Research on the housing choice voucher program and housing mobility interventions shows that even with assistance, it is difficult for poor minority families to relocate to, and remain in, low-poverty neighborhoods. Scholars suggest that both structural forces and individual preferences help explain these residential patterns. However, less attention is paid to where preferences come from, and how they respond to policies and social structure to shape residential decisionmaking. In this paper, we use data from fieldwork with 110 participants in the Baltimore Mobility Program (BMP), an assisted mobility voucher program, to demonstrate how residential preferences can shift over time as a function of living in higher opportunity neighborhoods. Since 2003, BMP has helped over 2,000 low-income African American families move from high-poverty, highly segregated neighborhoods in Baltimore City to low-poverty, racially mixed neighborhoods throughout the Baltimore region. Along with intensive counseling and unique program administration, these new neighborhood contexts helped many women to shift what we term residential choice frameworks: the criteria that families use to assess housing and neighborhoods. Parents who participated in the mobility program raised their expectations for what neighborhoods, homes, and schools can provide for their children and themselves. Parents report new preferences for the “quiet” of suburban locations, and strong consideration of school quality and neighborhood diversity when thinking about where to live. Our findings suggest that housing policies should employ counseling to ensure relocation to and sustained residence in low-poverty communities. Our work also underscores how social structure, experience, and policy opportunities influence preferences, and how these preferences, in turn, affect policy outcomes.