Behavioral determinants with the largest effects are often those related to the environments in which behaviors occur. This suggests the merits of a shift in focus of changing behavior at scale away from interventions based on deliberation and decision-making and toward interventions that involve changing cues – physical, digital, social, and economic – in environments. This chapter focuses on changing cues in small-scale physical environments – sometimes known as choice architecture or nudge interventions. Despite attracting much interest, these interventions have been little explored from a theoretical perspective. Exploring the mechanisms by which some of these interventions exert their effects provides a starting point. Examining evidence of three interventions – increasing availability of healthier food options, reducing glass size, and putting warning labels on food and alcohol products – suggests no single theory explains their effects. The mechanisms by which these interventions affect behavior change also necessitate different levels of explanation and demand a theoretical framework that applies at different levels. Recognizing the distinction between model-free and model-based learning and behavior may be central to this. Advancing knowledge on changing behavior by changing environments requires robustly designed field studies to estimate effect sizes, complemented by laboratory studies testing mechanisms to optimize interventions and develop theoretical understanding.