Precision grip | Ankur Gupta Research

Parkinson's Disease Though Precision Grip Lens

When Did Your Fingers Last Slip?

Have you ever tried picking up a wet glass or a slippery egg? Chances are, you know just how much force to use so you don’t drop it; or worse, crush it. Most people get this right without even thinking. That’s because our brains are constantly calculating how much “grip force” is needed, even as conditions change. But for someone living with Parkinson’s disease, this simple act can become a daily struggle: sometimes they grip too hard, and other times, their grip is wobbly or inconsistent.

An illustration of precision grip that is the grip formed by index finger and thumb.

The Brain’s Grip Calculation

So how do we “know” how to hold things properly? Before you pick up an egg, your eyes and hands take in clues: how heavy it might be, how smooth, what happened the last time you grabbed one. Your brain, especially the movement-controlling motor cortex, crunches these clues and sends commands to your fingers to use just the right amount of force. Too little grip and the egg slips; too much and it becomes breakfast in the wrong way.
If your fingers are dry, there’s plenty of friction (a sort of stickiness) between your finger skin and the egg. However, when your fingers get wet, friction drops dramatically and the egg becomes harder to hold. The sensors in your skin spot this immediately, sending a message up to your spinal cord: “Hey, we’re slipping!”, reflexes kick in and your grip tightens, often automatically. It’s a team effort between quick spinal reflexes and brain planning. But another team member is in play, the basal ganglia, a group of deep brain structures that quietly run quality control on your movements and adjust errors as you go.

Friction: Everyone’s Fingers Are Different

Not all skin is equally grippy. This is where things get interesting. During my Ph.D., I wondered what makes one person better at holding slippery stuff than someone else. I ran simple grip-and-lift experiments using a special sensor-equipped object, asking healthy volunteers to lift it under different conditions: with dry fingers (normal friction) and after soaking their fingers in water (low friction). Turns out, some people naturally have higher finger friction than others and dipping those fingers in water made them all lose grip, fast.
I measured the peak grip force (the strongest squeeze during a lift) and the steady grip force (how hard you’re squeezing after the object is off the table and you’re just holding it stable). Safety margin refers to how much stronger your grip is compared to the absolute minimum needed to keep from dropping the object. When fingers were wet, all subjects’ friction dropped, but the amount of force each person used to counteract this was surprisingly stable. The safety margin remained tightly controlled, even though the actual grip force had to be increased.

A simplified illustration of grip force profile.

Building a Computer Model: Simulating the Real Thing

To understand these finger forces better, I built a computer model consider it a digital version of the grip reflex circuit, using what’s known as a PID controller. The PID (Proportional-Integral-Derivative) controller is a simple but powerful concept: it “learns” from the past, reacts to the present, and anticipates the future. It’s like a thermostat that doesn’t just check today’s temperature, but also remembers yesterday and wants to keep the house from getting too cold or hot tomorrow.
In the model, if the simulated object slipped (say, when skin friction suddenly dropped after wetting), the simulated grip force quickly increased, just like the real volunteers’ did. But we wanted to test something more: What happens when you suddenly encounter a slippery surface, but you barely have time to adjust? Instead of retraining the entire model from scratch, we reused the dry-grip training and just made quick, partial adjustments. This “quick learning” approach matched the experiments: people adapted fast, with hardly any extra error or wasted energy.

From Fingers to Parkinson’s: The Dopamine Connection

Now, here’s where it links back to Parkinson’s. Deep inside the brain, the basal ganglia adjust those grip force calculations: checking for mistakes, correcting them, and deciding how big a movement correction is needed. Dopamine, a brain chemical that’s famous for many things, acts as a messenger here. In Parkinson’s disease, the cells that make dopamine slowly die. With less dopamine around, the brain’s error-correction system falters. Movements, including grip, become unpredictable sometimes far too strong, sometimes weak, often just plain inconsistent.
The second phase of my research was building a computer model that mimics this process. We created a simulated “brain” that not only generated grip forces but also included a “risk calculation” module because gripping something always involves a bit of risk. Grip too close to the minimum and you risk dropping the object. Grip much too firmly and you risk damaging it or tiring out your muscles. My model put together these value and risk calculations using reinforcement learning, utility theory (think of it as a cost-benefit calculator) and a neural network to simulate how grip force is chosen, both in healthy and Parkinson’s-affected individuals.
What did I find? The model showed that healthy people keep their grip far enough above the slip threshold for safety, but not so high that it becomes excessive. In Parkinson’s models, grip force increased and became erratic, especially when dopamine was simulated to drop. Medication raised the grip force even further but reduced its variability. These matched real-world experiments and helped us see why Parkinson’s patients grip things the way they do.

Stitching It All Together

In plain language: whether you’re picking up a fragile egg or a heavy pan, your brain is performing a quiet calculation of balancing friction, adjusting for surprises, and correcting errors in real time. Everyone’s fingers and brain circuits are a bit different, but we all rely on feedback, quick learning, and a bit of courage (risk!) to keep from dropping or crushing things. For people with Parkinson’s, these calculations are harder, but with the help of computer models and sensor-based experiments, we’re now better able to understand (and hopefully someday improve) how the brain keeps our grip just right.

To read more:

Gupta A., & Chakravarthy V. S. (2018). Modeling Precision Grip Force in Controls and Parkinson’s Disease Patients. Computational Neuroscience Models of the Basal Ganglia (pp. 131-151): Springer.
Moustafa A. A., Chakravarthy S., Phillips J. R., Crouse J. J., Gupta A., Frank M. J., Hall J. M., & Jahanshahi M. (2016). Interrelations between cognitive dysfunction and motor symptoms of Parkinson’s disease: behavioral and neural studies. Rev Neurosci, 27(5), 535-548.
Moustafa A. A., Chakravarthy S., Phillips J. R., Gupta A., Keri S., Polner B., Frank M. J., & Jahanshahi M. (2016). Motor symptoms in Parkinson’s disease: A unified framework. Neuroscience & Biobehavioral Reviews, 68, 727-740.
Gupta A., Balasubramani P. P., & Chakravarthy S. (2013). Computational model of precision grip in Parkinson's disease: a utility based approach. Front Comput Neurosci, 7, 172.
Gupta A., Manikanta A., Kandaswamy D., Murthy M., Devasahayam S., Babu S. K., & Srinivasa V C. (2013). Human precision grip performance under variable skin friction conditions: a modeling and experimental study. International Journal of Mind, Brain & Cognition, 4(1-2), 7-45.