The ability of acoustic characteristics, both separately and in linear combination with others, to differentiate among procedural pain-induced, hungry, and fussy crying was explored using audiorecordings of cries from healthy 2 to 4-month-old infants. Fussy cries were less tense than hungry or pain-induced cries and pain-induced cries had significantly stronger second formant amplitudes than fussy or hungry cries. Formants and tenseness were important contributors to a linear combination of acoustic measures, derived from discriminant function analysis, which correctly classified 74% of the procedural pain-induced crying specimens. The inability of this linear combination of acoustic measures to identify correctly approximately one-third of the cry specimens suggests that the discrete acoustic differences among the three situationally defined types of crying is not large.
© Lippincott-Raven Publishers.