Matlab uses a single random number generator for RAND, RANDI, and RANDN, for the command line or in called function. Setting a seed and using the same sequence of calls to the random functions results in the same random numbers. You can get the current state of the random number generator using
if sc is different after running the simulation with your different parameter sets, the number of calls has not been the same (if you set the rng to the same initial state before each simulation, of course). You could add some sc = rng; calls at crucial points of your code and compare the results to those values obtained from a previous run to narrow down the part where the simulation uses different number of calls to the random number generator for the different parameter sets.
