For years, scientists assumed psychedelics scrambled the brain's normal communication highways into random noise. A new study challenges that view with help from artificial intelligence. Researchers at Monash University used AI to analyze brain scans from 62 first-time psilocybin users and found that the activity, while distinct from the sober state, followed recognizable patterns. Neuroscientist Devon Stoliker, lead author of the study published in Nature, argues that the chaos model fails to explain why users often report meaningful insights and lasting psychological benefits.

What You Need to Know

The study, called PsiConnect, used machine learning to compare brain scans taken under the influence of 19 milligrams of psilocybin and scans taken while sober. The AI identified structured patterns that the researchers say represent a different kind of organization, not pure disarray. This finding could help explain why psychedelic therapy shows promise for conditions like depression and anxiety. The research adds to a growing field that uses AI to decode complex neural activity.

Rethinking the Chaos Model

Stoliker and his team recruited 74 participants, but only 62 completed the full scanning protocol. Each received a controlled dose of psilocybin, the active compound in certain fungi, and underwent functional MRI and EEG recordings. The resulting data was fed into an AI model trained to distinguish between drugged and sober brain states. The model did not just see noise. It detected consistent shifts in how brain networks coordinated their activity, suggesting a reconfiguration rather than a breakdown.

  • Network integration: Normally separate networks, such as those handling vision and self-awareness, showed increased communication.
  • Reduced segregation: The brain's default mode network, linked to ego and self-referential thought, became less dominant.
  • Nonrandom activity: The AI classified brain states with high accuracy, proving the activity was structured.

Implications for Psychedelic Therapy

If the psychedelic brain is not chaotic but restructured, that could explain why users often experience breakthroughs. The temporary dismantling of rigid network boundaries may allow new connections to form, leading to lasting changes in perspective. Stoliker said the chaos model "never really explained why an individual would have a meaningful experience" or how those experiences translate into clinical improvements. This new framework aligns with what therapists have observed: that a guided psychedelic session can help patients break free from entrenched thought patterns.

Why This Matters

Psychedelic research is experiencing a revival, with clinical trials exploring psilocybin for treatment-resistant depression, PTSD and addiction. Understanding the underlying brain mechanisms is critical for designing effective therapies and gaining regulatory approval. The AI-driven approach used by Stoliker's team offers a replicable method to quantify neural changes, potentially accelerating the development of personalized treatment protocols. If hidden order exists in the psychedelic state, it opens the door to targeted interventions that maximize therapeutic benefits while minimizing risks. Regulators and healthcare providers will need this level of precision to responsibly integrate psychedelics into mainstream medicine.