Concussions — essentially mild traumatic brain injuries — are among the most common injuries sustained on athletic fields. In the U.S., an estimated 3.8 million sports-related concussions occur every year. But new research by Stanford Medicine scientists and collaborators suggests that the prevailing understanding of the forces that lead to concussion may be missing the mark.
Their new study analyzed data from motion-sensing mouthguards worn by athletes in seven different sports and found that straight-line motion of the head, measured by linear acceleration, is more predictive of concussion than turning motion of the head, measured by rotational acceleration.
Linear acceleration tends to dominate when an impact force is directed to the head’s center of mass. In mixed martial arts, it could be a jab between the eyes; in lacrosse, a stick striking the side of the head; and in football, a collision to the top of the head when the head is lowered, or a fall backward and the head hitting the ground.
The new findings turn nearly 85 years of dogma in the field — which held that rotational motion was the primary driver of concussions — on its head. The results, if confirmed by more studies, could have implications for helmet design and safety guidelines in sports.
“Not only is the linear acceleration predictive of concussion, the threshold is lower than people had previously thought in terms of the levels that are unsafe,” said Jessica Towns, a graduate student in bioengineering and lead author of the study to publish Nov. 9 in the Annals of Biomedical Engineering .
An enduring hypothesis
The prevailing theory of head injury biomechanics can be traced to World War II-era work by the Scottish physicist A.H.S. Holbourn. He proposed that impacts that rotate the head, such as an uppercut to the chin in boxing, are particularly injurious to the brain as a result of shearing — the stretching and straining of the brain as different parts of the heavy but gelatinous organ rotate at different speeds. Linear motion does not cause shearing and was considered less dangerous.
In the following decades, some animal studies seemed to bear out what became known as the Holbourn hypothesis. And today, the most commonly used rating system for helmets in a wide range of sports prioritize reducing rotational acceleration.
“Holbourn’s reasoning was based just on mechanics; it wasn’t biology,” said David Camarillo , PhD, associate professor of bioengineering in the schools of medicine and engineering and senior author of the study.
Until recently, reliable measurements of head impacts in humans were difficult to come by.
“The only data that was available was from helmet-mounted sensing systems,” Camarillo said. “The problem with the helmet is it moves around a lot on your head, so it doesn’t give you a direct measurement of the head.”
Motion-sensing mouthguards
For more than a decade, Camarillo’s lab has helped develop motion-sensing technology that can be incorporated into mouthguards, where they can more accurately capture head motion. Variations of these mouthguards are now used in many sports to help monitor for concussions in athletes.
“The question we started out with many years ago was, ‘What are the forces required to cause concussion?’ If we understand the nature of those forces and the limits of the human brain, then it might be possible to prevent concussions,” he said.
The new study analyzes the largest and most diverse set of mouthguard data to date, including 203 male and female athletes spanning seven sports, from lacrosse to mixed martial arts to football. It includes 47 impacts that led to concussions and 3,804 impacts that did not. (The impacts were videotaped and concussion was determined either by on-site medical staff or by later review of video for visual signs of injury, such as disorientation or balance disturbance.)
When the team set out to determine the forces that contribute to concussion using the new data, they expected to confirm the Holbourn hypothesis.
“But we found the opposite,” Towns said. “We found that linear head motion was more predictive of concussion than rotational head motion.”
Specifically, in their mathematical model, the researchers found that peak linear acceleration was the single most predictive factor in whether an impact would lead to concussion. Adding in peak rotational acceleration did not improve their model, but adding a related variable known as peak rotational velocity slightly improved the model’s precision. (That may be caused by the short duration of the impacts in the study, the researchers said. In longer impacts, such as whiplash, the rotational acceleration, which is the change in rotational velocity over time, might be a more significant factor.)
Underestimated risk
In real life, almost every collision results in both linear and rotational head motion, Camarillo said. He noted that the new findings do not mean that rotational motion isn’t harmful, only that it was less predictive of concussion than linear motion. It’s possible that the impacts included in the study tended to have larger linear motion than rotational motion.
The findings do suggest that safety guidelines for helmets, from football to cycling, could be underestimating the risk of linear acceleration.
The researchers found that head impacts with a linear acceleration of 100 g (100 times the standard acceleration because of gravity) carry a 50% risk of concussion. The average head impact in football is about 25 g.
“If 100 people were each hit at 100 g, half of them would get a concussion,” Towns said. “That’s important because the thresholds that have been used for helmets to be sold on the market are usually much higher than 100 g.”
An often-cited threshold for 50% concussion risk, derived from helmet-mounted sensors, is 192 g. In contrast, the new mouthguard data showed that a collision of 150 g nearly guarantees a concussion, with a risk of more than 99%.
“There are still too many concussions happening,” Camarillo said. “Why are there so many concussions even when people are wearing helmets? Part of the answer could be in this paper, which is that the linear accelerations that we previously thought were safe, we now see are much more dangerous.”
The researchers also analyzed a few concussive impacts in athletes younger than 14 and found the threshold for concussion is likely much lower for children. “There’s an urgent need to get more of this data on kids because that might end up being the biggest problem,” Camarillo said.
New opportunities
Recent innovations in concussion safety, such as slippery helmet linings, have focused on reducing head rotation. An optimistic take on the new findings, Camarillo said, is that they offer a new lever to improve safety.
“If linear acceleration is its own distinct mechanism of concussion and it’s very common, then we’ve been missing a lot,” he said. “We haven’t tried as hard to optimize for this variable. It gives me some hope that there’s a new opportunity for preventing concussions.”
The researchers tested football helmets with new liquid-filled pads, comparing them against standard football helmets on crash dummies in the lab. At all impact levels, they found that the liquid technology cut the risk of concussion by reducing peak linear acceleration. At the highest impact levels tested, for example, concussion risk was 73.3% for standard helmets and 55.7% for helmets with liquid-filled pads. (Camarillo is a co-founder of SoftShox, the company that licensed the Stanford technology and makes the liquid-filled pads.)
The biomechanics of how linear acceleration damages the brain is still unclear. It could be the brain hitting the skull, or perhaps a process known as cavitation, in which pressure changes inside the brain causes tiny bubbles to form and collapse. The researchers are developing various physical and computational models to better understand what is happening to the brain.
Despite the large dataset in the new study, it still represents a relatively small number of concussions, Camarillo said. It will take more studies by his team and others to verify the results, convince skeptics and ultimately improve concussion safety in sports.
“I would say this is the beginning, not the end,” he said. “This is an 85-year-old theory, and this is the first paper that has really challenged it.”
Researchers from SoftShox, Monash University, Deakin University, Duke University and Alfred Health contributed to the work.
The study received funding from the National Institutes of Health (grants R24NS098518; R43NS119134; R44NS119134), the Taube Stanford Children’s Concussion Initiative, the Pac-12 Conference’s Student-Athlete Health and Well-Being Initiative, Stanford Medicine’s Maternal and Child Health Research Institute , the Office of Naval Research, and the National Science Foundation Graduate Research Fellowship Program.
# # #
About Stanford Medicine
Stanford Medicine is an integrated academic health system comprising the Stanford School of Medicine and adult and pediatric health care delivery systems. Together, they harness the full potential of biomedicine through collaborative research, education and clinical care for patients. For more information, please visit med.stanford.edu .
Annals of Biomedical Engineering
Experimental study
People
Limitations of Rotational Acceleration for Predicting Concussion: A New Injury Risk Function
9-Nov-2026