IsoForge surfaces four game modes. They share the same underlying hardware path (live Bluetooth force, sub-second feedback loop on the screen), but they are not interchangeable as clinical tools. Each game asks the patient to produce a different shape of force over time, and the published evidence on force-tracking tasks is clear that distinct shapes engage distinct motor-control strategies and distinct cortical computations (Amematsro et al. 2025). The choice of game is a clinical decision about which subskill you want to train, not a stylistic preference about which visualization the patient finds engaging.
This article walks the four modes, what each one trains, and a starting frame for picking between them.
The four modes
Hold Steady
Hold Steady (a sustained submaximal hold) puts a static force band on the screen. The patient pushes to hold their output inside the band for as long as possible. Time-in-zone is the headline score; the progress arc fills as time accumulates.
What it trains. Sustained submaximal force at a specific target. This is the closest IsoForge mode to a classical force-steadiness drill. The patient's task is to find the target, settle into it, and hold without drifting up or down or oscillating. The motor control challenge is precision at a fixed level.
Best for. Early-phase rehab when the patient needs to recover the ability to produce a consistent moderate force, especially after periods of unloading or surgical inhibition. Also the right mode when steadiness (CV, Yank) is one of the limiting factors on the patient's assessment LSI.
Make Waves
Make Waves (sinusoidal force tracking) puts an oscillating target on the screen that moves up and down on a sine wave. The patient pushes harder and softer in time with the wave. Frequency and amplitude are configurable; presets cover typical training ranges.
What it trains. Modulation of force, the ability to fluently move between higher and lower output without overshooting or losing track of the target. Slow sine targets train closed-loop reactive control (the patient watches the target and adjusts); fast sine targets push toward open-loop predictive control (the patient anticipates the wave because reaction is too slow). The crossover usually happens somewhere between 0.5 and 1.5 Hz, depending on the patient.
Best for. Mid-phase rehab where the patient has recovered enough capacity to modulate force smoothly but needs to rebuild the timing precision. Also useful for athletes returning to sports that require continuous force modulation (rowing, cycling, sustained gripping).
Take Flight
Take Flight (rapid force development) is an impulse game. The patient produces brief explosive pushes to clear gates that scroll across the screen. Difficulty presets adjust gate spacing, target force, and the impulse-rise-time requirement.
What it trains. Rapid force development. Each impulse is a small ballistic contraction; the patient is rebuilding the speed-of-force-production capacity that often lags peak strength recovery. Unlike a single MVIC assessment that produces one RTD number, Take Flight produces dozens of small RFD events per set, which is a different training stimulus than maximum-effort assessments.
Best for. Mid-to-late rehab when the assessment RTD windows are recovering but still below the uninvolved side. Also good for athletes whose sport demands frequent rapid force production (cutting, jumping, change-of-direction work). For the assessment-side context on what RTD captures, see what RTD Early actually means.
Rapid Fire
Rapid Fire (chirp pulse trains) is a more advanced impulse game. The target is a sequence of bell-curve pulses (singlets, doublets, or triplets) at predetermined frequencies (0.5, 1, or 2 Hz). The patient produces matching pulses; scoring is based on peak amplitude (within ±25% of target) and peak timing.
What it trains. Force template-matching under temporal constraint. The patient is asked not just to produce force, but to produce it in a specific shape at a specific moment. The chirp design (varying inter-pulse intervals) prevents the patient from settling into a metronome rhythm; each pulse has to be produced fresh against the current template.
Best for. Late-phase rehab and athletic populations where the goal is to rebuild precise temporal coordination of force production. The 2 Hz triplet condition is the hardest because it forces open-loop pulse generation; the patient cannot react fast enough to correct mid-pulse, they have to commit to the pulse template before they see the feedback.
Why the modes are not interchangeable
The published evidence on force-tracking tasks shows that what looks like the same muscle action on the surface can engage substantially different motor-control machinery underneath. High-density recordings from primary motor cortex during a continuous force-tracking task (methodologically very close to IsoForge's Make Waves and Rapid Fire modes) show that distinct force-profile families (static holds, slow ramps, slow sines at 0.25 Hz, fast sines at 1 to 3 Hz, chirps) activate distinct neural state-space regions and dimensions. Closed-loop reactive strategies suffice for slow-changing profiles; higher-frequency profiles require open-loop template-based predictive strategies (Amematsro et al. 2025).
The practical translation: you cannot substitute Make Waves at 0.5 Hz for Make Waves at 2 Hz and expect the same training adaptation. The slower version trains reactive control; the faster version trains predictive control. The same logic applies across the four modes. Hold Steady, Make Waves, Take Flight, and Rapid Fire are not four UIs over the same training, they are four different trainings.
This also intersects with the contraction-type taxonomy on the assessment side. Each game maps to a corresponding contraction-intent setting (see choosing the contraction type):
- Hold Steady → submaximal hold
- Make Waves → submaximal modulation (no direct assessment analog)
- Take Flight → ballistic / RTD
- Rapid Fire → ballistic / RTD with timing constraint
When you assess a patient with a contraction intent and then train them with the matching IsoForge mode, the assessment-and-training pair are probing and rebuilding the same underlying skill. When you mix modes (assess MVIC, train a submaximal hold), you're working on different facets, which is fine and often correct, but worth being deliberate about.
A starting framework
Three patterns that work for most rehab progressions:
- Patient with sustained-output deficit (high nRMSE / CV on assessment). Start with Hold Steady at 30 to 50% MVIC. Progress by increasing target percentage and hold duration.
- Patient with explosive deficit (low RTD on assessment). Start with Take Flight at easy difficulty. Progress by tightening the rise-time requirement and increasing target amplitude. Graduate to Rapid Fire once the patient is producing clean singlet pulses reliably.
- Patient with motor-control precision deficit (high Yank on assessment, or athlete late in rehab). Start with Make Waves at slow frequency (0.25 to 0.5 Hz). Progress by increasing frequency. Graduate to Rapid Fire for the temporal-template-matching layer.
None of these is exclusive. A single rehab block can rotate through multiple modes; what matters is that the choice is deliberate and tied to the assessment side.
What to do next
- Pick the mode based on the assessment deficit, not on patient engagement. Engagement matters, but it's a tiebreaker, not the primary criterion.
- Match the IsoForge mode to the contraction intent you assess in. Assessment and training should target the same skill for the patient to feel the connection.
- Progress within a mode before switching modes. Hold Steady at 30% MVIC and Hold Steady at 60% MVIC are different drills inside the same skill; mode-switching changes the skill being trained.
See also setting submaximal targets from a recent MVIC for how to pick the target force for Hold Steady and Make Waves modes, and dosing IsoForge sessions across a rehab block for the broader frequency-and-volume framing.
References
- Amematsro EA, Trautmann EM, Marshall NJ, et al. Motor cortex flexibly deploys a high-dimensional repertoire of subskills. bioRxiv. 2025. doi:10.1101/2025.09.07.674717
- Kozinc Ž, Pleša J, Djurić D, Šarabon N. Comparison of rate of force development between explosive sustained contractions and ballistic pulse-like contractions during isometric ankle and knee extension tasks. Applied Sciences. 2022;12(20):10255. doi:10.3390/app122010255