A 2026 quantitative MRI preprint reanalyzed 138 healthy adults and found that multivariate modeling detected coordinated age-related shifts across myelin-sensitive, iron-sensitive, and water-sensitive brain maps.1 The signal was broad, but split-half validation weakened the result enough to keep it in the measurement-method lane rather than the clinical-test lane. Research Highlights 138 adults were reanalyzed: The …
A 2026 NF1 mouse-model study found a gut-brain signal that was strongest in males: Nf1 +/- mice had altered gut microbiome composition, higher relative brain mass in 2 cohorts, reduced sociability, and combined autism-like phenotype shifts, with male microbiome beta diversity differing from controls at p = 0.0120.1 The result is a preclinical mechanism lead, …
A 2026 rat study found that foot-shock stress transiently reduced food-cue approach during the first 2 conditioned-stimulus trials (t = 2.807; P < 0.05), and inhibiting the nucleus of the solitary tract weakened that stress effect.1 The useful finding is selective: stress used an NTS-linked route to dampen cue-evoked VTA dopamine, while outcome-specific satiety reduced ...
A 35-person EEG study found that pain empathy did not collapse when live two-way video was replaced by prerecorded viewing: observer-target pain-rating correlations were nearly identical during video calls and recordings, r = 0.65 vs. r = 0.64, both p < 0.001.1 The calibrated read is narrow but useful: temporal immediacy was not required for …
A 97-dyad adolescent fMRI study found that trauma-exposed youth with high posttraumatic stress symptoms did not use parent-observed safety learning in the usual way: their arousal rose to a safety cue during vicarious extinction, F(1,159) = 4.81, p = 0.029, and parent-child physiological synchrony no longer predicted lower skin conductance during early safety learning.1 The …
A 2026 ENIGMA-OCD preprint tested a transformer model on resting-state fMRI from 1,706 participants and reached AUROC = 0.653 ± 0.039 for OCD classification.1 The more important result was the failure mode: held-out-site performance ranged from AUROC 0.427 to 0.819, which is exactly the gap clinical AI has to close before brain-scan diagnosis is credible. …
A 2026 neuroscience review argues that chemotherapy-induced peripheral neurotoxicity should not be treated as a purely peripheral nerve problem: CIPN affects 70-90% of patients during chemotherapy, persists chronically in 20-30% of survivors, and may interact with central sensitization, pain networks, mood, sleep, and chemobrain.1 The strongest claim is mechanistic connection, not proof that neuropathy alone …