Fig 1.
Sharpness of frequency tuning (Q10) increases with frequency and is higher in humans than in animal models tested.
Values are CAP-Q10 results obtained with an NNFM paradigm. (a) Humans (n = 9); (b) macaque monkeys (n = 5). Different colors indicate different subjects; black lines are trendlines (Robust-LOESS, see Materials and methods); dashed lines indicate 10th and 90th percentiles of the trendlines obtained by resampling (bootstrapping, see Materials and methods). (c) Comparison of CAP-Q10 trendlines in human and monkey (from panels a and b) with cat and chinchilla [22]; shaded areas indicate the area between the 10th and 90th percentile of panels a and b. Underlying data provided in S1 Data. CAP, compound action potential; NNFM, notched-noise forward-masking; Q10, 10 dB quality factor.
Fig 2.
Estimates of frequency tuning of AN single fibers in humans.
Estimates of sharpness of frequency tuning in single AN fibers of human (red) compared with single-fiber tuning of macaque [24] (blue) and other animal models (hatched area), and with other measures of human frequency tuning (green: dotted line, SFOAE [5]; dashed line, psychophysics [4]; both are converted from QERB with conversion factor 0.52). Red solid line: human estimate using conversion curve averaged across 3 species; red dashed line: estimate solely based on conversion curve of macaque. The hatched area outlines data for cat (archival data from our laboratory) and 4 other species [8]. Data provided in S1 Data. AN, auditory nerve; CAP, compound action potential; QERB, quality factor of Equivalent Rectangular Bandwidth; SFOAE, stimulus-frequency otoacoustic emissions.
Fig 3.
Phase-locking in humans extracted from neurophonic data.
(a) Averaged trendlines for human and macaque. (b) Same data anchored to and overlaid with trendline for cat [23]. Dotted line: human trendline normalized to maximum amplitude in cat. Data provided in S1 Data.