📄 bioengineering

The effects of muscle fibre type distribution on gait biomechanics: A predictive simulation study

This predictive simulation study reveals that muscle fibre type distribution significantly influences the metabolic cost and biomechanics of gait, with slow twitch fibres reducing energy costs at lower speeds but becoming less efficient than fast twitch fibres at higher running speeds, thereby driving adaptive changes in stride frequency and length to maintain mechanical efficiency.

Original authors: Daehlin, T. E., Ross, S. A., De Groote, F., Wakeling, J. M.

Published 2026-04-15

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Daehlin, T. E., Ross, S. A., De Groote, F., Wakeling, J. M.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine your body is a high-performance car, and your muscles are the engine. Inside that engine, there are two different types of pistons: Slow-Twitch and Fast-Twitch.

Slow-Twitch pistons are like the fuel-efficient hybrid engine in a city car. They are great for long, steady drives (like walking or jogging) but struggle to rev up quickly for a sprint.
Fast-Twitch pistons are like the roaring V8 in a race car. They are built for explosive speed and power but guzzle fuel quickly and get tired fast.

Usually, every person has a mix of these two "pistons" in their muscles. Some people are born with more "hybrids," while others have more "race cars."

This paper asks a fascinating question: Does the mix of these pistons change how we walk and run, and does it change how much energy (fuel) we burn?

Since we can't easily swap a person's muscle fibers in a lab (it's too invasive and complex), the researchers built a virtual human inside a computer. They created a "digital twin" of a human body and then played a game of "what if." They took this digital human and gave them different muscle mixes:

One version was almost 100% "hybrid" (slow) muscles.
One version was almost 100% "race car" (fast) muscles.
They tested everything in between.

Then, they told the computer to make this virtual human walk and run at different speeds, letting the computer figure out the most efficient way to move, just like a real human brain does.

Here is what they discovered:

1. The Walking Test (The City Drive)

When the virtual humans were walking, the "hybrid" (slow-twitch) models were the clear winners.

The Result: The more slow-twitch muscles a model had, the less energy it burned to walk a mile.
The Analogy: Think of walking as driving through a quiet neighborhood. The hybrid engine is perfect here; it sips fuel and keeps a steady pace. The race car engine (fast-twitch) is overkill and wastes fuel just trying to walk.
The Gait: Interestingly, the mix of muscles didn't change how they walked much. They all took roughly the same step length and frequency. The body just found a way to use the available muscles efficiently.

2. The Running Test (The Highway Sprint)

When the virtual humans started running, the story got complicated. It wasn't just about who had the best engine; it was about speed.

The Result: There is a "tipping point" speed.
- At slower running speeds, the "hybrid" (slow-twitch) models were still more efficient.
- But once they hit a fast sprint, the "race car" (fast-twitch) models suddenly became more efficient.
The Analogy: Imagine driving on a highway. At 40 mph, a hybrid is fine. But if you need to hit 100 mph, the hybrid engine might struggle or overheat, while the V8 race car thrives. The computer predicted that at very high speeds, having more "race car" muscles actually saves you energy because those muscles are designed to work efficiently at high speeds.

3. The Stride Change (The Dance)

The most surprising finding was how the body changed its dance moves to match the engine.

The Result: When the virtual humans with mostly "race car" muscles ran fast, they started taking shorter, quicker steps (high step frequency).
The Counter-Intuitive Twist: The researchers originally guessed that "hybrid" muscles would make people take shorter steps. They were wrong! The "race car" models took the shortest, fastest steps.
Why? It turns out that to run super fast, you need to cycle your legs rapidly. The "race car" muscles can fire and reset quickly, allowing for that rapid-fire stepping pattern. The "hybrid" muscles are a bit slower to reset, so they had to take longer, slower strides to keep up, which turned out to be less efficient at top speeds.

The Big Takeaway

Your muscle makeup isn't just about how strong you are; it dictates how you move and how much energy you use depending on how fast you go.

Walking: If you have more slow-twitch muscles, you are naturally more fuel-efficient.
Running: If you are a slow jogger, slow-twitch muscles help. But if you are a sprinter, having more fast-twitch muscles might actually be the secret to running faster without getting as tired.

The study concludes that our bodies are incredibly smart. They automatically adjust our step length and speed to match our muscle "engine type" to find the most efficient way to move, whether we are strolling in the park or sprinting for the bus.

1. Problem Statement

Muscle fibre type distribution (the ratio of slow-twitch to fast-twitch fibres) significantly influences the metabolic and contractile properties of muscles. While it is known that humans self-optimize gait to minimize the metabolic cost of transport (COT), the specific biomechanical mechanisms by which fibre type distribution influences gait patterns remain unclear.

Previous in vivo studies attempting to isolate these effects have been limited by confounding variables (e.g., age, sex, training status, and cardiovascular fitness) and the inability to measure fibre type distributions across all major locomotor muscles simultaneously. Furthermore, traditional Hill-type muscle models used in simulations typically represent a muscle with a single set of parameters, failing to capture the distinct recruitment and efficiency profiles of slow and fast fibres. This study aims to fill this gap by using predictive musculoskeletal simulations to isolate the effects of fibre type distribution on gait energetics and mechanics across walking and running speeds.

2. Methodology

Musculoskeletal Model Architecture:

Base Model: A 31-degree-of-freedom (DOF) OpenSim-based model was used, actuated by 92 muscles in the lower extremities and lumbo-sacral region, plus idealized torque motors for the arms.
Novel Muscle Model: The standard Hill-type models were replaced with a two-element parallel contractile model. Each muscle consists of two distinct contractile elements (CE) arranged in parallel: one representing slow-twitch fibres and one representing fast-twitch fibres.
- Parameters: The slow CE was assigned a maximal shortening velocity ( $v_{max}$ ) of 5 optimal fibre lengths/s and an activation time constant of 45 ms. The fast CE was assigned a $v_{max}$ of 10 lengths/s and a time constant of 25 ms.
- Force Generation: Total muscle force is the sum of forces from both CEs, weighted by the slow-twitch area fraction ( $\phi_s$ ), plus passive elastic forces.
- Metabolic Model: A smooth, continuous metabolic probe was modified to calculate energy consumption based on the separate activations of slow and fast fibres rather than a single aggregate activation.

Predictive Simulation Framework:

Optimal Control Problem (OCP): Simulations were formulated as OCPs to generate de novo movements. The system minimized a multi-objective cost functional ( $J$ $J$ ) comprising:
- Metabolic rate ( $\dot{E}$ )
- Muscle activations ( $a$ )
- Joint coordinate accelerations ( $\ddot{q}$ )
- Passive joint torques ( $\tau_{pass}$ )
- Arm actuator excitations
Solver: The problem was transcribed using direct collocation (cubic Radau quadrature) and solved using IPOPT and MUMPS solvers.

Experimental Design:

Fibre Type Variations: Nine models were created with average slow-twitch area fractions ranging from 6% to 96% (including a nominal model based on literature data).
Speeds: Simulations were run at 11 speeds ranging from 1.0 m/s to 4.5 m/s (covering walking and running gaits).
Validation: The nominal model's output (joint angles, torques, powers, ground reaction forces, and muscle activations) was validated against experimental data from Falisse et al. (walking) and Hamner & Delp (running).

3. Key Contributions

First Predictive Simulation of Mixed Fibre Types: This is the first study to implement a predictive simulation framework where slow and fast fibres are recruited independently within a single muscle model to optimize gait.
Isolation of Variables: By using in silico modeling, the study successfully isolated the effect of fibre type distribution from other physiological confounders (age, sex, fitness) that plague in vivo comparisons.
Identification of a Speed-Dependent Threshold: The study identified a critical interaction between fibre type and locomotion speed, challenging the assumption that slow-twitch dominance is universally more efficient.

4. Key Results

Model Validation:

The nominal model produced realistic gait patterns. Most joint kinematics and kinetics showed strong cross-correlations ( $R > 0.7$ ) with experimental data.
Discrepancies were noted in hip internal/external rotation and some peak joint moments, likely due to the use of within-individual variability for error weighting in walking validation.

Walking (1.0 – 2.0 m/s):

Cost of Transport (COT): COT decreased linearly as the proportion of slow-twitch fibres increased. The model with 96% slow fibres had a COT ~21% lower than the 6% slow fibre model.
Biomechanics: Stride length and frequency were minimally affected by fibre type distribution.
Mechanism: Slow fibres are more efficient at the lower contraction velocities typical of walking.

Running (2.5 – 4.5 m/s):

Interaction Effect: A threshold speed exists where the relationship between fibre type and COT reverses.
- At lower running speeds (2.5–3.5 m/s), models with higher slow-twitch proportions generally had lower COT.
- At the highest speed (4.5 m/s), models with higher fast-twitch proportions achieved the lowest COT.
Spatiotemporal Changes: At high speeds, models with more fast-twitch fibres adopted a strategy of higher stride frequencies and longer stride lengths compared to slow-twitch dominant models.
Efficiency: Fast-twitch fibres showed a steeper increase in mechanical efficiency as running speed increased compared to slow fibres. At high speeds, the active shortening velocities of major locomotor muscles entered the range where fast fibres operate near peak efficiency.

5. Significance and Conclusion

This study demonstrates that muscle fibre type distribution is a critical determinant of gait biomechanics and energetics, but its influence is speed-dependent.

Theoretical Insight: The findings support the hypothesis that humans (or simulated agents) optimize gait to operate muscles near their peak mechanical efficiency. At slow speeds, slow fibres are optimal; at high speeds, the recruitment of fast fibres becomes metabolically advantageous because they can perform work more efficiently at high shortening velocities.
Reinterpretation of Population Differences: The results suggest that observed differences in gait between groups (e.g., males vs. females, young vs. old) may not be solely driven by fibre type distribution, as the study found that fibre type had minimal effect on stride length/frequency during walking and the predicted trends for running contradicted some previous observational hypotheses.
Clinical and Athletic Implications: The existence of a "threshold speed" implies that training or pathology altering fibre type composition could have divergent effects on performance depending on the specific speed of locomotion. For instance, a shift toward fast-twitch dominance might improve sprinting economy but potentially reduce walking economy.

In summary, the study provides a robust computational framework for predicting how neuromuscular properties dictate movement strategies, revealing that the "optimal" fibre type is not static but is dynamically selected by the locomotion speed to minimize metabolic cost.