How chronic pain reshapes our brain activity

Share this post:

The process of understanding and managing chronic pain is complex, because everyone’s experience with pain is unique.  In the United States, at least 116 million U.S. adults are burdened by chronic pain, with an economic cost of well above 500 billion per year[1]. And yet, as pervasive as it is, we’ve never had a way to measure pain objectively, nor have we really been able to study all the facets of what it means to be a pain patient.

In a recent collaboration with researchers at the University of Rochester[2], we published new research showing that the activity of the nucleus accumbens – a deep structure in the brain – is altered as pain becomes chronic, providing an objective marker of how this disorder reshapes brain function. Strikingly, the same observation could be repeated in very different groups of back pain patients, from different hospitals. Such a robust brain marker is a first in chronic pain and could possibly guide future clinical trials aimed at helping this vulnerable population.

Chronic pain is a complex phenomenon as there is wide variability in its presentation, disabling effects, duration and severity, and it has a strong emotional component beyond clear biological causes. These factors make objective measures of chronic pain difficult to isolate. Recently, there has been some progress in understanding patient characteristics that predispose them for chronic pain[3],[4]. However, understanding what happens in the brain as pain persists over time and reorganizes nervous function has remained elusive.

This new study, published in the Proceedings of the National Academy of Sciences (PNAS), evaluated chronic back pain patients, subacute back pain patients (whose pain had lasted less than three months), and a healthy control group. Our colleagues from the Yale School of Medicine and the University of Rochester followed the subacute group over time; some recovered, but — for others — pain became chronic.

Chronic pain shapes nucleus accumbens activity. A: In a subacute patient who recovers, the volume of the nucleus accumbens tends to be larger, and activity oscillations remain stable at follow up after recovery. B: A subacute patient who becomes chronic, with a smaller accumbens volume and activity oscillations that weaken as pain persists. C: Power of slow accumbens oscillations at follow-up correlates with pain visual analog scale (VAS) in subacute patients who recover (SBPr) and those with persistent pain (SBPp).

The question was whether there is any systematic change in the brain that is seen specifically when pain becomes chronic. As a way to study brain function, we measured brain activity using magnetic resonance images during the so-called “resting state.” Even when we are “doing nothing” (the resting state) our brain is always active, and the structure of this activity can provide valuable information about brain disorders.

One way to analyze brain activity is by studying the tendency of different pairs of brain regions to become active in synchrony during the resting state — their “functional” connectivity. We checked structures of the limbic brain (amygdala, the nucleus accumbens, hippocampus, the thalamus), deep regions inside the brain that process emotional inputs from sensory systems. They relay information to the cortex and other structures. The accumbens in particular is part of the striatum and is involved in processing motivation, aversion, rewards, and addiction.

We knew from past studies that the connectivity of these structures, particularly the accumbens, could determine the early risk for chronic pain in the subacute patients. This pattern makes sense in light of the behavioral changes that can accompany chronic pain: decreased motivation, problems in value-based decision making, disrupted perception of pleasure and satiety signals of tasty foods. We wondered if the function of the nucleus accumbens would be further affected as pain becomes chronic and the brain learns in the presence of this ongoing signal. One of the findings in the paper was specific to the accumbens connectivity, with connectivity alterations spreading from the accumbens to other limbic brain areas as pain became chronic.

In light of this, we proposed a way to quantify how abnormal the activity of the nucleus accumbens is, irrespective of its connectivity to the rest of the brain, by separating it in different timescales – fast versus slow oscillations.

Slow oscillations, which indicate changes of the order of minutes (really slow, keep in mind a neuron can fire up to a hundred times in one second), can be thought of as a measure of the switching between different states. We observed that chronic pain patients have slow oscillations so severely reduced in the accumbens that one could distinguish them from controls based on this one aspect of brain function.

What was also shocking is that this reduced slow oscillations was also seen in the patients who started in the subacute state, but became chronic. Further, we examined data from other studies and saw the same pattern. This is remarkable because a lack of robustness is often a recurring problem in neuroscience. Measures of brain activity are often not reproduced when patients are selected following different criteria, or studied with different scanners, but — in this case — this pattern was shared across all chronic pain populations.

This contribution to our understanding of chronic pain is among the first of several efforts that IBM is making in this field. From elucidation of brain mechanisms to finding better methods for quantifying how pain can affect patients’ lives, our scientists are committed to creating a truly integrative picture of one the most complex health disorders that arise as a byproduct of the modern lifestyle.


[1]      Institute of Medicine (US) Committee on Advancing Pain Research, Care, and Education, “Relieving Pain in America: A Blueprint for Transforming Prevention, Care, Education, and Research,” 2011.

[2]      M. M. Makary, P. Polosecki, G. A. Cecchi, I. E. DeAraujo, D. S. Barron, T. R. Constable, P. G. Whang, D. A. Thomas, H. Mowafi, D. M. Small, and P. Geha, “Loss of nucleus accumbens low-frequency fluctuations is a signature of chronic pain.,” Proceedings of the National Academy of Sciences, Apr. 2020.

[3]     E. Vachon-Presseau, P. Tétreault, B. Petre, L. Huang, S. E. Berger, S. Torbey, A. T. Baria, A. R. Mansour, J. A. Hashmi, J. W. Griffith, E. Comasco, T. J. Schnitzer, M. N. Baliki, and A. V. Apkarian, “Corticolimbic anatomical characteristics predetermine risk for chronic pain.,” Brain, vol. 139, no. 7, pp. 1958–1970, Jul. 2016.

[4]      M. N. Baliki, B. Petre, S. Torbey, K. M. Herrmann, L. Huang, T. J. Schnitzer, H. L. Fields, and A. V. Apkarian, “Corticostriatal functional connectivity predicts transition to chronic back pain.,” Nat Neurosci, vol. 15, no. 8, pp. 1117–1119, Jul. 2012.


Inventing What’s Next.

Stay up to date with the latest announcements, research, and events from IBM Research through our newsletter.


Guillermo Cecchi

Principal Research Staff Member

Paul Geha

Neuroscience, University of Rochester Medical Center

More Healthcare stories

Finding new uses for drugs with generative AI

New research published in the journal "Frontiers" demonstrates advanced AI algorithms successfully and rapidly modeling clinical trials to find new uses for existing drugs and therapeutics.

Continue reading

IBM researchers use epidemiology to find the best lockdown duration

In our recent paper, "Optimal periodic closure for minimizing risk in emerging disease outbreaks," published in PLoS One, we developed a technique to calculate the optimal duration of a periodic lockdown during an outbreak of an infectious disease where there is no cure or vaccine. Our findings are different from the lockdown duration being widely applied, today.

Continue reading

Speech-to-text AI could help doctors prescribe placebo to ease chronic pain

In a newly published paper “Quantitative language features identify placebo responders in chronic back pain” in the peer-reviewed journal PAIN, we report the first proof-of-concept that uses AI to analyze patients’ clinical trial experiences. The AI quantifies a placebo response in patients with chronic pain and distinguishes those who respond to placebo from those who do not.

Continue reading