The present study aimed to investigate the effect of domain-specific and domain-general precursors of mathematical training on preschoolers' number sense and estimation. This was a quasi-experimental, pretest-posttest-follow-up study. The statistical population comprised all preschoolers in Kermanshah (Iran) in 2020. Using multi-stage cluster random sampling, a sample of 45 was selected and allocated to three groups (15 per group). The data were collected using the short form of the Wechsler Preschool and Primary Scale of Intelligence (WPPSI), the Number Sense Screener, and the PLUS test (number estimation). The data were analyzed using descriptive statistics and repeated-measures analysis of variance (ANOVA). The mean (SD) of IQ was 95.91 (4.21) in the working memory group and 97.78 (3.38) and 96.66 (4.03), respectively, in the number estimation and control groups. In the linear combination of number estimation variables in terms of group membership, the group x time interaction was significant on the pretest, posttest, and follow-up (p<0.01). The difference in the mean scores of the working memory, number estimation, and control groups was also significant on the variable of number sense (p<0.01). There was no significant difference in the effectiveness of the two training methods on the posttest and follow-up, and both interventions exerted the same effect on improving number sense and estimation.