INFANT BEHAVIOR AND DEVELOPMENT 9, 107-l 18 (1986) PESTI: A Procedure for Estimating Individual Thresholds in Infant Listeners* SANDRA E.TREHUB,DALE BULL,BRUCEA.SCHNEIDER, ANDBARBARAA.MORRONGIELLO University of Toronto An adaptive procedure far rapid estimation of adult thresholds (PEST: Parameter Estimation by Sequential Testing; Taylor 8 Creelman, 1967) was modified and applied to the estimation of auditory thresholds in 6-month-old infonts (PESTI). This procedure yielded similor absolute thresholds to those obtained in previous research with the method of constant stimuli. Advantages of the present pro- cedure are outlined. infants audition absolute thresholds methodology In the past decade, there has been considerable growth in our knowledge of in- fant auditory perception. This has led, on the one hand, to considerations of the nature of auditory processing in infancy (e.g., Jusczyk, 1981; Kuhl, 1983; Werker & Tees, 1984) and, on the other, to systematic efforts to ascertain the limits of their sound detection abilities (e.g., Berg & Smith, 1983; Sinnott, Pisoni, & Aslin, 1983; Schneider, Trehub, & Bull, 1980; Trehub, Schneider, & Endman, 1980). In the latter domain, the major impediment to progress con- tinues to be methodological-namely, how to obtain reliable information from nonverbal subjects who have limited response repertoires and even more limited attentional resources. Despite the inherent difficulties, several attempts to establish auditory detection thresholds for infant listeners have met with some measure of success for infants 6 months of age and beyond. Typically, head turning toward a sound source has been the target response, and this response has been reinforced by the presentation of an interesting visual event (e.g., the illumination and activation of a mechanical toy). Whereas some investigators have employed unidirectional head turning toward a single sound source and a fixed response interval (Berg & Smith, 1983; Sinnott et al., 1983; Wilson &Thompson, 1984), others have required the infant to turn toward one of two loudspeakers in con- junction with an unlimited response interval (Schneider et al., 1980; Trehub et al., 1980). The use of the two-alternative, forced-choice procedure avoids prob- lems of response bias that could potentially generate underestimates of threshold in the single-response procedure. Moreover, the unlimited response interval l This research was supported by grants from the Medical Research Council of Canada and The University of Toronto. The authors extend thanks to K.J. Kim for technical assistance. Correspondence and requests for reprints should be sent to Sandra E. Trehub, Centre for Research in Human Development, University of Toronto, Mississauga, Ontario, Canada L5L IC6. 107