Background: The increasing use of smartphones in schools presents challenges for teachers, school administrators, and policymakers. In response, school-wide bans on smartphones are becoming increasingly implemented. Empirical research on the effects of school-wide smartphone bans remains limited. Existing studies show small positive effects on social behaviour but no clear impact on academic performance. Moreover, the broader consequences of such bans, particularly their effects on home and parenting contexts, have not been explored.

Aims: The first aim of the study is to examine the implication of smartphone bans in Belgian secondary school on pupils’ cognitive, motivational, emotional and social functioning. The second aim is to look at two competing hypotheses exist: (1) the compensation hypothesis, where banning smartphones at school leads to increased usage at home, potentially causing conflicts with parents, and (2) the generalization hypothesis, where improved self-regulation at school translates to healthier smartphone use at home.

Methodology:
This study employs a longitudinal, multi-informant design (pupils ages 12-19, teachers, and parents) to examine the ban’s impact on four developmental domains of pupils: cognitive (e.g., concentration in class), motivational (e.g., motivation for learning), emotional (e.g., well-being) and social (e.g., bullying) functioning. It also investigates the conditions necessary for effective policy implementation (e.g., school climate) and the role of individual student characteristics (e.g., age). Data will be collected across three time points (May 2025, November 2025, May 2026) to analyse both short- and long-term effects. Results of the first data collection in May 2025 will be presented at the ECDP conference.

Conclusion:
By providing evidence-based insights, this study aims to support schools, parents, and policymakers in developing effective and motivating smartphone policies that balance academic performance, student well-being, and parental concerns.

In recent decades, social media has emerged as a crucial channel for adolescents and young adults to express personal emotions and thoughts, where some glues for suicidal intentions or actions can be observed. This study presents a comprehensive content analysis of 1,424 authentic posts from Dcard, Taiwan's leading social media platform among the youth, to methodically identify critical factors of a suicidal crisis using machine learning techniques. Human professionals classified each post into one of four crisis levels and labeled each sentence according to six risk factors. Subsequent application of discriminant analysis and the eXtreme Gradient Boosting method enabled our model to achieve over 96% accuracy in binary classification and above 84% in ternary classification for crisis-level prediction. Notably, the most significant risk factors identified are suicidal attempts, suicidal thoughts with depression, and positive emotions. Our work demonstrates that nearly complete results can be replicated using these three factors alone, highlighting their importance in identifying high-risk posts. This research contributes a new perspective to the field of adolescent and young adult suicide studies. It suggests practical applications for suicide prevention on social media through natural language processing technologies in artificial intelligence.

Social networking sites like TikTok are now seen as microsystems that proximally influence adolescents’ development. However, their effects depend on the activities and content teenagers encounter, which still need to be explored in-depth. Thus, given TikTok’s affordances and widespread use among adolescents, we employed a mixed-methods research design (N = 328, 60.4% male, Mage = 16.99) to shed light on relations between adolescents’ activities on TikTok and the content they consume. In this sense, cross-lagged panel analysis was used to explore the reciprocal relations between adolescents' distinct activities on TikTok and the different types of content they view. Over a three-month period, there was a positive unidirectional longitudinal association between viewing entertainment-related content and active TikTok participation through likes, comments, and shares. Similarly, there was a positive unidirectional longitudinal association between passively using TikTok and watching lifestyle-related content. To go beyond these broad content categories and gain a deeper insight into the actual messages adolescents are exposed to on TikTok, participants were asked to list the names of TikTok accounts they actually followed. Out of 789 TikTok accounts reported by participants, thematic analysis was performed on the accounts followed by two or more adolescents, yielding a final sample of 127 TikTok accounts. Out of this, 16 themes emerged, and, given the breadth and complexity of the TikTok accounts, most were attributed to more than one theme. Our results highlight the double-edged nature of TikTok. Themes such as science, awareness-raising, and family content speak to its positive side, while themes such as age-inappropriate content and consumerism relate to its negative side. Moreover, out of the 127 TikTok accounts thematically analyzed, more than half belonged to Romanian content creators. Our findings highlight the idiosyncratic and contextual nature of adolescents’ content consumption and its different relations with TikTok activities.

Key-words: adolescence, TikTok activities, content consumption, thematic analysis, mixed-methods

Numerous strategies and interventions have been designed to mitigate prejudice among adolescents and promote positive intergroup relations. Both direct and indirect forms of contact are pivotal in shaping young people's social dynamics and attitudes. Despite this, relatively few studies have taken a longitudinal approach to explore how adolescents' social networks influence their attitudes toward outgroups over time.

This study investigates the longitudinal relationship between adolescents' face-to-face and electronic contact with intergroup friends and their explicit and implicit attitudes toward Syrian immigrants. The sample consisted of high school and secondary school students (n=543) from Batman, Türkiye, a province near the Syrian border. Participants completed measures assessing face-to-face contact, electronic contact, and outgroup attitudes on paper, while the Implicit Association Test (IAT) was administered via computer. The study employed a two-time point design, with data collected at the beginning and end of the school term.

Cross-lagged panel model (CLPM) analyses revealed that adolescents' electronic contact with Syrian peers was longitudinally associated with increased explicit negative attitudes toward Syrian migrants. In contrast, face-to-face contact demonstrated an inverse longitudinal relationship with explicit attitudes toward the outgroup. Additionally, findings indicated a reciprocal longitudinal association between electronic and face-to-face contact, suggesting that these forms of interaction may mutually influence each other.

These findings highlight the significance of electronic contact in shaping adolescents' intergroup attitudes, as it was associated with more negative explicit attitudes toward Syrian migrants. In contrast, face-to-face interactions showed the potential to foster more positive perceptions. Interestingly, the results also suggest that adolescents’ electronic and face-to-face contact experiences may be mutually supportive, indicating that both forms of contact can influence one another over time. This reciprocal relationship highlights the complexity of how different types of intergroup contact interact and contribute to adolescents' attitudes toward outgroups.

This conference participation was supported by TUBITAK 2224-A-Grant Program.