Previous studies of theory of mind (ToM) in old age have provided mixed results. We predicted that educational level and cognitive processing are two factors influencing the pattern of the aging of ToM. To test this hypothesis, a younger group who received higher education (mean age 20.46 years), an older group with an education level equal to that of the young group (mean age 76.29 years), and an older group with less education (mean age 73.52 years) were recruited. ToM tasks included the following tests: the second-order false-belief task, the faux-pas task, the eyes test, and tests of fundamental aspects of cognitive function that included two background tests (memory span and processing speed) and three subcomponents of executive function (inhibition, updating, and shifting). We found that the younger group and the older group with equally high education outperformed the older group with less education in false-belief and faux-pas tasks. However, there was no significant difference between the two former groups. The three groups of participants performed equivalently in the eyes test as well as in control tasks (false-belief control question, faux-pas control question, faux-pas control story, and Eyes Test control task). The younger group outperformed the other two groups in the cognitive processing tasks. Mediation analyses showed that difficulties in inhibition, memory span, and processing speed mediated the age differences in false-belief reasoning. Also, the variables of inhibition, updating, memory span, and processing speed mediated age-related variance in faux-pas. Discussion focused on the links between ToM aging, educational level, and cognitive processing. Supported by Chinese National Natural Science Foundation (number: 30870766) and Anhui Province Natural Science Foundation (number: 11040606M166).