Within WP1, we have conducted several analyses on the large-scale datasets. Using data from the UKBiobank we have been able to demonstrate that the volume of the hippocampus is predicted by several indicators of air pollution at participants‘ home address, and interestingly, that the amount of green space around the home address is an additional predictor, adding further explained variance above and beyond the air pollution measures (Drewelies et al., 2024). Furthermore, we are in the process of publishing a paper which shows, on UKBiobank data, that the outcome measure of depressive symptoms (at similar neuroticism levels) are best predicted by risk factors such as living in a deprived neighbourhood and stressful life events; hours spent outdoors in summer, meanwhile, turned out to be a protective factor using random forest analyses (Fiedler et al., submitted).
Our paper on IMAGEN data, a European-wide neuroimaging dataset, was able to show that changes in brain structure during adolescence (from 14–22 years of age) are predicted by changes in environmental variables such as air pollution, tree cover density and presence of bodies of water (Quinones et al., submitted). The main analyses on the HCHS study are ongoing, but will lead to a submitted paper shortly. A manuscript that we will submit soon contains an analysis of a subsample of n=200 participants on whom we conducted a neurotoxicological screening of blood probes (Braun et al., 2024, Science). Here we linked the detection of different toxins to brain structure, observing that endogenous chemicals seem to be associated positively with brain structure, while human-made chemicals seem to be more negatively associated with brain structure (Kühn et al., submitted).
Within WP2, we will finish data collection of the first assessment wave in August 2025. Here we have tested 150 monozygotic twin pairs (n=300 individuals) and characterized their living environment as well as their brain structure and function. This is a major achievement, since contrary to our original plans, only a fraction of the twin participants came from the TwinLife cohort. We therefore had to find additional ways of recruiting twin pairs in Germany. As a byproduct of this effort, we have now started building the first German twin registry (www.gertrud.info).
Within WP3 we have already collected half of the sample (n=15 individuals), who underwent MRI testing for 25 sessions each, with environmental assessments 24h before entering the scanner. A written study design paper is now under submission to enable optimal data sharing once the sample is complete (n=30 individuals) (Falkenstein et al., submitted).
Within WP4, we tested the effects of the different modalities sight, sound and smell, and explored in particular whether any outcome measures show that the sum of experience is more than its parts. We observed exactly that, namely supra-additive effects of the joint exposure to sight, sound and smell of the forest in virtual reality (VR) on self-reported positive affect and nature connectedness (Ascone et al., 2025). Simultaneously, we ran an experimental study in which we explored the potential associations between the concept of forest and anxiety, because although we observed several potential positive effects of forest exposures, participants did mention some anxieties, in particular when we drove them to the real-life forest exposure. In an implicit study design we compared the picture categories forest, park and houses, yielding mixed results (Fischer et al., 2024) that practitioners in clinical settings should consider.
