Both mother and teacher are social agents in the child's environment and provide the foci for the two separate, but related, ongoing studies on reinforcement patterns described in this paper. The specimen record is the primary method used to obtain data. For the study of children at home, the sample includes 24 3-year-old children from lower income urban, middle income urban, and lower income rural families. Eight half-hour observations are planned for each child. Mealtime was chosen as the situational setting, but sampling has been difficult since some lower income families do not have organized meals. This general lack of scheduling or temporal patterning is one of the environmental factors affecting children's development; a second one is the behavioral inconsistency of the adults around children. The qualitative aspects of mothers' behavior and the frequency of unpredictable behavior towards the child are to be examined in this study. The pilot study of five teachers of culturally deprived 5-year-olds, involves gathering half-hour specimen records of teacher behavior in two classroom settings. Teacher behavior will then be divided into episodes (identifiable goal-directed actions). The long range study goal is to find clues on how to select teachers and help them act as constructive agents in the child's environment. (NH)