Early academic scores are strong and robust predictors of children’s later school and career performance (Duncan et al., 2007; Rose, 2006). However, the USA ranks well below other countries on math scores (27th out of 34; OECD, 2013), and have been marked as particularly inadequate at “mathematics tasks with higher cognitive demand(s)”. Thus, it is important to focus on the mechanisms which may contribute to differences in early mathematics problem solving and find tools that are uniquely suited to addressing this issue. One advantageous strategy young children use during math problem solving are hand gestures. Gestures are one of several overtly observable strategies in math contexts(e.g., counting on fingers vs. counting out loud without gestures), but have been specifically recognized as useful given their ability to reduce the user's working memory load during math contexts (Goldin-Meadow & Wagner, 2005). As children get older, the type and frequency of strategies used are reported to shift from basic to more advanced and efficient (Siegler, 1987). This pattern is often seen as younger children using more overtly observable strategies (e.g., finger counting), whereas older children rely on more implicit strategies (e.g., memory retrieval of math facts, Geary et al., 1991). However, less is known about how differences in children’s concurrent domain-general abilities (e.g., working memory, inhibitory control) and domain-specific knowledge (e.g., math specific) contribute to strategic use of gesture during arithmetic problem solving. This line of research is vital given that gestures may be especially advantageous based on their capacity to bolster mental resources needed for problem solving. Using the Gestures in Math Environments model (GME model; Gordon & Ramani, 2021) as a framework, the current study provides a comprehensive assessment of the factors underlying children’s domain-general and specific abilities, and provides evidence as to their relation to children’s use of gesture as a strategy during arithmetic problem solving. Furthermore, it tests a newly proposed adaptation to the GME model where inhibitory control plays a moderating role on the relation between children’s working memory and use of gesture. One-hundred-thirty-seven 4- to 7-year-old children and their parents participated in this study. All children completed two sessions; an autonomous online-game based assessment and a video recorded zoom session regulated by a trained research assistant. At each session, children completed measures of inhibitory control, early mathematical knowledge, and working memory. Their gesture use was video recorded during one measure where children partake in arithmetic problem solving. Parents completed a standardized measure assessing their child’s inhibitory control and working memory abilities. Using structural equation modeling, the relations between all measures and a consideration of how each corresponded to a set of comprehensible latent factors (one factor each for inhibitory control, working memory, and math) were examined. Further examination of how each factor related to children’s use of gesture was investigated. In line with the original GME model, working memory ability was a unique predictor of children’s use of gesture above and beyond impacts of age, math knowledge, inhibitory control, and gender. While there is not any evidence from the current study to support the proposed moderation between inhibitory control and working memory on gesture use, a modification to the GME model with the addition of gender is subsequently recommended.