Contingent and Noncontingent Reinforcement With Maternal Vocal Imitation and Motherese Speech: Effects on Infant Vocalizations

This paper examines the reinforcing effects of two maternal vocal topographies on infant vocalizations. Two groups of 3- to 8-month-old infants underwent a probe multi-element design with noncontingent reinforcement (NCR) as control condition, and two forms of contingent reinforcement (CR) as treatment: contingent maternal vocal imitation for the first group (n = 19), and contingent motherese speech for the second group (n = 16). Condition sequence was comprised of a baseline phase with no programmed consequences followed by CR1-NCR1-CR2-NCR2. Results revealed that both maternal vocal imitation and motherese speech reinforced infant vocalizations. For the imitation group, frequency of infant vocalizations during NCR1 was lower than CR1 in all but 5 participants and infant vocalization during NCR2 was lower than CR2 in all but 7 participants. For the motherese speech group, frequency of infant vocalizations during NCR1 was lower than CR1 in all but 8 participants and infant vocalizations during NCR2 was lower than CR2 in all but 4 participants. Repeated measures ANOVA indicated that infant vocal responding was significantly higher during CR as opposed to NCR in both groups. Effect magnitudes ranged from moderate to high (Cohen’s d = 0.33-0.99). Although maternal vocalization may have an eliciting effect on infant vocalization, the results of this study support the view that specific forms of contingent maternal vocal stimulation can reinforce early infant vocalizations effectively. These findings may provide a basis for early language acquisition programs in infants with and without delays.