Integrative Function in the Auditory Pathway for the Perception of a 4 kHz 10 dB Sound

Integrative Function in the Auditory Pathway for the Perception of a 4 kHz 10 dB Sound

Introduction
In the perception of an incoming 4 kHz 10 dB sound at 0 azimuth angle, the ears act as transducers for transformation of acoustic signals into electrical signals. The physical characteristics, location and meaning of the sound represented by these signals and are processed by our brain. This task requires contribution of various components along the auditory pathway including the outer ear, middle ear, inner ear and our brain. This essay will critically evaluate the role each component plays in the process.
Assumptions
Let us assume the sound is coming into the ears of an healthy hearing adult male with 25 mm ear canal length (Johansen, 1975), and intact tympanic membranes without cerumen occlusion. Shaw (1974a, b) suggested that these dimension are important for how sound pressure is distributed along the ear canal while Stinson and Lawton (1989) stated that precise definition of ear canal geometry is required to predict the sound pressure level at high frequencies. The angle of the incoming sound is also of importance because as Shaw (1974a, 1974b) stated that the amplification of high-frequency sound in the ear canal is directionally dependent. Therefore, we will assume that the incoming sound is at 0 azimuth angle to avoid complications regarding localization. Furthermore, as there are not many researches focusing on exactly 10 dBHL or 10 dBSPL sound, we will try to generalize our sound to be a soft sound and will use the research findings on low intensity sounds to support our argument. In our discussion, we will try to concentrate on studies derived from humans as it is most relevant. However, in order to explore beyond the limitations posted by research ethics, it is essential to refer to studies of other mammals for implications to the understanding of the human auditory system.
1. Outer Ear
We will summarize a series of findings showing consistently that the outer ear provides amplification of an incoming sound. We feel that these findings will be applicable to our 4 kHz 10 dB sound as Shaw (1966) stated that the outer ear naturally act as a coupler between any airborne sound and the middle ear in any natural listening situation, but the exact degree of amplification it provides is unknown.
After Shaw (1966) stated that the outer ear act as a coupler, years of devoted research has shown that the outer ear does provide approximately 10 dB gain to a 4 kHz sound (Wiener & Ross, 1946; Shaw, 1974a). This gain is suggested to be contributed by a combination of the ear canal and the concha (Shaw 1974a; Mehgardt and Mellert, 1977). Shaw (1974a) found that the outer ear acts as a low frequency filter that amplifies mid-frequencies. Specifically, he calculated that the ear canal resonance to be 3.4 kHz, contributing a gain effect to our sound. He also suggested that the concha plays a more definite role in amplifying high frequency sounds at 5 kHz. In agreement with Shaw (1974a), Mehrgardt and Mellert (1977) found an ear canal transfer function and calculated that the canal has around 12 dB gain at 4 kHz while the free-field-ear-canal-entrance contributes around 5 dB gain at 4 kHz. Together, we think that the outer ear structures will provide such gain to our 4 kHz 10 dB sound. To further investigate the amplification effect of the outer ear, it is essential to know whether the gain provided is intensity dependent whereby more gain might be provided with presentation of a soft sound.
2. Middle Ear
Findings show that our 4 kHz sound will be less reflected by the tympanic membrane when compared with sounds at other frequencies, and that the ossicles in the middle ear will move linearly for a soft sound. We also found that the middle ear provide further pressure gain to the soft 4 kHz sound.
As shown by Keefe and Levi (1996), the energy reflectance, defined as the ratio of the power reflected from the middle ear to the incident power, of the tympanic membrane is lower at 2-4 kHz when compared with sounds at other frequencies. This suggests that there is better transmission of energy of the 4 kHz sound into the middle ear. Similar finding was observed in Stinson (1990), Keefe, Bulen, Arehart and Burns (1993) and Voss and Allen (1994). A more recent research finding by Feeney and Sanford (2004) showed a minimum of the energy reflectance in both young and old human adults at around 4 kHz, meaning it is applicable to our sound.
After the 4 kHz sound travels pass the tympanic membrane, the ossicles are set into vibration and the movement of the ossicles is linear. Buunen and Vlaming (1981) found that umbo of the malleus vibrated linearly in proportion to the input sound pressure up to the highest intensities tested (11 dBSPL), to within the error of measurement (5% or 26 dB below the stimulus level). This linearity is specific to our soft sound, as later researches review that at higher intensity, the footplate motion is dominated by rotational motion rather than a linear piston-like movement (Heiland, Goode, Asai & Huber,1999; Hato, Stenfelt & Goode, 2003).
The pressure gain for 4 kHz sound is different from the theoretical amount proposed by many texts, which should be around 36dB (Moller, 2006). The reason is that this theoretical gain is calculated by considering the middle ear as an optimal impedance transformer. The difference in the area of the oval window and the tympanic membrane, the lever system of the ossicles, and the impedance of the fluid in the cochlea, are involved in the calculation of the theoretical gain. Interestingly, the human middle ear is slightly different. Factors include the mass and stiffness of the middle ear, the mode of vibration of the ossicular chain at different frequencies, and the effective area of vibration of the tympanic membrane will affect the actual gain. As illustrated in Khanna and Tonndorf (1972) study of the human tympanic membrane done in cadaver ears, the tympanic membrane has a smaller effective area at high frequencies than it has at lower frequencies. Therefore, the ratio of the effective vibration area of the tympanic membrane and the oval window will be frequency dependent.
The actual gain, supported by different researchers such as Puria, Peake and Rosowski (1997), who suggested the middle-ear pressure gain, a ratio of the pressure at the scala vestibuli near the stapes footplate to the ear-canal pressure near the tympanic membrane, to be 18 dB at 4 kHz. Merchant et al. (1997) had neatly summarized the findings of various studies including those of Kurokawa and Goode (1995). They arrived at gain values of approximately 20 dB between 250 Hz and 500 Hz with a maximum of 25 dB at 1 kHz above which the gain decreases at a rate of 6 dB/ octave. Refer to this study, we can estimate the gain of the middle ear at 4000 Hz is around 13 dB. Aibara and colleagues (2001) found a 12 dB pressure gain at 4.0 kHz. Although the magnitudes are slightly different, we regarded these findings as consistent with each other.
We suspect that the middle-ear air space may affect the impedance of the tympanic membrane in addition to the above factors. Some researchers (Gyo, Goode, & Miller, 1986) suggested that the aditus ad antrum has little effect on the processing of our 4 kHz sound. However, some researches believed that the volume of middle-ear air space does affect the impedance at tympanic membrane to our sound. In particular, Stepp and Voss (2005) found that at frequency above 1 kHz, variations in middle-ear air space can affect the middle-ear impedance at tympanic membrane by as much as 10 dB. With such controversy of the role of the middle ear air space on the impedance of the tympanic membrane, we feel that further investigation is possible to determine the interaction between them.
3. Inner Ear
So far, we saw that the outer and middle ear act as an essential transducer to our 4 kHz 10 dB sound. We believe that the cochlea is the first analyzer of the sound and the script writer for further analysis in the central auditory system.
3.1 Preparations in the cochlea
3.1.1 Basilar Membrane.
The 4 kHz sound will cause the basilar membrane (BM) to vibrate maximally towards the basal end of the cochlea. However, the exact location is unknown, but by referring to the human cadaver studies by Bekesy (1960) and the mathematical model proposed by Holmes (1987), we are confident to conclude that the BM vibrates maximally at the basal end.
Bekesy (1960) investigated the relative amplitude of passive vibration of the basilar membrane at various points (“the travelling wave theory”). He concluded that high and low frequency sounds will lead to a maximum deflection at the base and the apex of the cochlea respectively. For example, he found that the relative amplitude of vibration at 13 mm from stapes was the largest with a 2.5 kHz sound. However, the findings of Bekesy have two key limitations.
Firstly, his results were obtained using cadaver which only references to live humans. Another limitation of Bekesy’s research findings was pointed out by Pickles (1988), who suspected that the sound Bekesy used was of high intensity for such clear observations.
Nevertheless, the mathematical model of Holmes (1987) predicted similar location of BM vibration as Bekesy (1960). In addition, although Robles and Ruggero (2001) found that the peak of the active wave envelope in the cochleae was less basal (~1mm) with chinchilla because they believed that when a wave of a given frequency travels from the base towards the apex of the basilar membrane, it obtains energy from the nearby outer hair cells, the same conclusion was drawn that high frequency sounds produce the maximum vibration towards the basal end.
According the researches, we believe that our 4 kHz 10 dB sound will have maximum BM vibration around the basal area. However, we need to know more about the passive properties of the basilar membrane and the active contribution of outer hair cells (OHC) in order to define the exact location and the amplitude of the maximum deflection.
3.1.2 OHC
To further prepare the analysis specific to our 10 dB sound, the OHC enhance the vibration of the BM at the maximum location by active elongation and contraction. Evidences from various researches had shown that the cochlea act as an active transducer where the OHC gives a larger gain in low intensity than high intensity sound. This nonlinearity nature of the cochlea reveals to us that the OHC acts differently at low intensity providing a larger BM vibration for our 10dB sound than sounds at higher intensity.
According to Harte, Elliot and Rice (2005), the ratio of BM vibration to sound pressure level input was lower at high than low intensity. This nonlinearity nature is consistent with the concept of “Cochlear amplifier” found by Brownwell, Bader, Bertrand and Ribaupierre (1985) where the OHC contributes to amplify sound below 60dB over a 60dB range. Harte et al. (2005) concluded from other findings that below 30dB, the gain of the BM vibration level is linear (1dB/dB). However, if the sound intensity is higher than 30dB (relative to 20mPa), the increase in the BM vibration will become compressed (0.5dB/dB). When Dallos, Zheng and Cheatham (2006) reviewed the nature of the OHC, they found that the OHC has fast somatic electromotility and positive or negative changes in the membrane potential producing contraction or elongation respectively. Dallos (2003) explained that because the OHC hair bundle is attached to the overlying tectorial membrane, the movement of the OHC will increase the mechanical input to the IHC.
3.2 Coding of the Frequency of the 4 kHz 10 dB Sound
Through literature reviews of sound encoding, we found that there are two streams of arguments. Some believe that the frequency of a sound is coded by the place of the maximum BM vibration (“travelling wave theory”), while others argue that it is the phase locking property of the IHC that assist the coding. We believe that both of the above are important to the coding of our 4 kHz sound. As we have already discussed the travelling wave theory findings by Bekesy (1960), we will focus our discussion on the phase locking of the IHC and the auditory nerves.
Most experiments on the phase locking properties of the auditory nervous system had been based on the cochlear nerve fibers and ample evidences showed that the cochlear nerve fibers respond to sounds with precise such property. Relatively few researches had been done on inner hair cells (IHC); however, Palmer and Russell (1986)’s research on guinea pigs implied that the starting point of phase locking begins at the IHC.
Rose, Brugge, Anderson and Hind (1967) demonstrated phase locking in the cochlear nerve fibers of the squirrel monkey to sounds with frequencies up to at least 4.5 kHz to 5 kHz with less prominence above around 2.5 kHz. Similar findings had been found in cats for frequencies below 5 kHz (Kiang et al., 1965, as cited in Johnson, 1980) and in normal chinchillas for frequencies between 1.2 and 4 kHz (Woolf, Ryan & Bone, 1981).
We believe it is reasonable to postulate that phase locking of the IHCs and cochlear nerve fibers can be used for temporally coding our 4 kHz sound in the cochlea for the next stage.
3.3 Coding of the Intensity of the 4 kHz 10dB Sound
The intensity coding of our 10dB sound is less clear compared to frequency coding due to the mechanical properties of the cochlea and the complex interaction between frequency and intensity. For a given cochlear nerve fiber or neuron, same spike rate can be elicited by either a low-intensity sound at the optimal frequency, or a high-intensity sound at a non-optimal frequency. Thus, the spike rate tells us little information about either frequency or intensity of a sound (Covey, 2000). Moreover, frequency and intensity of a sound interactively affects the position of maximum BM vibration (Covey, 2000). More recent theory, such as the “population coding model”, may provide us with insight in this matter. We will further explore this model when after our discussion on the primary auditory cortex.
4. Auditory Cortex
In reality, our 4 kHz 10dB sound will travel from the inner ear to the brain through a complex network of auditory pathways. In the following discussion, we will focus on the analysis of our sound in primary cortex. Recent researches support for the tonotopic organization for the encoding of the frequency of a sound and we believe that this is reasonable and can apply to our sound. On the other hand, the coding of the intensity is still unclear and how this attribute is represented in the primary auditory cortex. We will explore the possibility of resolving this problem by looking into the concept of Population coding.
4.1 Coding of the Frequency through Tonotopic Organization
The tonotopic organization of the auditory nervous system towards pure tone stimuli with different intensities had been widely investigated by various researches using different techniques including the fMRI and PET scan. These researches had consistently pointed to the fact that a 4 kHz stimulus with different intensities maximally activated a more medial and posterior position on the transverse temporal gyrus (TTG) compared to low frequency stimuli (Lauter, Herscovitch, Formby & Raichie, 1985; Lockwood et al., 1999; Strainer et al., 1997; Yetkin, Roland, Christensen & Purdy , 2004). Out of these research findings, we believe that Lockwood et al. (1999) and Yetkin et al. (2004) provide the best insights for our 4 kHz 10dB sound. Specifically, Lockwood et al. (1999) had included 4 kHz stimuli as low as 30dBHL with PET scan and Yetkin et al. (2004), using silent fMRI, was able to investigate the area of maximum activation for 4kHz at an intensity as low as 20dBHL.
However, there are fewer consensuses on the extent of TTG activation for our 4 kHz sound. Some researchers found that 4 kHz stimulus was more potent in activating the primary auditory cortex (Lockwood et al., 1999). However, other researches, including Jacobson et al, (1992), Strainer et al. (1997) and Bilecen, Scheffler, Schmid, Tschopp and Seelig, (1998), found that TTG activation is lower for high frequency tones.
4.2 Coding of the intensity through ampitopicity
We know little about how our soft sound is analyzed in central auditory nervous system but it is generally believed that there is a decrease in activation volume with decreasing intensities but controversy exists, and the change in exact volume is unknown. Concerning the coding of the intensity, many of the recent researches had raised the idea of “amplitopicity” in human auditory cortex, showing that there is a particular pattern of activation for the encoding of intensity. We believe that this can explain how the intensity of our 4 kHz is coded. However, the exact pattern of activation specific to our sound needs to be further investigated.
The amplitopicity nature of the auditory system had been suggested by various researches. For example, both Pantev, Hoke, Lehnertz and Lutkenhoner (1988) and Bilecen, Seifritz, Scheffler, Henning and Schelte (2002) revealed an ampliotopic arrangement for 1 kHz sound to various intensities, though the exact location and direction of activation cannot be concluded from their findings. Specific to our 4 kHz 10 dB sound, Lockwood et al. (1999) found evidence for ampliotopic change in the medial-inferior part of the auditory cortex, showing that 4 kHz tones with low intensity (30 dBHL) had their maximum activation at a site 8 mm below the commissural plane, but the site rose to the plane level with increasing intensity.
Although in our discussion, we write as if the auditory system encode the frequency and the intensity of our sound in a separate fashion, current researches shed light on the interaction of both factors. Tanji et al. (2010) found a tonotopic organization dependent on the intensity level in the auditory system of unanesthetized monkey. They did not mention the exact area of activation of the 4 kHz tone, but it does provide suggestions that intensity level does affect the tonotopic organization, echoing the mechanism described in the inner ear.
Applying the Population Coding Model
We feel that Hanekom’s (1999) population coding model may be a possible solution to explain the effect of the interaction of intensity and frequency of our 4 kHz 10 dB sound from the inner ear onwards to the auditory cortex. In population coding, each neuron has a wide range of responses over the sound stimuli, and the responses of these neurons may be combined to determine a certain property of the sound. We feel that this can be applicable to the auditory system because Covey (2000) showed that population coding exists in the auditory system of bats. Therefore, both frequency and intensity information of a sound will be coded by a population of neurons at the same time. We believe that the population coding model paves the way for recent models such pattern matching and autocorrelation in the auditory system to explain the formation of perception for our sound (Cheveigne, 2010).
Conclusion
From our discussion, we see that various components along the auditory pathway including the outer ear, middle ear, inner ear and our brain, work together to provide an integrated representation of our 4 kHz 10 dB sound. Of most importance to our 4 kHz 10 dB sound, the inner ear plays the chief role in the process in the peripheral auditory system. The OHC in the cochlea is particularly crucial for the sound to further travel from the inner ear to the brain through a complex network of auditory pathways. How exactly the frequency and intensity are coded in the inner ear and the auditory system is not yet resolved and there is no authoritative explanation of how these are integrated in the auditory system. The best that we can say is that the tonotopic organization and the amplitopicity of the auditory system will likely contribute to the analysis of our 4 kHz 10 dB sound. Population coding may provide a good direction for further researches for our understanding of this process, a fascinating field where there is still much to be learned.