I will present the overall design and first results from the CovSocial projectthat in a first phase aimed to unravel the impact of the Covid19-pandemic on various aspects of mental health, psychological resilience and social cohesion among the Berlin population in 2020 and 2021 (with 7 measurement time points). In a second phase, the same Berliners were taking part in a 10-week online mental training study to explore whether mental health, resilience and (self)compassion could be fostered through either socio-emotional partner-based dyadic training or mindfulness-based short-term daily practice.

With respect to the longitudinal psychological vulnerability trajectories measured throughout 2 lockdowns in Germany, two distinct effects emerged during the course of the pandemic: an acute first lockdown effect and a second lockdown fatigue effect. For example, on average, anxiety, loneliness, stress and depression increased during the first lock-down in March 2020 and recovered during re-opening in summer 2020, but decreased again in autumn during the second lock-down and most indicators declined from month to month. In general, women and the youngest cohorts suffered most. The results of the intervention study in phase2 performed in 2021/22, suggest that both online mental trainings interventions were indeed effective in reducing depression, anxiety and emotion regulation difficulties and increase empathy and (self)compassion. While the mindfulness training was better in improving stress recovery, the socio-emotional partner-based dyadic training was more effective in reducing loneliness and negativity bias and boost resilience and social connectedness. The findings are discussed in light of translational approaches to scale these compassion-based mental health approaches to profit different domains of society from health-care to education.