1.Nuclei-specific hypothalamus networks predict a dimensional marker of stress in humans

DOI: 10.1038/s41467-024-46275-y

https://www.nature.com/articles/s41467-024-46275-y

Researchers utilized high-resolution resting-state fMRI scans from 498 participants to investigate the hypothalamus, a key player in the stress response system. Despite its small size and complexity, they managed to delineate seven distinct nuclei within the hypothalamus in living humans. By examining the connectivity patterns between these nuclei and other brain regions, particularly the amygdala, they successfully predicted stress levels in individuals. Their findings suggest that specific changes in connectivity within subcortical networks, rather than overall hypothalamic activity, are associated with stress. These insights may pave the way for targeted interventions in stress-related disorders.

2.Natural language instructions induce compositional generalization in networks of neurons

DOI: 10.1038/s41593-024-01607-5

https://www.nature.com/articles/s41593-024-01607-5

The study investigates how humans interpret linguistic instructions to perform new tasks, even without prior experience. The scientists develop a neural model using natural language processing techniques, trained on various tasks with instructions provided by a language model. Remarkably, their best model achieves an average accuracy of 83% on unseen tasks, relying solely on linguistic instructions (zero-shot learning). They discover that language influences sensorimotor representations, enabling shared activity patterns for related tasks, guided by semantic instruction representations. The model can generate task descriptions based solely on motor feedback, which can then guide another model to perform the task. Their findings suggest how language shapes cognitive flexibility and offer testable predictions for understanding language's role in facilitating general cognition.

3.The dynamic state of a prefrontal-hypothalamic-midbrain circuit commands behavioral transitions

DOI: 10.1038/s41593-024-01598-3

https://www.nature.com/articles/s41593-024-01598-3

The study looks at how the brain decides which behaviors to do next. The researchers find that a part of the brain called the lateral hypothalamus (LH) helps predict what behaviors might happen next by showing specific activity patterns during transitions between behaviors. When they block this brain activity, it disrupts how behaviors change. They also find that signals from another brain area called the prefrontal cortex help trigger these transitions. Dopamine neurons, which are important for motivation and reward, also play a role in signaling behavior changes. This process helps the brain quickly switch between different actions, like socializing or eating, in a smart and adaptable way.

4.Genetics of environmental sensitivity and its association with variations in emotional problems, autistic traits, and wellbeing

DOI: 10.1038/s41380-024-02508-6

https://www.nature.com/articles/s41380-024-02508-6

Greater environmental sensitivity is linked to higher susceptibility to mental health issues under stress and lower subjective wellbeing. However, it also associates with reduced emotional problems in favorable environments and correlates positively with autistic traits. This study examined over 2800 adolescent twins to explore the genetic factors behind sensitivity and its relationships with emotional problems, autistic traits, and wellbeing. Results showed that heightened sensitivity correlated with increased emotional problems, autistic traits, and decreased wellbeing, mainly due to shared genetic influences. Importantly, genetic factors related to sensitivity explained a significant portion of variations in emotional problems, autistic traits, and wellbeing, suggesting their importance in understanding mental health heterogeneity. These findings advocate for integrating sensitivity genetics into future studies to better comprehend the complexities of emotional problems, autistic traits, and wellbeing.

5.Microglial over-pruning of synapses during development in autism-associated SCN2A-deficient mice and human cerebral organoids

DOI: 10.1038/s41380-024-02518-4

https://www.nature.com/articles/s41380-024-02518-4

Autism spectrum disorder (ASD) affects 1 in 36 children in the US, and while neuron studies have dominated ASD research, altered neuro-immune responses may be crucial. Microglia, brain immune cells, regulate brain development, and synaptic function. While ASD is typically seen as polygenic, SCN2A deficiency emerges as a leading monogenic cause. A mouse model lacking Scn2a exhibits behavioral and neuronal issues, with impaired learning and memory, synaptic transmission, and spine density in the hippocampus. These mice show heightened microglial activity, excessively pruning synapses, particularly during critical developmental stages. Suppressing microglia partly restores synaptic function. Similar findings are observed in human cells carrying an SCN2A mutation. This study highlights microglial involvement in ASD pathogenesis across species, offering insights into potential therapeutic interventions.