Abstract Density peaking has been studied between an ICRH and NBI identity plasma in JET. The comparison shows that 8 MW of NBI heating/fueling increases the density peaking by a factor of two, being R/L n = 0.45 for the ICRH pulse and R/L n = 0.93 for the NBI one averaged radially over ρ tor = 0.4, 0.8. The dimensionless profiles of q, ρ *, υ *, β n and T i/T e ≈ 1 were matched within 5% difference except in the central part of the plasma (ρ tor < 0.3). The difference in the curvature pinch (same q-profile) and thermo-pinch (T i = T e) between the ICRH and NBI discharges is virtually zero. Both the gyro-kinetic simulations and integrated modelling strongly support the experimental result where the NBI fuelling is the main contributor to the density peaking for this identity pair. It is to be noted here that the integrated modeling does not reproduce the measured electron density profiles, but approximately reproduces the difference in the density profiles between the ICRH and NBI discharge. Based on these modelling results and the analyses, the differences between the two pulses in impurities, fast ions (FIs), toroidal rotation and radiation do not cause any such changes in the background transport that would invalidate the experimental result where the NBI fuelling is the main contributor to the density peaking. This result of R/L n increasing by a factor of 2 per 8 MW of NBI power is valid for the ion temperature gradient dominated low power H-mode plasmas. However, some of the physics processes influencing particle transport, like rotation, turbulence and FI content scale with power, and therefore, the simple scaling on the role of the NBI fuelling in JET is not necessarily the same under higher power conditions or in larger devices.