In the processes of magnitude comparisons or parity judgments, the left (right) hand responds to small (large) numbers faster than the right (left) hand. Such association is known as the Spatial–Numerical Association of Response Codes (SNARC) effect. This study first introduces the origin and theoretical explanations of the SNARC effect. These explanations arise mainly from visuo-spatial coding and verbal-spatial coding aspects. The SNARC effect stems largely from visuo-spatial coding of magnitude (e.g., the mental number line hypothesis). Dehaene et al. (1993) indicates that the numerical representation appears to be a line from left to right. Small numbers are represented on the left whereas large numbers on the right. The spatial position of a number is activated automatically when you see this number. The spatial position of numerical representation corresponds to the position of the left–right coordinates of external space in the visuo-spatial frame. The number on the left responds faster when pressing the left button. Meanwhile, the number on the right responds faster when pressing the right button. Recently, many works began to challenge the visual spatial coding theory (Fias, Dijck, & Gevers, 2011; Gevers, Verguts, Reynvoet, Caessens, & Fias, 2006; Proctor & Cho, 2006). They suggested that with respect to the verbal-spatial frame, the SNARC effect results from a congruency between verbally coding numbers as small or large and verbally coding responses as left or right. Indeed, it is exactly this categorization of external space through verbal concepts such as “left,” “right,” “above,” or “below” that defines our conceptual coding of space. Such explanation is entitled verbal-space interpretation. These models include polarity coding theory and dual route coding model. The fundamental difference between visuo-spatial and verbal-spatial coding arises from different views on whether the numerical mental representation is characterized by spatial nature. To solve the theoretical debate, scientists have attempted to examine the brain mechanisms of the SNARC effect to provide the neural basis over the last decade. Cutini et al. (2012) reported a hemodynamic feature of the SNARC effect in the bilateral intraparietal sulcus (i.e., a key region for numerical magnitude representation) and left angular gyrus (ANG) (i.e., a region responsible for processing verbal number). They therefore rejected the point that the SNARC effect is fully restrained to the stage in response selection. Moreover, they disagreed with the basic theories that overlook the numerical spatial representation. Further research is thus required from the perspective of neuropsychology. Second, this paper introduces the summarization of the character of this effect – popularity and flexibility. The popularity of the SNARC effect arises from the fact that the SNARC effect is completely determined by the numerical spatial nature, and is not influenced by experimental materials and response methods. This study discusses the relationship between the SNARC effect and the Simon effect, and between the SNARC effect and the MARC effect. The SNARC effect is a consistent and stable effect because it never alters in the range of response time. In contrast, the Simon effect weakens and even reverses in the range of response time. Furthermore, the brain mechanism of the SNARC effect is beneficial to solving the theoretical debate and on the stage response problems. Finally, three problems are proposed for further examination: (1) the processing mechanisms of the SNARC effect; (2) the theoretical exploration of the SNARC effect; (3) the nature of the SNARC effect.