Neural Tuning Curves Research Articles

The auditory cortex of the bat Molossus molossus: Disproportionate search call frequency representation

The extent of the auditory cortex in the bat Molossus molossus was electrophysiologically investigated. Best frequencies and minimum thresholds of neural tuning curves were analyzed to define the topography of the auditory cortex. The auditory cortex encompasses an average cortical surface area of 5 mm 2. Characteristic frequencies are tonotopically organized with low frequencies being represented caudally and high frequencies rostrally. However, a large interindividual variability in the tonotopic organization was found. In most animals, the caudal 50% was tonotopically organized. More anterior, a variable area was found. A distinct field with reversed topography was not consistently found. Within the demarcated auditory cortex, frequencies of 30–40 kHz, which correspond to the frequency range of search calls emitted during hunting, are overrepresented, occupying 49% of the auditory cortex surface. High minimum thresholds >50 dB SPL were found in a narrow dorsal narrow area. Neurons with multipeaked tuning curves (20%) preferentially were located in the dorsal part of the auditory cortex. In accordance with studies in other bat species, the auditory cortex of M. molossus is highly sensitive to the dominant frequencies of biosonar search calls.

A Map for Horizontal Disparity in Monkey V2

The perception of visual depth is determined by integration of spatial disparities of inputs from the two eyes. Single cells in visual cortex of monkeys are known to respond to specific binocular disparities; however, little is known about their functional organization. We now show, using intrinsic signal optical imaging and single-unit physiology, that, in the thick stripe compartments of the second visual area (V2), there is a clustered organization of Near cells and Far cells, and moreover, there are topographic maps for Near to Far disparities within V2. Our findings suggest that maps for visual disparity are calculated in V2, and demonstrate parallels in functional organization between the thin, pale, and thick stripes of V2.

Effects of activation of the efferent system on psychophysical tuning curves as a function of signal frequency

It has been shown that electrical stimulation of the efferent auditory system can influence neural tuning curves in animals. Here, we examined a psychophysical analog of this effect in humans. All of the 19 normally hearing subjects showed a reduction in the amplitude of otoacoustic emissions in one ear when contralateral broadband noise was presented, indicating a functioning efferent system. Psychophysical tuning curves (PTCs) were measured in simultaneous masking in the absence and presence of contralateral stimulation (CS). The CS was a continuous narrowband noise centered at the signal frequency and presented at a level of 50 or 60 dB SL. The CS had no consistent effect on the masker level at the tips of the PTCs. For the two highest signal frequencies (2000 and 4000 Hz), the CS reduced the masker level required for threshold on both the low- and high-frequency sides of the PTCs, and the sharpness of tuning, as measured by Q10, decreased significantly. For the two lowest signal frequencies (500 and 1000 Hz), the masker level required for threshold on the low-frequency sides of the PTCs increased with CS, and the Q10 values increased significantly. The general pattern of the results was consistent with that observed for electrical stimulation of the efferent system in animals.

Properties of distortion product otoacoustic emissions and neural suppression tuning curves attributable to the tectorial membrane resonance

Mechanically coupled cochlear structures are likely to form a resonator with several degrees of freedom. Consequently one can expect complex, frequency-dependent relative movements between these structures, particularly between the tectorial membrane and reticular lamina. Shearing movement between these two structures excites the cochlear receptors. This excitation should be minimal at the frequency of the hypothesized tectorial membrane resonance. In each preparation, simultaneous masking neural tuning curves and distortion product otoacoustic emissions were recorded. The position of the low-frequency minima in the tuning curves, frequency dependence of the emission bandpass structure, and level-dependent phase reversal were compared to determine if they were generated by a common phenomenon, for example the tectorial membrane resonance. The notch in the masking curves and the phase inversion of the emission growth functions at the auditory thresholds are both situated half an octave below the probe frequency and the high-frequency primary, respectively, and show similar frequency dependence. The emission bandpass structure is, however, likely to be generated by a combination of mechanisms with different ones dominating at different stimulus parameters.

A deafness mutation isolates a second role for the tectorial membrane in hearing

Alpha-tectorin (encoded by Tecta) is a component of the tectorial membrane, an extracellular matrix of the cochlea. In humans, the Y1870C missense mutation in TECTA causes a 50- to 80-dB hearing loss. In transgenic mice with the Y1870C mutation in Tecta, the tectorial membrane's matrix structure is disrupted, and its adhesion zone is reduced in thickness. These abnormalities do not seriously influence the tectorial membrane's known role in ensuring that cochlear feedback is optimal, because the sensitivity and frequency tuning of the mechanical responses of the cochlea are little changed. However, neural thresholds are elevated, neural tuning is broadened, and a sharp decrease in sensitivity is seen at the tip of the neural tuning curve. Thus, using Tecta(Y1870C/+) mice, we have genetically isolated a second major role for the tectorial membrane in hearing: it enables the motion of the basilar membrane to optimally drive the inner hair cells at their best frequency.

Comparison between distortion product otoacoustic emissions and nerve fiber responses from the basilar papilla of the frog

The basilar papilla (BP) is one of the three end organs in the frog inner ear that is sensitive to airborne sound. Its anatomy and physiology are unique among all classes of vertebrates. Essentially, the BP functions as a single auditory filter presumably arising from a mechanically-tuned mechanism. As such, both neural and distortion product otoacoustic emission (DPOAE) tuning may reflect a single mechanical filtering mechanism. Using the Duffing oscillator as a simple model for both neural and DPOAE tuning from the BP, two predictions can be made: [1] the characteristic frequency (CF) of neural tuning and the best frequency (BF) of DPOAE tuning will coincide and [2] the neural tuning curve and DPOAE-audiogram have a similar shape when the neural tuning curve is scaled by a factor of 4 along the y-axis. We recorded both neural tuning curves and DPOAE-audiograms from the BP of the leopard frog. These recordings show good agreement with the model predictions when the stimulus tones are related by relatively small stimulus frequency ratios. For larger stimulus frequency ratios, DPOAE recordings clearly deviate from model predictions. These differences are most likely caused by the oversimplified representation of the frog BP by the model.

Modeling the influence of the cochlear nonlinearity on estimates of psychophysical tuning

Most behavioral measures of human frequency selectivity have been made with simultaneous masking and the notched-noise technique. The resulting filter shapes may be influenced by the effects of cochlear nonlinearity, such as suppression. Forward masking may provide a measure that is more comparable to neural tuning curves, because it does not involve stimuli that interact with each other along the basilar membrane. This study investigated the extent to which cochlear nonlinearities can account for differences in results between forward and simultaneous masking. The model was constructed using a nonlinear auditory filter, a sliding temporal integrator, a logarithmic transform and a template mechanism. The effects of compression and suppression on psychophysical performance were simulated by varying the relevant parameters of the model auditory filter. The psychophysical results were simulated for both forward and simultaneous masking, using the same parameters and tracking procedure as in the behavioral studies. The results provide a detailed evaluation of the role of compression and suppression in the models predictions of psychophysical tuning and assist in the development of the refined nonlinear cochlear models for human. [Work supported by the ASA Hunt Fellowship and NIH R01DC03909.]

The effect of hair bundle shape on hair bundle hydrodynamics of non-mammalian inner ear hair cells for the full frequency range

The effect of the size and the shape of the hair bundle of a hair cell in the inner ear of non-mammals on its motion for the full range of frequencies is determined thereby extending the results of a previous analysis of hair bundle motion for high and low frequencies [Hear Res. 141 (2000) 39–50]. A hemispheroid is used to represent the hair bundle because it can represent a full range of shapes, from thin, pencil-like shapes to wide, flat, disk-like shapes. Boundary element methods are used to approximate the solution for the hydrodynamics. For physiologically relevant parameters, an excellent match is obtained between the model's predictions and measurements of hair bundle motion in the free-standing region of the basilar papilla of the alligator lizard [Aranyosi, Measuring sound-induced motions of the alligator lizard cochlea. Massachusetts Institute of Technology, PhD Thesis, 2002]. Neither in the model's predictions nor in experimental measurements is sharp tuning observed. The model predicted the low frequency region of neural tuning curves for the alligator lizard and bobtail lizard, but could not predict the sharp tuning or the high frequency region. An element that represents an active mechanism is added to the hair bundle model to predict neural tuning curves, which are sharply tuned, and an excellent match is obtained for all the characteristics of neural tuning curves for the alligator lizard, and for the low and high frequency regions for the bobtail lizard. The model does not predict well the sharp tuning of the shorter hair bundles of the bobtail lizard, possibly because it does not represent tectorial sallets.

Relating psychophysical and otoacoustic emission suppression tuning curves

Psychophysical tuning curves (PTCs) are a well established method for estimating frequency selectivity in humans. PTCs are closely related to the shape of neural tuning curves. Furthermore, these two measurements are believed to represent frequency resolution processes primarily occurring in the cochlea. More recently, otoacoustic emission suppression tuning curves (STCs) have been measured and found to be characteristically similar in shape and tip frequency to PTCs (Abdala et al., 1996). STCs may represent an objective, noninvasive, measure of initial stages of cochlear tuning. The goal of this study was to compare and quantify the difference between transient evoked STCs (TE-STC) and PTCs. TE-STCs and PTCs were obtained in each normal-hearing subject with 4000-Hz probe tones. Preliminary data show that Q10 values and tip levels follow the same trend for both tuning curves. However, actual values of these parameters differ. Implications of methodological differences between these two paradigms will be discussed.

Reorganization of the auditory cortex specialized for echo-delay processing in the mustached bat.

Focal excess sensory stimulation evokes reorganization of a sensory system. It is usually an expansion of the neural representation of that stimulus resulting from the shifts of the tuning curves (receptive fields) of neurons toward those of the stimulated neurons. The auditory cortex of the mustached bat has an area that is highly specialized for the processing of target-distance information carried by echo delays. In this area, however, reorganization is due to shifts of the delay-tuning curves of neurons away from those of the stimulated cortical neurons. Elimination of inhibition in the target-distance processing area in the auditory cortex by a drug reverses the direction of the shifts in neural tuning curves. Therefore, such unique reorganization in the time domain is due to strong lateral inhibition in the highly specialized area of the auditory cortex.

Coding with noisy neurons: stability of tuning curve estimation strongly depends on the analysis method

An important issue in the neurosciences is a quantitative description of the relation between sensory stimuli presented to an animal and their representations in the nervous system. A standard technique is the construction of a neural tuning curve, that is, a neuron’s average firing rate as a function of some parameter characterizing a family of stimuli. It is unavoidable that some of the response data are erroneously attributed to a cell, e.g., during spike sorting. However, the widely used method of statistical analysis based on the sample mean and least-squares approximation for the spike count can perform extremely badly if the noise distribution is not exactly normal, which is almost never the case in applications. Here, we present a method for constructing neural tuning curves that is especially suited for cases of high noise and the presence of outliers. Since it is usually not decidable if an outlier is faulty or not we limit the influence of far outlying points rather than try to identify and discard them. In contrast to traditional methods employing a point-by-point estimation of a tuning curve, we use all measured data from all different stimulus conditions at once in the construction. Given the measured data at only a finite number of stimulus conditions, a robust tuning curve is obtained that approximates the cell’s ideal tuning curve optimally in all stimulus conditions with respect to a given distance measure. A measure that assesses the quality of this fitting method with respect to the traditional least-squares fitting method and to a median-based fitting method is introduced. The reliability of inference with respect to the encoding accuracy that can be achieved by a population of neurons is demonstrated in both artificially generated and experimentally recorded data from rat primary visual cortex. While the data shown in this paper are responses to orientation stimuli, the method of tuning curve construction is also viable and maintains its optimality properties for the case in which the stimulus is defined on a finite interval.

Neural and perceptual discrimination of the spectral and temporal modulations in birdsong and speech

My laboratory is interested in the neural basis of complex sound perception, including vocalizations. Our neural studies have focused on the high-level auditory system of songbirds where neurons respond preferentially to birdsong. We quantified the selectivity of these auditory neurons by recording their responses to degraded versions of song. To do so, we added noise to this natural stimulus, affecting the time-frequency structure of the sound at different scales. We found that at particular time-frequency scales the noise has little effect on the response of the neurons, while at other scales it greatly reduces the neural response. We correlated the tuning of the neurons with the statistical structure in the birdsongs that allows the discrimination of songs produced by different birds. We found a good match between the optimal scale of the neurons sensitivity curve and the scale that best captures the variance in the acoustical structure of an ensemble of songs. We have done a similar analysis with speech, where the neural tuning curve in songbirds is replaced by a human perceptual tuning curve based on speech intelligibility. As in songbirds, the scale for optimal speech perception is also best for capturing the statistical structure in the speech signal.

Suppression tuning in noise-exposed rabbits.

Psychophysical, basilar-membrane (BM), and single nerve-fiber tuning curves, as well as suppression of distortion-product otoacoustic emissions (DPOAEs), all give rise to frequency tuning patterns with stereotypical features. Similarities and differences between the behaviors of these tuning functions, both in normal conditions and following various cochlear insults, have been documented. While neural tuning curves (NTCs) and BM tuning curves behave similarly both before and after cochlear insults known to disrupt frequency selectivity, DPOAE suppression tuning curves (STCs) do not necessarily mirror these responses following either administration of ototoxins [Martin et al., J. Acoust. Soc. Am. 104, 972-983 (1998)] or exposure to temporarily damaging noise [Howard et al., J. Acoust. Soc. Am. 111, 285-296 (2002)]. However, changes in STC parameters may be predictive of other changes in cochlear function such as cochlear immaturity in neonatal humans [Abdala, Hear. Res. 121, 125-138 (1998)]. To determine the effects of noise-induced permanent auditory dysfunction on STC parameters, rabbits were exposed to high-level noise that led to permanent reductions in DPOAE level, and comparisons between pre- and postexposure DPOAE levels and STCs were made. Statistical comparisons of pre- and postexposure STC values at CF revealed consistent basal shifts in the frequency region of greatest cochlear damage, whereas thresholds, Q10dB, and tip-to-tail gain values were not reliably altered. Additionally, a large percentage of high-frequency lobes associated with third tone interference phenomena, that were exhibited in some data sets, were dramatically reduced following noise exposure. Thus, previously described areas of DPOAE interference above f2 may also be studied using this type of experimental manipulation [Martin et al., Hear. Res. 136, 105-123 (1999); Mills, J. Acoust. Soc. Am. 107, 2586-2602 (2002)].

Effects of reversible noise exposure on the suppression tuning of rabbit distortion-product otoacoustic emissions.

Distortion-product otoacoustic emissions (DPOAEs) at 2f1-f2 can be suppressed by the introduction of a third "suppressor" tone. Plotting the suppression of the DPOAE level against the changing frequency and level of the suppressor produces frequency-tuning functions referred to as suppression tuning curves (STCs). The dominant features of STCs, including their shape, are similar to the features of neural tuning curves (NTCs) recorded from single auditory nerve fibers. However, recent findings using reversible diuretics suggest that STCs do not provide the same measure of cochlear frequency selectivity as provided by NTCs. To determine if STCs are also insensitive to the adverse effects of excessive sounds, the present study exposed rabbits to a moderate-level noise that produced temporary threshold shift-like (TTS) effects on DPOAEs, and examined the influence of such exposures on STCs. DPOAEs were produced using primary tones with geometric-mean frequencies centered at 2.8 or 4 kHz, and with L1 and L2 values of 45/45, 50/35, 50/50, and 55/45 dB SPL. STCs were obtained before and during recovery for a period of approximately 2 h immediately following, and at 1, 2, 3, and 7 d post-exposure to a 2 kHz octave band noise, at levels and durations sufficient to cause significant but reversible reductions in DPOAE levels. STC data included tip center frequency, tip threshold, and Q10dB measures of tuning for suppression criteria of 3, 6, 9, and 12 dB. Recovery was variable between animals, but all rabbits recovered fully by 7 d post-exposure. STC center frequencies measured during the TTS typically tuned to a slightly higher frequency, while tip thresholds tended to decrease and Q10dB increase. Together, the results indicate that, despite similarities in the general properties of STCs and NTCs, these two types of tuning curves are affected differently following reversible cochlear insult.

Attention modulation of neural tuning through peak and base rate.

This study investigates the influence of attention modulation on neural tuning functions. It has been shown in experiments that attention modulation alters neural tuning curves. Attention has been considered at least to serve to resolve limiting capacities and to increase the sensitivity to attended stimulus, while the exact functions of attention are still under debate. Inspired by recent experimental results on attention modulation, we investigate the influence of changes in the height and base rate of the tuning curve on the encoding accuracy, using the Fisher information. Under an assumption of stimulus-conditional independence of neural responses, we derive explicit conditions that determine when the height and base rate should be increased or decreased to improve encoding accuracy. Notably, a decrease in the tuning height and base rate can improve the encoding accuracy in some cases. Our theoretical results can predict the effective size of attention modulation on the neural population with respect to encoding accuracy. We discuss how our method can be used quantitatively to evaluate different aspects of attention function.

The effect of shape on the hydrodynamics of a hemispheroid projecting from a plate in irrotational fluid

In this paper, we examine the problem of hemispheroid that is compliantly hinged to a plate oscillating with infinitesimal motions in fluid that can be modeled as irrotational. We use this model to understand the effect of the size and shape of hair bundles of sensory cells in the inner ear on hair bundle hydrodynamics at high frequencies. In response to the oscillating plate, the hemispheroid translates and rotates in the plane of the oscillation. Since the equations of motion are linear, the solution can be described as the superposition of the response due to translation of the hemispheroid and the plate, and the response due to rotation of the hemispheroid. The solution can be found by solving Laplace’s equation in prolate and oblate spheroidal coordinates. The response due to translational motion is the same as that derived in Hydrodynamics, 6th ed. (Dover, New York, 1945) for a full spheroid undergoing translational motion. The response due to rotational motion of the hemispheroid about its hinge has not been previously derived. The hydrodynamic pressure, torque, and drag on the hemispheroid as a function of hemispheroidal shape are presented. A good match was obtained for results of the model and measurements of a neural tuning curve of the alligator lizard.

Marc D. Hauser (1996). The evolution of communication. Cambridge, Mass.: MIT Press. Pp. xiii+760.

This volume impressively synthesises vast literatures from the fields of linguistics, bioacoustics, psychology, neurobiology, evolutionary biology, ethology and anthropology, and in the process raises a number of provocative questions regarding the contentious issue of human language origins. Because the book is so far-reaching, both in terms of the breadth of communicative phenomena which it covers and the depth in which it discusses them, a short review such as the present one can only scratch the surface of the wealth of information and ideas which it contains.This book was written to fill a perceived need for a text covering a wide range of issues in comparative communication, for which it is certainly well suited. Those interested in the production and perception of auditory and visual signals, as well as in issues as diverse as evolutionary biology and cognitive psychology, will find it an easily readable – or browseable – piece of work. As Hauser notes, he has ‘attempted to write a book that is aimed primarily at the expert while being useful to those wishing to pick out pieces...for undergraduate and graduate instruction’ (p. 14). The book is successful along both lines. It is extremely well organised and well indexed, making it easy to select case studies relevant to specific communicative phenomena (e.g. audition, vocalisation, acquisition, signed languages, etc.) or particular species (humans, monkeys, anurans, birds, etc.). Particularly useful are the large number of graphics and illustrations, as well as conceptual ‘boxes’ which succinctly summarise key concepts or theoretical perspectives which may be unfamiliar to some readers (e.g. statistical information theory, neural tuning curves, source-filter theory and sexual selection theory, to name just a few).

Changes in the tonotopic map of the dorsal cochlear nucleus in hamsters with hair cell loss and radial nerve bundle degeneration

Hamsters were exposed to an intense tone (10 kHz) at levels and durations sufficient to cause hair cell loss and radial nerve bundle degeneration. A previous study reported changes in the tonotopic map of the dorsal cochlear nucleus (DCN) in hamsters with tone-induced stereocilia loss. Such changes appear similar to those observed by others in the auditory nerve following acoustic trauma, and suggest that the map alterations have a peripheral origin. However, the potential for tonotopic map reorganization after more severe lesions involving cellular degeneration in the cochlea has not yet been determined. The purpose of the present study was to determine how the tonotopic map of the DCN appears in animals with severe cochlear injury involving hair cell loss and radial nerve bundle degeneration. Neural population thresholds and tonotopic organization were mapped over the surface of the DCN in normal unexposed animals and those showing tone-induced lesions. The results indicate that cochlear lesions characterized mainly by radial bundle degeneration in a restricted portion of the organ of Corti cause changes in a corresponding region of the tonotopic map which reflect primarily changes in the shape and thresholds of neural tuning curves. In many cases the center of the lesion was represented in the DCN as a distinct characteristic frequency (CF) gap in the tonotopic map in which responses were either extremely weak or absent. In almost all cases the map area representing the center of the lesion was bordered by an expanded region of near-constant CF, a feature superficially suggestive of map reorganization (i.e., plasticity). However, these expanded map areas had abnormal tip thresholds and showed other features suggesting that their CFs had been shifted downward by distortion and deterioration of their original tips. Such changes in neural tuning following tone-induced loss of anatomical input to the central auditory pathway are similar to those observed in our previous study and by others in the auditory nerve following less severe acoustic trauma, and thus would seem to have a peripheral origin. Thus, changes in the DCN tonotopic map can be explained by peripheral modifications and do not seem to involve plastic changes (i.e., reorganization).

Evaluating human neural tuning curves from a mechanical model of the cochlea by relating them to psychophysical masking data

Human neural threshold tuning curves are estimated by scaling the parameters of Allen’s [J. Acoust. Soc. Am. 68, 1660–1670 (1980)] resonant tectorial membrane model of the cat cochlea. A way to evaluate the derived tuning curves using psychophysical data is developed, based on a psychophysical detection model which relates the physiological tuning curves to psychophysical masking data. A detection criterion, defined by a relationship among the bandwidth of the frequency tuning curves, expressed as an equivalent rectangular bandwidth (ERB), the width of the excitation patterns, expressed as an equivalent rectangular spread (ERS), and the psychophysical critical ratio, is explored and verified using cat data. The detection criterion is then used to test the derived human curves by making predictions of psychophysical masking and comparing the predictions to experimental data. The detection model may also provide a deeper understanding of the frequency resolving properties of the cochlea.

Deriving human neural tuning curves from a mechanical model of the cochlea

Human neural threshold tuning curves are derived using Allen’s [J. Acoust. Soc. Am. 68, 1660–1670 (1980)] resonant tectorial membrane model of the cat cochlea by appropriately scaling the parameters. Because there is no measured neural threshold data available for the human to test the derived curves, a detection criterion, defined by a relationship among the bandwidth of the frequency tuning curves, expressed as an equivalent rectangular bandwidth (ERB), the width of the excitation patterns, expressed as an equivalent rectangular spread (ERS), and the psychophysical critical ratio, is explored and verified using cat data. The detection criterion can then be used to test the derived human curves using psychophysical data, and may also provide a deeper understanding of the frequency resolving properties of the cochlea.