In a classic report, Murray and Trevarthen (Emotional regulation of interaction between two-month-olds and their mothers. In T. Field & N. Fox (Eds), Social perception in infants (pp. 101± 125). Norwood, NJ: Ablex, 1985) suggested that 6-to 12-week-olds were sensitive to contingency in maternal stimulation because they became upset during the replay of their mothers' behavior during an initial period of live interaction over closed circuit TV. However, a number of concerns were raised about the procedure and it was argued that infants may simply have become increasingly fussy over time. To clarify this very important issue, we adopted a three-condition design (live ±replay ±live) and modified Murray and Trevarthen's procedure by inserting a seamless shift from the initially live interaction period to a replay period of good maternal behavior. This was followed by a seamless shift back to a subsequent live interaction. Two-month-old smiling and gaze decreased while frowning and mouth closures increased during the replay period relative to the initial live interaction period, replicating Murray and Trevarthen's results. Fussiness over time did not account for these results because seven of the ten babies increased their positive behaviors and decreased negative behaviors during the second live interaction period, and there was a significant increase in their visual attention to their mothers, thus suggesting a renewed interest in the previously noncontingent partner.

No abstract