TY - JOUR
T1 - On the theoretical error bound for estimating psychometric functions
AU - Dai, Huanping
AU - Richards, Virginia M.
N1 - Funding Information:
We gratefully acknowledge the helpful comments and suggestions provided by Dr. C. Douglas Creelman, Dr. Simon Grondin, Dr. Neil Macmillan, and two anonymous reviewers. The paper also benefited from editing by Eva Maria Carreira. This work was supported by University of Arizona and NIH/NIDCD grants R29 DC01827 (HD) and R01 DC02012 (VMR).
PY - 2011
Y1 - 2011
N2 - The theoretical limits to the amount of error, or the Cramer-Rao bounds, were derived for estimating psychometric functions. These theoretical error bounds were compared with the variability of psychometric functions estimated from human as well as computer-simulated observers. For the simulated observers, due to the limited efficiency of the sampling strategies, including the placement of the signals and the distribution of the trials, the variances of the estimated parameters are seven times the theoretical bound for threshold and 22 times that for slope. For the human observers, the variance is 18 times the theoretical bounds for threshold and 80 times that for slope. Therefore, a major portion of the variances (60% for threshold and 73% for slope) for the human observers is associated with factors other than sampling strategies. Further improvement of the accuracy for estimating psychometric functions will depend on not only optimizing the sampling strategy, but also better understanding the various sources of error related to the behavior of human observers.
AB - The theoretical limits to the amount of error, or the Cramer-Rao bounds, were derived for estimating psychometric functions. These theoretical error bounds were compared with the variability of psychometric functions estimated from human as well as computer-simulated observers. For the simulated observers, due to the limited efficiency of the sampling strategies, including the placement of the signals and the distribution of the trials, the variances of the estimated parameters are seven times the theoretical bound for threshold and 22 times that for slope. For the human observers, the variance is 18 times the theoretical bounds for threshold and 80 times that for slope. Therefore, a major portion of the variances (60% for threshold and 73% for slope) for the human observers is associated with factors other than sampling strategies. Further improvement of the accuracy for estimating psychometric functions will depend on not only optimizing the sampling strategy, but also better understanding the various sources of error related to the behavior of human observers.
KW - Cramer-Rao bound
KW - Psychometric function
UR - http://www.scopus.com/inward/record.url?scp=79959697054&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79959697054&partnerID=8YFLogxK
U2 - 10.3758/s13414-010-0050-1
DO - 10.3758/s13414-010-0050-1
M3 - Article
C2 - 21264715
AN - SCOPUS:79959697054
SN - 1943-3921
VL - 73
SP - 919
EP - 926
JO - Attention, Perception, and Psychophysics
JF - Attention, Perception, and Psychophysics
IS - 3
ER -