical and mental health. Romantic partnerships are typically considered a key resource for stress buffering. However, whether these relationships fulfil their protective potential appears to depend on their inherent characteristics, including attachment. In two complementary studies, we examined how attachment relates to stress responses in romantic partners across subjective, physiological, endocrine, and neural markers.

Using a dyadic psychosocial stress paradigm, one partner (“target”) was exposed to an acute laboratory stressor while the other partner (“observer”) was present. In the first project, we examined the relationship between both partners’ attachment and targets’ subjective stress experience, autonomic reactivity, and cortisol responses. Attachment was assessed using the ECR-RD questionnaire. Results consistently indicated a potential role of attachment avoidance for stress regulation within couples. While targets’ avoidant attachment was negatively associated with subjective stress reports, suggesting stress suppression as part of a deactivating strategy, observers’ avoidant attachment correlated positively with autonomic stress reactivity in targets. Furthermore, attachment avoidance was negatively associated with psychoendocrine covariance, suggesting a decoupling between perceived and physiological stress responses.

In the second project, we extended this framework to neural stress processing using functional near-infrared spectroscopy (fNIRS). Targets showed increased prefrontal activation during the arithmetic part of the stressor task, reflecting engagement of higher-order regulatory processes. Importantly, targets’ neural stress reactivity was not associated with targets’ attachment, but was significantly positively related to observers‘ attachment avoidance, particularly in the left prefrontal cortex. This pattern suggests that targets’ cognitive control mechanisms may be inhibited under stress when their partner is potentially experienced as unsupportive.

Taken together, these findings suggest that romantic partnerships do not uniformly buffer stress. Rather, partners’ attachment characteristics critically determine whether relationships function as a resource or become an additional stressor, across multiple levels of stress responses.

Adult attachment can be assessed using conceptually distinct approaches, most notably narrative-based interviews and self-report questionnaires. While both approaches are widely used and reliable, their differential associations with brain structure remain poorly understood. The present preregistered study examines how two different adult attachment measures relate to brain anatomy in a large sample of mothers and fathers, and whether these associations show converging versus diverging patterns.

Structural magnetic resonance imaging (MRI) data were obtained from 91 parents (50 fathers, 41 mothers) who completed both the Adult Attachment Interview (AAI) and the Experiences in Close Relationships-Revised questionnaire (ECR-R). Cortical and subcortical brain structure was assessed from T1-weighted anatomical images. Cortical measures of surface area, thickness, and grey matter volume were derived using a surface-based approach, while subcortical volumes were extracted for regions previously implicated in attachment-related processes, including the hippocampus and amygdala.

To ensure comparability between the AAI and ECR-R, analyses were first conducted using a security-insecurity construct derived from each measure, followed by more fine-grained analyses distinguishing dismissive/avoidant from preoccupied/anxious attachment. To address the high dimensionality of neuroanatomical data relative to sample size, preregistered L1-regularized regression (LASSO) models are applied separately for each brain structure modality, followed by a non-parametric combination approach integrating results across modalities, while controlling for participants’ age, gender, and total brain volume.

By focusing on a parental sample and directly comparing a narrative-based versus a self-reported attachment measure, this study aims to clarify whether different measures of adult attachment are associated with overlapping or distinct patterns of brain structure. Analyses are currently ongoing, with preregistered LASSO models being applied to identify robust attachment-brain structure associations. By integrating attachment theory with multimodal structural neuroimaging, the project seeks to advance our understanding of how individual differences in attachment are reflected in adult brain anatomy.

Parents play a central role in children’s development, including during early adolescence and especially in supporting children's emotional experiences. Interpersonal neural synchrony (INS) is a new measure of the temporal alignment of brain activity between individuals. It is thought to reflect both context-dependent information flow during social interactions and broader interpersonal mechanisms such as attachment and bonding.

This study focused on mother-child dyads (n = 38; children aged 10-14 years) and employed an emotional imagery task. Mothers and children imagined children experiencing emotional situations (positive, negative, neutral), either with mothers being present or absent. They were also asked to rate their subjective experiences on a valence scale reflecting mothers’ and children’s feelings as part of imagining being together versus apart. Using functional near-infrared spectroscopy (fNIRS) hyperscanning, we simultaneously measured INS in the right prefrontal cortex, dorsolateral prefrontal cortex, and temporoparietal junction. Additionally, children’s attachment security to their mothers was assessed in terms of secure base support (SBS) and safe haven support (SHS) using the Security Scale Questionnaire.

Results from the subjective valence ratings showed that mothers and children reported feeling better when imagining being together for both positive and negative situations. fNIRS data showed that INS was overall higher during negative (versus positive) situations when mother-child pairs imagined the mother being present (versus absent). Moreover, in mother-present scenarios, INS in the frontopolar cortex was higher in dyads in which the child reported lower SBS security.

Increased self-reported positivity when mothers and children imagined being together (compared to being apart), regardless of situational context, suggests that maternal presence, even if only imagined, enhances emotional well-being. Our findings further suggest that mother-child INS may reflect dyadic emotional processing and that individual characteristics can shape its nature.