Peter Heil

Functional Organization of Auditory Cortex in the Mongolian Gerbil (Meriones unguiculatus). I. Electrophysiological Mapping of Frequency Representation and Distinction of Fields

The frequency representation within the auditory cortex of the anaesthetized Mongolian gerbil (Meriones unguiculatus) was studied using standard microelectrode (essentially multiunit) mapping techniques. A large tonotopically organized primary auditory field (AI) was identified. High best frequencies (BFs) were represented rostrally and low BFs caudally along roughly dorsoventrally oriented isofrequency contours. Additional tonotopic representations were found adjacent to AI. Rostral to AI was a smaller field with a complete tonotopic gradient reversed with respect to that in AI (mirror image representation) and was termed the anterior auditory field (AAF). BFs in the range from 0.1 to 43 kHz, apparently covering the hearing range of the Mongolian gerbil, were found in AI and AAF. Units in these two core fields responded to narrow frequency ranges with short latencies. Ventral to the common high‐frequency border to AAF and AI, a rapid transition to very low BFs suggested the presence of a ventral field (V). Caudal to AI two small tonotopically organized fields were identified, a dorsoposterior field (DP) and a ventroposterior field (VP). The VP showed a tonotopic organization mirror imaged to that of AI, i.e. low frequencies were represented rostrally near the caudal border of AI, and high frequencies caudally. The DP showed a concentric frequency organization with high BFs located in the centre. Units in DP and VP fired less strongly, with considerably longer latencies, and responded to a broader range of frequencies than units in AI and AAF. Dorsocaudal to AI a dorsal field (D) was identified, harbouring units that responded to very broad ranges of frequencies. A tonotopic organization of field D could not be discerned. In the border region of AI and D, low‐frequency responses were similar to those found in parts of AI and AAF, but without a clear‐cut tonotopic organization. This region was termed Aid. The two core fields AI and AAF appeared to be located within the koniocortex, while the remaining fields lay outside. Our data show that the organization of the gerbil auditory cortex is highly elaborate, with parcellation into fields as complex as in cat or primates.

Sensitivity of neurons in cat primary auditory cortex to tones and frequency-modulated stimuli. II: Organization of response properties along the ‘isofrequency’ dimension

The spatial distribution of neuronal responses to tones and frequency-modulated (FM) stimuli was mapped along the isofrequency dimension of the primary auditory cortex (AI) of barbiturate-anesthetized cats. In each cat, electrode penetrations roughly orthogonal to the cortical surface were closely spaced (average separation approximately 130 microns) along the dorsoventral extent of a single isofrequency strip in high frequency parts of AI (> 15 kHz). Characteristic frequency (CF), minimum threshold, sharpness of frequency tuning (Q10 and Q20), the dynamic range of the spike count-intensity function at CF, sensitivity to the rate of change of frequency (RCF) and to the direction of frequency-modulation (DS) were determined for contralaterally-presented tone and FM stimuli. Sharpness of tuning attained maximum values at central loci along the dorsoventral isofrequency axis and values declined towards more dorsal and more ventral locations. Minimum threshold and dynamic range varied between high and low values in a similar and correlated periodic fashion. Their combined organization yielded an orderly spatial representation of response strength, relative to maximum, as a function of stimulus amplitude. The distributions of the most common forms of FM rate sensitivity (RCF response categories) and best RCF along isofrequency strips were significantly non-random although there was a considerable degree of variability between cats. FM directional preference and sensitivity appeared to be randomly distributed. Sharpness of tuning may be related to the analysis of the spectral content of an acoustic stimulus, both minimum threshold and dynamic range are related to the encoding of stimulus intensity, and measures of FM rate and directional sensitivity assess the coding of temporal changes of stimulus spectra. The independent, or for minimum threshold and dynamic range dependent, topographic organizations of these neuronal parameters therefore suggest parallel and independent processing of these aspects of acoustic signals in AI.

Topographic representation of tone intensity along the isofrequency axis of cat primary auditory cortex

The sound pressure level (SPL), henceforth termed intensity, of acoustic signals is encoded in the central auditory system by neurons with different forms of intensity sensitivity. However, knowledge about the topographic organization of neurons with these different properties and hence about the spatial representation of intensity, especially at higher levels of the auditory pathway, is limited. Here we show that in the tonotopically organized primary auditory cortex (AI) of the cat there are orderly topographic organizations, along the isofrequency axis, of several neuronal properties related to the coding of the intensity of tones, viz. minimum threshold, dynamic range, best SPL, and non-monotonicity of spike count--intensity functions to tones of characteristic frequency (CF). Minimum threshold, dynamic range, and best SPL are correlated and alter periodically along isofrequency strips. The steepness of the high-intensity descending slope of spike count--intensity functions also varies systematically, with steepest slopes occurring in the regions along an isofrequency strip where low thresholds, narrow dynamic ranges and low best SPLs are found. As a consequence, CF-tones of various intensities are represented by orderly and, for most intensities, periodic, spatial patterns of distributed neuronal activity along an isofrequency strip. For low--to--moderate intensities, the mean relative activity along the entire isofrequency strip increases rapidly with intensity, with the spatial pattern of activity remaining quite constant along the strip. At higher intensities, however, the mean relative activity along the strip remains fairly constant with changes in intensity, but the spatial patterns change markedly. As a consequence of these effects, low- and high-intensity tones are represented by complementary distributions of activity alternating along an isofrequency strip. We conclude that in AI tone intensity is represented by two complementary modes, viz. discharge rate and place. Furthermore, the magnitude of the overall changes in the representation of tone intensity in AI appears to be closely related to psychophysical measures of loudness and of intensity discrimination.

Sensitivity of neurons in cat primary auditory cortex to tones and frequency-modulated stimuli. I : Effects of variation of stimulus parameters

In the primary auditory cortex (AI) of barbiturate-anesthetized cats multi-unit responses to tones and to frequency-modulated (FM) tonal stimuli were analyzed. Characteristic frequency (CF), sharpness of tuning, minimum threshold, and dynamic range of spike count--intensity functions were determined. Minimum threshold and dynamic range were positively correlated. The response functions to unidirectional FM sweeps of varying linear rate of change of frequency (RCF) that traversed the excitatory frequency response areas (FRAs) displayed a variety of shapes. Preferences for fast RCFs (> 1000 kHz/s) were most common. Best RCF was not correlated with measures of sharpness of tuning. Directional preference and sensitivity were quantified by a DS index which varied with RCF. About two-thirds of the multi-unit responses showed a preference for downward sweeps. Directional sensitivity was independent of CF and independent of best RCF. Measurements of latencies of phasic responses to unidirectional FM sweeps of different RCF demonstrated that the discharges of a given multi-unit over its effective RCF range were initiated at the same instantaneous frequency (effective Fi), independent of RCF. Effective Fis fell within the excitatory FRA of a given multi-unit. The relationships of effective Fis to CF show that responses were evoked only when the frequency of the signal was modulated towards CF and not when modulated away from it, and that responses were initiated before the modulation reached CF. Changes in the range and depth of modulation had only minor, if any, effects on RCF response characteristics, FM directional sensitivity, and effective Fis, as long as the beginning and ending frequencies of FM sweeps fell outside a multi-units FRA. Stimulus intensity also had only moderate effects on RCF response characteristics and DS. However, effective Fis were influenced in systematic fashions; with increases in intensity, effective Fis to upward and downward sweeps decreased and increased, respectively. Thus, for higher intensities FM responses were initiated at instantaneous frequencies occurring earlier in the signal. The results are compared with previous data on tone and FM sensitivity of auditory neurons in cortical and subcortical structures, and mechanisms of FM rate and directional sensitivity are discussed. The topographic representations of these neuronal properties in AI are reported in the companion report.

A unifying basis of auditory thresholds based on temporal summation

Thresholds of auditory-nerve (AN) fibers and auditory neurons are commonly specified in terms of sound pressure only, implying that they are independent of time. At the perceptual level, however, the sound pressure required for detection decreases with increasing stimulus duration, suggesting that the auditory system integrates sound over time. The quantity commonly believed to be integrated is sound intensity, implying that the auditory system would have an energy threshold. However, leaky integrators of intensity with time constants of hundreds of milliseconds are required to fit the data. Such time constants are unknown in physiology and are also incompatible with the high temporal resolution of the auditory system, creating the resolution–integration paradox. Here we demonstrate that cortical and perceptual responses are based on integration of the pressure envelope of the sound, as we have previously shown for AN fibers, rather than on intensity. The functions relating the pressure envelope integration thresholds and time for AN fibers, cortical neurons, and perception in the same species (cat), as well as for perception in many different vertebrate species, are remarkably similar. They are well described by a power law that resolves the resolution–integration paradox. The data argue for the integrator to be located in the first synapse in the auditory pathway and we discuss its mode of operation.

Invasion of visual cortex by the auditory system in the naturally blind mole rat

PREVIOUSLY we have shown that the dorsal lateral geniculate body (LGB), which is strictly visual in sighted mammals, receives a strong auditory input in the naturally blind mole rat (Spalax ehrenbergi). Here we show with the 2-deoxyglucose technique and with single-unit recordings that in this species the initially non-degenerated visual cortex, as defined by its connection with LGB, is also activated by the auditory modality. These findings suggest that cross-modal compensation may occur as a natural consequence of the degeneration of a sense organ.

On determinants of first-spike latency in auditory cortex.

THE first-spike latency of neurones at any level of the auditory pathway decreases with stimulus amplitude. As stimuli are generally shaped with rise functions to avoid spectral splatter, a common interpretation of the latency decrease is that the amplitude of the signal reaches the neurones firing threshold earlier during the rise time. We demonstrate here, for auditory cortex neurones and by varying the amplitude and rise time of tonal stimuli, that this threshold model is inadequate to account for the observed latency changes, particularly when adaptive processes are taken into account. The data raise the possibility that latency may be a function of other properties associated with a signals onset, such as rate of change of peak pressure.

Functional organization of the avian auditory cortex analogue. I. Topographic representation of isointensity bandwidth.

Bandwidth of auditory units in the chick forebrain (field L/Hv complex) was measured with isointensity tone stimuli. Isointensity bandwidth is topographically represented within the four-layered tonotopically organized structure. It declines continuously from rostrodorsal to caudoventral along the longitudinal axis of two-dimensional best frequency planes (frequency band laminae). Layer-specific differences along the radial axis are also obvious. In the input layer of field L and in Hv ON-response bandwidths are relatively broad. The narrower bandwidths of units in the two postsynaptic layers of field L are probably caused by lateral inhibition mechanisms, as derived from the different topographic representations of OFF-versus ON-response bandwidths. A quantitative comparison of the topographic representation of bandwidth is made with the geometry of the tonotopic organization of the chick auditory forebrain complex, as revealed by 2-deoxyglucose data in a former study. A number of possible input-output transformations are derived from this comparison.

Spontaneous Activity of Auditory-Nerve Fibers: Insights into Stochastic Processes at Ribbon Synapses

In several sensory systems, the conversion of the representation of stimuli from graded membrane potentials into stochastic spike trains is performed by ribbon synapses. In the mammalian auditory system, the spiking characteristics of the vast majority of primary afferent auditory-nerve (AN) fibers are determined primarily by a single ribbon synapse in a single inner hair cell (IHC), and thus provide a unique window into the operation of the synapse. Here, we examine the distributions of interspike intervals (ISIs) of cat AN fibers under conditions when the IHC membrane potential can be considered constant and the processes generating AN fiber activity can be considered stationary, namely in the absence of auditory stimulation. Such spontaneous activity is commonly thought to result from an excitatory Poisson point process modified by the refractory properties of the fiber, but here we show that this cannot be the case. Rather, the ISI distributions are one to two orders of magnitude better and very accurately described as a result of a homogeneous stochastic process of excitation (transmitter release events) in which the distribution of interevent times is a mixture of an exponential and a gamma distribution with shape factor 2, both with the same scale parameter. Whereas the scale parameter varies across fibers, the proportions of exponentially and gamma distributed intervals in the mixture, and the refractory properties, can be considered constant. This suggests that all of the ribbon synapses operate in a similar manner, possibly just at different rates. Our findings also constitute an essential step toward a better understanding of the spike-train representation of time-varying stimuli initiated at this synapse, and thus of the fundamentals of temporal coding in the auditory pathway.

Field-specific responses in the auditory cortex of the unanaesthetized Mongolian gerbil to tones and slow frequency modulations

Abstract Responses of multi-units in the auditory cortex (AC) of unanaesthetized Mongolian gerbils to pure tones and to linearly frequency modulated (FM) sounds were analysed. Three types of responses to pure tones could be clearly distinguished on the basis of spectral tuning properties, response latencies and overall temporal response pattern. In response to FM sweeps these three types discharged in a temporal pattern similar to tone responses. However, for all type-1 units the latencies of some phasic response components shifted systematically as a function of range and/or speed of modulation. Measurements of response latencies to FMs revealed that such responses were evoked whenever the modulation reached a particular instantaneous frequency (Fi). Effective Fi was: (1) independent of modulation range and speed, (2) always reached before the modulation arrived at a local maximum of the frequency response function (FRF) and consequently differed for downward and upward sweeps, and (3) was correlated with the steepest slope of that FRF maximum. The three different types of units were found in discrete and separate fields or regions of the AC. It is concluded that gross temporal response properties are one of the key features distinguishing auditory cortical regions in the Mongolian gerbil.