Psychoacoustic S-An Introduction PDF

PSYCHOACOUSTIC S—AN INTRODUCTION CMSD6310 Dalhousie University M. Markotjohn, MSc.  Topics 1. Overview of Psychoacoustics 2. Sensitivity and Limitations 3. Intensity Discrimination 4. Loudness 5. Masking 6. Frequency and Pitch 7. Temporal Processing 8. Bina...

PSYCHOACOUSTIC S—AN INTRODUCTION CMSD6310 Dalhousie University M. Markotjohn, MSc.  Topics 1. Overview of Psychoacoustics 2. Sensitivity and Limitations 3. Intensity Discrimination 4. Loudness 5. Masking 6. Frequency and Pitch 7. Temporal Processing 8. Binaural Hearing 1. Overview Definition of Psychoacoustics Psychoacoustics is the study of how we perceived sound, making it the psychological side of hearing science It forms the foundation of hearing science Psychoacoustics is closely related to psychophysics: – Psychoacoustics explores how sound affects behaviour and perception – Psychophysics studies how sound behaves and the properties of sound We investigate the relationship (F): how the behaviours () are resulted from the sound stimuli (S): = F(S), or what is F. This relationship can be described in different wants—qualitative, descriptive, or quantitative Position of Psychology in Neuroscience Behavior is both the starting point and the end goal of neuroscience. Initially, the brain is treated as a "black box" — we focus on inputs (stimuli) and outputs (behavioral responses). Later, we open the black box to identify which brain structures are responsible for specific behaviors. Hypotheses are created to explain these relationships and are tested quantitatively. These hypotheses are tested and refined through psychological, neurological, and molecular experiments, uncovering the mechanisms behind behavior. Over time, theories are developed to better explain behaviors and help improve people’s lives.  PSYCHOACOUSTIC FUNCTION:  = F(s) S: sound properties: : behaviors – Intensity – Loudness – Frequency contents – Pitch – Temporal pattern – Temporal patterns, F: functional – Timbre: a combination of relationships (between S frequency and intensity and ) changes over time – Detection – Discrimination – High Level processes: i.e., – Identification the recognition of speech – Judging (scaling) Analytical Studies What roles do How are acoustic individual acoustic parameters parameters like detected, intensity, discriminated, frequency, and identified and temporal patterns scaled (quantified)? play?  Information is embedded within multiple acoustic parameters Integrative The brain integrates responses to individual parameters to improve the Studies perception of information Integration involves neurons across different auditory channels (such as different locations in the same nuclei), various neuron types, and different nuclei, balancing excitation and inhibition, as well as afferent and efferent pathways  Timbre: The quality of a complex sound, integrating frequency components, intensity, and temporal patterns Examples of Integration in Sound localization: Using different cues like intensity, timing, and spectrum to Psychoacousti determine where a sound is coming from cs Auditory perception: Creating an “acoustic image” in a complex sound environment, involving higher-level processes like memory and prior experience  “Analytical versus integrative” refers to two different ways the brain processes information. We use two categories of Who Cares skills in perception Sometimes, we focus on individual details About Analytical Other times, we focus on the overall meaning vs Integrative We often use both types of skills unconsciously ? Training can change how we handle these signals  Absolute limen: The lower boundary of detection (i.e., minimum or sensitivity) DIMENSIONS Terminal limen: The upper boundary of OF AUDITORY detection (i.e., ABILITY maximum or limitation) Absolute limen— Difference limen: The detection smallest change in Difference limen— some aspect of the discrimination stimulus that can be detected 2. SENSITIVITY AND LIMITATION Chapter 9 in Gelfand textbook Concepts Sensitivity: Absolute thresholds for sound intensity/pressure — the basic goal of a hearing evaluation. Minimum Audibility Curves: Sensitivity across the frequency range, often shown in an audiogram. Frequency Range of Hearing: The limits at the low and high ends of the hearing spectrum. Terminal Threshold (upper limit): Defined by: – Threshold of discomfort – Threshold of pain Note: high sensitivity means low threshold  Dynamic Range of Hearing  Two thresholds define the dynamic range. Dynamic range  What is dynamic— output changes with input?  Loudness changes as the result of sound level change  Please define DR.  Dynamic range varies with frequency: largest Note: -Terminal threshold curve is flat, while audibility curve is curved. -In a relationship, variation of independent variable results in the change of dependent variable. This is dynamic. If the change of independent variable does not cause change of dependent one, this is NOT dynamic. -Dashed lines at the two ends of frequency indicate uncertainty of hearing.  COMPARIS ON TO THE DYNAMIC RANGE OF AUDITORY NERVES Simple Summary: High SR units: Great at picking up soft sounds but max out (saturate) quickly with louder sounds. Low SR units: Not as sensitive to soft sounds but can handle louder sounds without saturating as quickly. This is how our auditory system can handle both very quiet and very loud sounds, thanks to the balance between these two types of nerve units. Hearing Area (Defined by Sound Level and Frequency) Terminal thresholds can be defined in different ways. They are less clear or accurate and less studied due to ethical reasons Threshold is Determined Statistically Multiple trials are tested at each level to calculate the percentage of correct responses. The threshold is statistically found based on the percentage of correct responses at a chosen criterion. The 50% correctness point is where the threshold is set, indicating the sound level where hearing begins.  Note: For “yes/no” method, the criterion is 50%, which does not clearly show this principle. Better example will be seen. Variability and Reliability Many factors (other than actual ability) can affect performance – that’s why we use statistics Variability (differences between individuals): Hearing thresholds can vary by as much as 20dB across people with normal hearing (normal range: -10 to 10 dB HL) Test-retest reliability (differences within individuals): Hearing thresholds can vary by 10-15dB when the same person is tested multiple times A range is provided to account for both variability and reliability The meanings of the variations Normality is defined as a range. The range depends on the accuracy of the testing method and the skill level of the tester. In our clinical setting, a variation of 10-15 dB is considered normal for test-retest reliability. If performance is worse than that, additional training may be needed*. This range also helps determine if a hearing loss is present, as a 10-15 dB variation is normal.** Criteria based on reliability are used to judge if someone truly has hearing loss.  Note: For “yes/no” method, the criterion is 50%, which does not clearly show this principle. Better example will be seen. THE OPEN V SHAPE OF ABSOLUTE THRESHOLD  The external and middle ears play a role in shaping How is the the audibility curve Audibility Curve Consider how the Shaped? frequency response of the external and middle ears affects hearing sensitivity  The external and middle ears don’t fully explain hearing sensitivity—there’s also a contribution from the cochlea. Contributi on of the Cochlea Bone conduction bypasses the external and middle ear, allowing us to test cochlear function directly. Technical challenges Bone conduction with bone conduction doesn’t provide a flat can make it harder to frequency response. interpret results.  MINIMAL AUDIBLE FIELD (MAF) VS. MINIMAL AUDIBLE PRESSURE (MAP)  Minimum Audible Field (MAF): Measured with speakers in an open field, calibrated at 1 meter, 0 degrees incidence, with both ears (binaural). Sound pressure levels (SPLs) are measured with the listener absent. Minimum Audible Pressure (MAP): Measured in a closed field (with headphones), in one ear (monaural), and SPLs are measured with a MAF vs. coupler. MAP Difference: MAF thresholds are 6-10 dB better (lower) than MAP thresholds. ON MINIMUM AUDIBLE SOUND FIELDS J. Acoust. Soc. Am. Volume 4, Issue 4, pp. 288-321 (1933); L. J. Sivian and S. D. White  MAF VS. MAP Simple Explanation of the Graph: MAF (open ear): We hear better because both ears work together, and the sound is enhanced naturally. MAP (closed ear): Hearing thresholds are higher because sound is delivered directly into  Early Theories for the 6dB Difference Three reasons Studied from 1960 to 1970. for the 6 dB difference: Three possible reasons were proposed: MAP removes The MAP method removes the amplification provided the natural by the external ear’s resonance. boost from the MAF benefits from binaural summation (using both outer ear. MAF lets both ears together). ears work MAP may pick up physiological noises (e.g., internal together. body sounds). MAP can pick up internal body noises, raising Reviewed by Killion (1978) and Yost & Killion (1997). thresholds.  Current view: The "missing 6 dB problem" doesn’t exist. Yost and Killion (1997) reviewed the issue again, finding that the difference is due to calibration methods. When calibrated MAP=MAF using the real ear, there is no 6 dB difference The key is between MAP and MAF. ? that calibrating with a coupler or open field without the head does not account for the contribution of the outer ear. If calibration is done using real ear measurements (with a probe microphone at the eardrum), there is no 6 dB difference.  Real-ear calibration is Why often not practical in many cases. Should We Care We need to understand About the the difference and why it MAF and happens to interpret hearing data correctly. MAP Difference This helps in the ? establishment of sound level references. Reference Equivalent Threshold SPL (RETSPL) Sound pressure is measured in a coupler or in a space without human subjects. Coupler pressure is measured to establish a voltage-to-sound pressure relationship, ensuring consistent reference levels across labs. RETSPL is established and verified globally. RETSPL is created for couplers, not for real ears, as it's nearly impossible to create one for real ears. The head’s contribution (transfer function) is ignored in this process. Dummy heads are available for testing but are expensive and less commonly used. 6-CC COUPLER FOR SUPRA-AURAL EARPHONES 2-CC COUPLER FOR INSERT EARPHONES Left image: A 2-cc coupler simulates the ear canal for measuring sound from insert earphones. Right diagram: A cross-section showing how the coupler mimics the ear canal and  Tables 9.1 and 9.2 list RETSPLs for earphones and bone vibrators. RETSPLs represent thresholds or the sound pressure levels (SPLs) needed to reach RETSPLs: a hearing threshold. Data and Applications include: Applicatio ns Calibration references for speakers/earphones Establishing hearing levels (using RETSPLs as the zero point for 0dB HL) Setting noise allowance standards *The idea is that the noise below threshold does not matter, but if higher than the threshold, masking occurs. Why using HL? HL Makes Clinical Applications Easier SPL: Reference for physical sound pressure (20 µPa). HL: Reference for hearing thresholds of normal subjects using RETSPL. SL (Sensation Level): Reference based on individual hearing Left graph: Shows hearing thresholds in terms of thresholds. sound pressure (SPL). Right graph: Translates the same data into hearing level (HL), making it easier to understand clinically.  Applicatio For calibration in SPL: This explains how SPL is related to HL (hearing n of level) as indicated by your equipment RETSPL RETSPL (Reference Equivalent Threshold Sound Pressure Level) Control of ambient noise: Noise plays an essential role in below RETSPL won’t be heard, so it calibrating audiometric equipment. is allowed. If the noise level is For calibration, SPL is measured higher than RETSPL, masking will and then related to HL (hearing occur level) as indicated on your equipment. This ensures that the devices you’re using are properly When using earphones, the calibrated to reflect real hearing attenuation level must be thresholds in decibels. considered. More ambient noise is allowed with earphones, making closed-field tests more practical for bedside evaluations 3. INTENSITY DISCRIMINATIO N Chapter 9 of Gelfand Weber’s Law A well-known principle in sensory discrimination. Difference threshold is also called difference limen. For intensity, it's the smallest intensity difference you can detect, noted as ∆I (also known as the just noticeable difference or JDD). Weber’s law states that ∆I (the smallest detectable difference/change) is proportional to the original intensity (I), meaning ∆I/I is constant. This means the smaller the value of ∆I, the less of a change is needed for detection.  If the initial intensity (I) is 10 units, and the smallest detectable difference (∆I) is 0.5 units, what will ∆I be if I is 1000 units? Since 0.5/10 = 0.05, we can **Predictio multiply by 1000 to get ∆I = 50 n from units. Weber’s According to Weber’s Law, if it's perfectly accurate, ∆I/I would plot as Law a straight horizontal line on a graph. However, in real-life experiments, the data doesn’t always perfectly match Weber’s prediction.  ∆I/I (Weber’s fraction) measures how well we can tell differences in intensity. Converting We can also express this ability in decibels (dB): Weber’s ∆I in dB = 10 log (1 + ∆I/I) Fraction to According to Weber’s Law, both Weber’s fraction and the just detectable difference dB (JDD) in dB will not change based on the starting intensity. For pure tones, the smallest detectable difference is about 1 dB (or around 0.3 in ∆I/I).  Intensity Discrimination Result: A key point to remember is Not Quite a Horizontal Line that the intensity discrimination threshold for tones or narrow band signals is around 1 dB. The graph shows that intensity discrimination A single number to remember for isn’t exactly linear or intensity discrimination threshold is 1 dB constant for tones or narrow band signals! across intensities, but hovers  If Weber's law were entirely accurate, we'd see a completely flat horizontal line. Near Miss But for pure tones, there's of Weber’s a small slope instead. Law So, Weber's law is close, but it's not a perfect fit for pure tones. White noise follows Weber’s law more closely than tones Intensity discrimination works better with broadband signals (like white noise) than with narrowband ones. Note that the intensity discrimination threshold is smaller for white noise than for pure tones.  Rely upon short Method memory Considerati on Influenced by adaptation The results can vary depending on the method used Method B is also called the pedestal method or pedestrian detection method Explanation of the Picture: Panel A: Shows a gated pulse tone, meaning the sound is turned on and off. To detect a change, you rely on your short memory of what the tone sounded like before. 1.Panel B: This is the continuous tone method with an added increment in the middle. Here, your ear can adapt to the tone, which might affect your ability to notice the change. 2.Panel C: Involves a modulated tone, where the sound gets louder and softer in waves. The change you measure is the modulation depth, or how much the volume varies. Each of these methods tests your ability to detect changes in sound in slightly different ways. The choice of method matters because it can lead to different results in experiments or testing situations. GAP-DETECTION THRESHOLD AS A FUNCTION OF SENSATION LEVEL (SL) TWO METHODS: 1) CONTINUOUS PEDESTAL (SQUARE MARKERS, BETTER RESULTS) 2) GATED (CIRCLE MARKERS) Key Points: Continuous Pedestal (square) gives better results than Gated (circle). Higher sensation levels (louder sounds) improve gap detection.  Bandwidth Summary of Frequency Factors Affecting Sound Level Intensity Discriminati Duration (not addressed) on Testing Methods 4. LOUDNESS SENSATION Chapter 11 from Gelfand’s textbook  Loudness refers to the perception of sound intensity, which is mostly influenced by sound level but can also be affected by other factors like frequency, duration, and the Loudness presence of other sounds. Detection Loudness is subjective and can only be measured through behavioral tests.  Arbitrary Scale: 800 Hz tone at 100 dB SL ≈ 100 arbitrary units; Loudness Half the loudness = 50 arbitrary Scale** units, Double the loudness = 200 Aarbitrary units in level 10 dB change results in a 2x change in loudness. Pay Attention to Scaling Methods: Both Show Loudness-Sound Level Relationship  Explanation of the Picture: Left graph (a): This graph shows how loudness increases as the intensity (measured in dB SL) gets higher. The curve is steep, showing that as intensity grows, the perception of loudness increases rapidly. Right graph (b): Here, loudness is plotted against relative sound pressure, and we see a similar trend. As the sound pressure increases, the loudness goes up, but it’s a bit less steep compared to the left graph. Both graphs illustrate the relationship between loudness and sound level, but using different units.  SL refers to sensation level, which is calculated using an individual’s threshold (0 dB) as the reference point. Note for The two graphs display the the same data, but the X-axes Previous differ: the left uses SL in dB, and the right uses sound Slide pressure on a linear scale. Be cautious—visual memory of the graphs can be misleading!  Stevens (1957): A 1000 Hz, 40 dB tone is defined as 1 sone. Sone Scale Doubling or halving the loudness results in 2 sones or 0.5 sones. The relationship between sone Sone Loudness loudness and intensity (m = k * I^0.3 easured in physical units, (where k is a constant). not dB) follows a power function:  A 2x If we increase 10 log (I2/I1) = 10 dB log (I2/I1) = 1 Change in intensity level I2 = 10 x I1 (the intensity (IL) by 10 dB, Sone increases by 10 times) Correspon ds to a S2 = k(10I1)^0.3 = According to the 10dB Level power law: 10^0.3 * kI1^0.3 Change S2 = 2 * S1  Intensity (I) is related to pressure by: I = kp², or p = k’√I** Converting A 10x increase in intensity results in a √10 change in pressure: I₁/I₂ = 10, p₁/p₂ = Intensity (I) (I₂/I₁)¹/² = √10 to Pressure Therefore, a 10 dB pressure change is (p) required for a doubling of loudness (S): 20 log (p₁/p₂) = 10 dB A 10 dB change in sound pressure level (SPL) corresponds to a 2x increase in loudness. A 10 dB Change in Intensity Results in a 2x Change in Loudness Not applicable at low SPLs near the threshold. Key Points: A 10 dB change leads to a doubling in loudness, except at low SPLs near the hearing threshold. Equal Loudness Contours The phon scale uses 1000 Hz as a reference point. At this frequency, phon and SPL va lues are the same. For other frequencies, SPL is determined by comparing the loudness to a 1000 Hz tone at the same phon level. Key Points: The phon scale matches SPL at 1000 Hz. At other frequencies, the SPL needs to change to maintain equal loudness. How Contour Shapes Change with Sound Level At low SPLs (e.g., 10 dB), the SPL required for the same loudness (10 At high SPLs (e.g., 100 phons) changes dB), the variation across significantly with frequencies is smaller frequency: for example, (the contours are flatter). 30 dB at 100 Hz, 0 dB at 4000 Hz.  Dynamic range measures the SPL Impact of difference between the lowest and highest points (floor and ceiling). the The ceiling remains constant, but contour the floor varies. shape on dynamic The 1000-4000 Hz range is the most sensitive (lowest threshold), meaning it range and has the largest dynamic range. loudness Low frequencies have a smaller growth dynamic range and faster loudness growth; the same happens, though to a lesser degree, at the higher end of the frequency spectrum.  Note: -Loudness growth function: how loudness increase with sound level. -The flat contour at high sound level means the ceiling is the same, the curved contour at lower level means the floor is equal in height. Therefore, loudness grows “faster” with SPL. -Here the “faster” is not a time concept, per se; but mean how much change with the same change of SPL. How Loudness Contours Affect the Perception of Sound Boomy: Excessive low-pitch sound when playback is louder than the recording. Low-frequency sounds that were inaudible during recording become audible during playback. – Boomy: making a loud, deep, resonant sound. Tinny: Excessive high-pitch (crisp) sound when playback is quieter than the recording. Filter Settings of Sound Level Meters Key Points: Filter A is for quieter sounds and emphasizes human sensitivity to mid-range frequencies. Filter C is for louder sounds and treats all frequencies more equally. Filter B is less commonly used but works as an intermediate filter. Filter Networks Used in Sound Level Meters dB A: for measuring sound in quiet environments dB C: for measuring high-level noise Key Points: Filter A: De-emphasizes low frequencies, used for quiet environ Filter C: More balanced across frequencies, used for louder soun  Concept of the Critical Band: The Range of Sound Frequencies Detected by Neurons in a Channel Intensity -The critical band (CB) is the range of CB frequencies that a hair cell (HC) and its connected SGNs can detect -This is related to the tuning curves of ANFs Frequency  “To be heard” is equal to “to be masked”. The note "To be heard is equal to to be masked" refers to the idea that in the context of the critical band, sounds that can be "heard" by neurons within that band are also the ones that can be masked by other sounds within the same frequency range. Here's what that means: Hearing and masking: When a sound falls within a particular critical band, it can be detected (or "heard") by the neurons responsible for that range. However, if another sound—especially a louder one—also falls within that same critical band, it can mask the first sound, making it harder or even impossible to hear. Example: Imagine two tones very close in frequency. If both fall within the same critical band, and one is louder than the other, the louder tone can mask the quieter one, preventing it from being perceived. This is because the neurons responsible for processing that range of frequencies get "overloaded" by the louder sound. In other words, if a sound is within the range that can be "heard" by neurons, it is also in the range where it can be masked by other sounds of similar  Keep the total sound level (and total sound power) constant. The Effect Increase the bandwidth of the signal from very narrow to beyond the critical band (CB). of Signal Bandwidth Loudness remains the same as long as the signal is within the critical band. on Loudness Loudness increases when the bandwidth exceeds the critical band.** If the signal is well above the threshold, and its bandwidth extends beyond one critical band, more auditory channels are activated, making the sound seem louder. Impact of Bandwidth on Loudness Turning point The change of tells CB loudness for equal intensity equal intensity Loudness for Intensity for The change of loudness intensity for equal equal loudness Bandwidth  Explanation of the Picture: The graph illustrates how bandwidth (how wide the range of frequencies is) affects loudness and intensity. The turning point on the graph marks the critical band (CB), the point where a change happens in how sound is processed. When the bandwidth is within the critical band, loudness doesn’t change much, even if the bandwidth increases. This is shown by the flat line. Once the bandwidth exceeds the critical band, the loudness increases for the same intensity (as shown by the upward line). Similarly, if you want to keep loudness constant as bandwidth increases, you would need to reduce the intensitybeyond the critical band (shown by the downward slope). In essence, when the bandwidth is small (within the CB), changes in loudness and intensity aren’t significant. But once the bandwidth surpasses the critical band, the way we perceive loudness changes significantly. Adaptation vs Fatigue Adaptation: A decrease in sensitivity during a signal’s presentation. The response is reduced to repeated or non-novel signals, but the sensitivity remains high if a new signal appears Fatigue: A decrease in sensation after the signal has been presented. The response is reduced due to a partial failure, and there is a loss of sensitivity to all stimuli How to Measure Adaptation? Measure using loudness matching: In monaural presentation, play a continuous sound to one ear. Match the loudness by playing a pulsed signal to the other ear. Observe the changes in loudness over time. Note: The pulsed signal does not undergo adaptation.  Explanation of the Picture: The graph shows the adaptation period over time. In part a, the level of the comparison tone starts off steady. In part b, as time progresses, we see the adaptation begin, and the tone’s loudness decreases over time. In part c, after the adaptation period, the level has dropped by about 30 dB, showing the full effect of adaptation. Key Points: Adaptation reduces the loudness of a continuous sound over time. **The pulsed signal remains unaffected by adaptation, so it helps measure the change. Results: 1. Significant variation between individuals 2. **More noticeable at low sensation levels (SL) 3. More pronounced at higher frequencies (not shown here)  Note that in binaural presentation, matching is done when the continuous sound is stopped. Explanation of the Picture: The graph shows how loudness estimates change over time at different sensation levels (SL). The x-axis shows time in seconds, and the y-axis shows the loudness estimate. As time goes on, especially at lower SLs like 20 dB, the loudness drops more significantly, showing the effect of adaptation. At higher SLs, like 70 dB, the loudness stays more constant over time, meaning there’s less adaptation at higher levels. Key Points: Adaptation happens more at lower sensation levels (SL). There’s a variation in how people experience adaptation. Masking Changes the Growth of Loudness When masking occurs, the threshold (the floor) is elevated. The ceiling (maxim um loudness) Growth faster with masker remains the same, so the dynamic range becomes narrower. This results in faster loudness growth. This effect can also be seen  Explanation of the Picture: The graph shows how loudness grows with increasing sound levels both with and without a masking sound: Without a masker, the loudness increases steadily, as shown by the dashed line. With a masker (solid lines), the starting point or threshold is higher, meaning softer sounds are harder to hear. However, once the sound exceeds this threshold, the loudness grows more rapidly. The result is a faster loudness growth when masking is present. Key Points: Masking elevates the threshold (soft sounds become harder to hear). The ceiling doesn’t change, so loudness grows faster in a narrower dynamic range.  Refers to the abnormally fast Loudness growth of loudness as sound Recruitmen levels increase. This term is specifically used in relation t to sensorineural hearing loss (SNHL). Note: The loudness recruitment in SNHL The pattern of loudness is different from the fast growth of growth in SNHL is similar to loudness in partial masking. what is seen in masking for normal hearing individuals. Key Points: Loudness recruitment = abnormally fast loudness growth in SNHL. It behaves similarly to masking in However, there are normal hearing but has distinct important differences (to be differences we'll address later. discussed further in Audition II). CODING BY SPREAD OF EXCITATION  The graph illustrates how neurons in the ear respond to different sound intensities (dB SPL) at different frequencies. Key Points: High SR neurons react to a broader range of frequencies as intensity increases. Low SR neurons respond more selectively to louder sounds and in a narrower frequency range. Spread of excitation means that as sounds get louder, they activate more neurons over a wider frequency range. 5. MASKING Chapter 10 of Gelfand  Masking: When one sound interferes with the perception of another sound. What is Why study masking? Masking? It happens frequently in daily life and impacts how we hear It is a useful way to study how the hearing system works It has important clinical uses for diagnosing hearing issues  Masked threshold: The point at which a sound (probe or target tone) becomes audible in the presence of a masking sound. The masked threshold is usually higher than when listening in silence. Masking Lowest level you can hear target sound Changes Threshold Maskers can be any sound—such as noise or tones. Timing relationship: The masker can overlap, precede, or follow the target sound. Masking raises the hearing threshold The amount the threshold increases shows how much masking is happening  How Does Masking Occur? Excitation Patterns of the Basilar Membrane as the Basis of Masking BM vibration by a tone masker GRAPH: The graph shows how different sounds cause the basilar membrane BM vibration to vibrate. A loud sound (masker) creates bigger movements, while a by a target softer sound (target) creates smaller movements, making it harder tone, masked to hear the softer one when both are present. The target sound is heard only when a new and distinct vibration （ ΔE ） is produced The graphs show that the target sound can only be heard if it causes a noticeable difference (ΔE) in the vibration compared to the masking sound. Key Concepts: ΔE (difference in excitation) is crucial for hearing the target sound. The masking sound covers the target sound unless there's a distinguishable difference in how the basilar membrane vibrates. This difference allows the brain to  The x-axis shows the distance along the basilar membrane. The base (left side) responds to high frequencies, and the apex (right side) responds to low frequencies. Key Points (c) The test tone has a lower from the frequency than the masker. This is because the peak for both is closer Previous to the apex (low frequency) compared to the masker alone. Slide… (d) The test tone has a higher frequency than the masker.  This type of masking occurs when the masker’s excitation to the neurons overlaps with the excitation needed to sense the target signal. Energetic In the cochlea, both the signal and masker excitations partially (Peripheral overlap, making it harder to ) Masking distinguish the target sound. This is different from “informational masking” or “central masking,” which happens at higher cognitive levels.  Partial masking occurs when the target sound is still heard, but a masker is present. Partial Masking The masker doesn't completely block the target but reduces its perceived loudness. Using Masking as a Research Tool Masking tuning curves (TC) are used to measure frequency resolution. – A masking TC shows how the masked threshold (the quietest sound you can hear with a masker present) changes with signal frequency – Psychophysical TCs are obtained from behavioral tests Psychophysical Tuning Curve The psychophysical tuning curve (PTC) shows how the masked threshold varies depending on the masker frequency. A probe signal is presented at 10-20 dB above the listener’s threshold (marked by the asterisk). Explaining the Graph… The wavy line represents the masker threshold required to fully mask the probe tone at the asterisk (*). The masker’s threshold is measured at different frequencies As the masker frequency approaches the target tone, the masker threshold decreases The curve is asymmetrical—it’s broader on the low- frequency side, which shows that low frequencies more easily mask high frequencies  We test how other sounds affect the hearing of a target tone (masking). How the The ability to hear the target tone Psychologic represents the function of a single auditory channel. al Tuning Changes in masking across different Curve (TC) frequencies tell us the range where this auditory channel is most sensitive. Reveals The frequency range where masking is Frequency most effective shows what the channel can hear or is selective to. Selectivity This concept of frequency selectivity is covered by the idea of the “critical band.  Key Concepts: Masking reveals how one sound affects the hearing of another. Auditory channels each have a sensitive frequency range. The critical band is the range where the channel is most effective. Graphic Explanation of the Critical Band (CB) Intensity -The critical band (CB) is the frequency CB range where a sound is "heard" by a hair cell (HC) and its connected spiral ganglion neurons (SGNs), as measured through the psychophysical tuning curve (TC). Frequency  Key Concepts: Critical Band (CB): The frequency range a hair cell and its neurons are sensitive to. Psychophysical Tuning Curve (TC): Measures how different frequencies are heard by the ear.  The masking tuning curve shows that if the masker’s spectrum is too far from the tone, no masking occurs. Critical When broadband noise is used as Band (CB) a masker, only the energy in a certain frequency range is Defined by effective; this range is called the critical band. Masking The critical band acts as a filter, allowing sounds within it to affect the perception of the signal, while sounds outside the band don’t interfere.  Key Concepts: The masking tuning curve shows how masking works depending on frequency. The critical band is the frequency range where masking is effective. The critical band acts like a filter, blocking sounds outside the range from interfering  The critical band filter defines the bandwidth of an auditory channel, set at the auditory periphery and modified in the central auditory system (CAS). The CB can be measured in several ways. Critical Band The CB is proportional to the central frequency, around 20% or 1/3 of an (CB) cont’d octave. In clinics, we use a 1/3 octave band of noise as a masker because it is more efficient. Using broadband noise wastes energy outside the CB, and the masker becomes too loud.  Key Concepts: The critical band defines the auditory channel's bandwidth. It's proportional to the central frequency (about 1/3 octave). 1/3 octave noise is efficient for masking in clinical settings. The image shows three types of temporal masking: 1.Backward masking (a): The masker comes after the signal, but it still interferes with the perception of the signal that came before. 2.Forward masking (b): The masker comes before the signal, making it harder to hear the signal that follows. 3.Combined masking (c): This shows both backward and forward masking happening at the same time, with a signal in between two masker sounds. The timeline at the bottom shows how the masker and signal are spaced in time, measured in milliseconds. Threshold is higher when masking Dichotic: masker and probe are in different ear. Masking occurs via CAS: central masking Key Points About Previous Slide… The closer the signal is to the onset or offset of the masker, the stronger the masking effect Dichotic masking (central masking, masker and signal in different ears) is weaker than monotic masking (masker and signal in the same ear) Backward masking is likely due to memory interference, not cochlear interaction  Key Concepts: Closer timing between signal and masker = stronger masking. Monotic masking (same ear) is stronger than dichotic masking (different ears). Backward masking is likely due to memory interference, not direct interaction in the cochlea. Informational/ Central asking Informational masking doesn’t involve energetic interaction in the cochlea. It’s more about the content or "information" being confused. Central masking happens when the masker and signal are presented to opposite ears. Informational masking ≠ central masking, but they are related. – Informational masking can still occur even when the signal and masker are in the same ear In informational masking, the masker and signal don’t overlap in frequency  Key Concepts: Informational masking happens when the brain confuses the content of the masker and signal. Central masking occurs when the masker and signal go to opposite ears. No frequency overlap is needed for informational masking to occur.  Thank-You for a Great Semester!   Course Evaluation Complete now if you haven’t already!!! They should have been emailed to you  **Your feedback is VERY valuable to me!!! If you enjoyed this course and my teaching, course feedback is taken seriously in terms of determining if I will teach the course again **