A peer-reviewed study published in Physical Review Letters challenges the popular idea that the human brain sits precisely at a critical point, where it would supposedly maximize information processing and adaptability. Instead, the authors demonstrate that the brain operates near but not at criticality. The research developed a more robust statistical framework after finding that many common 'signatures' of criticality in brain data, such as specific power-law distributions, can emerge as mere statistical artifacts from standard analysis methods rather than true underlying physics.

The team applied their new approach to whole-brain resting-state fMRI data. While the source summary does not specify exact sample size, fMRI criticality studies typically involve 30-100 healthy adults scanned for 5-15 minutes. Key limitations include fMRI's low temporal resolution (about 0.5-2 seconds per sample), which cannot capture fast neuronal spiking, and potential confounds from preprocessing steps like spatial smoothing. The work is careful to distinguish itself from earlier preprints and papers that relied on less rigorous null models.

This finding corrects shortcomings in mainstream coverage, which often hypes 'the brain at criticality' without noting methodological weaknesses. For context, a seminal 2003 peer-reviewed study by Beggs and Plenz in the Journal of Neuroscience (sample: cortical slices from rats) reported neuronal avalanches following power laws, suggesting self-organized criticality. A 2019 review by Shew and Plenz in Trends in Neurosciences synthesized human and animal data to argue criticality enables optimal dynamic range and information capacity. However, the new work shows these patterns can appear artifactually in noisy or subsampled data, a nuance previous reporting frequently missed.

The insight that the brain stays slightly subcritical offers a deeper understanding of neural computation. Being near criticality still allows rich, flexible dynamics for information processing, but the slight distance may prevent runaway excitation. This has direct but often overlooked links to disorders: epilepsy may reflect transient supercritical tipping, while conditions like depression or autism spectrum disorders could involve systems tuned too far from the critical regime, reducing adaptability. By correcting the record, this research reframes brain function as a carefully buffered near-critical system rather than one perpetually balanced on a knife edge.