Darbe Schlosser
THOUGHT PAPER / MOTORVATION
BEFORE THE FREEZE
Why Parkinson’s rehabilitation may be asking the wrong question about freezing of gait
A dynamic systems perspective on movement, learning, and stability, drawn from more than 14,000 hours of one-on-one work with people who have Parkinson’s. This is not a finding. It is a hypothesis, and an invitation to ask better questions.
I The wrong question
When someone with Parkinson’s freezes, the instinct is to ask why. We look for the trigger: the doorway, the turn, the crowded room, the divided attention. These triggers explain less than they appear to.
The same person has walked through that same doorway a thousand times. Until one day, they cannot.
If the doorway were the cause, the freeze would have happened the first time. It did not. Which means the doorway is not the cause. It is the final straw on a system that had already changed.
So perhaps the more useful question is not why did this person freeze, but what changed inside the nervous system that made freezing possible in the first place.
A trigger only works on a system that is already vulnerable. If that is true, the freeze is not the start of the problem. It may be the first moment the problem becomes visible.
II The story almost everyone tells
I have heard a version of the same story more times than I can count. I was doing well. I exercised every day. I stayed active. Then suddenly, everything changed.
That moment of change is almost always when freezing began. And almost everyone describes it as sudden.
But “sudden” is a statement about perception, not about mechanism. A hurricane is not sudden. Long before anyone names a storm, thousands of small atmospheric interactions are already unfolding. By the time we see it, the storm is not beginning. It is simply the first thing large enough to notice.
What if freezing works the same way? What if the freeze is not the start of a decline, but the first visible sign of a movement system that has been slowly losing stability for months, or years?
I am not claiming Parkinson’s is a chaotic system in the formal mathematical sense. That has not been demonstrated. The hurricane is a metaphor, borrowed from nonlinear dynamics, and I am using it only to raise testable questions, not to assert a mechanism. It points at one real and testable question: if freezing is the storm, what was the weather doing for the year before?
III Movement as a dynamic system
To take that question seriously, it helps to stop thinking about movement as a set of muscles firing independently, and start thinking about it as a set of relationships that have to be continuously balanced. This is not a new way of seeing movement: motor coordination research already treats action as coupled variables shaped by feedforward and feedback control, rather than as isolated muscles working on their own.
In healthy movement, space and time are coupled. If you want to move bigger, you naturally slow down enough to cover the distance. If you want to move faster, the size of the movement usually has to change. Running and marching are not two separate activities so much as two solutions to the same underlying trade-off: fast with smaller excursions, or slower with larger ones.
A healthy nervous system organizes this automatically, thousands of times a day, adjusting joint angles, stride length, cadence, weight transfer, force, and timing together so the whole pattern stays efficient. This constant re-optimization may be one of the quiet foundations of movement stability.
Here is the hypothesis at the center of this paper: what if Parkinson’s gradually impairs this ability to re-optimize? Not the ability to move, but the ability to keep adjusting the relationships between the parts of a movement as context changes.
If that is true, the consequences would be subtle at first. The movement would still happen. The person would not fall. It might even look fine. But underneath, the parameters would no longer be adapting together the way they should.
IV What is the nervous system actually learning?
For decades, the dominant message in Parkinson’s rehabilitation has been clear and largely correct: move more, walk more, exercise harder, raise the heart rate, stay active. I believe in this. The evidence for exercise is strong, and nothing here should be read as an argument against it.
But underneath that message is a question we rarely ask. When a person is moving hard, what is their nervous system actually learning?
Consider a boxing-style session, the kind common in our field. I offer it as an illustration, not as a criticism of any specific program. The instruction is to punch bigger, punch faster, work harder. Heart rate climbs. Sweat pours. By every visible measure, it is a successful session. But did the person learn a more stable way to move? Or did they repeat, hundreds of times, a pattern in which timing, trunk rotation, joint angles, and effort no longer matched each other?
Healthy scaling
Wants a bigger punch. Slows the movement slightly, rotates the trunk more, shifts weight, and lets hips, shoulder, elbow, and hand scale together. The whole pattern stays coordinated.
Compensated effort
Is told to go bigger and faster. Drives from the arm alone, never adjusts the timing, effort rises while coordination falls. The punch still lands, but a less stable pattern is being rehearsed.
If the second pattern is repeated often enough, the nervous system is not practicing strength. It is practicing a less stable movement strategy, possibly without anyone noticing, including the person performing it.
I will admit something that is difficult to put in a paper. I see a version of this almost every day. Someone posts a video of a person with Parkinson’s exercising, and the comments fill with encouragement. That encouragement is understandable, and often deserved: the person is showing up, working hard, which for many is the biggest hurdle. Everyone is moving, everyone is celebrating, everyone is encouraging one another. But while others see progress, my eye keeps catching something else: timing that no longer matches amplitude, effort rising while coordination quietly falls away, a system that looks busy rather than stable. I cannot prove what I am seeing, and I hold it as a hypothesis rather than a verdict. But after enough hours, the pattern becomes hard to unsee.
I watch the early conditions of a hurricane forming, while everyone celebrates the sunny weather.
So I have to ask the uncomfortable version of the question. Am I the only one seeing this? Or has the field grown so enthralled with exercise, which is comparatively easy to study, easy to fund, and easy to promote, that we have stopped asking the harder questions beneath it? Movement quantity is simple to measure and simple to cheer for. Movement quality is neither. My worry is that we have optimized for the part that is easy to see.
V The error you never felt
There is a second hypothesis underneath all of this: that people with Parkinson’s may have a reduced ability to detect their own subtle movement errors. This is not a wild guess: people with Parkinson’s are known to show altered proprioception and reduced awareness of certain non-motor changes, so the idea that some subtle movement errors might go unregistered is at least consistent with what is already observed.
In healthy movement, you make small errors constantly and correct them just as constantly. The nervous system compares what you intended with what actually happened, registers the mismatch, and adjusts. This loop runs silently, beneath awareness.
Now imagine that loop is blunted, just slightly. Not enough to cause a fall. Just enough that small mismatches stop being noticed. Each uncorrected repetition is not a catastrophe. It is a tiny compensation, a micro-adaptation, barely anything on its own. But practiced thousands of times without correction, those compensations may slowly become the system’s new default.
Practice does not only reinforce good patterns. It reinforces whatever pattern is actually being repeated, monitored or not.
This reframes the central concern. The risk may not be exercise. The risk may be unmonitored repetition under impaired self-correction.
VII The nonlinear part
This is where nonlinear dynamics enters.
A system that accumulates small inefficiencies does not necessarily decline in a smooth, proportional way. It can tolerate a great deal for a long time while looking stable from the outside. The person keeps walking, keeps functioning, keeps doing well. And then, past some threshold, the same system can transition abruptly into a qualitatively different state.
In dynamical terms, freezing looks less like a symptom that switches on and more like a state transition: the movement system slipping from one mode of organization into another. Some existing models already frame freezing of gait as a network-level disturbance or a change of state rather than a purely local, peripheral event; what this lens adds is to carry that same logic out of the brain-network picture and into the practice environment, where training and daily movement may be nudging the system toward or away from those transitions.
Seen this way, a trigger like a doorway or a turn does not cause the freeze. It is a small perturbation that happens to arrive when the system is already close to a tipping point. The disproportion is what you would expect from a system operating near a threshold.
I will say again what I said before: this is an analogy and a lens, not a proof. The lens earns its place not by being true, but by changing the questions we ask.
VII What this would change
If even part of this is correct, it shifts the goal of rehabilitation. The question moves from only asking how hard is this person working to also asking is the nervous system learning a more stable way to move. These are not the same question, and they do not always have the same answer.
It would mean treating movement quality, not just movement quantity, as a primary target. And quality can be operationalized with observable parameters, such as how well space and time stay coupled, how consistent timing is from one repetition to the next, and whether a person shows signs of detecting and correcting their own errors. It would mean designing practice so that error-detection is supported rather than bypassed. It would mean thinking of cueing not only as a way to externalize control in the moment, but as a way to temporarily restore the coupling between space and time that the system is struggling to hold on its own. And it would mean paying attention to stability long before freezing appears, on the premise that the most treatable moment may be the one before the system reaches its threshold.
Quality of life, in the end, is determined by what a person can still access when they are not moving well. The aim is not only to help people recover from freezing once it arrives. It is to understand what happens in the months and years before, and to intervene while the system is still flexible enough to change.
THE BETTER QUESTIONS
1. What changed in the nervous system before the freeze, rather than what triggered it, and could we learn to detect those changes within practice sessions?
2. Is the system learning stability, or just accumulating effort?
3. Can the person still detect their own subtle movement errors, and what does practice reinforce if they cannot?
4. Can we measure the coupling of space and time, in both gait and upper-limb tasks, and does improving it change what happens next?
If these questions are even partly the right ones, then freezing is not simply an event to be managed. It is the visible edge of a slower story we have not yet learned to read. That story is where I intend to keep looking.
[Your name]
Founder, MOTORVATION · THINK and Move Well
This paper presents a hypothesis and a point of view, not a clinical guideline or a claim of proven mechanism. Shared to invite discussion among clinicians, researchers, and people living with Parkinson’s.
© 2026 MOTORVATION. All rights reserved.