Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard — Psychoacoustics- Facts and Models

Yield

659 of 3380 paragraphs kept (19.5%). Target for mining: 30%–50% — ⚠️ BELOW target.

Pre-Assessment

This is a well-established, standard academic reference text in psychoacoustics. Hugo Fastl was a professor at the Technical University of Munich and a leading figure in auditory perception research, co-authoring this work with Eberhard Zwicker, who essentially founded the German school of psychoacoustics with his classic models of loudness, masking, and critical bands. As a “Facts and Models” title, it is deliberately structured to deliver precisely what the reading stance promises: empirical data alongside the quantitative models (loudness, fluctuation strength, sharpness, roughness, masking) that organize those facts. The edition matters for currency — the Zwicker models are decades old and largely settled science; newer editions incorporate Fastl’s later work but the field’s core models haven’t radically shifted. Expected yield is heavily weighted toward concepts/frameworks (the models and their derivations) backed by a solid factual/empirical base, with almost none of the inspirational or speculative register. For a mining read, this is an ideal source: dense, systematic, and authoritative, though notation-heavy and best read with general linear algebra and signal-processing fundamentals in hand. I can’t verify which edition you hold without more detail, but the authority of the authors and the purpose-fit of the text are unambiguous.

Reading Goal

Prioritizing: Factual Claims (Reader wants facts about psychoacoustics, so empirical findings are central.), Definitional Content (Understanding concepts requires clear definitions of key terms and models.), Causal Relationships (Models often explain cause-and-effect between physical stimuli and perception.), Contrarian or Surprising Claims (Counterintuitive perceptual phenomena are part of mastering the subject.), Connection Points (Tying psychoacoustics to neuroscience and physics aids model understanding.) De-prioritizing: Craft and Method (Method is secondary to comprehension-focused reading.), Inspiration and Speculation (Reader seeks established knowledge, not speculative extrapolation.)

Thesis

Human hearing spans sound pressures from 10-5 to 102 Pa · ch. 0

Argument

1. White noise shows frequency-flat long-term spectra but frequency-dependent short-term spectra — ch. 0
2. An amplitude-modulated tone is a carrier plus two side lines — ch. 0
3. Gaussian-shaped tone bursts minimize spectral splatter in gated tones — ch. 0
4. A single tone burst yields a continuous spectrum while periodic pulses yield line spectra — ch. 0
5. Noise envelope fluctuation rate is inversely proportional to bandwidth — ch. 0
6. Periodic gating creates spectral line spacings set by the pulse repetition rate — ch. 0
7. Psychoacoustics is the quantitative science of the ear as an information receiver. — ch. 0
8. Filter bandwidth limits how fast a band-pass noise envelope can fluctuate — ch. 0
9. Gaussian shapes minimize the bandwidth-duration product — ch. 0
10. Psychoacoustic phenomena can be experienced directly through audio demonstrations. — ch. 0
11. Median and interquartile range stay scale-invariant where means fail — ch. 1
12. Active stimulus control by subjects can bias psychophysical measurements — ch. 1
13. Switching a tone on and off generates broadband transient clicks in distant frequency bands — ch. 1
14. Outer haircells act locally within the cochlea to modulate inner haircells — ch. 1
15. Outer haircells amplify quiet sounds only, while inner haircells respond directly to loud ones. — ch. 1
16. Human hearing spans 150 dB and three decades of frequency — ch. 1
17. The four otoacoustic emission types are distinguished by whether and how they are evoked — ch. 1
18. Physical stimuli map to separate hearing sensations only within audible physical ranges — ch. 1
19. Forced-choice procedures inform subjects after each trial which interval held the signal. — ch. 1
20. Abrupt tone onset and offset generate broadband transient energy at distant frequencies. — ch. 1
21. Hearing is most damage-prone in the 3-8 kHz frequency range — ch. 1
22. Normal hearing is most sensitive near 2 to 5 kHz — ch. 1
23. Transforming a stimulus scale shifts where means but not medians fall — ch. 1
24. The inner ear acts as a linear system offering both frequency and temporal resolution. — ch. 1
25. The inner ear amplifies quiet sounds by up to 30 dB through active feedback. — ch. 1
26. Tracking methods differ from adjustment by letting subjects control only direction, not magnitude — ch. 1
27. Psychoacoustic tuning curves directly reflect cochlear mechanical processing. — ch. 1
28. Young headphone users already show measurable hearing damage in the 3-12 kHz range — ch. 1
29. Halving exposure duration permits a 10 dB increase in safe sound level — ch. 1
30. A low-level-only dip in tuning curves reveals the ear’s active frequency-enhancing processes — ch. 1
31. The threshold in quiet is highly reproducible for individual subjects — ch. 1
32. Psychoacoustic results depend on experimental procedure and sensitivity rises with alternatives. — ch. 1
33. Psychoacoustic headphones distort sound unless free-field equalized to compensate their band-pass response — ch. 1
34. The cochlea can be modeled as a bank of overlapping asymmetric bandpass filters — ch. 1
35. Hearing damage risk halves allowed exposure time for every roughly 10 dB increase — ch. 1
36. The ear is most easily damaged between 3 and 8 kHz — ch. 1
37. The median resists stimulus-scale transformations, unlike arithmetic and geometric means — ch. 1
38. Music occupies a wider frequency and dynamic range than speech — ch. 1
39. Hearing sensations correlate with stimulus magnitudes but only within the audible range — ch. 1
40. Feedback in outer haircells enhances sensitivity while lateral spread shapes frequency selectivity. — ch. 1
41. Cochlear amplification is instantaneous and symmetrically saturating in its transfer. — ch. 1
42. Magnitude estimation lets listeners assign numbers proportional to perceived loudness. — ch. 1
43. Repeated noise overexposure converts temporary threshold shifts into permanent hearing loss — ch. 1
44. The inner ear can be modeled as an array of asymmetrical overlapping bandpass filters. — ch. 1
45. Threshold in quiet is individually reproducible to within plus or minus six decibels — ch. 1
46. Spontaneous otoacoustic emissions show a minimum frequency spacing that grows with frequency — ch. 1
47. The choice of averaging unit among level, pressure, or intensity critically shapes threshold data. — ch. 1
48. Inner hair cells form one row while outer hair cells form three rows — ch. 1
49. Absolute hearing threshold is a statistical construct, defined at 50% detection probability. — ch. 1
50. Inner and outer hair cells differ structurally and likely function differently — ch. 1
51. Auditory tuning curves derive from constant basilar membrane velocity rather than displacement — ch. 1
52. Outer hair cells actively amplify and sharpen sound before inner hair cells fire — ch. 1
53. Yes-No procedures force signal detection decisions without an unsure option — ch. 1
54. Bekesy-tracking measures threshold curves by letting the subject reverse intensity direction — ch. 1
55. Adaptive procedures adjust stimulus intensity based on prior responses and converge on thresholds — ch. 1
56. Basilar membrane displacements during speech are only tens of nanometers, near atomic scale — ch. 1
57. Spontaneous emission frequencies remain stable to within one percent — ch. 2
58. Present spontaneous emissions indicate good hearing, absent ones prove nothing — ch. 2
59. Suppressor tones reveal tuning curves matching neural frequency selectivity — ch. 2
60. SOAE levels fluctuate widely while their frequencies stay remarkably stable — ch. 2
61. Suppressor tones that reduce SOAEs map the ear’s fine frequency selectivity — ch. 2
62. Delayed evoked emissions are the ear’s response to short impulses — ch. 3
63. Spontaneous and evoked otoacoustic emissions likely share a single source — ch. 3
64. Delayed evoked emissions respond linearly to impulses below about 18 dB sensation level — ch. 3
65. Most human ears display synchronized spontaneous otoacoustic emissions when driven — ch. 3
66. Delayed otoacoustic emissions grow linearly then saturate 10-20 dB above threshold — ch. 4
67. Delayed otoacoustic emissions obey linear superposition at low amplitudes — ch. 4
68. Suppression of evoked emissions tracks the audibility of the evoking tone — ch. 5
69. Delayed evoked otoacoustic emissions show discrete spectral peaks rather than continuous energy — ch. 5
70. Delayed otoacoustic emissions arrive faster for higher stimulus frequencies — ch. 5
71. Inner hair cells only convert oscillations to spike rates in the model — ch. 6
72. Masking disrupts everyday speech and music whenever competing sounds differ in level — ch. 6
73. Uniform masking noise mirrors white-noise masking and gives flat thresholds — ch. 6
74. Density level converts to total level once bandwidth is known — ch. 6
75. Simultaneous masking raises a test tone’s just-audible level above threshold in quiet. — ch. 6
76. Outer and middle ear act linearly while inner-ear mechanics are nonlinear — ch. 6
77. Critical bandwidth stays constant near 100 Hz up to 500 Hz then grows — ch. 6
78. Spacing of hearing-threshold minima matches the 180-degree phase-shift frequency distance — ch. 6
79. Otoacoustic emission suppression mirrors hearing threshold masking patterns — ch. 6
80. Masking can occur even when the masker follows the test sound in time. — ch. 6
81. Masking pattern shapes differ by center frequency even at equal masker level — ch. 6
82. Spontaneous otoacoustic emissions align with peaks in stimulus-frequency fine structure — ch. 6
83. Masked threshold always exceeds quiet threshold except when masker and test frequencies differ greatly — ch. 6
84. Outer hair cells act as nonlinear amplifiers with bidirectional lateral feedback — ch. 6
85. Low-frequency suppressor impulses suppress cochlear emissions most at rarefaction peaks — ch. 6
86. Basilar-membrane models are validated by frequency responses comparable to tuning curves — ch. 6
87. Threshold in quiet has a fine structure of maxima and minima. — ch. 6
88. Masking is graded, with partial masking reducing loudness rather than audibility. — ch. 6
89. Analog hearing models let experimenters measure physical states real ears hide — ch. 6
90. A single source generates all three kinds of otoacoustic emissions. — ch. 6
91. Low-frequency suppressors reduce emissions by saturating cochlear feedback-loop nonlinearity — ch. 6
92. Distortion product emissions are objectively measurable in the ear canal — ch. 6
93. Frequency-ripple spacing and emission delay time are inversely related — ch. 6
94. Forward masking follows masker termination and is also called post masking — ch. 6
95. Two tones presented together produce audible combination tones in the ear — ch. 6
96. White noise masking of tones is linear but frequency-dependent above 500 Hz — ch. 6
97. Individual auditory fibres cover only 40-60 dB yet we hear over 100 dB range — ch. 6
98. Broad-band noise masking erases individual differences in quiet threshold — ch. 6
99. Delayed otoacoustic emissions arise from decaying basilar-membrane oscillations that sum in phase — ch. 6
100. Basilar-membrane resonance peaks shift downward as input level rises — ch. 6
101. Combination tones can be cancelled by adding matched opposite-phase tones — ch. 6
102. Distortion product emissions arise only in a live ear, not passive cavities — ch. 6
103. Masking measurements require noise slopes steeper than the ear’s own selectivity — ch. 6
104. Equal-level narrow-band maskers mask less at higher centre frequencies — ch. 6
105. Combination tones arise from the hearing system’s own nonlinearity — ch. 7
106. Masked threshold rise toward high frequencies is about 100 dB per octave. — ch. 7
107. Instrument timbre differences stem from harmonic spectra that shape masking breadth — ch. 7
108. Tonal maskers create distinct audible regions beyond simple masking — ch. 7
109. Odd-order difference tones explain why single tones mask differently than noise — ch. 7
110. Five tones replicate narrow-band noise masking while single tones distort it — ch. 7
111. Masking by bandlimited noise depends only on masker energy within the critical band. — ch. 7
112. Masking pattern curves are scale-invariant and predictable by horizontal shifts — ch. 7
113. The near-miss of Weber’s law means threshold slopes can reach 3 to 6 — ch. 7
114. Tonal masker spread inverts from low to high frequencies as masker level rises — ch. 7
115. Harmonic spacing on a log scale determines masking-pattern dip size — ch. 7
116. The masked threshold’s high-frequency slope steepness depends on masker level — ch. 7
117. Postmasking threshold decay is not exponential despite fitting one approximately. — ch. 8
118. Auditory threshold for broadband noise is set by the most sensitive frequency region only — ch. 8
119. Postmasking stays at simultaneous-masking level for 5ms then decays to quiet by 200ms. — ch. 8
120. Low-frequency maskers mask most during their rarefaction phase — ch. 8
121. A masker can mask sounds occurring before it is switched on. — ch. 8
122. Masking-period patterns require masker masking and test bursts under an eighth of the period — ch. 8
123. Temporal masking patterns comprise simultaneous, post, and premasking effects — ch. 8
124. Internal nonlinear distortion products, not poorer frequency selectivity, explain noise maskers’ extra low-frequency masking — ch. 8
125. Psychoacoustical tuning curves grow more asymmetric at higher test-tone levels — ch. 8
126. Masking-period patterns reveal how masker phase within its period affects audibility — ch. 8
127. Aperiodic noise maskers are studied using repeated random pulse sequences with long periods. — ch. 8
128. Very brief test tones stay within a critical band only at high centre frequencies. — ch. 8
129. Masked thresholds spike within milliseconds of masker onset before settling — ch. 8
130. Broadband maskers overshoot-effect short tones but narrowband maskers do not — ch. 8
131. Short masker gaps are bridged by postmasking effects — ch. 8
132. Post masking outlasts premasking and asymmetrically distorts threshold patterns. — ch. 8
133. Test bursts stay within an eighth of the low-frequency masker’s period — ch. 8
134. Plosive consonants get masked by preceding loud vowels even in free field — ch. 8
135. Masking occurs before, during, and after a masker is present. — ch. 8
136. Noise-burst masked thresholds rise more gradually than tone-burst thresholds with shorter duration — ch. 8
137. Strong condensation can suppress masking below the quiet threshold — ch. 8
138. Premasking measurements are only reproducible with trained subjects and remain noisy — ch. 8
139. Asymmetric pre- and post-masking durations distort symmetric reversals in FM masking patterns. — ch. 8
140. Frequency-modulated maskers produce two masking peaks where instantaneous frequency crosses the test tone twice — ch. 8
141. Repeated short tones reach steady threshold when repetition rate exceeds 5 per second — ch. 8
142. Temporal masking alone masks brief consonants even without room reverberation. — ch. 8
143. Reducing audible fluctuations in a masker lowers post-masked thresholds by up to 24 dB — ch. 8
144. Postmasking depends strongly on masker duration and is highly nonlinear. — ch. 8
145. Envelope-structured sounds produce time-variant temporal masking patterns. — ch. 8
146. Post masking outlasts premasking and can skew frequency-modulation measurements by nearly 20 dB. — ch. 8
147. The ear follows frequency-modulated masker dynamics only up to about 8 Hz modulation — ch. 8
148. Postmasking decay needs two separate time constants depending on masker duration — ch. 8
149. Shorter masker and test-tone durations flatten the lower slope of pulsation patterns — ch. 9
150. Positive second derivative of masker drives low-frequency masking-period patterns — ch. 9
151. Pulsation patterns map test-tone level against frequency like masking patterns do — ch. 9
152. A gated tone sounds continuous at a level called the pulsation threshold — ch. 9
153. Pulsation patterns can replicate masking patterns to under 1 dB accuracy — ch. 9
154. Alternately presented sounds can be perceived as continuous when actually gated — ch. 9
155. Masking depends on the masker’s acceleration, not its amplitude. — ch. 9
156. Rarefaction, not masker level, drives the audibility threshold maximum — ch. 9
157. Simulated Gaussian noise with too few components gives atypical masking results — ch. 11
158. Intensity-duration tradeoff in masking holds only for durations under 200 ms — ch. 11
159. Nonsimultaneous masking summation resists simple generalizations — ch. 11
160. Masking reduces to a comparison of excitation-level patterns over frequency and time — ch. 11
161. Two simultaneous maskers partition the spectrum where each dominates — ch. 11
162. Near threshold, masked threshold depends on both masker level and threshold-in-quiet via internal noise — ch. 11
163. Adding multiple maskers raises masked thresholds most when each contributes equally — ch. 11
164. Adding two post-maskers together can reduce rather than increase masking — ch. 11
165. Masking increments peak when the two individual maskers yield equal threshold shifts — ch. 11
166. Pairs of incoherent or coherent noises can be modeled as amplitude modulation — ch. 11
167. Combined maskers interact predictably in simultaneous but not nonsimultaneous masking — ch. 11
168. Post-masking depends on masker bandwidth as well as duration — ch. 11
169. The minimum masker excitation within its modulation cycle sets masked threshold — ch. 11
170. Pure-tone versus noise masking differences depend on masker frequency — ch. 11
171. A frequency-modulated masker’s excursion along the critical-band scale sets the test-tone threshold. — ch. 11
172. Masking relates closely to excitation level patterns. — ch. 11
173. Postmasking temporal decay is realizable by a two-resistor two-capacitor diode network — ch. 11
174. Critical bandwidth varies a hundredfold across the audible range. — ch. 11
175. Combined masking adds most when the two maskers are equally strong — ch. 11
176. Simultaneous masking originates in the cochlea while nonsimultaneous masking does not — ch. 11
177. Post-masking decreases when a second masker suppresses audible masker fluctuations — ch. 11
178. Masking-induced pitch shift is a product of three pattern-derived factors — ch. 12
179. Amplitude-modulated broadband noise yields only faint pitch despite flat spectrum — ch. 12
180. Bandpass noise pitch corresponds to spectral edge frequencies which fuse when close — ch. 12
181. A spectral gap in switched-off noise yields a tonal afterimage, not a noisy one — ch. 12
182. Low-pass noise can mask harmonics to create the same virtual-pitch stimulus as filtering. — ch. 12
183. Most everyday sounds are harmonic complex tones, not pure tones — ch. 12
184. The spectral pitch model was tuned by fitting published experimental data — ch. 12
185. Complex-tone pitch tracks the spectral pitch of lower components — ch. 12
186. Spectral pitch model corrections are segmented into adjustable level ranges — ch. 12
187. Removing lower harmonics barely changes the perceived pitch of a complex tone. — ch. 12
188. Maskers below a tone raise its pitch while maskers above lower it. — ch. 12
189. Medium-level band-stop noise around 43 dB most reliably produces Zwicker tones — ch. 12
190. Ratio pitch curve shape matches other psychoacoustic frequency-dependence curves — ch. 12
191. Rippled noise generated by feedback delay produces distinct pitch sensations. — ch. 12
192. Steep-filtered noise produces pitch matching the filter’s cut-off frequency — ch. 12
193. Incomplete complex tones show twice the pitch deviation of complete ones — ch. 12
194. Raising the level of a complex tone shifts its pitch downward. — ch. 12
195. Narrower noise bands produce single centre pitch rather than edge pitches — ch. 12
196. Virtual pitch arises only for certain fundamental-to-lowest-component combinations — ch. 12
197. Pitch of a tone is defined by the matching 40 dB reference frequency — ch. 12
198. Shifting all spectral components uniformly changes perceived virtual pitch in a sawtooth pattern — ch. 12
199. Uniformly shifting a harmonic complex tone’s components changes pitch non-monotonically with a sawtooth pattern — ch. 12
200. Spectral pitch of pure tones derives from three magnitudes of masking patterns — ch. 12
201. Pure tone pitch shift decomposes additively into level and masking components — ch. 12
202. Perceived pitch compresses against frequency above roughly 500 Hz — ch. 12
203. Level-induced pitch shift depends on masking height above the quiet threshold — ch. 12
204. The gap width producing a Zwicker tone depends on line spacing — ch. 13
205. Pitch strength is quantifiable by magnitude estimation across diverse sounds — ch. 13
206. Pitch as a sensory dimension has both height and perceptual strength, which vary independently. — ch. 13
207. Zwicker-tone pitch sits exactly 1 Bark above the gap’s lower edge — ch. 13
208. Pitch strength results were normalized per frequency region for comparison across panels. — ch. 13
209. Higher-frequency gaps need narrower spectral widths to trigger Zwicker tones — ch. 13
210. Pitch strength rises only modestly with level while loudness explodes. — ch. 15
211. Pitch strength grows linearly with log duration up to 300 ms. — ch. 15
212. Steep low-pass noise slopes yield pitch strength far below that of a pure tone — ch. 16
213. Peaked ripple noise pitch strength stays stable across a wide level range — ch. 16
214. Narrow spectral bands with steep slopes produce stronger pitch than narrow spectral gaps do — ch. 16
215. Ripple-noise pitch strength grows linearly with log spectral modulation depth beyond 10 dB — ch. 16
216. Amplitude discrimination depends on relative loudness increment, not absolute change — ch. 17
217. A 1 dB level change anchors the model of just-noticeable slow variations. — ch. 17
218. Auditory critical bandwidth stays constant at 100 Hz below 500 Hz then rises proportionally. — ch. 17
219. Excitation patterns can be computed from masking data in either direction — ch. 17
220. Frequency modulation detection is level-independent but thrice frequency-difference detection — ch. 17
221. Hearing detects AM better than FM only when sidebands fall within a critical bandwidth — ch. 17
222. Abrupt level changes create audible clicks via broad short-time spectra. — ch. 17
223. The ear’s upper excitation slope is essential for detecting amplitude modulation at high levels — ch. 17
224. White noise’s critical-band intensity rises above 5 Bark because bandwidth grows there — ch. 17
225. Frequency-discrimination and frequency-modulation thresholds are commonly conflated in the literature — ch. 17
226. Fletcher defined critical bands as noise regions whose power equals a just-masked tone’s — ch. 17
227. Comparing two tones improves frequency sensitivity beyond detecting a continuous modulation — ch. 17
228. Low-frequency loudness growth compresses as level rises — ch. 17
229. Tone-on-tone masking thresholds are measured via narrow-band noise maskers instead — ch. 17
230. Loudness stays constant below critical bandwidth then grows beyond it — ch. 17
231. Amplitude-modulated broadband noise yields only 15% of a pure tone’s pitch strength. — ch. 17
232. Pitch can be modeled as a positional sensation scaled by discrimination steps — ch. 17
233. The excitation pattern’s upper slope changes nonlinearly with level, unlike the main peak — ch. 17
234. Just-noticeable level variations always exceed just-noticeable level differences — ch. 17
235. Raising noise level paradoxically worsens cut-off frequency discrimination as upper slopes flatten. — ch. 17
236. Amplitude modulation and frequency modulation differ only by sideband phase at low indices — ch. 17
237. Constant 1-Bark windows yield triangular noise patterns and rectangular tone patterns — ch. 17
238. Pitch shift magnitude correlates inversely with pitch strength of masked tones — ch. 17
239. Instrument tuning relies on high harmonics because pitch changes there are easier to hear — ch. 17
240. Fletcher’s equal-power assumption implies a low-frequency critical band near 50 Hz — ch. 17
241. Detecting level variation is about 2.5 times harder than detecting level difference — ch. 17
242. Equal-loudness contours flatten at low frequencies as level rises — ch. 17
243. Postmasking can last up to 200 milliseconds after a masker ends — ch. 17
244. Uniform masking noise can substitute for threshold-in-quiet across all frequencies. — ch. 17
245. Main excitations make critical-band level equal excitation level at signal centers — ch. 17
246. Hearing becomes less sensitive to frequency variations as level increases — ch. 17
247. Just-noticeable amplitude modulation of narrow-band noise is near 6 percent at high centre frequencies — ch. 17
248. A 1 dB level change marks a recurring just-noticeable unit in hearing — ch. 17
249. Internal physiological noise from blood flow and heartbeat raises low-frequency thresholds — ch. 17
250. Two-tone maskers define critical bandwidth at the threshold’s transition point. — ch. 17
251. Changing a tone’s phase alters both its temporal envelope and its instantaneous frequency. — ch. 17
252. Masked threshold rises once noise bandwidth exceeds the critical bandwidth — ch. 17
253. White noise shows no sideband-mediated detection ceiling for AM. — ch. 17
254. Auditory filters respond asymmetrically, favoring frequencies below center — ch. 17
255. The ear ignores sideband phase once components fall outside one critical band — ch. 17
256. Amplitude modulation faster than the tone’s own steady pitch is heard as pitch change — ch. 17
257. Just-noticeable amplitude modulation of noise shrinks from 40% to 6% as bandwidth widens. — ch. 17
258. Frequency deviation measures one-sided swing while total variation doubles it — ch. 17
259. Short tone duration inflates JNDF, and level affects low-pass but not high-pass cutoff discrimination. — ch. 17
260. Binaural envelope-delay sensitivity drops once tone separation exceeds the critical band. — ch. 17
261. Just-noticeable frequency increments are constant along critical-band rate but rise as pitch strength weakens — ch. 17
262. Masking shifts detection thresholds and may affect just-noticeable sound changes — ch. 17
263. Critical bandrate scale stabilizes frequency discrimination measures across durations — ch. 17
264. Closing the ear canal makes internal body noise audible and measurable — ch. 17
265. Critical-band rate and excitation level are the intermediate transforms for hearing models — ch. 17
266. Masked tones recover pitch strength as their level rises above masked threshold — ch. 17
267. Raising low-pass noise level measurably raises its perceived pitch — ch. 17
268. A critical band spans roughly 1.3 mm, some 145 hair cells wide. — ch. 17
269. Pitch from gated noise vanishes when gaps shrink below temporal resolution — ch. 17
270. Excitation slopes become frequency-independent above 200 Hz center frequency — ch. 17
271. Loudness measures span quiet, masked, spectral, and temporal domains — ch. 17
272. Masking shifts thresholds upward and may alter just-noticeable sound changes at all levels. — ch. 17
273. Narrow-band noise keeps a Gaussian amplitude distribution that statistically modulates its perceived sound — ch. 17
274. JNDL level dependence unifies across frequency when scaled to loudness level — ch. 17
275. Maskers’ influence on modulation thresholds vanishes beyond 20 dB above their excitation pattern. — ch. 17
276. Low-pass noise cutoff discrimination degrades with level while high-pass cutoff does not — ch. 17
277. Slow amplitude modulation reduces pure tone pitch strength by ten to twenty percent — ch. 17
278. The phon scale defines loudness relative to equal-loudness matches at 1 kHz — ch. 17
279. Just-noticeable changes are tenfold worse at masked threshold then normalize 20 dB above — ch. 17
280. The steep lower slope of the excitation pattern determines frequency modulation detection threshold — ch. 17
281. Sound direction and field type change perceived loudness at a fixed level. — ch. 17
282. Level variations within a sound and differences between sounds activate distinct hearing processes. — ch. 17
283. Phase-change detection yields a critical bandwidth of 150 Hz at 1 kHz — ch. 17
284. Above a threshold, modulation is heard as separate sideband tones rather than as flutter. — ch. 17
285. Phase changes become audible in three-tone complexes — ch. 17
286. Narrow-band noise’s intrinsic amplitude jitter masks external amplitude modulation detection — ch. 17
287. Barkhausen created the loudness level measure in the 1920s — ch. 17
288. Band-pass noise audibly self-modulates far above its threshold — ch. 17
289. Partially masked sounds have a reduced loudness range in both time and spectrum — ch. 17
290. Rapid AM of noise is heard via overall level, not its rapid fluctuations — ch. 17
291. Narrow critical bandwidths at low frequencies reduce hearing sensitivity via noise fluctuations — ch. 17
292. Threshold perception integrates total sound intensity across a limited frequency bandwidth. — ch. 17
293. Level discrimination for broadband noise is level-independent above 40 dB sound pressure — ch. 17
294. Frequency discrimination stems from steep filter slopes plus 1 dB level sensitivity — ch. 17
295. Loudness is measured subjectively against 1 kHz reference tones — ch. 17
296. Equal-power masking assumption is wrong and only partly power detection holds — ch. 17
297. Slope excitation can transmit low frequencies to high critical-band regions. — ch. 17
298. Loudness is the sensation most directly tied to sound intensity — ch. 17
299. The audible range spans 24 bark-scaled critical bands totaling 0 to 24 Bark. — ch. 17
300. Shorter tone bursts drastically worsen level discrimination while longer ones do not help — ch. 17
301. Critical bandwidth at one kilohertz is about 150 hertz — ch. 17
302. The masking index varies with frequency, distinguishing UEN from uniform masking noise. — ch. 17
303. Critical bandwidth stays constant with masker level up to about 50 dB. — ch. 17
304. Frequency discrimination thresholds rise dramatically from 500 Hz to 5000 Hz — ch. 17
305. Frequency just-noticeable differences rise sharply at low sensation levels — ch. 17
306. Just-noticeable frequency shifts map to constant basilar distances — ch. 17
307. Internal bodily noise raises the hearing threshold at low frequencies — ch. 17
308. Frequency discrimination depends on pitch strength and worsens as pitch strength falls — ch. 17
309. For broadband noise, AM detection extends up to 500 Hz without sideband interference — ch. 17
310. Large cutoff-frequency sweeps boost low-pass noise pitch strength 2.4-fold — ch. 17
311. Noise’s inherent statistical fluctuations degrade frequency-change detectability relative to tones. — ch. 17
312. Hearing selectivity is modeled via critical-band and excitation intermediate values — ch. 17
313. At very high modulation rates, listeners detect AM only via average intensity change. — ch. 17
314. Wider noise bandwidths decrease detectable amplitude modulation via multi-band cooperation — ch. 17
315. A-weighted dB values misrepresent perceived loudness for complex sounds. — ch. 17
316. Threshold in quiet depends on total sound intensity within one critical band only — ch. 17
317. A just-noticeable frequency shift moves basilar excitation a constant 0.05 mm. — ch. 17
318. Frequency is a poor physical scale for describing inner-ear effects; critical-band rate works better. — ch. 17
319. Just-noticeable amplitude modulation drops steeply as tone level rises — ch. 17
320. One just-noticeable pitch step shifts excitation across about six inner hair cells — ch. 17
321. Critical-band rate correlates with pitch, frequency discrimination, and basilar-membrane position. — ch. 17
322. Sequential tone comparison yields finer discrimination than modulation detection. — ch. 17
323. Phase-change sensitivity in hearing is bounded by critical bandwidth, scaling with centre frequency. — ch. 17
324. Above 500 Hz, Bark scale maps logarithmically to frequency change — ch. 17
325. Level discrimination thresholds grow steeply near hearing threshold — ch. 17
326. The basilar membrane yields only about 640 just-noticeable frequency steps end to end. — ch. 17
327. Overlapping excitation slopes raise the valley minimum by at least 3 dB — ch. 17
328. Excitation detectors are modeled as independently acting channels with occasional coherence effects — ch. 17
329. Masking-pattern slope predicts low-pass noise pitch strength better than filter slope does — ch. 17
330. Adding tones lowers threshold only until about nine tones span the critical band — ch. 17
331. Temporal partial masking of loudness is robust across sound types and frequencies — ch. 18
332. Everyday speech communication usually occurs under partial masking conditions — ch. 18
333. Adjustment methods bias moderate sounds toward louder settings for good audibility. — ch. 18
334. Loudness perception takes time to develop and so can be interrupted by later sounds — ch. 18
335. Critical bandwidth at 1 kHz is 160 Hz and governs loudness behavior — ch. 18
336. Ten-phon increments double loudness, mapping phon to sone by powers of two — ch. 18
337. Partial masking steepens the loudness function in a way that mirrors hearing recruitment — ch. 18
338. Medium-separated tones add by intensity once the ear stops tracking envelope — ch. 18
339. Loudness falls to zero at threshold in quiet, forcing a vertical asymptote — ch. 18
340. Above 10 Hz, loudness grows 10 phon per tenfold rise in repetition rate. — ch. 18
341. Total loudness is the integral of specific loudness over critical-band rate — ch. 18
342. Tone-burst loudness rises 10 phon per tenfold duration up to 100 ms, then plateaus. — ch. 18
343. Phase differences affect loudness only at low modulation frequencies the ear can track — ch. 18
344. Uniform exciting noise delivers equal intensity to every critical band — ch. 18
345. Two tones within a critical band interact to reduce total loudness — ch. 18
346. Critical bandwidth marks the point where noise loudness begins to grow with bandwidth — ch. 18
347. Loudness doubles per 10 phon except below 40 phon where it drops steeply. — ch. 18
348. Widely separated tones add by loudness, not pressure or intensity — ch. 18
349. A following loud sound reduces the loudness of an earlier short sound — ch. 18
350. Noise loudness is level-independent until bandwidth exceeds the critical bandwidth. — ch. 18
351. Steady-state loudness models fail for speech, music, and impulsive technical noises. — ch. 18
352. Loudness perception shows distinct behavior in two critical-band-defined regions — ch. 18
353. Narrow-band noise produces substantial loudness in critical-band regions where no physical sound exists — ch. 19
354. Continuous loudness curves assume the ear has hundreds of movable critical-band filters — ch. 19
355. Total loudness requires linear scales before integration is valid — ch. 19
356. Free-field versus diffuse-field transmission gains differ by under eight decibels up to 8 kHz — ch. 19
357. Wider bandwidth increases perceived loudness at equal sound pressure — ch. 19
358. Any sound narrower than a critical band produces identical loudness regardless of its width. — ch. 19
359. Loudness grows as a power law of physical intensity so relative changes match — ch. 19
360. Threshold in quiet is attributed to internal physiological noise — ch. 19
361. Specific loudness derives from Stevens’ power law applied to excitation — ch. 19
362. Spectral summation precedes temporal summation in loudness processing — ch. 19
363. Specific loudness generalizes beyond loudness to describe sharpness — ch. 19
364. Uniform exciting noise loudness grows as intensity raised to the 0.23 power — ch. 19
365. A single exponent of 0.23 governs the loudness power law — ch. 19
366. Model-computed and measured loudness match across the full level range — ch. 19
367. Total loudness includes contributions outside the stimulus’s own band — ch. 19
368. Loudness grows with bandwidth via area tradeoff except at extreme levels — ch. 19
369. Specific loudness is better expressed against excitation level than against linear excitation — ch. 19
370. Excitation slopes are steeper toward lower frequencies than higher ones — ch. 19
371. Loudness rises with bandwidth beyond the critical band despite falling pattern height — ch. 19
372. The G subscript indicates loudness computed from critical-band levels — ch. 19
373. Specific loudness rises steeply then saturates toward a power-law asymptote — ch. 19
374. Threshold in quiet affects loudness mainly for faint sounds — ch. 19
375. Sharpness changes little with level and bandwidth below one critical band. — ch. 20
376. Broadband factory noise loudness in phons far exceeds its measured dB level — ch. 20
377. Adding sound intensities of equal level yields only a three-decibel increase — ch. 20
378. Spectral envelope, not fine structure, determines perceived sharpness — ch. 20
379. Loudness processing acts as a temporal low-pass filter with about 45 ms integration time — ch. 20
380. Loudness patterns approximate specific loudness by ignoring the lower slope entirely. — ch. 20
381. Timbre is a residual category masking distinct sensations like sharpness and sensory pleasantness. — ch. 20
382. Third-octave loudness computation requires steady sounds but handles any spectrum — ch. 20
383. High modulation frequencies increase loudness because spectral spread outweighs temporal fluctuation — ch. 20
384. ISO 532 B loudness charts span five maximal band levels and two sound-field types — ch. 20
385. Noise reduction should target the largest loudness pattern area — ch. 20
386. Vowels dominate perceived loudness while consonants carry intelligibility — ch. 20
387. Frequency-modulated tones grow louder only above roughly 20 Hz modulation — ch. 20
388. Timbre served as a residual basket for non-loudness non-pitch sensations — ch. 20
389. Third-octave band filter leakage flattens the lower slope of measured loudness patterns — ch. 20
390. Sensory pleasantness decreases as sound sharpness increases — ch. 21
391. Sensory pleasantness stays flat up to 20 sone then declines with loudness — ch. 21
392. Narrow-band noise sharpness grows proportionally with critical-band rate up to about 3 kHz — ch. 21
393. Sensory pleasantness cannot be measured as a single elementary auditory sensation — ch. 21
394. Noise bandwidth changes sharpness dramatically across a wide range — ch. 21
395. Fluctuation strength requires at least 3 dB modulation depth before it rises — ch. 21
396. Loudness reduces sensory pleasantness only beyond roughly 20 sone. — ch. 21
397. Sharpness is modeled as a weighted first moment of specific loudness — ch. 21
398. Adding lower-frequency sound can decrease perceived sharpness — ch. 21
399. A 40-dB level rise increases fluctuation strength by a factor of roughly 2.5 — ch. 21
400. Sensory pleasantness is computable from sharpness, roughness, tonality, and loudness. — ch. 21
401. AM noise masking depth is frequency-independent while FM tones are strongly frequency-dependent — ch. 21
402. Modulation fluctuation strength is governed by excitation patterns — ch. 21
403. FM-tone fluctuation strength declines with center frequency above 1 kHz while AM does not — ch. 21
404. Low modulation frequencies produce fluctuation sensation while higher ones produce roughness — ch. 21
405. The weighted-moment sharpness formula matches measured sharpness across diverse sounds — ch. 21
406. Listener attitude toward a sound, not just acoustics, shapes perceived pleasantness — ch. 21
407. Sensory pleasantness rises with tonality only up to a ceiling. — ch. 21
408. FM fluctuation strength scales with the number of critical bands crossed — ch. 21
409. The vacil unit anchors fluctuation strength to a 4 Hz modulated tone — ch. 21
410. Sensory pleasantness depends most on sharpness, then roughness and tonality. — ch. 21
411. Sensory pleasantness was measured across tones, narrow-band noise, and band-pass noise. — ch. 21
412. Periodic amplitude fluctuations are perceived as stronger than equivalent random fluctuations. — ch. 21
413. A closed equation approximates sensory pleasantness from sharpness, roughness, tonality, and loudness — ch. 21
414. Temporal masking depth is less than envelope modulation depth due to postmasking — ch. 21
415. Roughness depends on frequency and temporal resolution of hearing — ch. 22
416. Subjective and physical duration stay proportional only above about 100 ms — ch. 22
417. A 100 ms tone burst sounds as long as a 400 ms pause does — ch. 22
418. Roughness depends on modulation degree or FM index plus modulation rate — ch. 22
419. Subjective duration derives from excitation level, not physical duration — ch. 22
420. Roughness grows by a factor of three per 40 dB of level increase — ch. 22
421. Roughness reflects rate of change, so fast modulation roughens while slow change does not — ch. 22
422. Roughness is limited by frequency selectivity at low and temporal resolution at high frequencies — ch. 22
423. Plosive pauses of 60-150 ms go unnoticed in speech perception — ch. 22
424. Roughness only emerges beyond 50 Hz frequency deviation then scales logarithmically — ch. 22
425. Subjective duration, not physical duration, governs rhythm perception in music — ch. 22
426. Amplitude-modulated tones produce three distinct sensations depending on modulation frequency — ch. 22
427. Physically equal spacing only yields subjectively uniform rhythm for steep-envelope bursts — ch. 22
428. Computer models substitute specific loudness differences where masking-depth data are unavailable — ch. 22
429. One asper of roughness is standardized by a 70-Hz fully modulated 60-dB tone — ch. 22
430. Random envelope fluctuations in 100-Hz-wide noise bands produce strong roughness — ch. 22
431. The perception of uniform rhythm tracks envelope rises, not onsets — ch. 22
432. Duration perception distortion is largely level-independent above 30 phon — ch. 22
433. Wideband frequency modulation produces far more roughness than amplitude modulation — ch. 22
434. Musicians exaggerate physical note and pause durations about fourfold to convey written rhythm — ch. 22
435. Only about twenty audible roughness steps span the entire perceptual range — ch. 22
436. Subjective duration of sound bursts can diverge drastically from objective duration — ch. 22
437. Modulation frequency determines whether sound is heard as fluctuation, roughness, or multiple tones — ch. 22
438. Perceived duration of pauses differs from perceived duration of equivalent tone bursts — ch. 22
439. A minority of listeners show quadratic distortion that depends non-ideally on level and phase. — ch. 23
440. Regular distortion laws from electroacoustics describe hearing’s own nonlinear products — ch. 23
441. The hearing system becomes less nonlinear at higher input levels. — ch. 23
442. Cancellation tones alter inner-ear nonlinearity by only 1-2 dB at worst — ch. 23
443. Quadratic difference tones measured by cancellation originate primarily in the middle ear. — ch. 23
444. Hearing’s own nonlinearity produces audible difference tones even in distortion-free equipment — ch. 23
445. Human hearing is most frequency-selective around 2 kHz — ch. 23
446. Fourth-order distortion products are nearly inaudible and appear only above 85-90 dB. — ch. 23
447. Cancellation methods quantify the ear’s internally produced difference tones — ch. 23
448. Rhythm perception can arise from pitch variation alone without loudness change — ch. 23
449. Hearing-system quadratic distortion is largely independent of frequency separation between primaries. — ch. 23
450. Quadratic difference tones become audible only above roughly 70 dB primary levels. — ch. 23
451. Combination tones arise at a place well separated from both primaries. — ch. 24
452. Large primary frequency separations yield multiple cancellation-level notches — ch. 24
453. Phase-correct vector summation is essential in cochlear models — ch. 24
454. Large local excitations can cancel globally because their phases oppose — ch. 24
455. Individual variability in distortion-product cancellation grows with primary frequency separation — ch. 24
456. Vector phase differences, not amplitudes, determine cancellation tone levels — ch. 24
457. Individual differences make averaging meaningless in cancellation-level data — ch. 24
458. Phase shifts between the ears can substitute for onset time delays in detecting sounds. — ch. 25
459. Reduced BMLD stems from masking effects, not from signal frequency or masker level — ch. 25
460. Binaural masking-level difference measures hearing improvement from interaural stimulus asymmetries — ch. 25
461. Temporal integration up to 200 ms approximates masking for in-band signals — ch. 25
462. Interaural-delay-tuned channels exist only for frequencies below 1500 Hz — ch. 25
463. Binaural masking release can be tested under non-simultaneous masking conditions — ch. 25
464. Interaural phase differences can produce time-delay sensations measured at threshold. — ch. 25
465. Two ear sounds can differ in phase while differing little in amplitude. — ch. 25
466. Binaural masking level differences reach 25 dB only for overlapping masker and target — ch. 25
467. Monotic in-phase masking-level difference is frequency-independent from 250 to 4000 Hz — ch. 25
468. Binaural hearing enables sound-source direction perception that one ear cannot produce — ch. 25
469. Ears detect direction changes as small as 5 degrees via 50-microsecond delays — ch. 25
470. Binaural unmasking depends on masker duration but not signal duration — ch. 25
471. BMLD magnitude fluctuates wildly with signal timing within a noise burst — ch. 25
472. Interaural level differences of 1-2 dB produce just-audible lateral displacement — ch. 26
473. Industry pressure forced ISO to standardize A-weighted SPL before a better loudness-based method — ch. 26
474. How binaural loudness summation is measured changes the measured effect by several dB — ch. 26
475. Opposite-phase binaural signals stay louder than in-phase ones up to 40 dB above threshold — ch. 26
476. Localization precision peaks in front, degrades at the sides, improves behind — ch. 26
477. Binaural loudness is measured by matching against a monaural comparison tone in filtered noise — ch. 26
478. Headphone lateralization is one-dimensional whereas free-field localization is three-dimensional for sound images — ch. 26
479. Out-of-phase binaural tones can double perceived loudness — ch. 26
480. Interaural level differences and time shifts trade off to cancel each other — ch. 26
481. Two distinct loudness distributions are used in measurement — ch. 26
482. Carrier time shifts lateralize only low frequencies; envelope time shifts work at high frequencies — ch. 26
483. The first arriving sound wave dominates perceived direction of a source — ch. 26
484. Just-audible image displacement needs only 1 to 2 dB level change. — ch. 26
485. Computerized loudness models track the 1 kHz tone within 2 percent — ch. 26
486. Sounds in the median plane are localized to fixed directions based on frequency — ch. 26
487. Stereo inter-loudspeaker level differences steer perceived sound location between speakers. — ch. 26
488. Equal-level stereo speakers create a phantom image midway between them — ch. 26
489. A-weighted SPL was a compromise standard known to give misleading noise results — ch. 26
490. Lateralization is one-dimensional while localization spans three dimensions — ch. 26
491. Loudness measures can reveal real improvements A-weighting misses entirely — ch. 27
492. Annoyance from sounds is quantitatively described as psychoacoustic annoyance. — ch. 27
493. Inkjet printers are much quieter than needle printers — ch. 27
494. A-weighting’s underestimate of low-frequency energy explains about four dB of the railway bonus. — ch. 27
495. A nostalgic ‘Good old Times’ association could explain railway-noise preference — ch. 27
496. The annoyance formula predicts ratings for real technical sounds. — ch. 27
497. A-weighted sound power ratios overestimate perceived loudness differences in noise emissions. — ch. 27
498. Percentile loudness N5 predicts how listeners judge overall noise immission loudness — ch. 27
499. At equal A-weighted level, railway noise is judged quieter than road and aircraft noise — ch. 27
500. Perceived loudness of speech matches near-maximum not average loudness — ch. 27
501. A-weighted sound power exaggerates perceived loudness differences in marketing — ch. 27
502. Overall noise loudness is judged retrospectively rather than averaged moment to moment — ch. 27
503. Racing starts and low-gear diesels produce maximal psychoacoustic annoyance — ch. 27
504. Loudness evaluation may be unchanged by obscuring the sound’s source identity — ch. 27
505. Speech loudness equals the value exceeded only seven percent of the time — ch. 27
506. Speech loudness is captured by the 7th-percentile loudness value — ch. 27
507. The railway bonus persists even when sound sources are obscured, ruling out cognitive effects — ch. 27
508. Noise reduction claims should use loudness ratios to stay credible — ch. 27
509. Loudness meter readings predict subjective evaluations better than A-weighted measures — ch. 27
510. Spectral broadening can mask source identity while preserving loudness-time function — ch. 27
511. Cognitive source identification can be experimentally decoupled from loudness perception — ch. 27
512. Typical offices lack the acoustics needed to keep speech private. — ch. 27
513. A-weighted decibels are a rough practical approximation of perceived loudness — ch. 27
514. Train loudness depends on length and speed, not just type — ch. 27
515. Source identification affects annoyance but not loudness perception — ch. 28
516. Identifying a sound source can shift annoyance independent of loudness — ch. 28
517. Knowing a sound’s source barely affects how loud we rate it — ch. 28
518. Absent otoacoustic emissions signal abnormal inner or middle ear function — ch. 28
519. Speech discrimination requires five distinct auditory features — ch. 28
520. Nearly all normal-hearing babies produce delayed evoked otoacoustic emissions — ch. 28
521. Normal-hearing test methods rarely transfer to hearing-impaired listeners — ch. 28
522. Clinical hearing-loss diagnoses rest on subjective psychoacoustic data because human neurophysiology is rarely available. — ch. 28
523. Otoacoustic emissions and frequency resolution arise in the inner ear, peripheral to the synapse. — ch. 28
524. Conductive hearing loss shifts tuning curves upward without degrading frequency resolution. — ch. 29
525. Tuning curve measurements are plotted on a critical-band-rate scale representing basilar membrane distance — ch. 29
526. Hearing-impaired listeners vary more than normal hearers in quiet threshold, complicating averaging — ch. 29
527. A gap-based level-discrimination test detects acoustic neurinoma early — ch. 31
528. Hearing-impaired listeners fail to gain temporal resolution in background noise — ch. 31
529. Hearing-impaired specific loudness jumps to near-normal values 20 dB above threshold — ch. 31
530. Outer hair cell loss alone can account for the recruitment effect — ch. 31
531. Hearing aids restore loudness sensitivity but cannot restore damaged frequency selectivity. — ch. 31
532. Impaired frequency selectivity directly causes poor speech discrimination — ch. 31
533. Noise-induced hearing loss flattens temporal masking valleys and breaks the threshold triad alignment — ch. 31
534. Recruitment can explain why loudness summation is absent in noise-induced hearing loss — ch. 31
535. Temporal resolution is quantified by peak and valley masker thresholds — ch. 31
536. In hearing-impaired regions loudness grows far faster with level — ch. 31
537. Acoustic neurinoma distinctly elevates level-difference thresholds while most hearing impairments do not. — ch. 31
538. Noise-impaired listeners show reduced temporal integration, only 4 dB per decade — ch. 31
539. Noise-induced high-frequency threshold shifts can make previously inaudible difference tones audible — ch. 31
540. Outer hair cell damage explains recruitment by disabling active cochlear gain — ch. 31
541. Recruitment means steep loudness growth that catches up to normal at high levels — ch. 31
542. Normal listeners show a straight-line relation for the quiet-modulated-continuous threshold triad — ch. 31
543. Simulated hearing loss via masking preserves temporal integration, unlike real loss — ch. 31
544. Speech-noise test backgrounds were upgraded to match both spectral and temporal envelopes — ch. 31
545. Audibility thresholds rise about 10 dB per tenfold sound duration reduction — ch. 31
546. Temporal masking patterns are characterized by peak and valley values. — ch. 31
547. Hearing-impaired listeners gain far less from noise gaps than normal listeners — ch. 32
548. Inhibitory networks that sharpen spectral peaks can aid listeners with reduced frequency selectivity — ch. 32
549. Auditory Gestalt perception mirrors vision through spectral-pitch contours — ch. 32
550. Sensory consonance depends on roughness, sharpness, and tonalness — ch. 32
551. String instruments naturally produce inharmonic partial spectra. — ch. 32
552. Hearing’s salient spectral features can be matched by adjustable Fourier analysis — ch. 32
553. Adding noise can make listeners hear nonexistent high-frequency brilliance in music. — ch. 32
554. Musical consonance comprises harmony and sensory consonance from separate domains — ch. 32
555. Partial-tone strength thresholds offer graduated levels of audio data reduction — ch. 32
556. Psychoacoustic findings explain musician preferences for tympani heads — ch. 32
557. String instrument inharmonicity may be not merely unavoidable but perceptually desirable — ch. 32
558. Digital transmission makes meaningful reduction of information flow important. — ch. 32
559. Audio compression exploits masking to discard inaudible signal parts — ch. 32
560. The human ear remains the best speech recognizer and should anchor machine systems — ch. 32
561. Adding soft noise makes listeners perceive brilliance that is not actually present — ch. 32
562. Fluent speech in all languages fluctuates at about four elements per second — ch. 32
563. Human hearing adapts to speakers and rooms but the mechanism remains unknown — ch. 32
564. Combined cochlear implant and hearing aid restores sound localization ability. — ch. 32
565. Long recovery-time automatic gain control can erase trailing consonant sounds — ch. 32
566. Normal-hearing listeners need 7 dB lower SNR in fluctuating versus continuous noise — ch. 32
567. Guitar-string inharmonicity creates audible beats on low strings. — ch. 32
568. Perceived speech rhythm tracks the loudness-time function of the signal — ch. 32
569. Skin-stimulation devices and cochlear implants both serve as hearing prostheses for deaf people. — ch. 32
570. No evidence supports preferring 432 Hz piano tuning over 440 Hz — ch. 33
571. The 432 Hz grand-piano tuning preference claim fails blind testing — ch. 33
572. Historic recordings made around 432 Hz allow fair preference testing — ch. 33
573. Room acoustics should incorporate psychoacoustic not merely physical metrics — ch. 33
574. Reverberation multiplies perceived speech loudness well beyond measured level gains — ch. 33
575. Echo loudness decays nonlinearly so level peaks misrepresent perceived peaks — ch. 33
576. Speech intelligibility in rooms can be predicted from fluctuation strength — ch. 33
577. Pitch strength of pure tones varies with their frequency — ch. 49
578. Signals identical in loudness and pitch can differ markedly in pitch strength — ch. 49
579. Brief tone durations severely impair frequency discrimination — ch. 49
580. Narrow-band noise self-modulates at 0.64 times its bandwidth — ch. 49
581. Partial masking of a tone by background noise diminishes as tone level rises — ch. 49
582. Tone bursts are heard as longer than pauses of equal physical length — ch. 49
583. A fluctuating noise simulating speech features is a standard test stimulus — ch. 49
584. Narrow-band noise self-modulation masks external amplitude modulation — ch. 49
585. Only tones exceeding a level-dependent masked threshold are audible in noise — ch. 49
586. Harmonic versus inharmonic spectra of a note are plainly perceptible to a listener — ch. 49
587. The continuity illusion works even with complex signals like music — ch. 49
588. Magnitude estimation with an anchor uses a fixed 70 dB car sound as reference. — ch. 49
589. Partially masking a tone shifts its perceived pitch upward — ch. 49
590. Interaural level differences alone can shift the perceived position of a sound source — ch. 49
591. Humans can typically discriminate level differences of 1 to 2 dB — ch. 49
592. Low-pass noise is heard at a pitch matching its cut-off frequency — ch. 49
593. Narrower noise bandwidth raises pitch strength while higher center frequency also raises it — ch. 49
594. Two physically present tones produce an audible third tone inside the ear — ch. 49
595. Loudness halves for every tenfold decrease in tone duration below 100 ms — ch. 49
596. Healthy hearing detects 1 to 2 dB level changes at 60 dB — ch. 49
597. Cubic difference tones arise inside the ear, not in the equipment — ch. 49
598. Listeners perceive missing fundamentals from the harmonics that remain — ch. 49
599. Critical-band noise masking can be approximated by simple tone complexes — ch. 49

Annotation Roundup

Question

• Psychoacoustic phenomena can be experienced directly through audio demonstrations. — What are these particular demonstrations, and could some of their wav-files still be found online to hear the phenomena?
• Active stimulus control by subjects can bias psychophysical measurements — What specific biases does active stimulus control introduce in loudness comparison tasks? Worth chasing for experimental design.
• Basilar membrane displacements during speech are only tens of nanometers, near atomic scale — What ‘very special arrangements’ achieve this sensitivity given such tiny basilar membrane motion at threshold?
• A low-level-only dip in tuning curves reveals the ear’s active frequency-enhancing processes — Worth chasing: what exactly are the active processes — outer haircell electromotility and otoacoustic emissions — that sharpen frequency selectivity? Later sections presumably detail this.
• Spontaneous otoacoustic emissions show a minimum frequency spacing that grows with frequency — This suggests an underlying place-based mechanism on the basilar membrane — emissions correspond to fixed cochlear locations. Could tie into tonotopic organization models.
• The choice of averaging unit among level, pressure, or intensity critically shapes threshold data. — Why does the averaging unit so strongly shape results? Worth chasing the specific mathematical reason for choosing one measure.
• Psychoacoustic headphones distort sound unless free-field equalized to compensate their band-pass response — Which equalization standard, free-field vs diffuse-field, is considered correct, and how much does the choice bias loudness judgments in practice?
• Outer haircells act locally within the cochlea to modulate inner haircells — How do outer haircells influence inner haircells with no direct neural connection? Presumably mechanical via basilar membrane motion — verify in Sect. 3.1.3(b).
• Inner and outer hair cells differ structurally and likely function differently — What exactly makes outer haircell afferent synapses ‘atypical’? Worth chasing for the mechanistic detail.
• Spontaneous and evoked otoacoustic emissions likely share a single source — Worth chasing: what single cochlear mechanism would generate both spontaneous and evoked emissions?
• Suppression of evoked emissions tracks the audibility of the evoking tone — This ties emission suppression to auditory threshold, implying emissions may index the same cochlear nonlinearity behind masking — worth chasing how the threshold/emission link is established next section.
• Spontaneous and evoked otoacoustic emissions likely share a single source — Worth chasing: why is 0.4 Bark (less than half a critical band) the characteristic spacing — what cochlear mechanism sets it?
• Broad-band noise masking erases individual differences in quiet threshold — Why exactly does noise masking suppress inter-individual variability? The cited mechanism (frequency selectivity unifying masker and tone in one band) could be worth unpacking further.
• Equal-level narrow-band maskers mask less at higher centre frequencies — Why does a fixed 60-dB masker mask less effectively at higher frequencies? This seems to invoke something about excitation patterns and the auditory filter changing with frequency — worth chasing in later chapters.
• Threshold in quiet has a fine structure of maxima and minima. — Worth chasing: what mechanism produces the point-by-point threshold fine structure beyond cochlear emissions?
• Otoacoustic emission suppression mirrors hearing threshold masking patterns — Worth chasing: what is the mechanism linking spontaneous-emission frequencies to minima of the threshold-in-quiet curve, and do these dips correspond to the microstructure of the audiogram?
• Outer and middle ear act linearly while inner-ear mechanics are nonlinear — Is the linearity assumption for outer/middle ear well-established or an approximation the model simply adopts?
• Spacing of hearing-threshold minima matches the 180-degree phase-shift frequency distance — This is a striking numerical correspondence between microstructure of threshold-in-quiet and cochlear phase. Worth chasing how exact the match is across a wider frequency range and what mechanism binds them.
• Masking can occur even when the masker follows the test sound in time. — Backward masking where a later masker conceals an earlier stimulus is counterintuitive — worth investigating its neural mechanism.
• Tonal maskers create distinct audible regions beyond simple masking — Why is a 1-kHz 80-dB masker able to suppress test tones over a wide region where quiet threshold is far lower - this illustrates upward spread of masking.
• Odd-order difference tones explain why single tones mask differently than noise — What is the special procedure of Chap. 14 that estimates difference-tone levels subjectively?
• Masking-period patterns require masker masking and test bursts under an eighth of the period — Why exactly one-eighth of the period as the cutoff? Worth chasing the underlying reasoning for that specific bound.
• Premasking measurements are only reproducible with trained subjects and remain noisy — Why is premasking harder to reproduce than post-masking for trained listeners?
• The ear follows frequency-modulated masker dynamics only up to about 8 Hz modulation — Worth chasing: why precisely ~8 Hz? Does this correspond to a specific time constant of the cochlear filter or to neural envelope-following limits?
• Broadband maskers overshoot-effect short tones but narrowband maskers do not — Why does overshoot require a broadband masker specifically? Mechanism likely involves auditory transient processing differences, worth chasing in the model discussion.
• Pulsation patterns can replicate masking patterns to under 1 dB accuracy — What causes the beats, roughness and combination tones that confound pure-tone masking measurement, and do they have dedicated models elsewhere? Could chase the mechanism behind why narrow-band noise avoids them.
• Masking depends on the masker’s acceleration, not its amplitude. — Why would the second derivative dominate — is this pointing at nonlinear cochlear mechanics downstream of mere basilar-membrane displacement?
• Combined maskers interact predictably in simultaneous but not nonsimultaneous masking — The 3 vs 6 dB question for masker addition is a concrete empirical crux worth following into the cited measurements.
• Simulated Gaussian noise with too few components gives atypical masking results — This is a practical warning for anyone constructing experimental noise stimuli — worth confirming exact line counts and phase rules needed for faithful Gaussian approximation.
• Adding two post-maskers together can reduce rather than increase masking — Why does post-masking behave opposite to simultaneous masking when two maskers are combined? The mechanism described in the next paragraph (temporal fluctuation reduction) deserves further study.
• Pairs of incoherent or coherent noises can be modeled as amplitude modulation — The ‘sonnds’ typo suggests a proofing error; worth verifying the original text.
• Nonsimultaneous masking summation resists simple generalizations — Why do simultaneous-masking results summarize cleanly while nonsimultaneous do not? What role does bandwidth play with temporal parameters? Worth chasing literature on this asymmetry.
• Postmasking temporal decay is realizable by a two-resistor two-capacitor diode network — A concrete hardware-level model worth pursuing — how does the diode encode the rectification/asymmetry behavior of the auditory system?
• A spectral gap in switched-off noise yields a tonal afterimage, not a noisy one — What neural mechanism turns a spectral gap afterimage into a true tonal percept rather than narrowband noise?
• Virtual pitch arises only for certain fundamental-to-lowest-component combinations — Fig 5.10’s shaded existence region — what exactly bounds it physically? Worth chasing what spectral-spacing constraint defines the border.
• Shifting all spectral components uniformly changes perceived virtual pitch in a sawtooth pattern — Why does the virtual pitch become ambiguous specifically near a component shift of 150 Hz, then snap to a lower pitch? The mechanism behind this sawtooth discontinuity is worth chasing.
• Amplitude-modulated broadband noise yields only faint pitch despite flat spectrum — Worth chasing: exactly which short-term spectral information correlates with the faint pitch — unresolved here, deferred to Sect. 5.7.
• Medium-level band-stop noise around 43 dB most reliably produces Zwicker tones — What mechanism makes medium overall levels optimal for Zwicker tone perception rather than louder levels?
• Uniformly shifting a harmonic complex tone’s components changes pitch non-monotonically with a sawtooth pattern — The sawtooth, jump-discontinuity pitch pattern resembles ambiguities in other perception domains (e.g. binocular rivalry, bistable figures); worth probing what mechanism drives the sudden drop.
• Steep-filtered noise produces pitch matching the filter’s cut-off frequency — Why does high-pass noise only elicit pitch above ~400-800 Hz cut-offs while low-pass works at all frequencies? Section 5.7 promises elaboration on pitch strength.
• Large cutoff-frequency sweeps boost low-pass noise pitch strength 2.4-fold — A threshold-like dependence on frequency deviation suggests a nonlinearity in the pitch-strength mechanism worth chasing.
• The basilar membrane yields only about 640 just-noticeable frequency steps end to end. — Curious whether 640 JND steps is frequency-modulation specific or generalizes across loudness and other dimensions measured the same way.
• Above 500 Hz, Bark scale maps logarithmically to frequency change — Worth recalling this 20%-per-Bark relation when later chapters translate between Bark steps and just-noticeable pitch intervals.
• Overlapping excitation slopes raise the valley minimum by at least 3 dB — The mechanism by which slope excitations sum remains unknown; only an empirical rounding approximation exists.
• A 1 dB level change anchors the model of just-noticeable slow variations. — This 1 dB slow-variation anchor recurs through the models — worth tracking how it is reused across later chapters.
• Wider noise bandwidths decrease detectable amplitude modulation via multi-band cooperation — The unstated mechanism — coherent summation across bands vs. another strategy — is an open question worth chasing in later chapters or the literature.
• Just-noticeable level variations always exceed just-noticeable level differences — The claim that frequency modulation and frequency difference may involve different signal processing is a fascinating open question worth chasing.
• Frequency-discrimination and frequency-modulation thresholds are commonly conflated in the literature — Worth chasing which classic experimental papers mixed up absolute frequency discrimination with moduation-detection thresholds, and what that corrupted.
• Low-pass noise cutoff discrimination degrades with level while high-pass cutoff does not — Why would level dependence differ so systematically between low-pass and high-pass noise — an asymmetry worth chasing in later auditory-neurophysiology reading.
• Masking shifts detection thresholds and may affect just-noticeable sound changes — Whether partial masking affects just-noticeable modulations at supra-threshold levels remains an open question this chapter frames but does not fully resolve.
• Just-noticeable changes are tenfold worse at masked threshold then normalize 20 dB above — Why exactly 20 dB above threshold as the normalization point — is this tied to critical-band excitation patterns?
• Band-pass noise audibly self-modulates far above its threshold — Why does band-pass noise exhibit audible self-modulation? This seems to anticipate a limitation of the basic detection model worth understanding.
• Excitation detectors are modeled as independently acting channels with occasional coherence effects — The ‘addition of probabilities’ for coherent excitation variations is flagged as ‘not clear so far’ — a genuine open question in the model worth chasing down.
• Raising noise level paradoxically worsens cut-off frequency discrimination as upper slopes flatten. — The passage trails off at the end — worth chasing the actual measured curve and its quantitative values in Fig. 7.21.
• Closing the ear canal makes internal body noise audible and measurable — What exactly does the internal noise sound like to a subject with closed ear canals? Is it a hiss, roar, or pulse?
• Just-noticeable frequency increments are constant along critical-band rate but rise as pitch strength weakens — The claim that JND frequency is constant along the critical-band-rate scale regardless of frequency is a strong simplification worth checking against data across the audible range and for complex tones.
• Just-noticeable amplitude modulation drops steeply as tone level rises — The claim that loudness level, not pressure level, is the unifying abscissa across frequencies depends on Ch. 8.1’s loudness scaling — worth verifying when reaching that chapter whether this generalization fully holds.
• Comparing two tones improves frequency sensitivity beyond detecting a continuous modulation — Why does a time gap between compared tones increase rather than harm frequency sensitivity? The memory-limitation explanation for low modulation rates seems to conflict.
• Masking shifts thresholds upward and may alter just-noticeable sound changes at all levels. — Whether partial masking alters just-noticeable sound changes above the masked threshold is left as an open research question worth following across the later chapters.
• Just-noticeable amplitude modulation of narrow-band noise is near 6 percent at high centre frequencies — The paragraph trails off at ‘This correlates with a threshold factor’ — curious what the stated factor value is, context cuts it off.
• Hearing becomes less sensitive to frequency variations as level increases — Why does the excitation-pattern upper slope flatten with level? The model explains it mechanistically, but the perceptual consequence (hearing gets worse at higher levels) is worth chasing further.
• Critical bandwidth at one kilohertz is about 150 hertz — One numerical value for critical bandwidth at 1 kHz under this phase-detection method; worth cross-checking against other methods and classic ERB estimates.
• Widely separated tones add by loudness, not pressure or intensity — Worth chasing: the ear transitions among pressure addition, intensity addition, and loudness addition as frequency separation grows — a clean three-regime model of loudness summation worth fully internalizing.
• Loudness perception takes time to develop and so can be interrupted by later sounds — What mechanism underlies the temporal integration of loudness over roughly 100-200 ms? This connects to models of temporal summation of intensity.
• Noise loudness is level-independent until bandwidth exceeds the critical bandwidth. — How exactly does loudness integrate across critical bands above the critical bandwidth — is it a sum-of-parts model? This is likely developed in the next chapter on loudness summation.
• Narrow-band noise produces substantial loudness in critical-band regions where no physical sound exists — The 9.5 Bark cutoff seems specific — is there a physiological reason (basilar membrane asymmetry beyond that rate) or is it stimulus-dependent?
• Loudness grows with bandwidth via area tradeoff except at extreme levels — The level-dependence of bandwidth loudness integration begs explanation in terms of the cochlear excitation slope — worth chasing the mechanistic detail in later chapters.
• Threshold in quiet is attributed to internal physiological noise — Worth verifying how the same threshold factor bridges the masking case and the internal-noise estimation case.
• Free-field versus diffuse-field transmission gains differ by under eight decibels up to 8 kHz — Above 8 kHz the deviation presumably grows - worth chasing the diffuse-field vs free-field divergence in the higher critical bands.
• Broadband factory noise loudness in phons far exceeds its measured dB level — Why does a 73 dB noise calculate to 86 phon? The sone/phon relationship for broadband noise deserves closer study.
• Third-octave band filter leakage flattens the lower slope of measured loudness patterns — How does filter leakage at -20 dB compare across different filter bank designs (FFT vs analog)? Worth checking whether newer loudness meters account for this.
• Sharpness changes little with level and bandwidth below one critical band. — Is sharpness’s mild level dependence from 30 to 90 dB consistent across centre frequencies, or does it interact with the equal-loudness contours covered earlier?
• Adding lower-frequency sound can decrease perceived sharpness — The symmetry with loudness models is worth chasing: does shifting cut-off frequency affect sharpness just via the moment’s center-of-gravity location, or does g(z) do real work here?
• Sensory pleasantness is computable from sharpness, roughness, tonality, and loudness. — Tonality still lacks a calculation procedure and relies on subjective estimation, so the pleasantness formula is only partially objective.
• Modulation fluctuation strength is governed by excitation patterns — Interesting - the ear’s frequency analysis measured in critical bands/critical-band rates, or Bark, explains lower-level perception
• Adding lower-frequency sound can decrease perceived sharpness — Why does adding low-frequency sound reduce sharpness rather than leaving it unchanged? The moment-based model answers it, but the perceptual mechanism behind asymmetric spectral-weighting deserves chasing.
• Listener attitude toward a sound, not just acoustics, shapes perceived pleasantness — What exactly are the nonacoustic influences - learned associations, mood, context - and could any be studied experimentally rather than just eliminated?
• Subjective duration of sound bursts can diverge drastically from objective duration — What exactly differs between perceived duration of bursts versus pauses — ratio, compression? Worth chasing the supporting data.
• Perceived duration of pauses differs from perceived duration of equivalent tone bursts — The paragraph cuts off before giving the actual plotted results — worth chasing the direction of the effect (do pauses sound longer or shorter than bursts?).
• Subjective duration, not physical duration, governs rhythm perception in music — What exactly does ‘10 dB above the surrounding minima’ correspond to perceptually — is there a threshold rationale for equating lengths this way?
• Subjective and physical duration stay proportional only above about 100 ms — The text cuts off mid-point on what happens below 0.3 ms — worth locating the continuation.
• A 100 ms tone burst sounds as long as a 400 ms pause does — Why does filling an interval compress its perceived duration by a factor of four at short timescales, and why does effect depend on frequency?
• The hearing system becomes less nonlinear at higher input levels. — What mechanism makes the cochlea act less nonlinearly as input level rises — is this the compressive nonlinearity of basilar-membrane response?
• Individual differences make averaging meaningless in cancellation-level data — Why do only some subjects show phase steps correlated with level minima, and what mechanism separates the two subject groups?
• Large local excitations can cancel globally because their phases oppose — Worth chasing: how does this phase-cancellation depend on exact stimulus f2/f1 ratio and level? Consequence for predicting audible levels of combination tones.
• Combination tones arise at a place well separated from both primaries. — The difference tone’s characteristic place sits well below either primary — a concrete worked example (1757 and 2280 Hz primaries, 9.6 Bark output) worth revisiting for a worked psychoacoustics example.
• Phase shifts between the ears can substitute for onset time delays in detecting sounds. — Why exactly 105° out-of-phase addition — is that a sign convention or chosen for some threshold property?
• Reduced BMLD stems from masking effects, not from signal frequency or masker level — If it is the masking itself that shrinks BMLD, what exactly about the masking mechanism causes the degradation? Worth pinning down the mechanism.
• BMLD magnitude fluctuates wildly with signal timing within a noise burst — Why do MoSo and MoSrr threshold variations show no correlation? This seems to hint at separate binaural processing mechanisms for So versus S-pi conditions that deserve deeper investigation.
• Binaural masking level differences reach 25 dB only for overlapping masker and target — The claim that JNDL variation and delay-tuned channel slope differences may be correlated is left unresolved — worth checking whether later work addressed this correlation.
• Opposite-phase binaural signals stay louder than in-phase ones up to 40 dB above threshold — What mechanism keeps a masking-level-difference effect going 40 dB above threshold where masking should be irrelevant?
• A-weighted SPL was a compromise standard known to give misleading noise results — Worth chasing: what exactly did ISO acknowledge dBA gets wrong, and how does ISO 532B’s loudness-based method differ in practice?
• Overall noise loudness is judged retrospectively rather than averaged moment to moment — Does overall loudness judgment correlate with peak rather than average because of peak-end rule cognition? Worth comparing to Kahneman’s peak-end research.
• A nostalgic ‘Good old Times’ association could explain railway-noise preference — Worth chasing: how is the sound source obscured (Sect 16.1.5)? Does the cognitive effect replicate in subsequent studies?
• Typical offices lack the acoustics needed to keep speech private. — The threshold for 50% intelligibility being called ‘privacy’ is odd - real privacy usually needs far lower intelligibility. What intelligibility level do privacy standards actually require?
• The annoyance formula predicts ratings for real technical sounds. — Does the model hold for high-roughness sounds like tools or only moderately modulated technical noise?
• Spectral broadening can mask source identity while preserving loudness-time function — What exactly does ‘spectral broadening’ do perceptually — smear harmonic detail while keeping the spectral envelope intact?
• Knowing a sound’s source barely affects how loud we rate it — This contradicts intuition that source knowledge biases perception — worth chasing the underlying model of how top-down information is gated out of loudness processing.
• Clinical hearing-loss diagnoses rest on subjective psychoacoustic data because human neurophysiology is rarely available. — Why does objective animal neurophysiology not transfer cleanly to human diagnosis? The passage gestures at this gap without fully explaining it.
• Nearly all normal-hearing babies produce delayed evoked otoacoustic emissions — Why 98% rather than 100%? What fraction of normal-hearing adults and babies fail to produce emissions, and does that affect diagnostic specificity?
• Hearing-impaired specific loudness jumps to near-normal values 20 dB above threshold — This is the quantitative signature of recruitment — worth pairing with a specific amplitude-based explanation of why the growth rate steepens.
• Noise-induced high-frequency threshold shifts can make previously inaudible difference tones audible — Why does a high-frequency threshold shift make low-frequency difference tones audible? The mechanism connecting loss of masking at low frequencies to audibility of cubic/quadratic difference tones is worth chasing.
• Noise-impaired listeners show reduced temporal integration, only 4 dB per decade — The spectral splatter confound for short bursts in high-frequency-impaired listeners is intriguing - how are these data corrected, if at all, for normal low-frequency sensitivity?
• Outer hair cell damage explains recruitment by disabling active cochlear gain — The degree to which inner hair cell impairment contributes at more severe stages of hearing loss is acknowledged but unresolved here.
• Recruitment means steep loudness growth that catches up to normal at high levels — How does the DLM’s recruitment model handle the 20 dB-above-threshold catch-up claim in the next paragraph — does it match real patient data?
• Adding soft noise makes listeners perceive brilliance that is not actually present — Does the listener’s preference for the noise-free version despite its perceived loss of brilliance reflect a general pattern where objective quality and subjective appeal diverge?
• Human hearing adapts to speakers and rooms but the mechanism remains unknown — What is currently known about the neural basis of speaker/room adaptation in hearing?
• String instruments naturally produce inharmonic partial spectra. — The chapter presumably resolves whether inharmonicity is preferred by listeners — worth checking the experimental result at Figure 16.48.
• Long recovery-time automatic gain control can erase trailing consonant sounds — The paragraph cuts off mid-sentence describing multi-channel aids — worth checking whether loudness-time functions are still sufficient there or whether a different measure is needed.
• Reverberation multiplies perceived speech loudness well beyond measured level gains — The 1.78× figure is specific — would be worth verifying the equations linking reverberation time to loudness increment and where the ceiling before temporal masking kicks in lies.
• The 432 Hz grand-piano tuning preference claim fails blind testing — The music was originally registered near 432 Hz; curiosity about whether period-instrument advocacy faces similar empirical refutations. Xref
• An amplitude-modulated tone is a carrier plus two side lines — Connects to modulation-theory treatments of amplitude modulation used later in the book for fluctuation strength and roughness.
• Feedback in outer haircells enhances sensitivity while lateral spread shapes frequency selectivity. — Connects to the general principle of ‘criticality’/‘edge of chaos’ in biological systems (e.g. neural avalanches, self-organized criticality).
• Psychoacoustic results depend on experimental procedure and sensitivity rises with alternatives. — This tradeoff between accuracy and trial count parallels signal-detection theory’s treatment of criterion placement and measurement efficiency.
• Hearing sensations correlate with stimulus magnitudes but only within the audible range — The sone/Pa distinction parallels the figure-ground/signal distinction in soundscape studies and the perceptual-versus-physical scale point elsewhere in this reading set.
• Human hearing spans 150 dB and three decades of frequency — The 150 dB dynamic range and 20 Hz–20 kHz limits are the foundational numbers for all later models of loudness, masking, and frequency selectivity.
• Median and interquartile range stay scale-invariant where means fail — Directly extends the scale-transformation point of location 144 — together they establish that psychoacoustic data analysis should default to median-based statistics.
• Hearing is most damage-prone in the 3-8 kHz frequency range — Connects to Notch of noise-induced hearing loss around 4 kHz and occupational hearing-loss audiograms.
• Abrupt tone onset and offset generate broadband transient energy at distant frequencies. — Explains a spectral-splatter phenomenon behind click-evoked masking and onset transients.
• The inner ear can be modeled as an array of asymmetrical overlapping bandpass filters. — Precedes the auditory filter / critical band model with asymmetric shapes, relevant to masking research.
• The inner ear amplifies quiet sounds by up to 30 dB through active feedback. — This active cochlear amplifier idea links to broader neuroscience themes of peripheral sensory systems doing active signal processing, and to otoacoustic emissions research.
• Outer haircells amplify quiet sounds only, while inner haircells respond directly to loud ones. — The dual-mechanism explanation for wide dynamic range parallels how other sensory systems split processing between low- and high-intensity regimes.
• Suppressor tones that reduce SOAEs map the ear’s fine frequency selectivity — Connects to the neurophysiological tuning-curve material in Sect. 3.1.3 and to the broader cochlear-amplifier concept.
• Spontaneous and evoked otoacoustic emissions likely share a single source — The 0.4 Bark spacing between emissions echoes the critical band concept central to psychoacoustic masking models throughout this book.
• Otoacoustic emission suppression mirrors hearing threshold masking patterns — Ties otoacoustic emissions as a physical correlate of auditory threshold phenomena to the book’s broader theme of psychoacoustic models.
• Distortion product emissions arise only in a live ear, not passive cavities — Confirms the emission is physiologically generated in the cochlea, not an acoustic artifact — ties to cochlear amplifier / otoacoustic emission research.
• Outer hair cells act as nonlinear amplifiers with bidirectional lateral feedback — The lateral-feedback-with-decaying-strength structure mirrors lateral inhibition schemes in neural network modeling and on-center/off-surround receptive fields.
• Masking disrupts everyday speech and music whenever competing sounds differ in level — Connects naturally to the reader’s soundscape/ambient-noise work — the truck-passing example mirrors lo-fi soundscape masking in Schafer.
• Individual auditory fibres cover only 40-60 dB yet we hear over 100 dB range — Connects to the broader auditory level-coding problem in hearing science and to neural population-rate coding.
• Masking can occur even when the masker follows the test sound in time. — Connects to temporal processing theory and how delayed neural responses to the masker could overwrite the test tone’s trace.
• Tonal masker spread inverts from low to high frequencies as masker level rises — Builds directly on the level-dependent cochlear excitation nonlinearities familiar from masking literature (e.g. Moore’s ‘An Introduction to the Psychology of Hearing’).
• Instrument timbre differences stem from harmonic spectra that shape masking breadth — Connects to the reader’s psychoacoustics corpus on how spectral composition shapes perception; also relevant to instrument acoustics generally.
• Odd-order difference tones explain why single tones mask differently than noise — Connects to the mechanics of combination/distortion tones in the cochlea, mirroring nonlinear-systems behavior familiar from physics.
• Combination tones arise from the hearing system’s own nonlinearity — Connects to combination-tone and distortion-product otoacoustic emission phenomena in later psychoacoustics and audiology literature.
• Masking pattern curves are scale-invariant and predictable by horizontal shifts — This horizontal-shift universality of masking curves echoes the scale-invariance style of explanation (self-similarity across parameters) familiar from Fourier/time-frequency dualities in psychoacoustics.
• Psychoacoustical tuning curves grow more asymmetric at higher test-tone levels — Tuning-curve shape parallels cochlear nonlinearity: sharper, deeper dips at low probe levels resembles basilar-membrane compression and active gain.
• Reducing audible fluctuations in a masker lowers post-masked thresholds by up to 24 dB — A masker that makes a signal MORE audible than in silence is counterintuitive; ties to suppression phenomena and two-tone interactions in cochlear mechanics.
• Masking-period patterns reveal how masker phase within its period affects audibility — The period-histogram equivalence connects psychoacoustics to auditory nerve firing phase-locking, a core neuroscience concept.
• The ear follows frequency-modulated masker dynamics only up to about 8 Hz modulation — Connects to temporal resolution limits in the auditory system generally; the ~8 Hz ceiling parallels the sluggishness of the peripheral auditory filter’s ability to track rapid changes.
• Temporal masking alone masks brief consonants even without room reverberation. — Connects to the reader’s audio/psychoacoustics interest and to speech perception literature on consonant intelligibility.
• Repeated short tones reach steady threshold when repetition rate exceeds 5 per second — The 200 ms window and 5 Hz cutoff parallels temporal summation in other sensory/detection literature and neurophysiological integration time constants.
• Masking depends on the masker’s acceleration, not its amplitude. — Connects to the physical principle that acceleration (not displacement or velocity) is the effective stimulus for hair-cell mechanotransduction.
• Post-masking depends on masker bandwidth as well as duration — Extends earlier post-masking duration findings into the bandwidth dimension, connecting to the excitation-pattern framework.
• Intensity-duration tradeoff in masking holds only for durations under 200 ms — Connects to Bloch’s law / temporal integration in vision where similar intensity-duration reciprocity holds up to ~100 ms.
• Maskers below a tone raise its pitch while maskers above lower it. — The pitch-shift-away-from-masker rule echoes how loudness summation and excitation-pattern models treat spectral edges, tying to the broader cochlear excitation-pattern framework of this book.
• Incomplete complex tones show twice the pitch deviation of complete ones — Connects to the missing fundamental / residue pitch phenomenon and pitch perception models in other psychoacoustics texts.
• The spectral pitch model was tuned by fitting published experimental data — Illustrates the parameter-fitted, data-driven model-building approach common across quantitative psychoacoustics.
• Raising the level of a complex tone shifts its pitch downward. — Connects to prior chapter material on pure-tone pitch shift with level (Fig 5.3), and to the general finding that complex-tone pitch rests on lower spectral components.
• Incomplete complex tones show twice the pitch deviation of complete ones — Provides a quantitative contrast to the complete-complex-tone level-dependent pitch shift from Fig 5.9, highlighting that the lowest spectral component drives the effect.
• Steep-filtered noise produces pitch matching the filter’s cut-off frequency — Connects to the broader virtual pitch models (Ch. 5) being built up; also relates to why high-pass noise fails below ~400-800 Hz cut-offs, which recurs in later paragraphs.
• Perceived pitch compresses against frequency above roughly 500 Hz — Connects to auditory-place/cochlear mechanics and roughly parallels critical-band widening at high frequencies.
• Complex-tone pitch tracks the spectral pitch of lower components — Links to pure-tone pitch-level dependence (Fig 5.3) and harmonic interaction (Fig 5.6) discussed elsewhere in this book — the chapter’s own earlier material.
• Pitch as a sensory dimension has both height and perceptual strength, which vary independently. — Relates to how other auditory dimensions like loudness have distinct sub-dimensions (e.g., loudness vs. noisiness), a parallel worth noting.
• Pitch strength is quantifiable by magnitude estimation across diverse sounds — Magnitude estimation protocol recurs throughout Fastl & Zwicker for loudness and sharpness scaling.
• Peaked ripple noise pitch strength stays stable across a wide level range — Contrasts with the loudness-dependence of pure-tone pitch strength; a clean isolable perceptual dimension.
• Frequency is a poor physical scale for describing inner-ear effects; critical-band rate works better. — Connects to the audiology and pitch-perception material in earlier chapters on critical bands and place theory; also the mel pitch scale reappears across psychoacoustics texts.
• Internal bodily noise raises the hearing threshold at low frequencies — Connects to the reader’s noise/silence theme — the hearing system’s own quiet floor is set by the body’s internal sounds, internalizing the acoustic ecology concept.
• Narrow critical bandwidths at low frequencies reduce hearing sensitivity via noise fluctuations — Connects to the reader’s psychoacoustics reading on critical bands and just-noticeable-difference models.
• Abrupt level changes create audible clicks via broad short-time spectra. — Connects to the reader’s broader interest in psychoacoustic models and spectral analysis; relates to how transient phenomena map onto perception, echoing earlier chapters’ treatment of temporal and spectral loudness effects.
• Rapid AM of noise is heard via overall level, not its rapid fluctuations — Relates to temporal integration/limitations of auditory temporal resolution — connects to psychoacoustic models of duration and loudness integration.
• Narrow-band noise keeps a Gaussian amplitude distribution that statistically modulates its perceived sound — Connects to physics/statistics concepts of Gaussian distributions in random processes; relevant to later psychoacoustic models of envelope fluctuation.
• Just-noticeable level variations always exceed just-noticeable level differences — Observation tasks (detecting modulated sounds) vs. discrimination tasks (comparing two static sounds) call on different processing — connects to signal detection theory separation of detection from discrimination.
• Instrument tuning relies on high harmonics because pitch changes there are easier to hear — Explains a practical musical phenomenon (why players hear upper partials when tuning) through the psychoacoustic sensitivity curve — bridges the book’s models to real-world musicianship.
• One just-noticeable pitch step shifts excitation across about six inner hair cells — Bridges psychoacoustic pitch resolution to cochlear anatomy (inner hair cell spacing) — connects this chapter’s model to the neurophysiological substrate.
• Critical bandrate scale stabilizes frequency discrimination measures across durations — Links to the concept that critical band processing mediates perception — relates to discussions of auditory filters.
• Maskers’ influence on modulation thresholds vanishes beyond 20 dB above their excitation pattern. — Reinforces excitation-pattern framework connecting masking to loudness and other supra-threshold percepts.
• Auditory filters respond asymmetrically, favoring frequencies below center — Connects to critical band theory and the excitation pattern model common across the book’s masking chapters.
• The ear’s upper excitation slope is essential for detecting amplitude modulation at high levels — Connects to the excitation-pattern model of hearing described across earlier chapters of this book.
• Internal physiological noise from blood flow and heartbeat raises low-frequency thresholds — Connects to the physiology of hearing and to noise-masking research; resonates with the ‘silence’ theme books already in the library about how we adapt to always-present sound.
• Loudness is measured subjectively against 1 kHz reference tones — Relates to sound pressure level and dB scaling from earlier chapters; connects to the overall decibel framework of psychoacoustics.
• Partially masked sounds have a reduced loudness range in both time and spectrum — Connects to masking theory and critical bands from earlier chapters, and foreshadows temporal/simultaneous masking details.
• Barkhausen created the loudness level measure in the 1920s — Connects the Bark unit’s etymology to critical-band rate concepts from earlier chapters.
• Just-noticeable frequency shifts map to constant basilar distances — Constant physical displacement for a just-noticeable difference strongly parallels Weber’s law and place-coding tonotopy in the cochlea.
• Constant 1-Bark windows yield triangular noise patterns and rectangular tone patterns — The triangular-vs-rectangular excitation shapes set up masking-pattern analyses and the self-masking spread seen in loudness models.
• Excitation slopes become frequency-independent above 200 Hz center frequency — The lower excitation slope being a constant 27 dB per Bark independent of center frequency is a canonical psychoacoustic regularity worth holding alongside masking-pattern data in Fig. 4.3 that the text compares against.
• Tone-on-tone masking thresholds are measured via narrow-band noise maskers instead — This methodological substitution — narrow-band noise standing in for pure-tone maskers because exact tone-on-tone measurement is too hard — parallels how soundscape work substitutes measurable proxies for elusive targets.
• Postmasking can last up to 200 milliseconds after a masker ends — Connects to the reader’s broader psychoacoustics reading and Neuroscience of timbre/temporal processing knowledge.
• At very high modulation rates, listeners detect AM only via average intensity change. — Relates to the book’s own models of temporal integration/roughness and to the earlier square-root law; connects to the reader’s broader psychoacoustics interest in detection criteria switching.
• Frequency modulation detection is level-independent but thrice frequency-difference detection — Suggests separate neural mechanisms for detecting modulation vs. static differences — relevant to models distinguishing temporal coding from rate-place coding.
• Detecting level variation is about 2.5 times harder than detecting level difference — Amplitude modulation thresholds (“variations”) vs. level-difference JNDs contrast across two measurement paradigms, echoing Milgrom-style distinctions about how measuring context changes the outcome.
• JNDL level dependence unifies across frequency when scaled to loudness level — Loudness-level rescaling removing frequency differences anticipates later chapters’ pitch and loudness models; ties into why loudness level�not SPL�is the perceptual currency.
• Phase-change sensitivity in hearing is bounded by critical bandwidth, scaling with centre frequency. — Connects to the reader’s psychoacoustics reading on critical bands and the Bark scale established in earlier models.
• Low-frequency loudness growth compresses as level rises — Recruitment-like nonlinear loudness growth at low frequencies connects to hearing-impaired loudness phenomena.
• The phon scale defines loudness relative to equal-loudness matches at 1 kHz — The Bark/frequency correspondence ties to critical-band rate models mentioned elsewhere in psychoacoustics.
• Sound direction and field type change perceived loudness at a fixed level. — Connects to room acoustics and head-related transfer function research from psychoacoustics chapters on spatial hearing.
• Threshold in quiet depends on total sound intensity within one critical band only — This unifies threshold-in-quiet and masking under a single critical-band model, which is the organizing concept of the chapter.
• Threshold in quiet depends on total sound intensity within one critical band only — Connects to the broader model that the ear integrates sound energy within critical bands before detection, foundational to psychoacoustics and to masking theory.
• The ear ignores sideband phase once components fall outside one critical band — Ties to the critical-band model of hearing being developed across these chapters.
• Amplitude modulation and frequency modulation differ only by sideband phase at low indices — Connects to signal theory / communications-engineering knowledge of AM-FM equivalence.
• Critical bandwidth at one kilohertz is about 150 hertz — Compares with the ~130 Hz critical bandwidth at 1 kHz often cited from Zwicker-style measures.
• The audible range spans 24 bark-scaled critical bands totaling 0 to 24 Bark. — The Bark scale parallels the ERB (equivalent rectangular bandwidth) scale used in modern psychoacoustics and cochlear models.
• A just-noticeable frequency shift moves basilar excitation a constant 0.05 mm. — This constant-distance result is what makes the Weber-like behavior of frequency limens intelligible; ties earlier chapters’ DL findings to place theory.
• Uniform exciting noise delivers equal intensity to every critical band — Ties directly to the critical-band model central to psychoacoustics; connects to cochlear filtering and auditory peripheral processing.
• Loudness doubles per 10 phon except below 40 phon where it drops steeply. — Connects to Stevens’ power law scaling in psychophysics; also relevant to any audio engineering discussion of loudness normalization (LUFS standards).
• Steady-state loudness models fail for speech, music, and impulsive technical noises. — Frames why temporal loudness models are needed — ties to the reader’s interest in psychoacoustics of real-world sounds vs. laboratory steady tones.
• Tone-burst loudness rises 10 phon per tenfold duration up to 100 ms, then plateaus. — Connects to the onset/temporal-integration literatures in psychoacoustics and to how temporal summation of loudness is modeled (e.g., for sound reproduction systems).
• Loudness perception takes time to develop and so can be interrupted by later sounds — Temporal integration of loudness connects to the general psychophysical phenomenon of temporal summation seen in other sensory and perception literatures.
• Two tones within a critical band interact to reduce total loudness — Connects to the reader’s existing knowledge of the critical band concept introduced earlier in this book.
• A following loud sound reduces the loudness of an earlier short sound — The ~200 ms gap at which masking disappears corresponds to the window of temporal integration/loudness summation in psychoacoustic models.
• Loudness grows as a power law of physical intensity so relative changes match — Connects to psychophysics foundations (Fechner/Stevens scaling) familiar from sensation and perception material.
• Specific loudness generalizes beyond loudness to describe sharpness — Links specific loudness to sharpness and other psychoacoustic metrics in the wider hearing-modeling literature; the reader’s own model-building on psychoacoustics would reuse N’ as the common derived quantity.
• Model-computed and measured loudness match across the full level range — Model-validation result echoes the tone-versus-noise loudness matching theme from earlier chapters on excitation patterns.
• Threshold in quiet is attributed to internal physiological noise — Connects to the earlier discussion (other chapters) of masking thresholds and the ‘threshold factor’ used to relate masking sound to masked threshold.
• Wider bandwidth increases perceived loudness at equal sound pressure — Relates to the general principle that loudness depends on critical-band spread - same theme as why narrow filters do not sound as loud.
• Spectral summation precedes temporal summation in loudness processing — The cochlear-vs-central place-of-processing distinction ties into the broader peripheral-to-central processing hierarchy one meets in psychoacoustics and neuroscience reading.
• High modulation frequencies increase loudness because spectral spread outweighs temporal fluctuation — Connects to the loudness model’s handling of amplitude-modulated tones and to how the brain integrates spectral space versus time — parallels the figure-ground-field distinctions in The Soundscape.
• Vowels dominate perceived loudness while consonants carry intelligibility — Relevant to speech intelligibility research and to the ADDENDUM on trees/soundscape themes; also informs audio engineering about why syllabic separation tracks vowels.
• Spectral envelope, not fine structure, determines perceived sharpness — Extends the earlier chapters’ treatment of critical-band loudness models to sharpness — same family of percepts scaled by critical-band measures.
• Spectral summation precedes temporal summation in loudness processing — Sequencing of perceptual stages along the auditory pathway parallels the earlier basilar-membrane to central processing ordering discussed in the book.
• Frequency-modulated tones grow louder only above roughly 20 Hz modulation — The ~20 Hz threshold where FM loudness begins tracking modulation ties to the roughly 45 ms temporal integration window (1/45 ms ~ 22 Hz).
• Adding lower-frequency sound can decrease perceived sharpness — Ties to the signal-vs-ground asymmetry in the soundscape chapters: low-frequency ‘masking’ content reshapes perceptual judgments rather than merely adding to them.
• Periodic amplitude fluctuations are perceived as stronger than equivalent random fluctuations. — Relates to general perceptual principle that periodic/rhythmic structure is processed more strongly than stochastic variation, echoing work on rhythm perception and auditory scene analysis.
• Sharpness is modeled as a weighted first moment of specific loudness — Connects to the earlier loudness model (specific loudness vs critical-band rate pattern from Chap. 8) and the excitation-pattern framework common across psychoacoustic models, parallel to how loudness and roughness are each formalized as integrals over excitation patterns.
• Temporal masking depth is less than envelope modulation depth due to postmasking — Connects to how a low-pass masking characteristic is transformed into a band-pass fluctuation-strength characteristic — relevant to filter models in signal processing.
• Computer models substitute specific loudness differences where masking-depth data are unavailable — This exemplifies the pragmatic-modeling strategy of substituting a measurable quantity (specific loudness) for a hard-to-obtain one — parallels computational substitution tactics in other perceptual models.
• Amplitude-modulated tones produce three distinct sensations depending on modulation frequency — Connects to the reader’s interest in how fluctuation/roughness/perceived-tones form a continuous perceptual spectrum from temporal to spectral processing — parallels Zahorian’s temporal modulation spectrum work.
• One asper of roughness is standardized by a 70-Hz fully modulated 60-dB tone — Analogous to how ‘sone’ and ‘phon’ anchor loudness scales — the reader has been building a mental catalog of psychoacoustic units.
• Roughness is limited by frequency selectivity at low and temporal resolution at high frequencies — Connects to the critical-band and temporal-resolution sections of the same book’s earlier chapters, and to the auditory-system chapter.
• Roughness reflects rate of change, so fast modulation roughens while slow change does not — Connects to the roughness/fluctuation-strength models and to masking-depth measures; also parallels how fluctuation strength vs roughness are distinguished in this same book.
• Subjective duration, not physical duration, governs rhythm perception in music — Connects to prior chapter material on subjective duration and the excitation-level model of hearing sensations.
• The perception of uniform rhythm tracks envelope rises, not onsets — The idea that perceived timing tracks the perceptually salient part of a sound (rise) rather than physical onset connects broadly to work on perceptual centers (P-centers) in speech timing research.
• Hearing’s own nonlinearity produces audible difference tones even in distortion-free equipment — Connects to the loudness/specific-loudness models and the theme of the inner ear’s nonlinear mechanics, and to why combination tones arise physiologically (cochlear nonlinearity).
• The hearing system becomes less nonlinear at higher input levels. — Connect to the compressive nonlinearity / active cochlear amplifier literature, where low-level sounds get more gain than high-level ones.
• Ears detect direction changes as small as 5 degrees via 50-microsecond delays — The order-of-magnitude contrast between 50-microsecond interaural threshold and 2-ms monaural temporal resolution connects to the earlier soundscape work on temporal discrimination.
• Binaural masking level differences reach 25 dB only for overlapping masker and target — Connects to the reader’s broader psychoacoustics reading on binaural processing and how shared frequency content enables binaural unmasking.
• Out-of-phase binaural tones can double perceived loudness — Connects to interaural phase/timing processing and binaural unmasking phenomena in hearing science.
• A-weighted SPL was a compromise standard known to give misleading noise results — Connects to the broader noise-regulation literature and ISO standard history in One Square Inch of Silence; ISO’s admission that dBA is a practical-but-inaccurate compromise is exactly the kind of regulatory fact that book gestures at.
• How binaural loudness summation is measured changes the measured effect by several dB — Relevant to stereo/multichannel audio engineering and loudness of reproduced music.
• Industry pressure forced ISO to standardize A-weighted SPL before a better loudness-based method — Connects to broader theme of measurement conventions being politically/industrially negotiated rather than purely scientific — parallels the reader’s work on how standards encode compromises.
• Loudness meter readings predict subjective evaluations better than A-weighted measures — Connects to psychoacoustic loudness models from earlier chapters (e.g., Zwicker’s loudness), showing their applied predictive validity.
• Overall noise loudness is judged retrospectively rather than averaged moment to moment — Connects to Kahneman’s peak-end heuristic on retrospective evaluation of experiences.
• A-weighting’s underestimate of low-frequency energy explains about four dB of the railway bonus. — Connects to the reader’s established interest in A-weighting limitations, tying to the soundscape and noise chapters already read.
• The railway bonus persists even when sound sources are obscured, ruling out cognitive effects — Contradicts the common assumption that noise annoyance is mostly top-down/psychological; parallels findings on A-weighting limits in hearing science.
• Racing starts and low-gear diesels produce maximal psychoacoustic annoyance — Relates to noise-design tradeoffs in sound engineering; connects to the reader’s ongoing thread on urban noise annoyance.
• Loudness measures can reveal real improvements A-weighting misses entirely — Connects to the broader theme in the reader’s knowledge base of how A-weighted SPL inadequately predicts perceived annoyance versus true loudness models.
• A-weighted decibels are a rough practical approximation of perceived loudness — Connects to the well-known psychoacoustic finding that A-weighting is a poor predictor of loudness, related to earlier discussion in this book of loudness models vs. sound level meters.
• Identifying a sound source can shift annoyance independent of loudness — Relates to how top-down cognitive factors shape sonic perception, echoing broader auditory-cognition material.
• Conductive hearing loss shifts tuning curves upward without degrading frequency resolution. — Connects to the distinction between conductive vs. sensorineural hearing loss and how cochlear mechanics vs. middle-ear transmission affect frequency analysis.
• Hearing aids restore loudness sensitivity but cannot restore damaged frequency selectivity. — Connects to the broader theme of hearing loss and the limits of corrective devices — relevant to the reader’s psychoacoustics model-building as a boundary condition on what amplification alone can fix.
• A gap-based level-discrimination test detects acoustic neurinoma early — The design contrast - discrete pulses with 200 ms silent gaps versus continuous modulations - parallels the temporal-resolution findings central to the reader’s interest.
• Acoustic neurinoma distinctly elevates level-difference thresholds while most hearing impairments do not. — Connects to clinical audiology and differential diagnosis of hearing disorders, distinct from the psychoacoustic theory chapters.
• Hearing-impaired listeners fail to gain temporal resolution in background noise — Connects to the cocktail-party problem and to selective-attention research on how normal ears exploit masking release in noise.
• Hearing-impaired listeners fail to gain temporal resolution in background noise — Connects to the reader’s broader interest in speech-in-noise deficits and why hearing aids often fail in noisy restaurants — a familiar real-world phenomenon now given a mechanistic basis.
• Outer hair cell loss alone can account for the recruitment effect — Connects to earlier chapter material on cochlear model and outer/inner hair cell roles in cochlear amplification.
• Noise-induced hearing loss flattens temporal masking valleys and breaks the threshold triad alignment — Reduced temporal resolution with flattened masking valleys aligns with loss of cochlear compression — connects to basilar membrane nonlinearity topics elsewhere in the book.
• Normal listeners show a straight-line relation for the quiet-modulated-continuous threshold triad — The straight-line relation at ~40 dB above quiet threshold suggests a scaling/economy in temporal resolution worth probing — connects to ideas about compressive basilar-membrane nonlinearity.
• Audibility thresholds rise about 10 dB per tenfold sound duration reduction — Connects to the intensity-duration tradeoff seen in loudness and masking material from earlier chapters on temporal summation.
• Simulated hearing loss via masking preserves temporal integration, unlike real loss — Distinguishing simulated from real hearing-impairment mechanisms parallels the chapter’s broader theme that masking noise reproduces threshold shifts but not all perceptual consequences of cochlear damage.
• Outer hair cell damage explains recruitment by disabling active cochlear gain — Connects to the nonlinear-active-feedback cochlear model and outer hair cell compressive gain familiar from auditory neuroscience literature.
• Recruitment means steep loudness growth that catches up to normal at high levels — A clinical sign of cochlear damage long described in audiology texts; this gives the psychoacoustic mechanism behind it.
• Outer hair cell loss alone can account for the recruitment effect — Links hearing-loss mechanism to the cochlear amplifier theory; outer hair cells as the active gain stage.
• Hearing-impaired listeners gain far less from noise gaps than normal listeners — Connects to the reader’s interest in hearing loss mechanisms and why impaired listeners fail to benefit from temporal dips — a known hallmark of cochlear hearing loss.
• Hearing’s salient spectral features can be matched by adjustable Fourier analysis — Connects to the auditory scene analysis and harmonic pitch extraction literature.
• Adding soft noise makes listeners perceive brilliance that is not actually present — Connects to auditory illusions and expectation-driven perception; the perceived ‘brilliance’ from added noise parallels how expectation shapes perception elsewhere in perception studies.
• The human ear remains the best speech recognizer and should anchor machine systems — Ties to engineering/neuroscience bridges — cochlear-model speech recognition systems.
• Psychoacoustic findings explain musician preferences for tympani heads — Connects acoustic-instrument material preferences to measurable pitch-strength and loudness variance — relevant to the trees-as-materials angle of ‘Grafting & Growth’.
• Sensory consonance depends on roughness, sharpness, and tonalness — Tonalness as opposite of noisiness and related to spectral pitch echoes earlier chapters’ treatment of sharpness as a sensation metrically defined in acum.
• Auditory Gestalt perception mirrors vision through spectral-pitch contours — The vision-audition Gestalt parallel connects to the reader’s interest in neuroscience behind psychoacoustics; virtual pitch as ‘illusory contour’ recalls pitch perception theories in chapter 2 of this same book.
• Fluent speech in all languages fluctuates at about four elements per second — The 4 Hz center of speech fluctuation coupling to the 4 Hz band-pass of fluctuation strength ties directly to the reader’s chapter 10 material on fluctuation strength and to the physics of hearing.
• Adding noise can make listeners hear nonexistent high-frequency brilliance in music. — Connects to auditory scene analysis and masking theory concepts familiar from music production and audiophile debates.
• No evidence supports preferring 432 Hz piano tuning over 440 Hz — Connects to the ‘Complaints About Immodesty’ theme of debunking revivalist myths with controlled evidence; also to the broader human love of numerology-based tuning claims.
• Reverberation multiplies perceived speech loudness well beyond measured level gains — Connects to loudness summation modeling in earlier chapters of this book (specific loudness vs time functions used throughout).
• Room acoustics should incorporate psychoacoustic not merely physical metrics — Parallels the reading-purpose framing of psychoacoustics as linking physical stimuli to perception, and connects to loudness-models chapters elsewhere in this book.
• Echo loudness decays nonlinearly so level peaks misrepresent perceived peaks — Ties to loudness models where total loudness integrates specific loudness across critical bands — the rounded decay and shifted peak contradict naive level-time assumptions.
• The continuity illusion works even with complex signals like music — Same gap-filling mechanism as phonemic restoration; broadens the continuity illusion from simple tones to complex real-world signals.
• Critical-band noise masking can be approximated by simple tone complexes — Links the internal auditory filter model — noise approximated by discrete tones — to the filter-bank view of the cochlea.
• Only tones exceeding a level-dependent masked threshold are audible in noise — Illustrates the masking threshold concept formalized as the fundamental psychoacoustic relation between signal level and noise-induced audibility limits.
• Narrower noise bandwidth raises pitch strength while higher center frequency also raises it — This is testable explanation material — a clean experimental design fragment for teaching psychoacoustics.
• Narrow-band noise self-modulation masks external amplitude modulation — Connects to noise fluctuation and modulation perception literature; self-modulation of narrowband noise is a classic sonar/bioacoustics phenomenon.
• Brief tone durations severely impair frequency discrimination — Ties to duration-dependent auditory spectral resolution; relevant to pitch perception models.
• Frequency discrimination thresholds rise dramatically from 500 Hz to 5000 Hz — Demonstrates the classic Weber-fraction-in-frequency finding central to psychoacoustics.
• Tone bursts are heard as longer than pauses of equal physical length — Relates to time-perception literature, e.g. filled vs empty interval duration illusions studied in psychology.
• Harmonic versus inharmonic spectra of a note are plainly perceptible to a listener — Relates to the reader’s broader pitch/perception reading: virtual pitch and spectral pitch studies in the same book rely on this harmonic-versus-inharmonic distinction. Seed
• Young headphone users already show measurable hearing damage in the 3-12 kHz range — Angle: ‘the first audible casualty of the headphone era’ — measurable threshold shifts in young frequent listeners.
• Spontaneous otoacoustic emissions align with peaks in stimulus-frequency fine structure — An essay angle: the ear’s own emitted sounds mark its most sensitive frequencies — a kind of active resonance where the cochlea ‘shows’ where it hears best.
• Short masker gaps are bridged by postmasking effects — Angle for a piece on why MP3-style audio compression masking models exploit temporal smoothing of brief gaps.
• A-weighted dB values misrepresent perceived loudness for complex sounds. — Article angle: why ‘dB(A)’ on every product spec sheet fails to predict how loud things actually feel.
• Just-noticeable frequency shifts map to constant basilar distances — A candidate essay angle: the ear’s resolution is uniform in place, not in frequency — a counterintuitive ‘every note gets equal room’ framing.
• A 1 dB level change marks a recurring just-noticeable unit in hearing — The recurring ‘1 dB is just noticeable’ constant could anchor an accessible explainer on how hearing’s discriminative limits translate into concrete perceptual units.
• Specific loudness generalizes beyond loudness to describe sharpness — Angle: specific loudness as the primitive currency of audition - a single derived quantity that predicts loudness, sharpness, and roughness alike could anchor a piece on how hearing models unify.
• Sensory pleasantness decreases as sound sharpness increases — Angle: pleasantness of everyday sounds scores predictable by psychoacoustic sharpness.
• A-weighted sound power exaggerates perceived loudness differences in marketing — Article angle: why advertised ‘90% quieter’ claims for products don’t match what people actually hear — a primer on why metric choice (A-weighting vs. loudness) misleads consumers.
• A-weighted decibels are a rough practical approximation of perceived loudness — Angle: how ‘90% quieter’ marketing claims mislead consumers because A-weighted power ratios exaggerate perceived differences by ~2.5x — a consumer-advice piece with real data.
• Absent otoacoustic emissions signal abnormal inner or middle ear function — Article angle: the physics-to-biology bridge of newborn hearing screening — how a purely mechanical echo in the ear became the standard objective screen for infant deafness, precisely because it fails only past ~20 dB loss.
• Acoustic neurinoma distinctly elevates level-difference thresholds while most hearing impairments do not. — Angle: an old-school, low-tech psychoacoustic test that outperforms elaborate clinical equipment for detecting acoustic neurinoma — a case study in diagnostic elegance.
• Adding soft noise makes listeners perceive brilliance that is not actually present — An article angle: why the ‘clean’ remasters of classic recordings often disappoint audiophiles — a psychoacoustics explanation for nostalgia-driven sound preferences.
• Adding noise can make listeners hear nonexistent high-frequency brilliance in music. — Angle for ‘remastering old recordings and what noise actually does to perceived fidelity’ — the tradeoff between audio cleanliness and perceived brilliance.
• No evidence supports preferring 432 Hz piano tuning over 440 Hz — A debunking piece: ‘The 432 Hz myth, tested on a real Steinway’ — a clean myth-busting essay against a popular wellness/musician claim.
• Historic recordings made around 432 Hz allow fair preference testing — A nice methodological detail — old recordings were naturally at 432 Hz, making them ideal test material; could feed into the debunking essay angle.
• Tone bursts are heard as longer than pauses of equal physical length — Could feed an essay on how musicians’ notated rhythms systematically misjudge perceived pause lengths. Fiction
• Hearing detects AM better than FM only when sidebands fall within a critical bandwidth — Could seed a character detail around a sound engineer or perception researcher fixating on whether the ear can tell AM from FM. Define
• Psychoacoustics is the quantitative science of the ear as an information receiver. — Defines psychoacoustics as the science of the hearing system as a receiver of acoustical information, emphasizing the quantitative stimulus-sensation relation.
• Human hearing spans sound pressures from 10-5 to 102 Pa — Defines absolute threshold, threshold of pain, and sound pressure level as the standard dB scale used throughout psychoacoustics.
• White noise shows frequency-flat long-term spectra but frequency-dependent short-term spectra — Defines white noise in psychoacoustic terms and flags the long-term versus short-term spectrum distinction.
• Periodic gating creates spectral line spacings set by the pulse repetition rate — Establishes the reciprocal relationship between time-domain pulse timing and frequency-domain line spacing.
• Gaussian shapes minimize the bandwidth-duration product — Defines the Gaussian envelope as the optimum under the bandwidth-duration uncertainty tradeoff.
• Forced-choice procedures inform subjects after each trial which interval held the signal. — Defines the two-interval forced-choice paradigm, a core psychoacoustic measurement method.
• Inner and outer hair cells differ structurally and likely function differently — Clarifies the structural distinction between inner and outer hair cells, foundational anatomy for psychoacoustics.
• Psychoacoustic tuning curves directly reflect cochlear mechanical processing. — Clarifies what psychoacoustical tuning curves measure and links them to frequency resolving power.
• The four otoacoustic emission types are distinguished by whether and how they are evoked — Enumerates and defines the four kinds of otoacoustic emissions: SOAEs, SEOAEs, DEOAEs, and DPOAEs.
• Adaptive procedures adjust stimulus intensity based on prior responses and converge on thresholds — Clearly defines adaptive/up-down procedures and how threshold estimation works.
• Hearing sensations correlate with stimulus magnitudes but only within the audible range — Introduces the core distinction between physical stimulus magnitudes (Pa, dB) and psychophysical sensation magnitudes (sone) — the foundational vocabulary of psychoacoustics.
• Human hearing spans 150 dB and three decades of frequency — Establishes the reference values (2×10⁻⁵ Pa, 10⁻¹² W/m²) and the dB/Pascal conventions used throughout psychoacoustics.
• The median resists stimulus-scale transformations, unlike arithmetic and geometric means — Explains why median is the robust averaging choice for psychoacoustic data — a methodological principle of the field.
• The median resists stimulus-scale transformations, unlike arithmetic and geometric means — Explains why median is the robust averaging choice for psychoacoustic data — a methodological principle of the field.
• Tracking methods differ from adjustment by letting subjects control only direction, not magnitude — Defines the method of adjustment and distinguishes it from the tracking method.
• Magnitude estimation lets listeners assign numbers proportional to perceived loudness. — Explains magnitude estimation and its anchored variant using ratio judgments.
• Yes-No procedures force signal detection decisions without an unsure option — Clarifies the Yes-No procedure’s forced-choice character.
• The inner ear can be modeled as an array of asymmetrical overlapping bandpass filters. — Introduces the filterbank model of cochlear frequency analysis underlying auditory filter concepts.
• The inner ear acts as a linear system offering both frequency and temporal resolution. — Frames the cochlea’s dual linear-system behavior: spectral place analysis plus temporal delay.
• Auditory tuning curves derive from constant basilar membrane velocity rather than displacement — Defines what a tuning curve is and why velocity, not displacement, is the presumed haircell-driving stimulus.
• The inner ear amplifies quiet sounds by up to 30 dB through active feedback. — Establishes the concept of ‘active nonlinear feedback’ in the auditory periphery as a mechanism for low-level gain.
• Outer haircells amplify quiet sounds only, while inner haircells respond directly to loud ones. — Lays out the functional division of labor between inner and outer haircells in the cochlea.
• Outer hair cells actively amplify and sharpen sound before inner hair cells fire — Clarifies the functional role distinction between outer and inner haircells in the cochlea — the amplifier model.
• Physical stimuli map to separate hearing sensations only within audible physical ranges — Introduces the core psychoacoustic framework: stimulus magnitudes (Pa, dB) versus hearing sensations (loudness, pitch), with ‘sone’ as the sensation unit for loudness — foundational vocabulary for this whole book.
• Absolute hearing threshold is a statistical construct, defined at 50% detection probability. — Defines absolute threshold as a probability-based concept (50% detection), not a fixed level.
• The threshold in quiet is highly reproducible for individual subjects — Defines the threshold in quiet as the just-audible pure-tone level as a function of frequency.
• Most human ears display synchronized spontaneous otoacoustic emissions when driven — Clarifies the distinction between spontaneous emissions (SOAE) and stimulus-frequency-evoked emissions (SEOAE) as the same oscillator driven above its level.
• Delayed evoked emissions are the ear’s response to short impulses — Defines DEOAE as an impulse-driven otoacoustic emission with suitable repetition rates.
• Delayed otoacoustic emissions obey linear superposition at low amplitudes — Establishes that DEOAEs behave as a linear system under low-level stimulation, a defining property used to characterize them.
• Delayed otoacoustic emissions arrive faster for higher stimulus frequencies — Clarifies that the delay time of an evoked emission depends on its frequency, with a quantitative rule (lower frequencies slower).
• Otoacoustic emission suppression mirrors hearing threshold masking patterns — Explains masking-period pattern and its mirror correspondence with suppression-period pattern.
• Uniform masking noise mirrors white-noise masking and gives flat thresholds — Defines uniform masking noise as the frequency-shaped inverse of white-noise masking used to produce a flat masked threshold.
• Forward masking follows masker termination and is also called post masking — Clarifies the term ‘post masking’ (a.k.a. forward masking), which appears as a distinct phenomenon from simultaneous masking.
• Combination tones can be cancelled by adding matched opposite-phase tones — Defines ‘hearing cancellation’ — the perceptual method for cancelling audible distortion product tones.
• Distortion product emissions are objectively measurable in the ear canal — Distinguishes objective emission measurement from perceptual hearing cancellation — the two complementary methods.
• Outer hair cells act as nonlinear amplifiers with bidirectional lateral feedback — Explains how lateral coupling (feedback to neighboring sections at half and quarter strength) is operationalized in the hardware model.
• Masked threshold always exceeds quiet threshold except when masker and test frequencies differ greatly — Defines the masked threshold, the central operational measure of masking.
• Masking is graded, with partial masking reducing loudness rather than audibility. — Introduces ‘partial masking’ as distinct from total masking — a loudness reduction without full inaudibility.
• A single source generates all three kinds of otoacoustic emissions. — Introduces the Bark scale spacing and links delay time to spectral spacing, clarifying the shared-source logic behind otoacoustic emissions.
• Critical bandwidth stays constant near 100 Hz up to 500 Hz then grows — Defines narrow-band noise relative to critical bandwidth with the quantitative rule.
• Two tones presented together produce audible combination tones in the ear — Defines hearing cancellation versus objectively measured distortion product emissions as distinct phenomena.
• Delayed otoacoustic emissions arise from decaying basilar-membrane oscillations that sum in phase — Explains the mechanism behind delayed evoked otoacoustic emissions (DEOAEs) in causal terms.
• Low-frequency suppressors reduce emissions by saturating cochlear feedback-loop nonlinearity — Proposes that the second derivative of the suppressor waveform drives masking and suppression.
• Masking can occur even when the masker follows the test sound in time. — Defines pre- or backward masking — masking when the masker is not simultaneous with the test sound.
• Simultaneous masking raises a test tone’s just-audible level above threshold in quiet. — Defines ‘masked threshold’ — the sound pressure level at which a test sound becomes just audible in the presence of a masker.
• Tonal maskers create distinct audible regions beyond simple masking — Maps the distinct perceptual regions produced when a tonal masker suppresses a test tone, contrasting with the cleaner case of noise masking.
• The near-miss of Weber’s law means threshold slopes can reach 3 to 6 — Formally names and defines the ‘near-miss of Weber’s law’ phenomenon for tone level increments, with quantified slope magnitudes.
• Combination tones arise from the hearing system’s own nonlinearity — Clarifies that audible difference tones near 600 Hz arise from nonlinear distortion inside the listener’s own auditory system, not from the signals presented.
• Internal nonlinear distortion products, not poorer frequency selectivity, explain noise maskers’ extra low-frequency masking — Clarifies the mechanism by which internal difference tones and difference noises widen the effective internal stimulus for noise versus tone maskers.
• Masking-period patterns reveal how masker phase within its period affects audibility — Distinguishes masking-period patterns (sound-pressure timefunction) from temporal masking patterns (envelope) — a key conceptual distinction.
• Gaussian-shaped tone bursts minimize spectral splatter in gated tones — Explains why Gaussian envelopes are the standard stimulus — the time-bandwidth tradeoff drives the methodological choice.
• Postmasking stays at simultaneous-masking level for 5ms then decays to quiet by 200ms. — Explains how a brief Gaussian impulse can substitute for white noise as a test sound, isolating temporal from spectral masking.
• Envelope-structured sounds produce time-variant temporal masking patterns. — Defines which sound classes generate temporal masking patterns that vary over time.
• Short masker gaps are bridged by postmasking effects — Explains the mechanism by which postmasking ‘bridges’ brief gaps in amplitude-modulated maskers, smoothing perceived level modulation.
• Temporal masking patterns comprise simultaneous, post, and premasking effects — Defines the three-part structure of a temporal masking pattern — core psychoacoustic concept.
• Frequency-modulated maskers produce two masking peaks where instantaneous frequency crosses the test tone twice — Clarifies the mechanism: a two-peak temporal masking pattern arises precisely when the F-M masker crosses the test-tone frequency twice per modulation cycle.
• Post masking outlasts premasking and asymmetrically distorts threshold patterns. — Clarifies the pre-masking vs post-masking asymmetry as the cause driving the two curves’ divergence.
• Aperiodic noise maskers are studied using repeated random pulse sequences with long periods. — Explains the methodological trick of creating quasi-aperiodic noise with a long period.
• Masking occurs before, during, and after a masker is present. — Defines the three temporal masking regions: premasking, simultaneous masking, and postmasking, along with the key term sensation level.
• Broadband maskers overshoot-effect short tones but narrowband maskers do not — Names and delimits the overshoot effect: broadband-only, largest at tiny onset delays, gone by ~200 ms.
• Masked thresholds spike within milliseconds of masker onset before settling — Introduces the temporal-onset effect in simultaneous masking that motivates the overshoot terminology.
• Auditory threshold for broadband noise is set by the most sensitive frequency region only — Explains why broadband threshold in quiet is dominated by the ~3 kHz most-sensitive region rather than the whole spectrum.
• Aperiodic noise maskers are studied using repeated random pulse sequences with long periods. — Defines the experimental technique of repetition noise for measuring masking by aperiodic maskers — the method converts non-periodic noise into a measurable periodic signal.
• A gated tone sounds continuous at a level called the pulsation threshold — Defines the pulsation threshold technique and the continuity phenomenon operating behind it.
• Pulsation patterns map test-tone level against frequency like masking patterns do — Pulsation patterns are operationalized as a function of test-tone frequency and critical-band rate, mirroring masking-pattern methodology.
• Masking increments peak when the two individual maskers yield equal threshold shifts — Defines the two-masker framework: the increment f1LT and the threshold-difference index f1Dr(1-2), along with the dominance prediction.
• Post-masking decreases when a second masker suppresses audible masker fluctuations — Explains the mechanism: adding a second masker removes audible temporal fluctuation of a narrow-band masker, which lowers post-masked threshold by 13-24 dB.
• Pairs of incoherent or coherent noises can be modeled as amplitude modulation — Shows how two-noise masking sequences can be reinterpreted as rectangular amplitude modulation, linking masking models to models of just-noticeable variations.
• Critical bandwidth varies a hundredfold across the audible range. — Concrete anchor for critical bandwidth magnitude — 100 Hz at low vs over 2 kHz at high frequencies — worth remembering as a reference figure.
• Near threshold, masked threshold depends on both masker level and threshold-in-quiet via internal noise — Clarifies that threshold in quiet is modeled as masking by an internal noise source — a conceptual framework worth remembering.
• Simultaneous masking originates in the cochlea while nonsimultaneous masking does not — Distinguishes where each masking type originates in the auditory pathway.
• Nonsimultaneous masking summation resists simple generalizations — Defines ‘addition of masking’ precisely via the question of whether combined-masker thresholds follow intensity addition (3 dB) or sound-pressure addition (6 dB) or neither.
• Post-masking decreases when a second masker suppresses audible masker fluctuations — Explains the mechanism: post-masking suppression by adding a second masker operates by reducing audibly fluctuating temporal patterns of the narrow-band first masker.
• A spectral gap in switched-off noise yields a tonal afterimage, not a noisy one — Introduces and defines the Zwicker tone phenomenon and its counterintuitive tonal quality.
• Spectral pitch of pure tones derives from three magnitudes of masking patterns — Lays out the three-magnitude framework (MPTM, steepness components, MTHQ) that the pitch model is built on.
• Level-induced pitch shift depends on masking height above the quiet threshold — Shows how eliminating the masker simplifies the model to just the test tone’s masking pattern and the quiet threshold’s steepness.
• Pure tone pitch shift decomposes additively into level and masking components — Establishes the two-component decomposition of pitch shift used throughout the chapter.
• Most everyday sounds are harmonic complex tones, not pure tones — Defines harmonic complex tone as integer multiples of a fundamental frequency.
• Removing lower harmonics barely changes the perceived pitch of a complex tone. — Defines virtual/residue/low pitch — the classic phenomenon that deleted fundamentals are still heard via higher harmonics.
• Low-pass noise can mask harmonics to create the same virtual-pitch stimulus as filtering. — Explains the experimental technique of creating incomplete complex tones via masking rather than filtering, linking masking to pitch perception.
• Shifting all spectral components uniformly changes perceived virtual pitch in a sawtooth pattern — Defines an ‘incomplete inharmonic complex tone’ — a harmonic complex whose components are uniformly shifted so they are no longer integer multiples of the fundamental.
• Amplitude-modulated broadband noise yields only faint pitch despite flat spectrum — Establishes that pitch can arise from temporal (amplitude) modulation absent spectral peaks, a key distinction between temporal and spectral pitch mechanisms.
• Bandpass noise pitch corresponds to spectral edge frequencies which fuse when close — Defines how bandpass noise pitch maps to spectral edges and the fusion rule for close edges.
• Pitch of a tone is defined by the matching 40 dB reference frequency — Defines the operational reference used to quantify pure tone pitch — the 40 dB SPL matching criterion.
• Rippled noise generated by feedback delay produces distinct pitch sensations. — Explains how rippled/peaked noise is generated and its spectral parameters.
• Pitch as a sensory dimension has both height and perceptual strength, which vary independently. — Introduces the pitch-strength scale as distinct from pitch height — a foundational conceptual distinction in psychoacoustics.
• Pitch from gated noise vanishes when gaps shrink below temporal resolution — Case where temporal resolution failure explains pitch disappearance — pitch requires resolvable gaps between bursts.
• Frequency is a poor physical scale for describing inner-ear effects; critical-band rate works better. — Establishes critical-band rate (Bark) and ratio pitch (mel) as perceptual scales that map linearly onto the basilar membrane, unlike physical frequency.
• Above 500 Hz, Bark scale maps logarithmically to frequency change — Captures the log relation operating above 500 Hz where 1 Bark equates to roughly a 20% frequency change.
• Hearing selectivity is modeled via critical-band and excitation intermediate values — Distinguishes critical-band intensity/level (idealized steep filters) from excitation/level (real filter slopes), both used logarithmically.
• White noise’s critical-band intensity rises above 5 Bark because bandwidth grows there — Clarifies why the triangular critical-band-intensity plot appears and what its 2-Bark base and peak coincidence tell the experimenter about the stimulus.
• Main excitations make critical-band level equal excitation level at signal centers — States when the two levels are identical (main excitations only) and what must be added otherwise.
• The masking index varies with frequency, distinguishing UEN from uniform masking noise. — Defines the masking index as the difference between critical-band level and masked threshold, and explains why frequency dependence of that index separates UEN from UMN.
• Slope excitation can transmit low frequencies to high critical-band regions. — Clarifies the distinction between main excitation (frequency mapped via critical-band rate) and slope excitation (tied to the main excitation’s frequency), a key model concept.
• Narrow critical bandwidths at low frequencies reduce hearing sensitivity via noise fluctuations — Explains what the masking index (av) is and why it shifts from -2 dB to -6 dB.
• Narrow-band noise keeps a Gaussian amplitude distribution that statistically modulates its perceived sound — Explains why narrow-band noise differs perceptually from a pure tone despite similar spectral width: the Gaussian amplitude statistics persist.
• Narrow-band noise’s intrinsic amplitude jitter masks external amplitude modulation detection — Explains that narrow-band noise carries its own inherent amplitude fluctuation that interferes with detecting externally imposed modulation — a key concept in modulation perception.
• Wider noise bandwidths decrease detectable amplitude modulation via multi-band cooperation — Clarifies what ‘cooperation of critical bands’ means for amplitude-modulation detection — the mechanism, though unidentified, improves detection as more bands engage.
• Level discrimination thresholds grow steeply near hearing threshold — Defines JNDL (just-noticeable difference in level) and its range of roughly 0.2-0.4 dB at moderate-to-high levels.
• Critical bandrate scale stabilizes frequency discrimination measures across durations — Clarifies how the Bark/critical bandrate scale flattens duration-dependent frequency discrimination data.
• Maskers’ influence on modulation thresholds vanishes beyond 20 dB above their excitation pattern. — States main vs. slope excitations distinction for broad vs. narrow-band maskers.
• Just-noticeable changes are tenfold worse at masked threshold then normalize 20 dB above — Defines the quantitative relationship between masking level and just-noticeable difference/frequency/level perception.
• Auditory filters respond asymmetrically, favoring frequencies below center — Gives quantified slopes for auditory critical-band filter asymmetry, a core psychoacoustic model parameter.
• The excitation pattern’s upper slope changes nonlinearly with level, unlike the main peak — Clarifies how the excitation-level-versus-critical-band-rate pattern behaves under amplitude modulation: the upper slope is the nonlinear, level-dependent part.
• The steep lower slope of the excitation pattern determines frequency modulation detection threshold — Explains how a 1 dB excitation-level change at the detector array defines the just-noticeable frequency variation.
• Noise’s inherent statistical fluctuations degrade frequency-change detectability relative to tones. — Introduces the concept of statistical self-modulation of noise as a masking factor for detecting changes.
• Frequency discrimination depends on pitch strength and worsens as pitch strength falls — Clarifies that a JND model must include pitch and loudness memory submodels, so JND measurement is comparative across two stored sounds.
• Loudness is measured subjectively against 1 kHz reference tones — Defines the phon unit and the operational definition of loudness level via a 1-kHz reference.
• Partially masked sounds have a reduced loudness range in both time and spectrum — Introduces partial masking as a core concept distinguishing masked from unmasked loudness.
• Loudness is the sensation most directly tied to sound intensity — Establishes loudness as the intensity-correlated sensation and introduces ratio-scaling measurement.
• Threshold perception integrates total sound intensity across a limited frequency bandwidth. — Establishes the critical-band integration rule: overall level constancy up to a bandwidth limit, beyond which extra components don’t add to threshold.
• Two-tone maskers define critical bandwidth at the threshold’s transition point. — Explains the ‘masking in frequency gaps’ method, where the critical bandwidth is operationally defined as the crossing point where masked threshold begins to decay.
• Equal-power masking assumption is wrong and only partly power detection holds — Explains the ‘equal power assumption’ and shows why it fails, clarifying the actual ratio condition.
• Critical bandwidth stays constant near 100 Hz up to 500 Hz then grows — Establishes the critical-band approximation rule (constant 100 Hz below 500 Hz, 20% relative bandwidth above) that future chapters rely on.
• Critical-band rate correlates with pitch, frequency discrimination, and basilar-membrane position. — Establishes that the critical-band rate is a unifying scale linking the hearing system’s several frequency-encoding measures.
• Critical-band rate and excitation level are the intermediate transforms for hearing models — This locates the theoretical role of critical-band rate and excitation level as model-ready intermediate representations sitting between physical stimulus and percept.
• Level variations within a sound and differences between sounds activate distinct hearing processes. — Establishes the crucial distinction between just-noticeable variations (within-sound modulation) and just-noticeable differences (between-sound comparison), foundational for loudness and pitch measurement methodology.
• Fletcher defined critical bands as noise regions whose power equals a just-masked tone’s — Clarifies precisely how Fletcher operationalized the critical band concept, distinguishing it from later measurement approaches.
• JNDL level dependence unifies across frequency when scaled to loudness level — Shows why ‘loudness level’ rather than raw sound-pressure level is the orderable perceptual axis for detection thresholds.
• Frequency deviation measures one-sided swing while total variation doubles it — Clarifies a system-theory term that otherwise carries ambiguity between one-sided deviation and full range.
• Pitch can be modeled as a positional sensation scaled by discrimination steps — Explains the theoretical assumption that underlies building a pitch scale from just-noticeable frequency steps.
• Frequency modulation detection is level-independent but thrice frequency-difference detection — Explains what 2L1f means (twice the frequency-modulation depth at threshold) and why it has a U-shaped dependence on modulation rate.
• Changing a tone’s phase alters both its temporal envelope and its instantaneous frequency. — Illustrates how a phase parameter shift produces two concurrent acoustic effects - envelope variation and instantaneous frequency variation - clarifying the physical basis for phase perception experiments.
• Frequency discrimination stems from steep filter slopes plus 1 dB level sensitivity — Clarifies that auditory frequency sensitivity is derived from excitation-level slopes, not dedicated frequency-change mechanisms.
• The phon scale defines loudness relative to equal-loudness matches at 1 kHz — Defines phon and equal-loudness contours and ties Barkhausen (Bark) to their history.
• Loudness measures span quiet, masked, spectral, and temporal domains — Establishes the domain structure of loudness sensation: partial masking is both spectral and temporal.
• Threshold in quiet depends on total sound intensity within one critical band only — Captures the critical-band integration principle: energy within one critical band alone determines threshold.
• Threshold in quiet depends on total sound intensity within one critical band only — Establishes the operative definition of basic threshold behavior: integration within a bandwidth, exclusion outside it.
• Amplitude modulation and frequency modulation differ only by sideband phase at low indices — Explains the technical distinction between AM and FM in terms of sideband phase, central to the psychoacoustic phase-detection method.
• The audible range spans 24 bark-scaled critical bands totaling 0 to 24 Bark. — Defines the Bark unit and that 24 abutting critical bands cover the audible range.
• Auditory critical bandwidth stays constant at 100 Hz below 500 Hz then rises proportionally. — Defines the useful working approximation for critical bandwidth as a function of frequency.
• Uniform exciting noise delivers equal intensity to every critical band — Defines the ‘uniform exciting noise’ concept — a stimulus normalized to the ear’s own critical-band resolution rather than to physical spectra.
• Loudness doubles per 10 phon except below 40 phon where it drops steeply. — Establishes the phon-to-sone conversion (10 phon = 2x loudness) and its breakdown at low levels — a core psychoacoustic scaling fact.
• Critical bandwidth at 1 kHz is 160 Hz and governs loudness behavior — Introduces critical bandwidth as the characteristic bandwidth separating two loudness rules, quantified at 160 Hz for a 1 kHz centre frequency.
• Partial masking steepens the loudness function in a way that mirrors hearing recruitment — Explains the mechanism of partial masked loudness and links it to the audiological concept of recruitment.
• Total loudness is the integral of specific loudness over critical-band rate — Names the central construct ‘specific loudness’ (N’) and defines total loudness as its integral over the Bark scale.
• Total loudness is the integral of specific loudness over critical-band rate — Establishes the dimensional logic of loudness as an integral over critical-band rate, yielding sone as the unit.
• Uniform exciting noise delivers equal intensity to every critical band — Defines uniform exciting noise — equal intensity per critical band, not per physical hertz — as the spectral opposite extreme of a pure tone.
• Loudness falls to zero at threshold in quiet, forcing a vertical asymptote — Explains why the loudness function renders threshold-of-quiet as a vertical asymptote, a modeling constraint that would otherwise look like an artifact.
• Ten-phon increments double loudness, mapping phon to sone by powers of two — Crisp statement of the phon-to-sone conversion rule (10 phon equals a factor of 2 in sone), the conceptual bridge between the two loudness scales.
• Critical bandwidth marks the point where noise loudness begins to grow with bandwidth — Establishes the operationally distinct definition of critical bandwidth via loudness growth that cross-validates other measurement methods.
• Partial masking steepens the loudness function in a way that mirrors hearing recruitment — Defines the ‘partial masked loudness curve’ and explains why masking must produce a steeper loudness function.
• Total loudness is the integral of specific loudness over critical-band rate — Establishes the modeling rationale connecting excitation patterns to loudness integration.
• Total loudness is the integral of specific loudness over critical-band rate — Formal definition of specific loudness N’ and its integral relationship to total loudness N.
• Uniform exciting noise loudness grows as intensity raised to the 0.23 power — Gives the specific power-law exponent linking physical intensity to perceived loudness for uniform noise.
• Loudness grows as a power law of physical intensity so relative changes match — States Stevens’ power law and how the model substitutes excitation for intensity and specific for total loudness.
• Total loudness requires linear scales before integration is valid — Clarifies the methodological constraint that loudness integration is only valid on linear-scale plots of specific loudness versus critical-band rate.
• Loudness rises with bandwidth beyond the critical band despite falling pattern height — Explains the spectral effect, the mechanism by which loudness integrates acoustic energy across critical bands.
• Spectral summation precedes temporal summation in loudness processing — Clarifies temporal summation as distinct from (and weaker than, in everyday sounds) spectral summation in loudness development.
• Total loudness includes contributions outside the stimulus’s own band — Clarifies that total loudness = area under the whole specific-loudness pattern, including slope contributions outside the physical band.
• Specific loudness derives from Stevens’ power law applied to excitation — Explains how Stevens’ general psychophysical law is particularized into the specific-loudness variable used in the model.
• Specific loudness is better expressed against excitation level than against linear excitation — Introduces the normalization parameter LTQ - a0 that (subtracting the free-field transmission factor) removes frequency and critical-band-rate dependency from loudness relations.
• The G subscript indicates loudness computed from critical-band levels — Explains the meaning of the subscript G in soneG and distinguishes critical-band-level loudness from other measures.
• Threshold in quiet affects loudness mainly for faint sounds — Explains why threshold in quiet, despite being large in frequency terms (three octaves), is compressed to only 1.5 Bark on the critical-band-rate scale — hence its small practical effect.
• Any sound narrower than a critical band produces identical loudness regardless of its width. — Clarifies that the critical band is the width within which loudness equals that of a pure tone.
• Spectral summation precedes temporal summation in loudness processing — Clarifies the ordering of two integration mechanisms (spectral at the basilar membrane, temporal more centrally) that a loudness model must respect.
• Loudness patterns approximate specific loudness by ignoring the lower slope entirely. — Defines the ‘loudness pattern’ construction and its simplifying three-part approximation, ignoring lower-slope specific loudness.
• Timbre served as a residual basket for non-loudness non-pitch sensations — Defines timbre as the residual category of non-loudness, non-pitch sound sensations.
• Loudness processing acts as a temporal low-pass filter with about 45 ms integration time — Names the model mechanism: total loudness as a sum over critical-band rate passed through a low-pass system whose impulse response is the temporal loudness response.
• Timbre is a residual category masking distinct sensations like sharpness and sensory pleasantness. — Defines sharpness, sensory pleasantness, and reframes timbre as a residual category.
• Sharpness is modeled as a weighted first moment of specific loudness — Defines the sharpness model as a weighted first moment of specific loudness over critical-band rate, with a weighting factor g(z) rising above 16 Bark.
• Sensory pleasantness cannot be measured as a single elementary auditory sensation — Defines sensory pleasantness as a composite sensation requiring relative anchoring measurement.
• Temporal masking depth is less than envelope modulation depth due to postmasking — Precisely distinguishes temporal masking depth from modulation depth, a definitional subtlety central to the model.
• Sharpness is modeled as a weighted first moment of specific loudness — Explains the formal definition of sharpness as a weighted first moment — the conceptual model structure behind the acum scale.
• Adding lower-frequency sound can decrease perceived sharpness — Establishes acum as the unit of sharpness and shows the asymmetric effect of spectral widening — a concrete empirical anchor for the sharpness concept.
• The vacil unit anchors fluctuation strength to a 4 Hz modulated tone — Defines the unit ‘vacil’ with its calibration reference tone, the standard anchor for fluctuation strength.
• Low modulation frequencies produce fluctuation sensation while higher ones produce roughness — Distinguishes fluctuation strength from roughness by modulation frequency, clarifying two psychoacoustic constructs divided near 20 Hz.
• Temporal masking depth is less than envelope modulation depth due to postmasking — Defines ‘temporal masking depth’ as the difference between masking pattern maximum and minimum.
• Amplitude-modulated tones produce three distinct sensations depending on modulation frequency — Defines the three perceptual regimes: fluctuation (peak near 4 Hz), roughness (peak near 70 Hz), and separately audible tones (above about 150 Hz), a core psychoacoustic map.
• One asper of roughness is standardized by a 70-Hz fully modulated 60-dB tone — Standardizes the asper unit, anchoring it to a specific reference stimulus the way sones anchor loudness.
• Roughness depends on modulation degree or FM index plus modulation rate — Two-parameter characterization: amplitude modulation uses degree of modulation, frequency modulation uses the FM index.
• Only about twenty audible roughness steps span the entire perceptual range — Introduces the asper as the unit of roughness and the threshold of 0.07 asper.
• Subjective duration of sound bursts can diverge drastically from objective duration — Introduces ‘subjective duration’ as a distinct perceptual sensation, in contrast to objective physical duration.
• Physically equal spacing only yields subjectively uniform rhythm for steep-envelope bursts — Distinguishes physical uniform spacing from subjectively uniform rhythm, a key conceptual distinction in psychoacoustics of time perception.
• Temporal masking depth is less than envelope modulation depth due to postmasking — Temporal masking depth is defined as the maximum-minus-minimum of the masking pattern across the masker period, located at delays 0 and 0.5.
• Modulation frequency determines whether sound is heard as fluctuation, roughness, or multiple tones — Establishes the operational boundary between the three temporal sensations (fluctuation ~4 Hz peak, roughness ~70 Hz peak, separate tones above ~150 Hz) that recurs through the roughness chapter.
• Subjective duration derives from excitation level, not physical duration — Introduces the model that subjective duration is extracted from the excitation-level time course 10 dB above surrounding minima.
• Hearing’s own nonlinearity produces audible difference tones even in distortion-free equipment — Clarifies what difference tones are and why masking lets fundamentals but not difference tones go undetected.
• Cancellation methods quantify the ear’s internally produced difference tones — Explains the cancellation method for measuring internally produced distortion products.
• Quadratic difference tones measured by cancellation originate primarily in the middle ear. — Clarifies that the site of quadratic difference-tone generation (middle vs. inner ear) can be inferred from frequency independence of cancellation parameters.
• Binaural hearing enables sound-source direction perception that one ear cannot produce — Introduces the psychophysical stimulus-versus-sensation distinction as applied to spatial hearing — actual source direction vs. perceived direction.
• Binaural masking release can be tested under non-simultaneous masking conditions — Contrasts simultaneous vs. non-simultaneous masking as experimental conditions for BMLD measurement.
• Interaural phase differences can produce time-delay sensations measured at threshold. — Clarifies that interaural time delay can be created not only by onset delays of impulsive sounds but by steady-state phase shifts between the ears — a core methodological concept for binaural psychoacoustics.
• Binaural masking-level difference measures hearing improvement from interaural stimulus asymmetries — Defines BMLD operationally as the just-audible level difference between monaural and binaural stimulus presentations.
• Interaural-delay-tuned channels exist only for frequencies below 1500 Hz — Defines the interaural-delay-tuned channel array model, the core structure being proposed to explain binaural masking.
• Lateralization is one-dimensional while localization spans three dimensions — Distinguishes the one-dimensional lateralization from full three-dimensional localization and defines localization blur.
• Sounds in the median plane are localized to fixed directions based on frequency — Introduces the concept of ‘determining frequency bands’ — narrow-band sounds map to fixed perceived directions in the median plane regardless of true source location.
• The first arriving sound wave dominates perceived direction of a source — Defines the precedence effect / “law of the first wave front” and its echo threshold.
• Headphone lateralization is one-dimensional whereas free-field localization is three-dimensional for sound images — Makes the technical distinction between ‘lateralization’ (one-dimensional, inside the head) and ‘localization’ (three-dimensional spatial) explicit.
• Industry pressure forced ISO to standardize A-weighted SPL before a better loudness-based method — Clarifies the origin of dB(A) as a deliberately imperfect first-step standard versus the later, more psychophysically grounded ISO 532 loudness methods.
• Speech loudness is captured by the 7th-percentile loudness value — Introduces percentile loudness (N7) — loudness reached or exceeded 7% of the time — as the working descriptor for fluctuating sounds.
• A-weighted sound power exaggerates perceived loudness differences in marketing — Contrasts A-weighted sound power with loudness measurement as two different metrics for noise emission.
• Percentile loudness N5 predicts how listeners judge overall noise immission loudness — Defines the percentile loudness N5 metric as the loudness reached or exceeded 5% of measurement time.
• Annoyance from sounds is quantitatively described as psychoacoustic annoyance. — Defines psychoacoustic annoyance (PA) as a quantitative composite of hearing sensations that predicts laboratory annoyance ratings.
• Spectral broadening can mask source identity while preserving loudness-time function — Describes the FTT-based perceptual processing technique used to separate source identification from loudness content.
• Speech loudness equals the value exceeded only seven percent of the time — Introduces percentile loudness N7 as the matching statistic for time-varying sounds.
• At equal A-weighted level, railway noise is judged quieter than road and aircraft noise — Defines the railway bonus and aircraft malus phenomena directly.
• Speech discrimination requires five distinct auditory features — Defines the five fundamental features needed for speech discrimination in clinical audiology.
• Otoacoustic emissions and frequency resolution arise in the inner ear, peripheral to the synapse. — Defines the peripheral/retro-cochlear boundary at the inner haircell synapse, a useful information-processing framing distinct from clinical audiology.
• Tuning curve measurements are plotted on a critical-band-rate scale representing basilar membrane distance — Explains that the x-axis scaling for tuning curves uses critical-band rate, making equal distances correspond to equal basilar membrane distances.
• Conductive hearing loss shifts tuning curves upward without degrading frequency resolution. — Distinguishes conductive from sensorineural hearing loss: conductive loss preserves frequency selectivity, which points to where filtering occurs.
• Temporal masking patterns are characterized by peak and valley values. — Establishes the peak and valley as the two defining values of temporal masking patterns — core psychoacoustic terminology.
• In hearing-impaired regions loudness grows far faster with level — Recruitment is defined precisely: accelerated loudness growth in regions where threshold in quiet is elevated 30-60 dB.
• Noise-induced high-frequency threshold shifts can make previously inaudible difference tones audible — Distinguishes cubic and quadratic difference tones - distortion products generated in the cochlea - as something that can contaminate tuning curve measurements.
• Temporal resolution is quantified by peak and valley masker thresholds — Defines the two metrics (peak and valley) that characterize temporal resolution in masked-hearing measurement.
• Audibility thresholds rise about 10 dB per tenfold sound duration reduction — Defines temporal integration via the 200 ms window and the 10 dB-per-decade duration rule.
• Recruitment means steep loudness growth that catches up to normal at high levels — Defines recruitment precisely: elevated threshold with abnormally steep growth that rejoins normal loudness at high levels.
• Hearing’s salient spectral features can be matched by adjustable Fourier analysis — Explains the spectral-pitch/partial-tone pair as a hearing-relevant feature model.
• The human ear remains the best speech recognizer and should anchor machine systems — Outlines a full psychoacoustic speech-recognition architecture: nonlinear peripheral preprocessing, auditory sensation extraction, then linguistic rules.
• Perceived speech rhythm tracks the loudness-time function of the signal — Explains how the speech rhythm a listener perceives is derived from segmentation of the loudness-time envelope, an explicit processing model.
• Sensory consonance depends on roughness, sharpness, and tonalness — Defines sensory consonance through its three measurable psychoacoustic components: roughness, sharpness, and tonalness (opposite of noisiness).
• Adding noise can make listeners hear nonexistent high-frequency brilliance in music. — Illustrates the perceptual phenomenon where absent high-frequency content is inferred to be masked by added noise — a hallmark of psychoacoustic inference.
• Musical consonance comprises harmony and sensory consonance from separate domains — Lays out the two-part decomposition of musical consonance and names harmony’s three components (tonal affinity, compatibility, root).
• Listeners perceive missing fundamentals from the harmonics that remain — Introduces ‘virtual pitch’ — a central concept in pitch perception models.
• Signals identical in loudness and pitch can differ markedly in pitch strength — Pitch strength is a third perceptual axis beyond loudness and pitch height.
• Partially masking a tone shifts its perceived pitch upward — Concrete instance of how masking alters pitch perception, a key psychoacoustic concept.
• Narrow-band noise self-modulation masks external amplitude modulation — Introduces the concept of self-modulation in narrowband noise — its intrinsic fluctuation acts like its own amplitude modulation.
• Narrow-band noise self-modulates at 0.64 times its bandwidth — The 0.64-times-bandwidth factor is a precise, citable quantitative relationship linking noise bandwidth to perceived fluctuation.
• Two physically present tones produce an audible third tone inside the ear — Demonstrates the ear’s nonlinear distortion product (2f1-f2) as a concrete listening demonstration.
• Humans can typically discriminate level differences of 1 to 2 dB — Operationally defines just-noticeable level difference via anchor-comparison pairs, anchoring the reader’s model of the JND construct.
• Cubic difference tones arise inside the ear, not in the equipment — A self-performed whistling experiment proving the cubic difference tone’s site of generation.
• A fluctuating noise simulating speech features is a standard test stimulus — Explains what Fastl-Noise is and why it was created.