The literature on the psychological construct of state is reviewed, and it is proposed that state be defined in terms of an infant-environment interaction. Interactive behavior of 32 mother-infant dyads was observed in the home for a total of 2 hours for each pair, in order to explore various types of interactive processes and analyses. A checklist divided into 10-second intervals included various observed behaviors, for example; infant fret/cry, vocalize, play, smile, and eat; and mother touch, hold, vocalize, play, change, and feed. The data seemed to support the proposed model of state, namely that infant condition (behavior) alone was insufficient to describe state since often the same condition had widely different consequences which in turn should affect future conditions. The data also revealed individual differences as a function of the sex of the infant. Briefly, girls received more distal responses to the same behavior for which boys received proximal responses. This finding was discussed as an important source of individual variance and its effect on subsequent cognitive functioning. (Author/NH)