Temperature memory effect (TME) refers to the ability of shape memory polymers (SMPs) to memorize the temperature at which pre-deformation was conducted. In the past few years, this TME was experimentally demonstrated by comparing the applied programming temperature (T_d) with a characteristic recovery temperature (T_c), which corresponds to either the maximum recovery stress or free recovery speed. In these well-designed experiments, T_c was observed to be close to T_d, which is consistent with the intuitive understanding of ‘memorization’. However, since the polymer recovery behavior has been proved to be strongly dependent on various programming and recovery conditions, a new question that whether T_c is always equal to T_d in any thermo-temporal conditions remains to be addressed. In this paper, we answered this question by examining the free recovery profile of an acrylate based amorphous SMP. The recovery T_c, which is the temperature with the maximum recovery speed, versus the recovery temperature is shown to be strongly dependent on both programming and recovery conditions. Their detailed influence could be explained by using the reduced time. During a thermomechanical working cycle of SMPs, in addition to the T_d, any other thermo-temporal conditions, such as the holding time (t_h), cooling rate, recovery heating rate (q), etc., can affect the observed T_c by changing the reduced programming or recovery time. In this manner, the relationship between T_c and T_d is not uniquely determined. Besides, the TME in SMPs can only be achieved within a given temperature range. Both onset and offset of this temperature range are shown to be influenced by the programming history, but are independent of the recovery conditions.