Computerized glow curve convolution (CGCD) has been used to study the thermally induced room-temperature fading of both annealed and unannealed LiF:Mg, Ti following gamma ray, thermal neutron and alpha particle irradiation. We confirm the anomalous growth of peak 4 due to trap conversion in both types of material. However, we find no evidence for the growth of peak 5 in either material for any of the types of radiation studied. Our results similarly do not confirm the details of the differences in the fading previously observed following thermal neutron and gamma ray irradiation in unannealed material and point to the great difficulty encountered in arriving at a universal characterization of LiF:Mg, Ti when peaks 2 and 3 and 4 or any fraction thereof are included in the thermoluminescent signal. On the other hand when peak 5 only is employed for dosimetric measurements (via CGCD) a well behaved fading rate of (-5.5 ± 2.6)% (ISD) per month for all radiation types in both annealed and unannealed material is obtained. We believe that these results convincingly demonstrate the advantages of using only peak 5 in dosimetric measurements where fading corrections are of importance.