This commentary discusses the articles that comprise this special issue on attachment in middle childhood. Central to this discussion is the distinction between verbal, strategic, and conscious responses to questionnaires as compared to verbal and nonverbal, automatic and largely unconscious responses to interviews. Both methods have been developed to study attachment in middle childhood, often leading to divergent results. Research going forward should include both types of methods in order to maximize the potential for meaningful research and effective clinical work. Ongoing issues of importance for the field include the need to take account of differential susceptibility theory, culture-specific influences, and possible gender differences. This commentary is organized into three sections--looking within, looking behind, and looking ahead--a tripartite distinction that reflects developmental as well as theoretical and applied perspectives vital to hold in mind when researching attachment in middle childhood.