Fork the consciousness, or download the project and create your own.

Nicholas Schiff and Central Thalamic Stimulation Years After Brain Injury

Can a brain that was badly injured years ago be helped to work better by stimulating the circuits that remain? Neurologist Nicholas Schiff of Weill Cornell Medicine and neurosurgeon Jaimie Henderson of Stanford tested this in people with lasting deficits after moderate to severe traumatic brain injury. Deep brain stimulation of the central lateral nucleus of the thalamus improved processing speed on a standard test by 15 to 52% in all five people who completed the trial, 3 to 18 years after their injuries. The paper is Schiff, Giacino, Butson and colleagues, Thalamic deep brain stimulation in traumatic brain injury, a phase 1, randomized feasibility study, Nature Medicine 29, 3162-3174, December 2023 (doi:10.1038/s41591-023-02638-4).

The trial does not restore consciousness in an unresponsive patient. It shows something narrower and still important. Circuits that remain after an injury can work better when they are driven from outside, long after recovery is thought to have stopped.

From one minimally conscious patient in 2007

The trial builds on a single-patient study that Schiff and colleagues published in Nature in 2007 (doi:10.1038/nature06041). A man had remained in a minimally conscious state for six years after a traumatic brain injury. Disorders of consciousness that last more than a year after such an injury were generally considered permanent. In a six-month double-blind crossover design, bilateral stimulation of the central thalamus increased specific cognitively mediated behaviors, limb control and oral feeding during periods with the stimulator on compared with periods with it off.

The hypothesis behind both studies is that some patients keep large networks that are chronically underactive and could still be recruited. The central lateral thalamus sends projections to the frontal cortex and the striatum and helps regulate arousal and executive control. If those projections can be driven, the remaining networks may work closer to their capacity.

The CENTURY-S trial

The 2023 trial enrolled six adults who had a moderate to severe traumatic brain injury at least two years before, persistent cognitive problems, and moderate disability that kept them from returning to their previous work or study. The team used each person’s own imaging and a biophysical model to place electrodes so that stimulation would activate the central lateral nucleus and an associated fiber pathway, the medial dorsal tegmental tract.

Phase Design
Baseline Staggered baseline periods across participants
Titration 14 days to find each person’s stimulation settings
Treatment 90 days of open-label stimulation, 12 hours per day
Withdrawal 21 days, randomized and double-blind, stimulation continued or turned off

All six people were implanted safely. One developed a scalp infection after surgery, had the system removed and left the study, and adverse events resolved without lasting effects.

Results

The primary measure was completion time on Part B of the Trail Making Test, a timed task that requires switching between numbers and letters. It measures processing speed and executive control, the problems that most limit people after this kind of injury.

  1. All five participants who completed the trial passed the preset benchmark of 10% improvement. The average improvement was 31.75%, with a range of 15% to 52%.
  2. The two participants with the largest deficits improved most, by more than 40%, and two participants whose baseline scores were already in the normal range improved by more than 20%.
  3. Related measures also improved, including Part A of the test by an average of 30.7%, and self-reported executive function by an average of 32.7%.
  4. In the withdrawal phase, three participants agreed to take part. The one person randomized to have stimulation turned off showed a 34% decline in test performance, consistent with the effect depending on the stimulation.
  5. On the secondary measure of fatigue, two people improved, two were stable and one declined.

A separate interview study with the same participants and their families, led by ethicist Joseph Fins, describes better emotional regulation and changed family relationships, along with concern about who pays for devices and battery changes after a trial ends (Cambridge Quarterly of Healthcare Ethics 2024, doi:10.1017/S0963180123000518).

What this tells us about consciousness

Consciousness after brain injury is often described in terms of how much tissue survived. This trial and the 2007 case point to a second factor, how strongly the surviving circuits are driven. The central thalamus regulates arousal and the engagement of frontal networks. Raising its output changed what the remaining cortex could do years after injury, without adding any new tissue.

That fits theories that place the thalamus at the center of conscious access. Our article on thalamic gating and criticality in conscious perception covers intracranial evidence that early high-order thalamic activity gates whether a stimulus reaches awareness. Stimulation of the central lateral nucleus is a way to act on that gate.

It also bears on restoring consciousness, the question behind our consciousness fingerprint page. Restoring a lost person would need two things, the repertoire of dynamics that made them who they are, and the drive to move through it. Thalamic stimulation addresses the second part. It cannot rebuild a repertoire that the injury destroyed. The trial participants were conscious and could consent, so the study does not show recovery of consciousness itself, only better use of the capacity that remained.

Comparison to The Consciousness AI

The Neutral Core slow handover section names two routes for restoring function after brain damage. One gives the brain new tissue, as in the organoid and neuron graft studies. The other drives the circuits that remain, and this trial is the human evidence for it. The Core is a simulation and has not been connected to a living brain. The trial still offers a design lesson. A system can have the right structure and still fail for lack of drive from the regions that regulate arousal, so any synthetic replacement would need that drive as well as the right connections.

Limits of the thalamic stimulation trial

Nicholas Schiff and colleagues showed in a phase 1 trial that central thalamic deep brain stimulation is feasible and may improve executive control years after moderate to severe traumatic brain injury. The authors name the main limit themselves. The effect on the primary measure is large, but only five people completed the trial. The treatment phase was open label, only one person was randomized to stimulation off in the blinded withdrawal, and the study could not compare lesion types or mechanisms. The 2007 study was a single patient. Larger controlled trials are needed before any clinical claim, and neither study restores consciousness in a patient who has lost it. Within those limits the work belongs with the measurement debates in how scientists race to define and test AI consciousness in 2026, since judging a restored brain needs measures of both function and awareness.

Researchers covered here