The current study examined whether the effect of spatial training transfers to the math domain. Sixty-two 6- and 7-year-olds completed an at-home 1-week online training intervention. The spatial-training group received mental rotation training, whereas the active control group received literacy training in a format that matched the spatial training. Results revealed near transfer of mental rotation ability in the spatial-training group. More importantly, there was also far transfer to canonical arithmetic problems, such that children in the spatial-training group performed better on these math problems than children in the control group. Such far transfer could not be attributed to general cognitive improvement, since no improvement was observed for non-symbolic quantity processing, verbal working memory (WM), or language ability following spatial training. Spatial training may have benefitted symbolic arithmetic performance by improving visualization ability, access to the mental number line, and/or increasing the capacity of visuospatial WM.
