Jennifer S. Buchwald

Middle- and long-latency auditory evoked responses recorded from the vertex of normal and chronically lesioned cats

Abstract A prolonged sequence of auditory evoked potentials with latencies ranging from 1 to 250 msec was recorded from the vertex of the awake restrained cat. This sequence was reproduceable within and across subjects and was not altered by complete neuromuscular paralysis. The effects of click rate, pentobarbital, and chronic lesions of a number of different brain areas were evaluated for each of the potentials. Vertex waves 1–5, previously shown to originate from generators in the primary auditory pathway of the brain stem, were followed by smaller and less well defined waves 6 and 7, with peak latencies in the 6–8 msec and 10–12 msec range respectively. These potentials were not abolished by fast click rates (i.e. up to 50/sec) nor by moderate levels of pentobarbital. Correlative extra- and intracranial studies indicated that wave 6 occurred in the same latency range as the medial geniculate body, pars principalis potential, and that wave 7 occurred in the same latency range as the primary ectosylvian cortical potential. The intracranial potentials showed click recovery functions and barbiturate resistance which were similar to those of waves 6 and 7, and wave 7 disappeared following aspiration of ectosylvian cortex. These data suggest that waves 6 and 7 reflec generators in medial geniculate body and ectosylvian gyrus. In contrast to the stability of potentials 1 through 7, the longer latency waves were relatively unstable. Wave A occurred in a latency range of 17–25 msec, wave B, 35–45 msec, wave C, 50–75 msec, and wave D, 150–200 msec. All of these waves showed marked amplitude fluctuations, disappeared as click rates increased to 10/sec, and were abolished by moderate levels of pentobarbital. After bilateral aspiration of middle suprasylvian gyrus, ectosylvian gyrus, or frontal lobes, wave A continued to appear. After hemispherectomy, which removed all cortex, basal ganglia and limbic lobes, wave A was not abolished and appeared enhanced in one animal. Thus, the generator system of wave A appears to be largely independent of auditory cortex and adjacent association cortex, but may be modulated by other forebrain systems. Wave C continued to appear after aspiration of suprasylvian and ectosylvian gyri and after frontal lobectomy, but disappeared after hemispherectomy. Thus, wave C reflects a generator system which differs from that of wave A, but which also appears to be largely independent of the primary geniculo-cortical auditory pathway. These data suggest the following conclusions: waves 1 through 7, which show high fidelity, rate-resistant, barbiturate-insensitive acoustic transmission, appear to reflect activation of the primary auditory system from acoustic nerve to auditory cortex. Subsequent, longer-latency vertex potentials seem to be generated through other forebrain systems which receive auditory information in parallel from the brain stem, rather than serially from the primary geniculo-cortical pathway and association cortex relays. The relevance of data in the cat model to the human vertex potentials is discussed.

Interpretation of the vertex short-latency acoustic response: A study of single neurons in the brain stem

Summary Short-latency acoustic responses were recorded from the vertex of 14 adult cats concurrently with the recording of single unit activity in the brain stem auditory relay nuclei, i.e., cochlear nucleus, superior olivary complex, nuclei of the lateral lemniscus and inferior colliculus. The latency of each vertex-evoked response component, P2–5, was determined for each of a series of frequencies ranging from 500 Hz to 10 kHz. These latencies did not differ significantly and the pooled data showed a mean latency of2.3 ± 0.4msec for component P2, 3.1 ± 0.3msec for P3, 4.2 ± 0.4msec P4 and 5.4 ± 0.4msec for P5. Unit recordings indicated a subpopulation of constant latency units, i.e. with 0.2 msec or less standard deviation in response latency to 10 successive best frequency tone presentations, in each brain stem relay studied. Other units within the same nuclei showed up to one msec or more standard deviation in response latency. As the standard deviation increased, the mean latency of the units also increased. Thus, the shortest latency units were also the units with the most constant response latency. The proportion of constant latency units decreased at more rostral levels of the brain stem, particularly in the inferior colliculus. The group of constant latency units in the cochlear nucleus showed a mean latency of 2.6 ± 0.4msec, in the superior olivary complex 3.2 ± 0.8msec, in the nuclei of the lateral lemniscus4.5 ± 0.9msec, and in the inferior colliculus these latencies were 8.0 ± 3.1msec. These data indicate a close relation between the latencies of the vertex-evoked potential components and the subpopulation of constant latency units within the brain stem relay nuclei. It is suggested that the vertex short-latency acoustic responses P2–P5 reflect the successive activation of this subpopulation of brain stem units.

Midlatency auditory evoked responses: differential effects of a cholinergic agonist and antagonist

The effects of a cholinergic antagonist (scopolamine) and agonist (physostigmine) on the auditory middle latency evoked responses (MLRs) were studied in 7 normal male volunteers. Scalp recordings were made from a central (Cz) electrode referenced to linked ear lobes on one channel and to a non-cephalic, sternovertebral reference on a second channel. Three components were statistically analyzed for changes in latency and amplitude: Pa, with peak positivity in the 25-40 msec latency range, Nb, with peak negativity 40-50 msec, and P1, with peak positivity 50-65 msec. Control recordings included responses to click rates of 1, 5, 8 and 10/sec; as has been previously reported, P1 showed a marked decrease and disappeared at the faster rates of stimulation whereas Pa showed no change in amplitude. Intravenous injections of scopolamine resulted in a rapid and complete disappearance of P1 and a slight increase in Pa; concurrently, the subjects reported feeling drowsy but were awake with eyes open through the recordings. Subsequent injections of physostigmine resulted in a rapid reversal of the scopolamine effects so that the subjects became alert, Pa decreased, and P1 reappeared and increased to control amplitudes. Rapid click rates caused P1 to diminish, as in the control period, indicating a common P1 recovery cycle in both the control and physostigmine conditions. These data are discussed in terms of the hypothesis that the P1 generator system is comprised of a cholinergic brain-stem-thalamic component of the ascending reticular activating system.

Changes in cortical and subcortical unit activity during behavioral conditioning

Abstract Prior to conditioning a hindleg flexion response, electrodes were implanted in the primary projection pathway of the CS, in the primary projection pathway of the US and in the reticular formation in each of 9 chronic cats. Recordings of multiple-unit activity through these electrodes proceeded during habituation to the tonal CS, training in which the CS was paired with shock to the hindpaw, extinction and subsequent retraining. In the CS projection pathway unit activity patterns shifted from habituation to dishabituation after US reinforcement and with successive trials these responses progressively increased. Similarly, early in training responses appeared in the reticular formation. In the US projection pathway little or no unit response to the CS developed prior to the appearance of the overt response. Thus, potentials recorded from a population of neurons may show specific response patterns during behavioral conditioning directly related to activity at the recording site. The present data suggests that activation of the CS and reticular systems are primary events in conditioning.

Brain stem auditory evoked response development in the kitten

The development of brain stem auditory evoked responses (BAERs), recorded from a surface electrode as short-latency, volume-conducted potentials, was studied in a series of kittens over a postnatal period ranging from birth to 60 days. Repeated, longitudinal observations on particular kittens were supplemented with observations on additional kittens during the first and second postnatal week to determine age of onset of the BAERs. The position of the animal and sound source within the recording chamber were held constant across recording sessions, as was click intensity except during recordings in which intensity effects were specifically studied. Click rates of 1, 10, 50 and 100/sec were routinely presented. Reference electrodes at the tongue, pinna and neck showed volume-conducted responses to the click stimuli and resulted in considerable distortion of the activity recorded by the vertex electrode; the forepaw, in contrast, showed no activity and a vertex-forepaw electrode configuration provided good resolution of the BAERs across development. A number of new observations were made. BAERs were first observed at 4 days of age, approximately the same age at which depth evoked potentials are first recorded in brain stem auditory nuclei. Initially the BAERs were diffuse, high threshold and fatigued rapidly, characteristics shared with depth evoked potentials in the early postnatal period. Over the first two weeks, the potentials showed marked decrease in threshold, increased resistance to fast click rates, and better definition of wave forms. All BAER components showed exponential decreases in latency. Because all of the brain stem evoked potentials could be recorded concurrently and longitudinally in the same subject a number of developmental comparisons were possible among the BAER components. Wave 1, related to the acoustic nerve in the adult cat, showed a developmental time course and adult latency similar to that reported for N1. Wave 2, related to the cochlear nucleus in the adult, showed a marked bimodality over the first month; wave 2a was a large amplitude clearly separated wave which gradually fused as an inconspicuous leading shoulder on wave 2b. Wave 2b developed with a time course and adult latency similar to that reported for the ventral cochlear nucleus. Wave 3, related to the region of the superior olivary complex in the adult, showed a clear but transient bimodality during the third week of development. Wave 5, related to the inferior colliculus in the adult, appeared later than waves 1-4 and showed a significantly slower rate of development than waves 1-4. These data indicate that differential developmental changes occur within the brain stem auditory pathway and that the BAERs provide a dynamic probe of concurrent maturational interactions.

Cholinergic neurons of the feline pontomesencephalon. II. Ascending anatomical projections.

Immunoreactivity for choline acetyltransferase (ChAT) was analyzed in unoperated cats and in cats in which stereotaxic lesions were made in the pedunculopontine and laterodorsal tegmental nuclei. The fine reaction product revealed moderate to dense ChAT-immunoreactive fiber plexuses throughout the telencephalon, diencephalon, and midbrain. A pontomesencephalic origin of cholinergic innervation to virtually every nucleus of the diencephalon, as well as to various midbrain and basal telencephalic sites was indicated in the cats with lesions, in which the optical density of ChAT-immunoreactivity was significantly decreased as compared to controls. Pontomesencephalic lesions produced no changes, however, in the density of ChAT staining in the cerebral cortex, basolateral amygdala, or caudate nucleus. In addition to ChAT-positive terminal fiber arborizations which were widely distributed, cholinergic fibers-of-passage were traced in the unoperated and operated feline brains. The general course of ChAT fibers cut in cross-section was followed in successive transverse levels, and although pathways originating from the pedunculopontine nucleus demonstrated orientations in every direction, many demonstrated a rostral course. A particularly dense aggregate of ascending ChAT-positive fibers was localized in the dorsolateral sector of the pedunculopontine area which could be followed at more rostral levels into the central tegmental fields and the compact part of the substantia nigra. From the central tegmental fields, numerous ChAT-immunopositive fibers cut in cross-section continued to course rostrally in the intralaminar, reticular and lateroposterior nuclei of the thalamus, and a distinct bundle of ChAT fibers coursing dorsolaterally was observed medial to the optic tract ascending to the lateral geniculate. ChAT fibers with dorsolateral orientations were additionally observed in the zona incerta, ventral anterior thalamus, and ansa lenticularis on route to the reticular thalamus, the globus pallidus, and the substantia innominata. Pathways consisting of fibers traced from ChAT-containing cells in the laterodorsal tegmental nucleus could be traced to medial structures such as the periaqueductal gray, ventral tegmental area and dorsal raphe. Medially placed ChAT fibers were additionally followed through the ventral tegmental area, the midline thalamus, and the hypothalamus, up to the medial and lateral septal nuclei. The trajectories of the ascending cholinergic pathways from the pontomesencephalon are discussed in relation to locally generated electrophysiological responses in the cat.

Factors that affect the amplitudes and latencies of the vertex short latency acoustic responses in the cat.

The latencies and amplitude of the short latency (less than 10 msec) acoustic evoked responses recorded from the vertex of the cat have been studied as a function of acoustic stimulus parameters. A change in the stimulus intensity, duration, or rate of repetition resulted in parallel changes in the first 5 evoked responses, i.e., the latencies shifted the same amount and the amplitudes were modified in the same proportion. Comparison of responses to monaural versus binaural stimuli indicated an occlusive effect of binaural stimulation only in potential 4.

A technique for recording and integrating multiple unit activity simultaneously with the EEG in chronic animals

Abstract Multiple unit activity, with concurrent EEG recordings, has been recorded and integrated from various brain sites in chronic, unrestrained animals. The output of the integration circuit follows rapid changes in both frequency and amplitude of multiple unit activity without being disproportionately influenced by isolated, large amplitude spikes. Thus, action potentials from a large population of units can be quantitated and the resultant activity patterns compared with concurrent slow wave EEG activity recorded from the same site.

Classical eyeblink conditioning in the bilaterally hemispherectomized cat

Abstract The classically conditioned eyeblink response was studied in cats with various forebrain lesions. The most extreme of these were three chronically prepared “diencephalic” cats in which the cerebral hemispheres were separated from the thalamus and removed bilaterally. Other animals studied included two animals with frontal cortex ablation, one neodecorticate cat, two animals with bilateral lesions of the caudate nucleus, and one cat with extensive bilateral damage to the thalamic nucleus ventralis posteromedialis. Conditioning was established using a 400-msec duration auditory conditioning stimulus (CS) followed by a brief cutaneous shock to the eyelid as the reinforcing stimulus (US). The EMG was recorded bipolarly from orbicularis oculi, and integrated EMG for the CS interval served as an additional objective measure of conditioned response (CR) magnitude. The conditioned responses were characterized by a burst of EMG activity toward the end of the CS interval (latency 100 msec) which occurred as a result of systematic pairing of CS and US, and did not occur during random presentation of the CS and US, or CS alone. All of the experimental animals learned a conditioned blink response which appeared qualitatively similar to that obtained in the normal animals. All preparations acquired a response, showed extinction when reinforcement was discontinued, discriminated between two stimuli within the same sensory modality, showed discrimination reversal, and retained the acquired response over days. Our results are compatible with a growing body of evidence that telencephalic structures are not necessary for the acquisition and maintenance of a classically conditioned response. Further studies in this laboratory are aimed at investigating conditioning in the chronic decerebrate cat.

Cholinergic neurons of the feline pontomesencephalon. I. Essential role in ‘Wave A’ generation

Wave A in the cat appears to be analogous to P1 in the human. Both are positive middle-latency auditory-evoked potentials, present at slow click rates during wakefulness and REM sleep but absent during slow-wave sleep. Wave A has been recorded in the parabrachial and medial tegmental areas of the midbrain and in thalamic target projections of the reticular activating system. Two nuclei in this system, the pedunculopontine tegmental (PPT) and laterodorsal tegmental (LDT) nuclei, contain cholinergic cells; the cholinergic antagonist scopolamine eliminates Wave A. To test whether PPT and LDT were important in Wave A generation, we attempted to lesion these nuclei bilaterally in 11 cats. Wave A was markedly diminished or absent in all but 2 cats, in which the lesions did not include PPT. Loss of choline acetyltransferase-positive cells in PPT, but not LDT, was correlated with effects on Wave A, i.e. greatest cell loss occurred in cats in which Wave A disappeared, and least cell loss in cats with no change in Wave A. We conclude that the PPT nucleus, and particularly its cholinergic cell component, is essential for Wave A generation and suggests that a similar substrate may be significant for generation of the human P1.