Marshall L, Jesteadt W
J Speech Hear Res. 1986 Mar;29(1):82-91. doi: 10.1044/jshr.2901.82.
Audibility thresholds were measured at 500 and 4000 Hz with a standard clinical procedure and a two-interval, forced-choice (2IFC) adaptive procedure for 72 normal-hearing and hearing-impaired listeners, age 17 to 83. Psychometric functions were obtained for clinical, 2IFC, and Yes-No procedures. A measure of response bias was obtained from the Yes-No procedure. The 2IFC adaptive thresholds were 6.5 dB lower than audiological thresholds. The psychometric functions for the forced-choice procedures were generally shallower than those for the clinical procedure and were shifted to lower sound pressure levels. Response bias played a small role at best in accounting for the magnitude of the difference in threshold estimated by the adaptive and clinical procedures or for the differences among the psychometric functions.
采用标准临床程序和双间隔、强制选择（2IFC）自适应程序，对72名年龄在17至83岁之间的听力正常和听力受损的受试者，测量了500赫兹和4000赫兹的听阈。获取了临床、2IFC和是/否程序的心理测量函数。从是/否程序中获得了反应偏差的测量值。2IFC自适应阈值比听力学阈值低6.5分贝。强制选择程序的心理测量函数通常比临床程序的函数更浅，并向较低的声压级偏移。在解释自适应程序和临床程序估计的阈值差异大小或心理测量函数之间的差异时，反应偏差充其量只起了很小的作用。
