Heavy vs Light
Is muscle growth muscle fiber specific?
Your muscles are made up of different types of fibres. The type I fibres are slow and fatigue-resistant. The type II fibres are fast, powerful, and tire quickly. For simplicity's sake, we’ll only focus on these two.
For decades, lifters have wondered how the weight on the bar determines which fibre types actually grow. A new systematic review and meta-regression, published in March 2026 on SportRxiv, is the most statistically rigorous attempt yet to offer some clarity. Eight studies, 195 participants, and a purpose-built multilevel model later, they found that load probably does matter at the fibre level, but the evidence is far from concrete.
Aim
Researchers set out to map the relationship between training load (expressed as a % of 1-repetition maximum, or 1RM) and muscle fibre hypertrophy, treating load as a continuous variable rather than a simple “heavy vs. light” binary. A secondary goal was to examine whether factors like training status, volume, and proximity to failure influenced that relationship.
Methods
The team searched PubMed/MEDLINE, Scopus, and SPORTDiscus for relevant studies, ultimately including eight that met their criteria. To qualify, a study had to measure actual muscle fibre cross-sectional area via biopsy (not just whole-muscle size via ultrasound or MRI), use a randomised design lasting at least four weeks, and compare at least two groups training at different loads.
The pooled sample across all eight studies was 195 participants, mostly untrained young adult males, with study samples ranging from 12 to 49 people and training interventions averaging around 9.5 weeks. All biopsies were taken from the vastus lateralis (the outer quad).
Rather than simply comparing "low load" to "high load" as previous meta-analyses had done, the researchers modelled load as a continuous variable using multilevel meta-regression with cluster-robust variance estimation, which better accounts for the complexity and dependencies within and between studies.
Results
When both fibre types were lumped together in a single model, there was no clear relationship between load and hypertrophy. That finding aligns with the existing body of work showing that whole-muscle growth is broadly similar across a range of loads when sets are taken to failure.
The more telling picture emerged from the fibre-type specific model. At low loads (20% 1RM and 30% 1RM), type I fibre hypertrophy was favoured over type II. At moderate loads (40 to 50% 1RM), the two fibre types grew at roughly similar rates. Above 50% 1RM, the pattern flipped, with the effect estimates increasingly favouring type II fibre growth as load rose from 60% through to 90% 1RM.
The load-by-fibre-type interaction estimate was consistent across leave-one-out sensitivity analyses, ranging from 0.247 to 0.351 across the eight models, suggesting the finding was not driven by any single study.
That said, the confidence and prediction intervals were wide throughout, and for all load anchors above 30% 1RM, the confidence intervals crossed zero, meaning the differences between fibre types at those loads were not conclusive. The authors are explicit about this: these findings are exploratory, not definitive.
Key Takeaways
Light loads (roughly 20 to 30% 1RM) may preferentially grow type I fibres. These are your slow-twitch endurance fibres, and they appear to respond best to lighter, higher-rep work.
Heavy loads (above 50% 1RM, and especially toward 80 to 90% 1RM) lean toward type II fibre growth. Type II fibres are your primary drivers of strength and power, and heavier loading appears more aligned with their development.
For whole-muscle size, load likely does not matter much, as long as you train close to or to failure. The fibre-level differences appear to wash each other out when viewed from the outside.
All biopsies were from the quads (vastus lateralis), all participants were mostly untrained young males, and there are zero data points between 53% and 70% 1RM in this dataset. Generalising these findings to trained lifters, women, or upper-body muscles is not yet supported.
Proximity to failure matters. When the one study that did not train participants to failure was removed, the load-by-fibre-type interaction grew stronger, hinting that training hard enough to recruit all available motor units may be a prerequisite for this effect to show up.
The evidence base is thin. Eight studies and 195 participants are a small pool. The findings are a signal worth watching, not a rule worth programming around with confidence.
Reference
Varovic, D., Larsen, S., & Grgic, J. (2026). Heavy or Light: Is Muscle Fiber Growth Load-Specific? A Systematic Review and Meta-Regression. SportRxiv. https://osf.io/wgze9/overview?view_only=cb5a10d25a5e485c9329ace3580c97ab
You can also find me at dannyleejames.com for stories, personal training insights, and coaching.