ALBERT

Notes: Abstract Periodic envelope or amplitude modulations (AM) with periodicities up to several thousand Hertz are characteristic for many natural sounds. Throughout the auditory pathway, signal periodicity is evident in neuronal discharges phase-locked to the envelope. In contrast to lower levels of the auditory pathway, cortical neurons do not phase-lock to periodicities above about 100 Hz. Therefore, we investigated alternative coding strategies for high envelope periodicities at the cortical level. Neuronal responses in the primary auditory cortex (AI) of gerbils to tones and AM were analysed. Two groups of stimuli were tested: (1) AM with a carrier frequency set to the unit's best frequency evoked phase-locked responses which were confined to low modulation frequencies (fms) up to about 100 Hz, and (2) AM with a spectrum completely outside the unit's frequency-response range evoked completely different responses that never showed phase-locking but a rate-tuning to high fms (50 to about 3000 Hz). In contrast to the phase-locked responses, the best fms determined from these latter responses appeared to be topographically distributed, reflecting a periodotopic organization in the AI. Implications of these results for the cortical representation of the perceptual qualities rhythm, roughness and pitch are discussed.

Notes: Abstract Timbre and pitch are two independent perceptual qualities of sounds closely related to the spectral envelope and to the fundamental frequency of periodic temporal envelope fluctuations, respectively. To a first approximation, the spectral and temporal tuning properties of neurons in the auditory midbrain of various animals are independent, with layouts of these tuning properties in approximately orthogonal tonotopic and periodotopic maps. For the first time we demonstrate by means of magnetoencephalography a periodotopic organization of the human auditory cortex and analyse its spatial relationship to the tonotopic organization by using a range of stimuli with different temporal envelope fluctuations and spectra and a magnetometer providing high spatial resolution. We demonstrate an orthogonal arrangement of tonotopic and periodotopic gradients. Our results are in line with the organization of such maps in animals and closely match the perceptual orthogonality of timbre and pitch in humans.

Notes: Summary 1. Neurones in the auditory midbrain nucleus (MLD) of Guinea Fowl were examined for specific encoding of vocalizations. Single units were recorded in chronic awake preparations of birds. The animals were presented with various artificial stimuli and tape recordings of conspecific calls. 2. Several Guinea Fowl vocalizations are characterized by multiple-line spectra or by inhomogeneous noise components (Fig. 4). Such broad energy distributions were called ‘frequency complexes’. Neuronal preference particularly of calls which contained frequency complexes was examined by three methods: (a) Use of natural variations of calls; (b) selective filtering of calls; (c) technical synthesis of complex components of calls. 3. Of the neurones in the material, 60% showed responses to complex stimuli not simply predictable from pure tone responses (complex neurones). None of the neurons gave exclusive responses to a particular call but some preferred certain calls. Several types of complex neurones were distinguished which might be considered suitable for the detection of frequency complexes in calls. 4. Units with broad inhibitory bands in response to pure tones were excited by harmonic (pulse) spectra or other frequency combinations which fell into the inhibitory band (Figs. 4–6). Similarly units with weak but wide-band excitation by pure tones often responded to multiple-line spectra (Fig. 6). 5. Neurones with preference for several tone frequencies favoured spectra which overlapped those optimal frequencies (Fig. 3). 6. FM sensitivity was seen in combination with preference for frequency complexes (Fig. 8).

Notes: Summary 1. One auditory area of the neostriatum (field L) in the guinea fowl is defined both by autoradiography of terminal labelling after injection of tritiated amino acids into the nucleus ovoidalis and by systematic mapping of the activity of small ensembles of units in the frontal and horizontal plane. 2. Anatomical and physiological maps are congruent and show three zones in dorso ventral sequence L1, L2, L3, which together constitute the neostriatal field L (Fig. 2). L2 receives most of the input fibers from the nucleus ovoidalis and shows high activity of units with sharply peaked responses to tones (best frequencies). It probably corresponds to the former field L of Rose. 3. L1 and L2 show less terminal labelling and lower activity of units. Peak responses to tones are present but weaker (Fig. 2). 4. Field L of the neostriatum shows a tonotopic organization with peak frequencies between 300 Hz caudally and 6 kHz rostrally and isofrequency contours running from rostral-lateral to medial-caudal (Fig. 3). Continuous isofrequencies are also present in the vertical plane, i.e. across the boundaries between L1, L2 and L3 (Fig. 2). 5. The area of CFs between 1 and 2 kHz reveals a high proportion of units which respond to Iambus-like calls and distinguish these calls from other calls of the guinea fowl. Systematic penetrations with single unit recording confirm the subdivision of neostriatal field L into three zones. In addition to the different tone responsiveness, selectivity for calls provides a criterion for the subdivision (Fig. 5). 6. Units in L2 are responsive but do not distinguish Iambus-like calls from other calls. Units in L1 and L3 are highly selective. Some respond only to natural variations of the Iambus call (Figs. 6, 7, 8).

Notes: Summary 1. The tonotopically organized auditory neostriatum (Field L1,L2,L3) was examined in 7 animals for single units which respond to Iambus-like calls and other calls (responsive units), or which clearly distinguish Iambus-like calls of the guinea fowl from other calls of the species (selective units). A high proportion of responsive (168 units: 46%) and of selective units (58 units: 16%) was localized in an area with best frequencies of units between 1 and 2 kHz. Only selective units are described in this paper. 2. Stimulation with 16 natural variations of the Iambus call showed that the 58 selective units formed a population of comparable neurons, in that 80% responded to more than 9 Iambus variations. High response coincidences of neurons to particular Iambus variations indicated that the neuronal preference relied on characteristic lines in the Iambus spectrum and not on accidental properties (Figs. 1–6). 3. In most Iambi the energy is concentrated in spectral lines between 1 and 2.3 kHz. This band is called the doininant spectral band. The highest response score for Iambus variations contained those units which responded to tones of 1.8 and 1 kHz within the dominant band (Figs. 8, 11). One kHz is the fundamental frequency of the harmonic Iambus spectrum, and 1.8 kHz comes close to the formant, i.e. the energy maximum of an average Iambus spectrum. Selective units rarely responded to 1.5 kHz, and Iambus variations with a strong 1.5 kHz component least frequently elicited responses from units (Figs. 8, 11). 4. Half of the selective units showed multiple response maxima when stimulated with tone bursts. One or several maxima were between 1 and 2 kHz. Most other units showed suppression of activity between 1 and 2 kHz or no tone responses (Figs. 9–11). The absence of a tone response was no indication of a higher call selectivity of units. 5. Pulse trains of different periodicity (P) generate harmonic spectra with varying fundamental frequencies (Figs. 13, 14). These stimuli frequently yielded neuronal response maxima at 1 kHz fundamental frequency (the fundamental of Iambus-like calls) as well as maxima at 1.8 kHz and below 700 Hz. Such maxima were also observed in units with no pure tone response or with no tone response in this band (Figs. 9–11). 6. Pulse trains which are amplitude modulated (PAM) generate harmonic spectra with sidebands (Fig. 12). A majority of selective units gave prominent responses to PAM with a combination of 900 Hz to 1.2 kHz fundamental frequency and sidebands of 50 Hz to several hundred Hz distance from the harmonics (Figs. 9, 10). This corresponds to the characteristic spectral lines of Iambus calls (Figs. 13, 14). 7. In other units the spectral envelope of PAM stimuli had to be adjusted to the formant of Iambi by bandpass filtering the signal in order to obtain prominent responses (Fig. 11). These stimuli (Figs. 13, 14) mimic focal properties, i.e. the complete spectral construction of Iambi. 8. One of the most important findings appears to be that selective units were far more specialized to distinguish Iambi from other wide-band calls with spectral overlap in the dominant spectral band than they were to distinguish Iambi from their simple or complex spectral components.

Notes: Summary Pairs of very high frequency electric fish from South America (Sternarchorhynchus sp. andSternarchorhamphus sp.) synchronize their discharge and finally engage in phase coupling which is maintained over minutes (Fig. 2). This manoeuvre, called Active Phase Coupling (APC), may be as precise as a few microseconds phase jitter over half a min (Fig. 5). Phase coupling is also accomplished to other fish and to stimuli the frequency of which is several hundred Hz away (Fig. 3). In this case the EOD of the responding fish and the stimulus finally are higher harmonics of a common fundamental and phase coupling is to every nth wave of the stimulus. The APC is more complex a behavior than the known Jamming Avoidance Response. Its social significance is yet to be determined. It involves a neuronal control loop with at least 4 synapses some of which are probably electrotonic.

Notes: Abstract Auditory neurons typically respond to a restricted range of frequencies and amplitudes of pure tone stimuli. These findings have led to the concept of the classical frequency receptive field. Over the last few years evidence has accumulated that stimuli outside the frequency and amplitude boundaries of a neuron's receptive field can influence responses to stimuli inside the classical receptive field. We could recently show that sinusoidally amplitude-modulated pure tones could excite cortical neurons although all of their spectral components were above the spectral range of pure tones effective to excite the neuron. This result demonstrated that neurons in the auditory cortex integrate over spectral ranges that are much wider than is evident from responses to pure tones. Here, using sinusoidally amplitude-modulated pure tone stimuli we determine electrophysiologically the high-frequency boundaries of the spectral integration capabilities of auditory cortical neurons in anaesthetized Mongolian gerbils under normal conditions and under the influence of the microiontophoretically applied GABAA-receptor antagonist bicuculline. Our results demonstrate that some auditory cortical neurons integrate over the gerbil's entire audible spectrum. Therefore, the classical excitatory frequency receptive field of an auditory cortical neuron, as determined with pure tone stimuli, cannot provide a satisfactory description of its spectral integrative properties.

Notes: Summary The auditory pathway of the Guinea Fowl was labeled with [C14]2-deoxy-D-glucose after stimulation with pure tones, harmonic tones and species-specific calls. In addition to other auditory nuclei, which showed more or less uniform labeling with the present technique, the n. mesencephalicus lateralis dorsalis (MLD) of the midbrain, as well as field L and parts of the hyperstriatum ventrale in the telencephalon, showed a stripe-pattern of labeling after stimulation with a pure tone. The position and orientation of the tone-activated striped areas in field L, observed after stimulation with different tones, correspond to isofrequency contours obtained with microelectrode recordings. The labeling of the three congruent tonotopically organized layers of field L (L1, L2, and L3) was not uniform along the anterior-posterior axis of the field. Harmonic tones produced multiple reactive stripes each of which corresponded to the stripe characteristic of a particular harmonic presented as a pure tone. The species-specific Iambus-call labeled the tonotopic area of field L that corresponds to the frequency band with the highest energy of the call. The hyperstriatum ventrale generally showed a weaker pattern of labeling that, however, resembled the labeling in field L.