Subscribe to eTOC

Lesion Network Mapping: Discovering the Common Circuits Underlying Neurologic Mysteries

Article In Brief

A leading investigator describes how lesion network mapping—a new approach to understanding brain circuits—helped solve four mysterious neurologic cases.

Acute-onset visual hallucinations, an unfamiliar home, sudden criminality, and essential tremor that disappeared overnight are four neurologic mysteries, each of which was solved with “lesion network mapping,” a new approach that uses the human brain connectome to understand brain circuits and discover new targets for therapy.

The pioneering case studies that led to the development of this approach were recounted by Michael D. Fox, MD, PhD, associate professor of neurology at Harvard Medical School and director of the Laboratory for Brain Network Imaging and Modulation at Beth Israel Deaconess Medical Center in Boston, at the Hot Topics Plenary Session at the AAN Annual Meeting.

“Much of neurology is based on the concept that we can map symptoms to damaged brain regions,” Dr. Fox said, citing the canonical examples of left frontal cortex lesions causing Broca's aphasia, and bilateral hippocampal lesions leading to memory loss. “The vast majority of lesions that cause Broca's aphasia are not in Broca's area, and the vast majority of lesions that cause memory deficits are not in the bilateral hippocampus.

“So what do you do when you try to map symptoms to a damaged region, but the region that is damaged doesn't actually make sense for the symptoms the patient is presenting with?”

Visual Hallucinations

The patient that offered the first clue, “the one that prompted this entire field”, Dr. Fox said, experienced sudden onset of visual hallucinations, which an MRI showed was due to a stroke in the medial thalamus. But Dr. Fox, and his fellow Aaron Boes, MD, could not figure out how damage in this area could account for the patient's symptoms.

Dr. Boes turned to the literature and found 22 other cases of acute-onset visual hallucinations tied to thalamic or brainstem lesions, and found a focal spot where damage overlapped in some of the cases, including his patient.

“But two things really bothered Dr. Boes,” Dr. Fox said. First, only six of the 23 cases shared this lesion location, and second, “even if all 23 cases had overlapped in the medial thalamus, as far as we know, this part of the thalamus has nothing to do with vision and visual imagery. In fact, this is a common problem in lesion mapping.”

That was when Dr. Boes turned to the Human Connectome. “For the last decade the field of neuroimaging has been collaborating to build a wiring diagram of the human brain,” Dr. Fox explained, based on high-resolution MRI images of thousands of healthy subjects. This has allowed construction of detailed maps of brain networks, “identifying which regions are anatomically connected, and more important, which regions are functionally connected.”

Dr. Boes's insight was that, instead of trying to find a common single spot among the 23 cases, “maybe we should be trying to map lesion locations to a common brain network.” And so the team asked for each case, what is this patient's specific lesion connected to? “And all of a sudden, things started to make a lot more sense,” Dr. Fox said. Dr. Boes saw that 22 of the 23 lesions were all connected to a specific area in the extrastriate visual cortex, “the exact part of the brain we thought should be responsible for visual hallucinations, because it is known to be involved in visual imagery.”

Delusions of Familiarity

Dr. Fox's second case involved a woman who was four months post-stroke and had made an almost full motor recovery, but complained that her house was not her own, to the point that she would sometimes pack her bags and prepare to move back to her “real home.”


“The vast majority of lesions that cause Brocas aphasia are not in Brocas area, and the vast majority of lesions that cause memory deficits are not in the bilateral hippocampus. So what do you do when you try to map symptoms to a damaged region, but the region that is damaged doesnt actually make sense for the symptoms the patient is presenting with?”—DR. MICHAEL D. FOX

Ryan Darby, MD, searched the literature to find 17 lesions that caused similar “delusions of familiarity,” but again there was no single brain region of overlap. But when he looked at the network of connection, “things lined up,” said Dr. Fox. “All 17 cases were connected to a specific spot in the retrosplenial cortex, the exact spot that lights up whenever you see a familiar versus unfamiliar object.”

But while that made sense to account for the loss of familiarity, it didn't explain the delusions, the disconnection from reality that made that loss so troubling. The team found that, in 16 of 17 cases, the lesions were also connected to a common spot in the right frontal cortex, which is responsible for reality monitoring, “the area that is supposed to light up any time something doesn't make sense to you,” Dr. Fox said. “So a single brain lesion that is connected to both your familiarity detector and your reality monitor, appears capable of disrupting both functions, and causing a complex delusion of familiarity.”

Dr. Darby also pursued a study of the network localization of criminality, stemming from the 1966 case of Charles Whitman who shot 31 people from a bell tower on the campus of the University of Texas at Austin, and who was discovered at autopsy to have a tumor in the medial temporal lobe. Dr. Darby identified 16 cases of focal brain damage temporally linked to new-onset criminal behavior (not including Mr. Whitman, for whom neuroimaging data was not available).

“While no single brain region was lesioned in all sixteen cases, all sixteen were linked to a network involving the right frontal cortex, “the exact set of brain regions that light up when you do a moral decision-making task,” said Dr. Fox, and the association was “quite specific to lesions associated with new-onset criminality,” not to other types of neuropsychiatric syndromes.

No Essential Tremor Post-Stroke

Finally, Dr. Fox described a patient with longstanding and slowly progressive essential tremor, who awoke one morning with a stroke that had cured his movement disorder. Juho Joutsa, MD, PhD, in Dr. Fox's lab, found 10 other similar cases.

“These eleven cases should be able to tell us something about where we should intervene to help other patients with essential tremor,” Dr. Fox said, but the appropriate target was unclear from looking at lesion overlap alone. But they found that all the lesions were involved in a network that included a single spot in the thalamus, “and in fact that spot was exactly the target we had converged on as a target for deep brain stimulation, after a decade of trial and error.

“Our hope is that this approach, making sense of spontaneous brain lesions that provide therapeutic benefit, can help us identify new therapeutic targets in a more systematic manner.” Lesion network mapping is probably not applicable to conditions of diffuse brain damage, such as traumatic brain injury, but appears to be a useful way to understand symptoms that are caused by focal brain damage, he said.

Expert Commentary

“Most neurologists go into the field because we like to solve puzzles,” commented Natalia Rost, MD, FAAN, director of the Acute Stroke Services at Massachusetts General Hospital and professor of neurology at Harvard Medical School. “The real practical application, as in the essential tremor case, would be to find a place you can intervene to have a therapeutic effect. I think what Dr. Fox has discovered is a very exciting way to remap some of the mysteries of neurology.”


Dr. Fox has intellectual property on using brain connectivity to guide brain stimulation but collects no royalties.

Link Up for More Information

• Fox MD. Mapping symptoms to brain networks with the human connectome N Engl J Med 2018; 379:2237–2245.
• Boes AD, Prasad S, Liu H, et al. Network localization of neurological symptoms from focal brain lesions Brain 2015;138(Pt 10):3061–3075
• Darby RR, Laganiere S, Pascual-Leone A, et al. Finding the imposter: brain connectivity of lesions causing delusional misidentifications Brain 2017;140(2):497–507.
• Darby RR, Horn A, Cushman F, Fox MD. Lesion network localization of criminal behavior Proc Natl Acad Sci U S A 2018;115(3):601–606.
• Joutsa J, Shih LC, Horn A, et al. Identifying therapeutic targets from spontaneous beneficial brain lesions Ann Neurol 2018 84(1):153–157.