Background: The aim of this work was to establish how well gel dosimeters performed, as substitutes for brain tissue compared with standard phantom materials such as water, polymethyl-methacrylate (or PMMA), A150 plastic and TE- liquid phantom material for dosimetry of neutron beams in boron neutron capture therapy. Materials and Methods: Thermal neutron fluence, photon dose and epithermal neutron dose distributions were computed for the epithermal neutron beam of the optimized linac based BNCT. Results: Amongst all investigated phantom materials, TE-liquid was shown to be a better substitute for brain tissue than other phantom materials. The differences between TE- liquid and brain at the depth of 6.1 cm for thermal neutron fluence, gamma dose and epithermal neutron dose distributions was calculated 2.80%, 2.40% and -13.87% , respectively. In comparison with the other gel dosimeters, LMD2 provided a better simulation of radiation transport in the brain. It's results differed from the real brain, at the depth of 6.1 cm, for thermal neutron fluence, gamma dose and epithermal neutron dose distributions, by -1.27%, 4.20% and 21.05% respectively. Conclusion: Even though, in gamma dose distribution the LMD2 has large deviation from brain tissue distribution, the deviation is approximately independent of depth, so the results can be multiplied by a constant coefficient to be more consistent with reality. Even though, TE- liquid showed satisfactory results for brain tissue substitution in BNCT, but some properties of gel dosimeters such as three dimensionality, make LMD2 a potentially good dosimeter for dosimetric verification in BNCT.