For a DLS experiment only the RI of the solvent is required for the intensity distribution. For any conversion to volume, the RI of the particle is required. A robustness check is always mandatory as the end result is determined by the size, distribution, and relative refractive index. In static light scattering or laser diffraction above around 25um (ISO13320:2009) the result is usually robust to RI changes. In the large (> 25um) and small (< 60 nanometer)regions, approximations are often used (Fraunhofer and Rayleigh respectively). Look for postings by Ulf Nobbmann on this forum and on Materials Talks for more information.
Both DLS and SLS are scattering based techniques where one measures the scattered light intensity (autocorrelation function) as a function of the scattering angle. This scattering intensity depends on the RI of the medium and the size of the scatterer. Thus from these intensity data one can estimate the size if we know the medium RI.
The refractive index of the dispersant is required in order to calculate the correct scattering vector q. If you want to extract for example a size (Rh or Rg) out of light scattering data (DLS or SLS) then the scattering vector must be known. The short answer is: it is required for both.