###Theoretical Background

Entrainment refers to the phenomenon in which processes align and synchronize with each other, gradually adjusting to a shared rhythm (Phillips-Silver et al., 2010). This concept is evident in activities such as clapping, dancing, or rowing in unison (e.g., Freeman & Acena, 2021).

Entrainment Theory emphasizes that interpersonal motor synchronization (e.g., Rinott, & Tractinsky, 2021; Shaposhnikov et al., 2024) is crucial for successfully completing tasks, such as rowing a boat in one direction to propel it forward, or fostering social connections, such as in partnered dancing. Considering this, entrainment is relevant not only in real-world environments but also plays an increasing role in virtual reality (VR; Tarr et al., 2018). For instance, partnered dancing in VR requires the same synchronization and cooperation, making entrainment essential for achieving a seamless and immersive experience. However, VR introduces unique challenges that can disrupt the process, such as latency, sensory mismatches, and avatar embodiment discrepancies, and thus hinder the entrainment experience (Becher et al., 2020).

###Research Aim

Despite the critical role of entrainment in both real-world and virtual environments, a validated tool for measuring this phenomenon across these different environments is lacking. Therefore, this study aims to develop a scale for measuring entrainment experiences in both real-world and VR environments.

###Methodological Approach

To develop the Entrainment scale, we deductively developed a comprehensive Entrainment model with four main dimensions and ten subdimensions. The model was operationalized using both established and self-developed items. To test the instrument, participants will complete the Entrainment scale twice (once after a real-world entrainment experience and once after a VR entrainment experience), along with validation measures. The study adheres to established scale development procedures (e.g., Carpenter, 2017), was approved by the university’s ethical committee, and was conducted as part of a funded research project [both blinded for peer review].

Voice assistants (VAs) like Alexa and Siri are increasingly embedded in family life, providing convenience and accessibility through voice-command functionality. Families use VAs for tasks such as managing schedules, playing music, and answering questions, with their intuitive interface making them accessible to all members, including children. Positioned in shared spaces like kitchens or living rooms, VAs encourage family-wide interaction and fulfill diverse roles, including personal assistant, entertainer, teacher, and even a quasi-family member. Their ability to emulate human-like social behaviors leads users to anthropomorphize them, attributing traits such as friendliness and honesty. Despite their growing presence, research on VAs has largely focused on short-term use, leaving gaps in understanding their long-term impact on family dynamics.

This study addresses these gaps through a 2.5-year longitudinal survey of 128 families with children and at least one VA at home. Conducted biannually, the study investigates factors influencing relationship development between families and their VAs. Relationship development is operationalized through two indicators: usage frequency (behavior-based) and connectedness (perception-based). Influencing factors are categorized into social feelings and utility. Social feelings include connectedness as well as enjoyment, relationship closeness, and satisfaction. Utility comprises communication errors as well as parents’ perceived truthfulness and usefulness of the VA. Anthropomorphization variables, including trust, perception as social entity, and mind attribution, are also assessed for their impact on parents’ connectedness to VAs. Parents were asked to evaluate the usage frequency, connectedness, and potential influencing factors for themselves as well as for their children, when needed.

Findings aim to provide insights into the evolving role of VAs in family life, highlighting potential opportunities and risks associated with their integration. By focusing on long-term interactions, this research contributes to understanding which factors influence how families build relationships with VAs over time.

Recently, virtual reality (VR) addressing historical topics has become a part of history education by allowing the immersion into the past (Lewers, 2022). However, how VR can be beneficial for education is still discussed. While the immersion through VR motivates learners to learn more about historical events (Frentzel-Beyme & Krämer, 2023), the accompanying emotional involvement in past atrocities may lead to emotional reactions and limit critical reflection (Bunnenberg, 2020). By referring to the Transportation-Imagery Model proposed by Green & Brock (2002), research suggests that transportation in VR, describing the immersion into narration, may support uncritical, intensive processing leading to story-consistent beliefs and emotional responses (Green & Brock, 2000; 2002). Researchers assume that by being immersed in a story, recipients may become less aware of real-world facts and, consequently, more likely adopt story-consistent beliefs described by the term narrative persuasion (Green & Appel, 2024). Hence, when it comes to learning with VR, understanding how emotional arousal is involved and affects not only knowledge but also the learners’ beliefs remains a pressing question.

Therefore, the present study aims to investigate the narrative persuasion of historical VR by also focusing on democratic beliefs under consideration of learners' pre-existing political value orientation. In more detail, we ask whether transportation in historical VR affects users’ political beliefs and what role their political orientations play in this. To investigate these research questions, we use a laboratory experiment (N = 106) in which participants used Anne Frank House VR and answered questions assessing transportation, emotional arousal, story-consistent political beliefs, (perceived) knowledge and their preexisting political orientation.

In sum, this study aims to illustrate the conditions under which novel media such as VR can influence opinion formation in politically relevant settings.

With the advances in speech recognition and natural language processing, voice assistants (VAs) are now used as a source of information beyond simple tasks. Previous research showed that information read by a VA is more credible than the same information formatted like a Google search snippet (Gaiser & Utz, 2023). One reason may be that voice encourages people to anthropomorphize VA (Krämer et al., 2012). This might trigger users to assign roles to VAs similar to human interactions, viewing VAs as more than just tools (Schweitzer et al., 2019).

Previous research examined what kind of roles users assign to their VAs and how these roles are related to the credibility of VAs. Survey findings indicate that users generally view VAs as either servants or partners. Those who view their VAs as partners find them more credible than those who see them as mere servants. Moreover, given that VAs provide a diverse range of information, people use them for various tasks such as retrieving declarative (e.g., asking about historical events) or procedural knowledge (e.g., providing cooking instructions). Therefore, we aim to examine these dynamics experimentally. We expect that participants perceive VAs as more credible and intelligent in a partner role than in a servant role. Furthermore, we anticipate that VAs delivering declarative knowledge are regarded as more credible and intelligent than those providing procedural knowledge, as factual information is often viewed as more objective (Berge & Hezewijk, 1999).

Participants were assigned to both conditions, with the VA adopting either a servant or a partner role in random order. Subsequently, the information type was presented in either a declarative or procedural format, covering a variety of topics. After each presentation, participants rated the VA's perceived credibility and intelligence. Additionally, we examined perceived warmth and competence as potential moderators.

This study investigates how the ethnicity of the VR coach and the VR coach type (AI-driven vs. human-controlled) influence users' overall psychological wellbeing, encompassing dimensions such as loneliness and stress. Grounded in Computers Are Social Actors (CASA) theory, which proposes that individuals instinctively respond to computers as social entities, this study seeks to understand the intricate mechanisms shaping user perceptions and emotional responses to virtual coaches. Drawing from Similarity-Attraction theory, we anticipate that same-ethnicity avatars will be perceived as more similar. Furthermore, we expect human-controlled avatars to be perceived as more realistic compared to AI-driven counterparts. Stereotype Content Model pointed out competence and warmth are the two possible stereotype contents people may form towards virtual agents. SIT further suggests that in-group preferences based on ethnicity can amplify these perceptions, potentially leading to more favorable warmth and competence ratings for same-ethnicity avatars. This study also investigates how users’ perceived warmth and competence differ for VR coach of different ethnicity and type.

The primary research question is: To what extent do avatar ethnicity and avatar type (AI-controlled vs. human-controlled) impact users' perception towards VR coach and subsequently their psychological wellbeing, including loneliness and stress levels? It also explores the mediating role of perceived similarity and realism in the relationship between avatar characteristics, perceptions of warmth and competence of the VR coach, and subsequently, how warmth and competence relate to loneliness and stress levels.

A 2x2 between-subjects online experiment will be conducted. Participants will be randomly assigned to one of four conditions, manipulating avatar ethnicity (same vs. different) and control type (AI-driven vs. human-controlled). Within a simulated VR coaching scenario, participants will interact with VR avatars as life coach. Pre and post-test will be conducted.