TY - JOUR
T1 - Spectral weights in level discrimination by preschool children
T2 - Synthetic listening conditions
AU - Willihnganz, Melodie S.
AU - Stellmack, Mark A.
AU - Lutfi, Robert A.
AU - Wightman, Frederic L.
PY - 1997/5
Y1 - 1997/5
N2 - On most auditory discrimination and detection tasks young children perform more poorly than adults. The current experiment applies a technique which potentially can reveal the extent to which the adult-child performance difference results from suboptimal attentional strategies or simply greater internal noise in the children. In this experiment preschool children and adults were asked to discriminate between complex tones comprised of three random-amplitude sinusoidal components. A trial-by-trial correlational analysis [R. A. Lutfi, J. Acoust. Soc. Am. 97, 1333-1334 (1995)] provided an estimate of the weight listeners placed on the level information from individual spectral components in making the discrimination. The patterns of weights were interpreted as measures of 'attentional strategy.' Both children and adults produced reliable patterns of weights. This is an especially important result since measuring a single weighting pattern requires large numbers of trials and hence multiple sessions with the children. While individual weighting patterns were reliable, weighting patterns differed both within and across groups. Moreover, neither the children nor the adults produced weighting patterns that would maximize percent correct in the task. A substantial proportion of the responses from both children and adults could be predicted from their weighting patterns even when performance was near chance. However, differences in overall performance between children and adults could not be accounted for by differences in their weighting functions.
AB - On most auditory discrimination and detection tasks young children perform more poorly than adults. The current experiment applies a technique which potentially can reveal the extent to which the adult-child performance difference results from suboptimal attentional strategies or simply greater internal noise in the children. In this experiment preschool children and adults were asked to discriminate between complex tones comprised of three random-amplitude sinusoidal components. A trial-by-trial correlational analysis [R. A. Lutfi, J. Acoust. Soc. Am. 97, 1333-1334 (1995)] provided an estimate of the weight listeners placed on the level information from individual spectral components in making the discrimination. The patterns of weights were interpreted as measures of 'attentional strategy.' Both children and adults produced reliable patterns of weights. This is an especially important result since measuring a single weighting pattern requires large numbers of trials and hence multiple sessions with the children. While individual weighting patterns were reliable, weighting patterns differed both within and across groups. Moreover, neither the children nor the adults produced weighting patterns that would maximize percent correct in the task. A substantial proportion of the responses from both children and adults could be predicted from their weighting patterns even when performance was near chance. However, differences in overall performance between children and adults could not be accounted for by differences in their weighting functions.
UR - http://www.scopus.com/inward/record.url?scp=0030930533&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0030930533&partnerID=8YFLogxK
U2 - 10.1121/1.419478
DO - 10.1121/1.419478
M3 - Article
C2 - 9165734
AN - SCOPUS:0030930533
SN - 0001-4966
VL - 101
SP - 2803
EP - 2810
JO - Journal of the Acoustical Society of America
JF - Journal of the Acoustical Society of America
IS - 5 I
ER -