The integration of laughter and speech in vocal communication: a dynamic systems perspective.

Laughter in infant-directed speech was examined in 13 mother-infant pairs to investigate the possible co-occurrence of speech and laughter. Contrary to previous findings in adult-adult social interaction, all mothers produced speech simultaneously with laughter in up to 50% of laughs. In most of these speech-laughs the onset of laugh and speech was simultaneous. Laughter occurred on both function and content words and was more likely to occur on approximately 2 words and on utterances that were statements rather than questions or exclamations. Laughter and speech are different outcomes produced from a reorganization of the same vocal/anatomical parameters. A 3rd outcome is possible in the form of speech-laughs utilizing features from both laughter and speech. In speech-laughs, the duration of the vocalization was more likely to increase, and the changes in the utterance were likely to include 1 or more of the features of vowel elongation, syllabic pulsation, breathiness, and pitch change. These findings and individual variations in the resulting vocal output are discussed from a dynamic systems perspective. It is argued that neither speech nor laughter is dominant when both are combined, but that this is a more complex vocal outcome produced with idiosyncratic flexibility within stable temporal and physiological constraints.

Vocal affect in three-year-olds: a quantitative acoustic analysis of child laughter.

Recordings were obtained of the laughter vocalizations of four 3-year-old children during three sessions of spontaneous free-play between mother and child in a laboratory playroom. Acoustic analysis was used to determine laughter durations, laughter events, F0, and harmonic characteristics, and to suggest a taxonomy of laughter types. Melodic contours were assessed from patterns of F0 change during laughter. Mean duration of laughs ranged from 200 ms to 2.0 s, but events within a laugh were usually about 200-ms duration. Laughs were intuitively classified into four major types, and, following the acoustic analyses, were further defined and classified into types and subtypes of exclamatory and dull comment; chuckle; basic, variable, and classical rhythmical; and squeal. Melodic contours included more rising contours than previously reported for cry, but there was great variability in the types of contours produced especially for rhythmical laughs. The results of the acoustic analyses are discussed in relation to (a) the development of a taxonomy of laughter and (b) different features of the vocal affect characteristics of high-intensity emotion.