A spectrotemporal analysis of DCN single unit responses to wideband noise in guinea pig

A temporal integration mechanism enhances frequency selectivity of broadband inputs to inferior colliculus

Accurately resolving frequency components in sounds is essential for sound recognition, yet there is little direct evidence for how frequency selectivity is preserved or newly created across auditory structures. We demonstrate that prepotentials (PPs) with physiological properties resembling presynaptic potentials from broadly tuned brainstem inputs can be recorded concurrently with postsynaptic action potentials in inferior colliculus (IC). These putative brainstem inputs (PBIs) are broadly tuned and exhibit delayed and spectrally interleaved excitation and inhibition not present in the simultaneously recorded IC neurons (ICNs). A sharpening of tuning is accomplished locally at the expense of spike-timing precision through nonlinear temporal integration of broadband inputs. A neuron model replicates the finding and demonstrates that temporal integration alone can degrade timing precision while enhancing frequency tuning through interference of spectrally in- and out-of-phase inputs. These findings suggest that, in contrast to current models that require local inhibition, frequency selectivity can be sharpened through temporal integration, thus supporting an alternative computational strategy to quickly refine frequency selectivity.

Nonlinear temporal receptive fields of neurons in the dorsal cochlear nucleus

Studies of the dorsal cochlear nucleus (DCN) have focused on spectral processing because of the complex spectral receptive fields of the DCN. However, temporal fluctuations in natural signals convey important information, including information about moving sound sources or movements of the external ear in animals like cats. Here, we investigate the temporal filtering properties of DCN principal neurons through the use of temporal weighting functions that allow flexible analysis of nonlinearities and time variation in temporal response properties. First-order temporal receptive fields derived from the neurons are sufficient to characterize their response properties to low-contrast (3-dB standard deviation) stimuli. Larger contrasts require the second-order terms. Allowing temporal variation of the parameters of the first-order model or adding a component representing refractoriness improves predictions by the model by relatively small amounts. The importance of second-order components of the model is shown through simulations of nonlinear envelope synchronization behavior across sound level. The temporal model can be combined with a spectral model to predict tuning to the speed and direction of moving sounds.

Sparse codes for speech predict spectrotemporal receptive fields in the inferior colliculus

We have developed a sparse mathematical representation of speech that minimizes the number of active model neurons needed to represent typical speech sounds. The model learns several well-known acoustic features of speech such as harmonic stacks, formants, onsets and terminations, but we also find more exotic structures in the spectrogram representation of sound such as localized checkerboard patterns and frequency-modulated excitatory subregions flanked by suppressive sidebands. Moreover, several of these novel features resemble neuronal receptive fields reported in the Inferior Colliculus (IC), as well as auditory thalamus and cortex, and our model neurons exhibit the same tradeoff in spectrotemporal resolution as has been observed in IC. To our knowledge, this is the first demonstration that receptive fields of neurons in the ascending mammalian auditory pathway beyond the auditory nerve can be predicted based on coding principles and the statistical properties of recorded sounds.

Context dependence of spectro-temporal receptive fields with implications for neural coding

The spectro-temporal receptive field (STRF) is frequently used to characterize the linear frequency-time filter properties of the auditory system up to the neuron recorded from. STRFs are extremely stimulus dependent, reflecting the strong non-linearities in the auditory system. Changes in the STRF with stimulus type (tonal, noise-like, vocalizations), sound level and spectro-temporal sound density are reviewed here. Effects on STRF shape of task and attention are also briefly reviewed. Models to account for these changes, potential improvements to STRF analysis, and implications for neural coding are discussed.

Multilinear models of single cell responses in the medial nucleus of the trapezoid body

The representation of acoustic stimuli in the brainstem forms the basis for higher auditory processing. While some characteristics of this representation (e.g. tuning curve) are widely accepted, it remains a challenge to predict the firing rate at high temporal resolution in response to complex stimuli. In this study we explore models for in vivo, single cell responses in the medial nucleus of the trapezoid body (MNTB) under complex sound stimulation. We estimate a family of models, the multilinear models, encompassing the classical spectrotemporal receptive field and allowing arbitrary input-nonlinearities and certain multiplicative interactions between sound energy and its short-term auditory context. We compare these to models of more traditional type, and also evaluate their performance under various stimulus representations. Using the context model, 75% of the explainable variance could be predicted based on a cochlear-like, gamma-tone stimulus representation. The presence of multiplicative contextual interactions strongly reduces certain inhibitory/suppressive regions of the linear kernels, suggesting an underlying nonlinear mechanism, e.g. cochlear or synaptic suppression, as the source of the suppression in MNTB neuronal responses. In conclusion, the context model provides a rich and still interpretable extension over many previous phenomenological models for modeling responses in the auditory brainstem at submillisecond resolution.

Robust spectrotemporal reverse correlation for the auditory system: optimizing stimulus design

The spectrotemporal receptive field (STRF) is a functional descriptor of the linear processing of time-varying acoustic spectra by the auditory system. By cross-correlating sustained neuronal activity with the dynamic spectrum of a spectrotemporally rich stimulus ensemble, one obtains an estimate of the STRF. In this article, the relationship between the spectrotemporal structure of any given stimulus and the quality of the STRF estimate is explored and exploited. Invoking the Fourier theorem, arbitrary dynamic spectra are described as sums of basic sinusoidal components--that is, moving ripples. Accurate estimation is found to be especially reliant on the prominence of components whose spectral and temporal characteristics are of relevance to the auditory locus under study and is sensitive to the phase relationships between components with identical temporal signatures. These and other observations have guided the development and use of stimuli with deterministic dynamic spectra composed of the superposition of many temporally orthogonal moving ripples having a restricted, relevant range of spectral scales and temporal rates. The method, termed sum-of-ripples, is similar in spirit to the white-noise approach but enjoys the same practical advantages--which equate to faster and more accurate estimation--attributable to the time-domain sum-of-sinusoids method previously employed in vision research. Application of the method is exemplified with both modeled data and experimental data from ferret primary auditory cortex (AI).

Linear and nonlinear spectral integration in type IV neurons of the dorsal cochlear nucleus. II. Predicting responses with the use of nonlinear models

Two nonlinear modeling methods were used to characterize the input/output relationships of type IV units, which are one principal cell type in the dorsal cochlear nucleus (DCN). In both cases, the goal was to derive predictive models, i.e., models that could predict the responses to other stimuli. In one method, frequency integration was estimated from response maps derived from single tones and simultaneous pairs of tones presented over a range of frequencies. This model combined linear integration of energy across frequency and nonlinear interactions of energy at different frequencies. The model was used to predict responses to noisebands with varying width and center frequency. In almost all cases, predictions using two-tone interactions were better than linear predictions based on single-tone responses only. In about half the cases, reasonable quantitative fits were achieved. The fits were best for noisebands with narrow bandwidth and low sound levels. In the second nonlinear method, the spectrotemporal receptive field (STRF) was derived from responses to broadband stimuli. The STRF could account for some qualitative features of the responses to broad noisebands and spectral notches embedded in broad noisebands. Quantitatively, however, the STRFs failed to predict the responses of type IV units even to simple broadband noise stimuli. For narrowband stimuli, the STRF failed to predict even qualitative features (such as excitatory and inhibitory frequency bands). The responses of DCN type IV units presumably result from interactions of two inhibitory sources, a strong one that is preferentially activated by narrowband stimuli and a weaker one that is preferentially activated by broadband stimuli. The results presented here suggest that the STRF measures effects related to the broadband inhibition, whereas two-tone interactions measure mostly effects related to narrowband inhibition. This explains why models based on two-tone interactions predict the responses to narrow noisebands much better then models based on STRFs. It is concluded that a minimal stimulus set for characterizing type IV units must contain both broadband and narrowband stimuli, because each stimulus class by itself activates only partially the integration mechanisms that shape the responses of type IV units. Similar conclusions are expected to hold in other parts of the auditory system: when characterizing a complex auditory unit, it is necessary to use a range of stimuli to ensure that all integration mechanisms are activated.

A statistically based method to generate response maps objectively

One scheme to classify the physiological response properties of single units in the cochlear nucleus is based on the average discharge rate of the unit and is reflected in the distribution of excitatory and inhibitory regions in a frequency-level map (response map) that spans the unit's receptive area (e.g., Evans and Nelson, 1973; Young and Brownell, 1976; Young and Voigt, 1982; Shofner and Young, 1985, Spirou and Young, 1991). Typically, discharge rate versus level curves are acquired at many frequencies and the investigator determines that a unit is excited or inhibited at a given level if the driven rate is above or below a spontaneous rate estimate by a specified criterion (for example, 20%). The investigator then encloses regions of excitation and inhibition where responses over adjacent frequencies and levels are consistent. In the present report, we describe an objective 3-step computer-based method to generate response maps: raw driven and spontaneous rate estimates are smoothed with a low-pass spatial filter; a unit is said to be excited or inhibited at a given level if the filtered driven rate is above or below the mean filtered spontaneous rate for that frequency by a specified criterion (percentage or statistical); and resultant response maps are median spatial filtered to eliminate spurious regions. The results shown here demonstrate that use of a statistical criterion provides a more reliable detection of excitation and inhibition than a 20% criterion, particularly when the variance of the rate estimates is high. Further, the statistically based method permits unit classification based on response map data that are more rapidly acquired with shorter duration stimuli (32 vs. 200 ms). Although this method is applied to units recorded in the dorsal cochlear nucleus, the technique may be applicable to studies of receptive fields and their plasticity in other systems.

Wiener and Volterra analyses applied to the auditory system

The application of a particular branch of non-linear system analysis, the functional series expansion or integral method, to the auditory system is reviewed. Both the Volterra and Wiener approach are discussed and an extension of the Wiener method from its traditional white-noise stimulus approach to that of Poisson distributed clicks is presented. This type of analysis has been applied to compound and single-unit responses from the auditory nerve, cochlear nucleus, auditory midbrain and medial geniculate body. Most studies have estimated only first-order Wiener kernels but in recent years second-order Wiener and Volterra kernels have been estimated, particularly with reference to dynamic non-linearities. A particular form of second-order analysis, the Spectro Temporal Receptive Field, offers an alternative to first-order cross-correlation when phase-lock is absent. The correlation method has revealed that neural synchronization is less affected by intensity changes and damage to the hair cells than is neural firing rate. Although the presence of the static cochlear non-linearity could be demonstrated on the basis of the intensity dependence of the first-order Wiener kernel, the identification of the exact form of the nonlinearity of the peripheral auditory system on basis of higher-order Wiener kernels has so far been inconclusive. However, successes of the method can be found in the description of the dynamic non-linearities and non-linear neural interactions.

Shared-stimulus driving and connectivity in groups of neurons in the dorsal cochlear nucleus

Extracellular spike discharges were recorded from ensembles of up to five neurons simultaneously in the DCN of guinea pig using solid-state, thin-film, multichannel electrodes having up to five recording sites spanning up to 600 microns. Responses from 73 unit pairs were collected of which 54 had both units responding to pseudorandom wideband noise stimulation. Shared-stimulus driving was present in 78% (42/54) of the unit pairs and could be attributed to an overlap in their spectral sensitivities. Effective connectivity was indicated for 87% (47/54) of the unit pairs. Wideband noise proved more useful than tonebursts for investigating shared-stimulus driving and connectivity because it evoked widespread, but not overly synchronous, responses in the ensembles.

Spectrotemporal receptive fields of neurons in cochlear nucleus of guinea pig

Spectrotemporal receptive fields (STRFs) [Hermes et al., Hear. Res. 5, 147-178, 1981] for neurons in the cochlear nuclei (CN) of guinea pig were estimated. Sixteen periodic segments of bandlimited, synthesized noise evoked replicable, distinctive period histograms for spike discharges. All driven units in the major divisions of the CN having their characteristic frequency (CF) within the noise bandlimits had unique STRFs for a given intensity of noise stimulation. The STRF maximum corresponded to the unit's CF, and details of the STRF patterns differed over CN divisions and response classes derived from tonebursts. The sizes of features in STRFs from this mammal appeared significantly smaller in their temporal and spectral extents than those reported in the torus semicircularis of an amphibian and were roughly comparable to the few units reported from cat ventral CN [Eggermont et al., Quart. Rev. Biophys. 16, 341-414, 1983]. STRFs, as they are presently obtained, provide useful insight into some aspects of afferent processing and perhaps connectivity, but their interpretation is specific to the level of stimulation and limited by the need to choose a specific energy distribution to represent the stimulus.