Conference Agenda

Background: Interhemispheric functional connectivity (ihFC) reflects coordinated activity between brain hemispheres and is essential for cognitive and socio-emotional functions. Altered ihFC has been linked to psychopathological difficulties in various DSM-5 neurodevelopmental disorders. This study examines the relationship between ihFC, psychopathology, and genetic variation using a multidimensional, data-driven approach. Methods: Multi-modal data were drawn from the Philadelphia Neurodevelopmental Cohort (PNC; Satterthwaite et al., 2026, N=877, ages 8–21), a large-scale neurodevelopmental dataset including resting-state fMRI, genetics and clinical assessments. Resting-state fMRI data (TR = 3000 ms, TE = 32 ms, voxel size = 3×3×3 mm, 6 min acquisition) were preprocessed using fMRIPrep (Esteban et al., 2019). Functional connectivity (FC) matrices were computed using hMRF 400 atlases and Pearson’s correlation. Graph neural networks, such as GCN and ARMA layers, were used to model 7 psychopathological factor scores from ihFC features. An ablation analysis was performed to determine which inter- or intra-hemispheric components of the FC matrix, at both the graph and functional network levels, had the greatest impact. For functional networks found to be good predictor of psychopathology, a candidate gene association study was performed using PLINK, targeting different genes previously linked to FC. Results: FC matrices predicted general psychopathology with modest accuracy (mean r²=0.1; RMSE=0.98) with psychosis factor showing the strongest predictive signal (r²=0.21). ihFC in default mode, visual, dorsal attention, and limbic networks was especially important for prediction, and SNPs in candidate genes were associated with connectivity in these networks. Conclusion: Our findings demonstrate that ihFC plays a central role in psychopathology, with relevance for psychosis-related symptoms.

Background and Aims: The corpus callosum (CC), the brain’s largest white-matter commissure, enables interhemispheric communication. As a key projection structure, CC alterations can affect both callosal fibres and the cortical regions they connect to, influencing cognitive functioning. This study aims to investigate structural CC characteristics as potential biomarkers of cognitive development. Methods: This study used cognitive, diffusion-weighted imaging, and genetic data from 1342 participants (ages 8–21) of the Philadelphia Neurodevelopmental Cohort. Cognitive functions were assessed via the Penn computerized neurocognitive battery, and five cognitive factors were derived using factor analysis. Single-shell diffusion-weighted images were preprocessed (QSIprep), diffusion-tensor modelling generated whole-brain maps of fractional anisotropy (FA) and mean diffusivity (MD), fiber orientation distributions peaks were estimated (QSIrecon). Using tractography-based segmentation (TractSeg), 7 CC streamlines were reconstructed. FA and MD were computed along 100 equidistant segments for each of the CC streamlines. Penalised functional regressions were used to assess associations between spatial variation in diffusion properties along CC tracts with age and cognition. For CC metrics significantly related to cognition, a candidate gene association study was performed using penalised functional regression, targeting 4 genes—DCC, CDH2, AKT3, GLI3— previously linked to developmental neurobiological mechanisms regulating CC formation. Results: Functional data analyses revealed spatially varying associations between along-tract CC characteristics, age, and cognitive factors. As hypothesized, specific SNPs within the candidate genes were found to be associated with CC characteristics. Conclusion: Spatial variability in CC characteristics is associated with age and cognition, with genetic factors contributing to the diversity of CC development.

Amblyopia is the most common developmpental visual disorder, where due to abnormal sensory experience the brain ignores input from the eye(s) despite their physical health. This influences the development of higher-level visual, attentional and motor skills of patients. Traditional treatment of patching is long and effective in only a proportion of the patients. Thus, new, more efficient approaches are necessary. Our crossover randomized controlled trial AMBER assessed potential benefits of a stereoptic video game in virtual reality (VR) headsets versus standard of care (SC) in people with residual amblyopia. The primary outcome was best corrected visual acuity (BCVA).

Of 14 participants, mean age 21.7±9.8y (8–35y) with residual amblyopia (strabismic, anisometropic or mixed), 13 completed the 2 treatment arms: 1) VR training (30min/d, 5d/week) and 2) Continuation of SC. Each arm lasted 2 months, followed by a 2-month washout, before the second treatment started (8 months/patient). Treatment order was randomized. In all 5 visits, BCVA was assessed with Landolt and Sloan charts (logMAR), stereoacuity with the TNO, Titmus, Lang II and Asteroid tests (arcsec), and reading speed with the MNread test (words/min). We analyzed the short-term (screening to end of treatment at 2 months) effects and VR compliance. Outcomes were quantified per patient in absolute values and % change relative to the first visit.

BCVA (Landolt test) post-VR improved in 8/13 people (3 children, 5 adults; 3 anisometropic, 3 strabismic, 2 mixed; mean improvement 39.6±58.6%) and in Sloan test in 3/13 (18.7±1.5%). BCVA (Landolt) post-SC improved in 4/13 (41±40%) and in Sloan in 4/13 (55±32%). Stereoacuity changes were highly variable in both arms, depending upon both the participant and the specific test: Some patients improved and others either unchanged or declined. Titmus test showed improvements in 6/13 participants (by 68.7±21.2%) post-VR vs. in 1/13 post-SC (20%). TNO and Lang II tests showed improvements in 1/13 each (by 50% and 88.9%, respectively) post-VR and in 1/13 (by 50% TNO) and 4/13 (by 68±16% Lang II) post-SC. Asteroid test showed no improvement post-VR and in 1/13 post-SC (4.2%). Reading improved post-VR in 1 mixed amblyopic adult (28%) and there were no improvements post-SC. VR compliance was high (13.9±5.2h = 69.5±25.9%).

The results suggest that VR-induced visual skill rehabilitation depended on patient age and compliance. Younger participants improved with only medium compliance (34-66%), while a few adults with high compliance (67-100%) also improved. Together, the results refine our understanding of gamified, technology-based novel interventions into sensory developmental disorders beyond the “one-size-fits-all” approaches, offering future targets for better understanding the mechanisms of VR-based vision treatment responsiveness.

Humour represents both a socio-cognitive mechanism emerging early in development and a resource for emotion regulation across the lifespan. In our works, we focused on slapstick humour, characterised by physical comedy and misfortunate situations. Previous ERP studies in adults indicated that victims’ facial expressions act as specific triggers of amusement (Manfredi et al., 2014, 2017, 2020). We investigated whether preschoolers show similar neurocognitive responses when recognising humorous misfortune and whether these responses are linked to the ability to recognise emotional expressions. Thirty 4- to 5-year-old children completed a facial emotion recognition task (EmoRec) and a slapstick humour recognition task (HumRec). Results revealed distinct neural correlates of humour (N170, LP) and face processing (N170, P300), and a positive correlation between early face-sensitive processing (N170 in EmoRec) and later humour-related processing (LP in HumRec). These findings suggest that the appreciation of slapstick humour in preschoolers is closely tied to the early processing of emotional facial expressions.

More broadly, humour is also a well-documented strategy for emotion regulation in neurotypical populations, supporting resilience through reappraisal, distraction, and the amplification of positive affect. In autism, however, humour use shows qualitative differences, including reduced adaptive functions and heightened gelotophobia, yet emerging evidence points to its potential as a selective but valuable resource for well-being. Taken together, our results highlight the dual role of humour: as a neurocognitive mechanism integrating emotion and social information early in development, and as a promising tool to foster emotion regulation and resilience across neurotypical and neurodiverse populations.