Formation of spike response to sound tones in cat auditory cortex neurons: Interaction of excitatory and inhibitory effects

Neta,oscience Vol. 43, No. 2/3, pp. 307-321, 1991

03064522/91 $3.00+ 0.00 PersamonPreu pk C) 1991mR<)

Printed in Great Britain

F O R M A T I O N OF SPIKE RESPONSE TO S O U N D TONES IN CAT A U D I T O R Y CORTEX N E U R O N S : I N T E R A C T I O N OF E X C I T A T O R Y A N D I N H I B I T O R Y EFFECTS I. O. VOLKOV* and A. V. G^LAZJUK Department of Physiology of Cerebral Cortex and Subcortical Structures, A.A. Bogomoletz Institute of Physiology, Ukrainian Academy of Sciences, Kiev, U.S.S.R. Almemet--Reg~ome8of the auditory cortical neurons to sound tones were studied extra- and intrscellularly in anaesthetized cats. The pattern of response to tone stimuli could most differ in neurons tuned to the same sound frequency and forming a vertical cortical column. Phasic reactions were found in 69% of the nem.ons studied. Such neurons were encountered in all cortical layers but about 50% of them were localizedat a depth of 0.4-1.0 ram, which corresponds to layers HI and IV of the auditory cortex. Neurons with phasic reactions were able to respond to a relatively narrow frequency band that demonstrates high discriminative ability of these cells to the frequency analy~ of sound signals. InhibitOry p_roces~___realized via both forward afferent and recurrent intracortical inhibition mechanL~nRplay particular roles in the formation of phas~ reaction of such neurons to different frequency tones. Twenty-six per cent of neurons generated tonic responses to the sound. The majority of such cells (94%) were localized at a depth of 1.0-2.2 ram+ wh~h c o - - n t i s to cortical layer$ V and VI. Inhibitory processes exert a much kuer influence on formation of tonic responses in compm'imn with p h a ~ ones. Neurons of the tonic type, in contrast to phasic neurons, respond to a wider frequency band; their lower ability to discriminate sound frequency is obvious. Parameters of the responses of tonic neurons strictly correlated with the duration and intensity of the acoustic signal. The possibil/ty of some tonic neurons playing an inhibitory role in auditory cortex is discussed [Volkov I. O. et al. (1989) Neurophysiology, Kiev 21, 498-506, 613-620 (in Rns,6an)]. A small portion of the auditory area AI neurons (2%) demonstrated the suppmudon of background activity during tone stimulation. They were localized mainly in deep cortical layers (V and VI). lntracortical inhibition is supposed to play a dominant role in the formation of this type of response. About 3% of the studied auditory cortex neurons with background activity generated no response to tonic stimuli. Such cells were usually encountered in the superficial auditory cortex layers (I and II).

Experiments on una~a~thetized animalst,9,2t.ss~ revealed that auditory cortex neurons respond to acoustic stimulation in a more complex and various manner than in animals under barbiturate anaesthesia. sa~s4 The authors presented a detailed description of different forms of neuronal impulse reactions to tone stimuli and data about the proportion of cortical neurons with different response patterns. Two main neuronal populations were shown to exist in the auditory cortex of unanaesthefized animals. Units belonging to one population generated phasic responses (ON, ON--OFF or O F F reactions) to tone stimuli, while the other group responded by a steady impulse discharge during the whole period of the sound stimulation. The latter units were called "through ''~ or "tonic ''ss neurons. Auditory cortex neurons with phasic responses are more numerous. They form approximately two-thirds of soundreactive auditory cortex neurons; the rest respond to tone stimuli by sustained amplification or suppression of their background impulse activity.lJ, ILss~ *To whom correspondence ahould be addreaed. BF, best frequency; ~ , excitatory postsynaptic potential; IPSP, inhibitory postsynaptic potential.

Abbreviations:

307

The phenomenology of responses of auditory cortex neurons to tone sound stimuli is described in detail; it must be mentioned that some contradictions can be found in this information. At the same time, intracortical synaptic mechanisms of different auditory cortex neurons responsible for the formation of the discharge deserve further investiSation. The present study is devoted to the elucidation of this question by analysing the extra- and intracellular reactions of the neurons of auditory cortex area AI to sounds of different frequency and intensity. It is hoped that such an approach will promote a better understanding of neuronal and functional organization of the mammalian auditory cortex. EXPEIUMgNTAL PiIOCEDUiIIES Preparation

Twenty adult cats were used for acute experiments. Surgery was performed under ketamine

(20ms/kg,intrmu,eu~r ~,je~oa). The ~an w~ ~ped

and trepanation was made above the middle ectmylvian gyrta contralateral to the mimulatiun ide. A I ~ removal of the dura the expoted cortex wits ~ with Kxebs' solution warmed up to 37~C to pfeveat drying und cooling. To decrense brain l~Imtion a larlle occip/ttl cislern wns opened and drained. A plastk tube about 3 a n Ions was fixed into the external auditory meatus. The contralateral

308

I. O. VOLKOV and A. V. GALAZJUK

ear drum was destroyed. When the operation was over, the animal was immobilized with myorelaxin and artificially ventilated. A thorough infiltration anaesthesia of soft tissues in the operation field was provided with repetitive injections of 0.5% novocaine solution. General anaesthesia was maintained during the experiment with small supplemental doses of ketamine (10 mg/kg, i.m.).

Stimulation The acoustic stimuli consisted of sound tones of different frequencies and intensities; dicks were also used if necessary. The standard rise and fall period of tone presentation was 5 ms; it could be varied from 5 to 50 ms in special cases. The duration of tone varied from 10ms to I s. A set of sound emitting devices (dynamic loudspeakers 0.5GD-37 and 2GD-36 with appropriate frequency-response characteristics) and a generator of sound-frequency sinusoidal waveforms (GZ-33) were used to provide a broad frequency range (I-22 kHz) of sound reproduction. Calibration of the stimulus intensity was carried out in an acoustic chamber (volume, 80 m3; resonance frequency, 36 Hz; factor-of-merit equal-I; self-noise level less than 17 dB) with a precision sound indicator (PSI-202, Type-00001, DDR).

with firing index 0.7 was considered a~ threshold. 'Ihe frequency-threshold curve ("turning curve") was plotted by testing 20-25 different stimulus frequencies and the best frequency (BF) of a sound-responsive neuron was determined. When the neuron generated background activity. reactions to sound stimuli were estimated by analysing peristimulus histograms. In some experiments histological control was pertormed to trace a microelectrode trajectory. A three-channel micropipette was used for this purpose. One channel was filled with 10% primuline aquatic solution (pH 7.0) while two others, filled with 4 M NaC1, were used for recording and compensation. Two ionophoretic primuline injections (I-5,aA, 2-5m) were usually made at depths of 0.1 and 2.0 mm. At the end of the experiment the animal was perfused under deep nembutal anaesthesia (50 mg/kg) with 4% paraformaldehyde solution in phosphate buffer (pH 7.4). Frontal sections (50tam) of the studied cortical zone were made using a freezing microtome, and were then observed through a luminescent microscope at a wavelength of 380-420 nm. The microclectrode track direction was reconstructed on the basis of primuline label locations RESULTS

Recording Recording from cortical neurons was performed with glass microelectrodes. Electrodes filled with 4.0 M sodium chloride with an impodanc¢ of about 10-12 MfJ were used for extracellular recording of spike activity. For intracellular recording microelectrodes filled with 2 M potassium citrate and having an impedance of approximately 200 Mfl were used; their tip was sharpened on an abrasive disk to decrease the impedance to 50-60 M~, which allowed more effective perforation of the cell membrane. The intensity of an acoustic stimulus when a neuron generated spike response

Ten to fifteen n e u r o n s responding to tone presentations were usually f o u n d along a vertical microelectrode track t h r o u g h the cortex layers. Most o f them ( a b o u t 9 0 % ) generated b a c k g r o u n d impulse activity, the average rate being 4 . 0 s --~. N e u r o n s recorded within the individual microelectrode p e n e t r a t i o n had, as a rule, identical or very close BFs (Fig. 1B). At the same time such n e u r o n s located in the same cortical column could d e m o n s t r a t e different r e s p o n ~

B

dB

'°it 20

./-

C

V

°t \

20 01 . . . . .

1 Id'lz 0

s

8 16 Fig. 1. Characteristics of responses of A l cortical columnar neurons to tone stimulation. (A) Scheme of fragment of frontal section of cerebral cortex with one reconstructed micr~lcctrode penetration throughout the auditory cortex. Dark circlts represent recording sites for a part (1/2) of the neurons identified in a given track. Arrows indicate frequmacy- threshold curves (B) and peristimutus histograms (C) of neurons located in the column studied. The first, second and fourth rows of (B) and (C) illustrate the frequency-threshold curves and ptristimulus histograms of neurons with phasic responses to tone; the third and fifth rows illustrate the tonic reactions. Abscissa: (B) frequency (kHz); (C) time (ms). Ordinates: (B) intensity (dB); (C) number of impulses in a bin. Bin is I0 ms. Histograms are normalized to number of stimuli presented. Down: sound stimulation mark.

Formation of spike response in auditory cortex neurons patterns to the BF tone delivered at an intensity of 10 dB over the threshold (Fig. 1C). Two-hundred-andforty-seven (69%) out of 357 neurons studied exhibited phasic impulse responses ("phasic neurons"). Many of these neurons (140 units or 57%) generated a short latency response to tone onset consisting of a very fimited number (one, more rarely 2-3) of spikes (ON response). Thirty-six per cent of phasic neurons responded to both tone onset and offset (ON-OFF response). Approximately 7% of phasic neurons discharged only to tone offset (OFF response). The majority of phasic neurons had a narrow-tip tuning curve (Fig. 1B). Typical phasic neurons could be found in all cortical layers but the largest density (116 units or 47%) was observed at a depth of 0.4-1.0 mm, which corresponds to layers III and IV of the auditory cortex (Fig. 2). Ninety-three (26%) of the neurons studied exhibited tonic impulse responses that sustained during sound stimulus ("tonic neurons"). In most cases these neurons generated impulse responses to tones in almost the whole sound frequency range used. As a rule, tonic neurons bad a flat-tuned and considerably wider tuning curve than phasic ones (Fig. I B). Tonic neurons also showed an uneven distribution in the auditory cortex. Thus, six out of 93 tonic neurons were located at a depth of 0.4-0.8 mm, whereas the rest (87 units or 94%) were located at a depth of 1.0-2.2mm (Fig. 2). The maximal density of tonic neurons was observed at a depth of 1.2-1.8 ram, which corresponds to cortical layers V and VI. No tonic neurons were found in the superficial (up to 0.4 wan) and middle (from 0.8 to 1.0 ram) auditory cortex layers. A few neurons (seven units or 2%), usually seen at a depth of 1.4-1.6 mm (layer VI), showed a steady suppression of their impulse activity to tone stimulus. A comparatively small amount of neurons (3%) yielding no response to tone stimuli were found in layers I and II of the auditory cortex.

v

t0

oc ~ @

e-

1(] e-

0.4 0.8 1.2 1.8 D e p t h (ram)

2.0

Fig. 2. Depth distribution of neurons with phasic (1) and tonic (2) responses to best frequency tone in auditory cortex. Abscissa: cortical depth of neuron location (nun). Ordinate: relative number of neurons in corresponding groups (%).

309

Peculiarities of phasic neuronal responses As a rule, phasic ON responses consisted of 1-2 spikes, whose latency varied from 9 to 15 ms in different units (M + m -- 10.0 "t" 0.3 ms). The duration of the ON response varied from 2 to 5 ms, after which the neuron exhibited a suppression of impulse activity during presentation of tone stimulus (Fig. 3A). The strongest suppression of spike generation was observed 100--150ms after the ON response and later it gradually diminished. The tone offset led to cessation of the inhibitory effect and to recovery of the neuronal background activity. In many cases the tone offset was accompanied by an overshoot (by about 1.5-2.0-fold) of the initial background discharge rate typical of the unit (rebound effect). Intracellular recording showed that phasic neurons responding to BF tones for 10dB exceeding the threshold for spike generation produced excitatory postsynaptic potentials (EPSPs) of 5-10mV in amplitude (Fig. 3B). One or occasionally several spikes can be generated on the rising phase of EPSP or its peak 0.4-0.8 ms after the onset of the postsynaptic potential development. In 96% of the phasic neurons the excitatory ON response was followed by an inhibitory postsynaptic potential (IPSP), resulting in a deep suppression of impulse activity of the neuron studied. The pattern of response in the neurons depolarized by membrane injury caused by microelectrode impalement enabled us to determine "true" IPSP latency. Its average value was 0.8 ms longer than the average EPSP latency. The duration of the discharge depression following the ON response correlated with the duration of tone presentation, but the efficiency of this depression was unequal at different time intervals. During prolonged action of tone stimulus (over 200 ms) the most powerful inhibitory effect was observed for the first l(D--150ms after the ON response. This interval coincided with the duration of maximal hyperpolarization in these neurons. Then a gradual recovery of membrane potential to the resting potential level took place, which was accompanied by a decrease of inhibitory influence. Peristimulus histograms demonstrated that in the absence of hyperpolarization spike generation was possible but neuronal responsiveness remained lower till the end of the tone presentation. The tone offset evoked neuronal depolarization accompanied by spike discharges with the frequency considerably higher than that of the background activity level (Fig. 3A, B). Stable spike generation to both tone onset and offset ( O N - O F F responses) could be found in 25% of phasic units. It must be mentioned that in many cases the OFF discharge was larger than the ON due to the longer duration of the former. The analysis of intracellular records from these neurons showed cessation of postsynaptic inhibition and onset of the EPSP with a stable latency in respect to the sound offset (Fig. 3D).

310

I.O. VOLKOVand A. V. GALAZJUK

A

B

0

8O 40 l .

.

.

.

~

lOmV

0 . . . . .,

=1' .

.

~-,...

,_ Time

1000

" --'

--

100

ms

(ms)

Fig. 3. Reactions of AI phasic neurons to best frequency tones. (A), (C), (E) Poetatimulus histograms of different reactions of AI neurons to tone stimuli (ON, ON-OFF and OFF responses, respectively). Abscissa: time (ms). Ordinate: number of imput~ in bin. Bin is loins. Rcimtition number = 80. (B), (D), (F) Intracellular reactions to tone stimulus in the form ON, ON-OFF and OFF responses. Down: sound stimulation mark. Few neurons (5%) generated phasic spike reaction only to the tone offset (Fig. 3E). lntraceUular records showed that in such units the EPSP evoked by BF tone onset was effectively suppressed by the IPSP before reaching a threshold for action potential generation, but after the tone offset and cessation of inhibition these neurons developed a typical OFF response EPSP with short and relatively stable latency and successive spike generation. The pattern of phasic neuronal responses was largely dependent on the tone frequency used. The latency of the ON discharge was shortest when the BF tone was used. In this case the ON response was followed by an IPSP with the maximal amplitude and the slowest decrease in comparison with those when non-BF tones were used (Fig. 4). Gradual shift of tone frequency (with preservation of stable intensity) produced a progressive increase in the latency of ON action potential generation and a decrease of responsiveness of the neuron studied. The reason for discharge latency increase was the decrease of EFSP ampfitude and slope. At the same time the EPSP latency remained unchanged. IPSP amplitude and duration decreased with the shift of the tone frequency from the BF value. It was possible to select such a tone frequency when a neuron lost its ability to respond by spike generation; only a subthreshold EPSP-IPSP sequence remained. When tone frequency was shifted further the amplitude and

duration of both EPSP and IPSP were reduced. It shouM be noted that in a great portion of cortical neurons the process of frequency-dependent reduction of IPSP was much faster than the parallel EPSP depression (Fig. 4). The tone froquency band evoking synaptic (EPSP-IPSP) response in phasic neurons was two to three times wider than that evoking spike generation. The pattern of phasic neuron reaction was also dependent on the intensity of the tone presented. Response of such neurons to the threshold BF tone consisted of one or several depolarization deflections (EPSPs) of minimal amplitude (Fig, 5A). Increase of the tone intensity led to an increase of EPSP ampfitude and shortening of its latency. As a rule, a small IPSP of relatively short duration appeared just after the EPSP (Fig. 5B). Thus, the threshold for IPSP development was somewhat higher than that for EPSP generation. The difference between the thresholds for EPSP and IPSP development was most evident in the neurons with a high level of membrane potential. When the tone of 4-5 dB over threshold was presented, the action poumtial ~ ¢ a r e d on the rising phase of EPSP followed by IPSP (Fig. 5C). The ampfitude of postspike IPSP was considerably larger than that observed for EPSP-IPSP reaction. The intensity of impulse responses of phasic neurons incre~_~:lwith the increase of tone intensity: the probability of discharge (firing index) became

Formation of spike response in auditory cortex neurons

15

ilOmV 50m$

14

311

higher and additional spikes could be generated in many units. Nevertheless, a further increase of sound stimuli (about 50--60dB over impulse response threshold) led to the opposite effect: - a p p r e m o n of spike generation in the neurons (Fil~ 5D). The pattern of phasic neuronal reactiom was also strongly dependent on the rate of rise of acoustic stimulus. The steeper the rising p h = ~ of tone stimulus, the more intensive were the responses of the phasic neurons (Fig. 6A). A decrease of the rate of rise of acoustic stimulus led to a gradual decrease of the firing probability and stability of the ON respome of the phasic neuron. The annlym of ~ u p t i c events showed that such a decrease of neuronal responsiveness was accompanied by reduction of both EPSP and IPSP amplitude and duration. As a rule, a decrease of excitatory processes oocurred relatively faster than that of inhibitory ones. With a comparatively slow rate of rise of acoustic stimulus (20-50ms), the majority of phasic neurons were unable to generate action potential to sound onset. In this case the tone stimulus evoked only E I ~ P - I P S P successions or pure IPSPs in these units (Fig. 6C, D). The duration of neuronal response correlated well with the duration of tone presentation (Fig. 7). The O F F response corresponded closely to the offset of sound stimulation; the duration of suppression of

Fig. 4. Intracellular reactions of AI neuron to best frequency tones (14kHz) and to the tones of non-best frequency (15-21 kHz). Left: tone frequencies (kHz). Down: sound

stimulationmarl

A

1

jii'i 05

10me

dB

5O

DLJ

5ome

m8

Fig. 5. Intrazellular reactions of AI neuron to tone of different intemity. (A) Neuron response to tone itimulm of threahold intemity (15dB). (B-D) Neuron ~ to suprathRahold stimuli (20-65 dB). Left: intemity of tone stimulus (dB). Down: sound ~6mulation mark.

Fig. 6. l n ~ reactions of AI neurons to changes in ridng alope of b a t frequency tone ~imulm. (A-D) Neuron reactions to tone stimuli with different r~iug ph~e duration of tone stimulus (5, 10, 25 and 50 ms, ~ v e l y ) . Down: sound stimulation mark.

312

I.O. VOLKOVand A. V. GALAZJUK A

B

w ¢p

.m

o

Z

2'~

,

Time

(ms)

Fig. 7. Reactions of two AI phasic neurons to best frequency tones of differem duration. (A) Peristimulus histograms of the first neuron phasic reactions to different duration tones (30, 100 and |000 ms). Abscissa: time (ms). Ordinate: number of impulses in a bin. Bin is 10ms. Histolp'ams arc normalized to number of stimuli presented. (B) ]ntraccil~ar reactions of the second phasic neuron to tones of different duration (80, 430 and 800 ms). Down: sound stimulation mark.

neuronal impulse activity coincided approximately with the tone duration. Peristimulus histograms have clearly shown that when duration of the tone stimulus was increased from 30 ms up to I s, the duration of the ON reaction remained constant, whereas the duration postspike inhibition was prolonged with sound stimulus increase (Fig. 7A). 1001

u~

Peculiarities o f tonic neuronal reactior~ Eighty-six out of 93 (92%) tonic neurons exhibited complex spike reactions, including early and late components (Fig. 8A). The first, early component correlated with tone onset when the neuron generated a short-latent (8-10ms) series of 3--5 impulses. The

A

C

50

---K

)----

B t..

z

-~

Time

'~-

'IUO0

-P---

50ms

(ms)

Fig. 8. Reactions of four AI tonic neurons to tone stimuli of best frequency. (A), (B) Peristimulus histograms of complex and "pure" tonic neurons' reactions to tone stimuli. A b a t i s : time (ms). Oedinate: number of impulses in a bin. Bin is 10ms. Repetitions number=g). (C), (D) lutraceUular reactions of complex and "pure" tonic neurons to tone stimuli. Down: sound stimulation mark

Formation of spike response in auditory cortex neurons second, late component, which formed the main part of the neuronal response, consisted of a relatively steady impulse discharge as long as tone presentation lasted. The early component of tonic response was often followed by a short (from l0 up to 30ms) inhibition of impulse activity whose efficiency diminished with repeated activation of the cell studied. Eight per cent of tonic neurons discharged with relatively stable impulse frequency during sound stimulation (Fig. 8B). At high and stable membrane potentials (40-60 mV) most tonic neurons gave a high amplitude EPSP (up to 10 mV) to tone onset followed by impulse discharge (Fig. 8C). The early excitatory component was usually followed by a short-term (10-30 ms) IPSP during which impulse activity was suppressed. This short suppression of spike response was accompanied by a powerful depolarization as long as tone presentation lasted. This prolonged a plateau-like EPSP, had an amplitude of 5-10 mV and was accompanied by effective spike generation. Tone offset evoked fast cessation of the EPSP and consequently cell discharge. At the end of tonic reaction, inhibition of background impulse activity was usually observed unaccompanied by pronounced membrane hyperpolarization. In neurons with "pure" tonic reactions EPSP evoked by BF tone were not interrupted by IPSP and lasted during sound stimulation (Fig. 8D). The spikes were generated on the plateau of such fiat EPSPs. The tone, offset, led to a rapid membrane repolarization and cessation of impulse responses of the neuron. 'Changes of the neuron membrane potential caused by cell injury led either to substantial decrease of neuron excitatory response intensity or even to reversion of reaction sign. Thus, the decrease of the neuron membrane potential from 60 to 20 mV after microelectrode injury was accompanied by gradual inactivation of the spike generation mechanism (Fig. 9). The excitatory processes were blocked and amplitude and duration of the IPSP increased. Such conditions enable us to measure the proper values of IPSP latencies, being in this instance a few milliseconds longer than EPSP latencies. When tone stimulus frequency was shifted above or below BF (with stable intensity) the tonic neurons showed a gradual decrease of responsiveness. Figure 10A illustrates the dependence between tone frequency and intensity of the impulse reaction of the neuron studied. The strongest and most stable reaction was evoked by BF tone. Impulse frequency in the early component of tonic neuron response was the highest and could often reach 200 s-i. Gradual decrease of impulse intensity was obvious when nonBF tone stimulations were used, but the common response pattern was not dramatically changed. The frequency hand evoking spike generation by tonic neurons was about 2-3 times wider than that for the phasic neurons belonging to the same vertical column. Intracellular recordings from tonic neurons showed that changes in tone frequency (tone intensity NSC 43-2J~--B

2o~

mV

313

I~°mv SOme

Fig. 9. IntraceUular reactions of tonic neurons to best frequency tone for different membrane potential levels. Left: membrane potential levels (mV). being constant) led to a decrease of EPSP amplitude, naturally followed by a decrease of the number of impulses generated on the plateau of such an EPSP (Fig. 10B). Successive increase of BF tone intensity resulted in frequency growth of the response of tonic neurons (Fig. I1A). Maximal spike reaction was usually observed when the intensity of acoustic stimulus reached values 30-50 dB over response threshold. Intracellular records proved that increase of sound stimulation intensity was accompanied by the increase of EPSP amplitude and frequency of the neuronal discharge (Fig. 11B). Simultaneous combination of different acoustic stimuli, e.g. tone signals, paper rustling, whistles and so on, usually evoked stronger responses of tonic neurons than when these stimuli were used separately. Such inherent ability of tonic neurons to sum up the excitatory effects represents their significant specific features. The dynamics of responses of tonic neurons are also strongly dependent on the rate of rise of acoustic stimulus. The shortening of tone rising phase from 25 to 5 ms resulted in significant growth of impulse frequency (from about 100 to 250s -j) in the early, most pronounced dynamic reaction component, whereas the impulse frequency of the later strictly tonic component increased less (Fig. 12A-C). Substantial raising of impulse frequency in the early part of the reaction, due to the increase of excitatory synaptic action, was accompanied by parallel growth of the inhibitory effect on the neuron studied. This

314

I.O. VOLKOVand A. V. GALATJUK

A

B

.13

E Z

13 ~ - ~

"

-': . . . . . . . . . .

~

......

1

kHz

Time

-~"

~--

~)Oms

(ms)

Fig. 10. Reactions of two tonic neurons to tones of different frequencies and the same intensity (tO dB over threshold). (A) Pcristimulus histograms of the first reaction neuron to tones of best frequency (2[ kHz) and non-best frequency (8-18 kHz). Abscissa: time (ms). Ordinate: number of impulses in a bin. Bin is I0 ms. Histograms are normalized to number of stimuli presented. (B) Intraacllular reactions of second tonic neuron to tones of best frequency (9 kHz) and non-best frequency (I I -15 kHz). Down: sound stimulation mark.

A 2'

2

30clB

10dB t.

1,

@

B,

"5

~ J

O.

.E

a) .D

0 .......

"~,-,

'.

0

tO

2

40

2O

--~

r--

E

Z •

1000

?-

IOO0

i

~om4

dB

Time (ms) Fig. 11. Reactions of two tonic neurons to best frequency tones of different intensity. (A) PerisUmulus histograms of the first neuron reactions to best frequency tones of different intensity (10, 20, 30 and 40 dB over response threshold). Abscissa: time (ms). Ordinate: number of impulses in a bin. Bin is 10ms. Histograms are normalized to number of stimuli presented. (B) Intracellular reactions of second w~uron to best frequency tones of different intensity (10 and 20dB over response threshold). Down; sound stimulation mark.

Formation of spike response in auditory cortex neurons parallel effect could be detected especially easily in the case of a very sharp (click) stimulus, when a typical tonic neuron usually responded only by a short burst of impulses comprising 3-6 spikes (Fig. 12D-F). Instant frequency in response to click stimulus in some cases could be greater than 800 s-i. The first spike in such a burst was of shorter duration and higher amplitude than successive action potentials observed within 2ms (Fig. 12E). The changes of amplitude and shape of subsequent action potentials can probably be regarded as expressions of relative refractoriness. The complete recovery of spike amplitude and shape was observed about 2.0-2.2 ms after the first spike generation; therefore this interval may present the sum of absolute and relative refractory periods for a tonic neuron after impulse generation. After excitatory response to the click tonic neurons were under relatively prolonged (over 100 ms) suppression caused by IPSP development (Fig. 12D-F). As was mentioned above, the duration of impulse response of tonic neurons was determined by the duration of sound stimulation (Fig. 13A). Intracellular records proved that changes in tone duration evoked corresponding changes in the duration of EPSP and spike response of the neuron (Fig. 13B). Prolonged sound stimulation (over 200 ms) usually evoked gradual decrease of EPSP amplitude and impulse response frequency. Two per cent of neurons in the AI area expressed steady suppression of impulse activity sustained throughout the duration of tone stimulation, and no

A

j

excitatory effects. Intra~llular recording showed that such types of response are determined by the primary synaptic inhibition developing in these neurons (Fig. 14B). EPSPs generated in such neurons by tone onset were interrupted by high-amplitude and short-latent IPSPs during which (and, moreover, throughout the BF tone stimulation) the impulse activity of the unit studied was inhibited.

DISCUSSION

Columnar pattern of AI neuronal organization Results of the present study indicate that neurons encountered along the microelectrode track through different layers of auditory cortex are tuned to the same sound frequency (or very close frequencies). According to this criterion they can be attributed to the same cortical column. Such an interpretation agrees completely with the results obtained by other authors concerning the functional organization of auditory, ~'~s~'~'3°'35 visual ]~ and somatosensory 7,32 cortical areas. The existence of neuronal columns tuned to a certain BF in the auditory cortex is probably due to organization of sensory channels linking these neurons to a limited receptive field on the cochlear basilar membrane. At the same time, neurons apparently belonging to one column largely differ by the pattern of their responses to the BF tone. More than two-thirds of these neurons (69%) only responded

C

B 2.

I,

0

0 •

--~

)-

lores

25n~e ®

D

tO

E O

5ms E

F

2'

1.

~ o.

2ms

z

Time

315

lOOms

(ms)

Fig. 12. Dependence of two tonic neurons' response on the rising phase slope of best frequency tone. (A-D) Peristimulus histograms of first neuron reactions. (A-C) To tone stimulation with different rising phase slope (25, 10 and 5 ms, respectively). (D) To sound click (arrow points to the click onset). (E) Fragment of the record of extrae~ulax reaction of the given neuron to dick; reaponse latency is 9 ms. Abscism: time (ms). Ordinate: number of impulses in a bin. Bin is 10 ms. Histograms are normalized to number of stimuli presented. Down: sound stimulation mark. (F) Other neurons' intracellular reaction to click.

316

1. O. VOLKOVand A. V. GALAZJUK

A

101

B

0.5 ¢) m

¢,j

1.0

E 0

$ E Z

0.5

1.0] O'

l

0.5

Time (ms)

250ms

Fig. 13. Reactions of two AI tonic neurons to best frequency tones of different duration. (A) Petistimulus histograms of first neuron tonic reactions to tone stimuli of 150, 300 and 600 ms duration. Ab~ssa: time (ms). Ordinate: number of impulses in a bin. Bin is 10 ms. Histograms are normalized to number of stimuli presented. (B) Intracellular reactions of second neuron to tone stimuli of 150, 400 and 800 ms duration. Down: sound stimulation mark. to tone onset and/or offset while others (26%) generated spikes throughout the action of sound stimulation. The distribution of neurons with these types of response to tone in different cortical layers is obviously specific. The highest density of phasic neurons (about 50%) corresponds to cortical layers III and IV, whereas the overwhelming majority of tonic neurons (94%) are located in the deeper layers V and VI. Neurons with different types of response were found to have different abilities to discriminate sound

"t

80' "5 t-i

E_

A

B

L_L

t-I

E -i z 0

frequency. The most pointed tuning curves were determined for phasic units, whereas the curves for tonic neurons in most cases overlapped a wide range of sound frequencies used. These data indicate that neurons combined in one cortical column according to their frequency (or cochleotopical) characteristics can play different roles in auditory function. This is consistent with morphological and anatomical studies showing neuronal heterogeneity of the cat's auditory cortex and complex organization of synapses and interneuronal connections) '~°3~4~~

Time(ms)

10

lOOma

Fig. 14. Inhibitory response of AI neuron to best frequency tone. (A) Peristimulus histograms of neuron extracelhilar reactions to tone stimulus. Abscissa: time (ms). Ordinate: number of impulses in a bin. Bin is I 0 ms. Repetition number = 150. 0B) Intraoellular reaction of the same neuron to tone stimulus. Down: sound stimulation mark.

Formation of spike response in auditory cortex neurons

Formation of phasic pattern of reactions to tone stimuli Responses of phasic neurons to BF tone mediated by activation of receptors located in the centre of their receptive cochlear field in most phasic cortical units (93%) evoked short spike responses to the sound onset. Analysis of intracellular events shows that such a response pattern is due to the interaction

of complex p r o c e ~ occurring on the postsynaptic membrane of these cells caused by synaptic actions resulting from activation of excitatory and inhibitory synapses by afferent impulses coming from the thalamus and other subcortical structures of the auditory pathway. Excitation-inhibition relationships in a neuron at different time intervals of sound presentation are not uniform. At the initial period of neuronal reaction to BF tone the excitatory influence was more powerful and led to high-amplitude EPSPs (up to 10 mV) and successive spike output activity; the delay of action potential generated on the top of EPSP or on its rising phase in relation to the EPSP onset in the case of BF stimulation did not exceed 0.4-0.8 ms. Mechanisms responsible for such dynamics of the initial phase of neuronal response to BF tone should be investigated further. It is obvious that organiTation of synaptic inputs from the centre of the receptive field must supply very powerful excitatory action upon AI cortical neurons: 5.'~ At the same time the intensity of the inhibitory drive from this central zone of the receptive field is maximal, 43"47'¢ but the existence of a time lag between the onset of EPSP and I I S P leads to a period of overwhelming excitatory domination at the beginning of the neuronal reaction to the sound presentation. Comparison of the latencies of EISPs and IISPs evoked in auditory cortical neurons in response to acoustic stimulation 38'4t'43 and electrical stimulation of geniculate cortical pathway fibres'm'~9 shows that inhibitory drive following afferent burst arrival always shows an additional delay of about 0.8 ms. Some authors ~'4~'3 consider that such a difference between IPSP and E I S P latencies can be due to the presence of additional interneurons in the excitatory chain. However, the question as to whether such an insignificant difference between the arrival of excitatory and inhibitory synaptic influences upon AI neurons is the only reason for the predominance of the excitatory process over the inhibitory one at the initial phase of reaction to the BF tone is still open. The second part of the response of phasic AI neurons is based mainly on postsyuaptic inhibition developing after spike generation. The IPSP is usually intensive and prolonged enough to block further spike generation; such a succession of syuaptic events effectively forms spike response only to the tone onset. Thus, it must be emphasized that inhibitory processes play a decisive role in the formation of a phasic pattern of activity of the auditory cortex neurons. Interaction between excitation and inhibition allows a phasic cortical neuron to respond

317

by stable impulse reaction to the sound onset when sound frequency is the best, or close to it, for this neuron. "~-a Shift of the tone frequency in both directions from the BF corresponds to the displacement of the activated site on the ba~lar membrane from the centre to the periphery of the neuronal rc~'ptive field: 5 This is accompanied by a rapid decrease of firing index of the neuron studied. Intracellular recordings prove that such changes of the neuronal responsiveness result from a reduction of the E I S P amplitude and the slope. A decrease of synaptic excitatory drive upon cortical neurons in this case is probably due to a significant reduction of the quantity of activated cochlear receptors. At the same time, a relative intensity of inhibitory synaptic influence on this unit becomes greater. Further shift of the tone frequency resulted in a complete loss of the neuronal ability to generate spikes in response to such stimulation. In this case the neuron demonstrated an EPSP-IPSP succession only when subthreshold EPSP preceded the inhibitory synaptic component. The range of tone frequencies that evoked noticeable inhibitory responses in phasic neurons was about 2-3 times wider than that of the evoking spike generation. ~'47 Thus, when the periphery of the cochlear receptive field of phasic neurons is activated by non-BF tones, the balance between excitation and inhibition processes is shifted towards the inhibitory process. Such inhibitory drive, evoked by peripheral activation of the receptive field, does not result from the preceding spike excitation and therefore is not the postactivation inhibitory drive neuron. This type of inhibitory action has already been described in many studies concerning the principles of receptive field organization of cortical neurons belonging to different sensory projection systems. 14"m7'3~'47'~ This phenomenon is called lateral inhibition. 1'~3 The effectiveness of inhibition developing in phasic neurons after spike generation or in its absence varied during the period of presentation of prolonged tone stimuli. The most profound inhibition was observed during the initial 100-150 ms of sound stimulus and, as a rule, coincided in time with IPSP developing on the postsynaptic membrane. Later the membrane potential reduced to about its resting potential level and was accompanied by some decrease of inhibitory influence on the neuron. However, during this period of sound stimulation the neuronal responsiveness was much lower as compared with the initial one. Ionic and synaptic mechanisms of inhibition developing in cortical neurons in the absence of membrane hyperpolarization are still poorly understood. Some authors consider that this type of inhibition can be related to the inhibitory processes occurring on dendrites: 2'43'5~'s9It is thought that IPSPs generated via inhibitory synapses on the dendrites, especially on their distal parts, cannot appreciably affect the membrane potential level because of the considerable distance of the dendrites from the soma.

318

1. O. VOLKOVand A. V. GALAZJUK

At the same time it cannot be excluded that in this case inhibitory synaptic events are not necessarily localized in the cortical neuron itself; postsynaptic inhibition may occur in subcortical auditory structures, mainly in relay neurons of the medial geniculate body. It is known that in the majority of thalamic relay cells the ON reaction to afferent stimulation is followed by inhibition, which can effectively prevent the passage of subsequent afferent impulses to the cerebral cortex. 6"4~'44 A great part of the phasic neurons (43%) responded to offset of sound stimulus by O F F reactions. The start as well as the functional meaning of this component of the reaction of auditory cortical neurons are not clear so far. Since the OFF response is usually observed at the end of the inhibitory pause, one can assume that this event represents a passive consequence of postinhibitory exaltation (rebound). -~ This can be confirmed by our data obtained when studying the reactions of auditory cortical neurons under deep barbiturate anaesthesia. ~ Under these conditions neurons of acoustic cortex reversibly lose their ability to generate typical short-latent O F F responses to the tone offset. Instead, a specific reverberatory type of spike reaction to sound stimuli appears, consisting of alternating EPSP-spike--IPSP complexes. This type of activity is operated completely by thalamic rhythmic neuronal generators. 2'~s'44 The first component of late reaction that appears in the neurons of the anaesthetized animal after the ON response is sometimes classified as OFF response; in fact, this component is not a result of tone offset and can be regarded as the manifestation of postinhibitory rebound. 2'4~ Such a conclusion is supported by the pattern of reactions to tones of different duration; in this case the latency of the late discharge remains unchanged because of the stable duration of the IPSP after ON discharge. Of interest is a high sensitivity of phasic neurons to the changes in the rate of rise of the acoustic signal. The responsiveness of these neurons decreased with the prolongation of the tone rising phase. When BF tone was applied the rate of rise was 30-50 ms and the majority of phasic neurons were unable to yield an impulse response to sound onset. Such a reduction of neuronal responsiveness is probably the effect of summation for two main reasons. First, presentation of such a slow-rising acoustic stimulus results in the reduction of the quantity of cochlear receptors generating spikes because of their adaptation. This, in turn, leads to the reduction of the afferent burst coming via relay structures to the auditory cortex. Second, the intensity of the inhibitory influence declines, in this case more slowly than the excitatory drive, and therefore the relative contribution of inhibitory actions becomes higher. Thus, the reactions of phasic neurons to tones of different parameters allows us to assume that this population of auditory cortical neurons with high ~nsitivity to changes in sound frequency and

amplitude provides a rapid perception of sound appearance and analysis of its frequency characteristics. These cells give this information to other neurons of the same and neighbouring neuronal columns in the auditory cortex and to other cerebral structures. Neurons with similar features have been described in different subcorticai formations of auditory afferent pathway by other authors. 4"jl 13"20'~-4'3~'396° They represent a widely distributed specific cell type in the acoustic system. The higher the organization level of auditory centres the greater is the proportion of neurons with such properties. This indicates a special role for these nerve cells in the analysis and processing of auditory information. Formation o f tonic pattern oJ reactions to tone stimuli

Tonic neurons, in contrast to phasic ones, generate impulses throughout the period of sound presentation. Such a pattern of response is mainly based on a high amplitude (up to 10mV) plateau-like depolarization that can be recorded intracellulary from these neurons in response to tone presentation. Variation of tonic stimulus duration led to the corresponding changes in both EPSP time course and spike generation durations. A decline of discharge frequency in tonic neurons during prolonged sound stimulation is probably due to adaptation processes developing in these units. However. if sound stimuli lasted 1 s or more, the frequency of impulse activity in tonic neurons at a later stage of reaction was always significantly higher than that of the background discharge; this leads to the consideration of such neurons as slowly adapting units. This is in conformity with results reported by other authors ~2~'23~7for tonic neurons of lower parts of the auditory afferent pathway (cochlear nuclei, inferior colliculus and medial geniculate body). So, slov~ adaptation is the common feature of tonic neurons in all levels of the acoustic pathway. It is likely that tonic response rmght be determined by the pattern of afferent bursts coming from the medial geniculate body neurons or other subcortical auditory centres, since these structures have been shown to have neurons responding by tonic reaction to sound stimulus. 13~2~':3"24'3"In faw~ur of this suggestion is the fact that tonic impulsation entering the somatosensory cortical area in response to natural stimulation of cutaneous receptors has been found in many (about 50*/o) afferent fibres of the thalamocortical tract. ~ Thus, tonic input is a common property of other cortical projection areas. Although tonic reactions require obligatory participation of both excitatory and inhibitory processes, but at stable membrane potential and without deep barbiturate anaesthesia, excitation predominates considerably over inhibition in tonic reaction. 54 The study of the organization of the receptive fields of tonic neurons confirms the prevalence of excitatory inputs over inhibitory ones. Our previous investigations showed that tonic neurons of

Formation of spike response in auditory cortex neurons the cat auditory area AI have wider cochlear receptive fields and lower values of Q~0 for BF as compared with phasic neurons, thus indicating their lower capability to discriminate sound frequency, sv The differences in frequency selectivity of the two types of cortical neurons are probably due to the peculiarities of morpho-functional organization of their receptive fields. Thus, a wide convergence of excitatory inputs, the ability to generate spike discharges during the whole time of tone presentation, high sensitivity to the changes in the intensity and duration of the acoustic stimulus and a relatively short refractoriness (about 2 ms) of tonic neurons are evidence for their high excitability and functional lability. As we have shown before, the tonic neurons are characterized by lower thresholds of spike responses and shorter durations of action potentials as compared with phasic neurons.S7 Neurons with similar properties have been described by other authors in different projection zones of the cortex,zs's4'~'S°'~Neurons characterized

by "thin" spikes and predominance of excitatory inputs over inhibitory ones have also been detected in different layers of the auditory 3s and somatosensory cortex of monkeys and cats. ~.s° It has been shown in experiments using intracellular staining of physiologically identified neurons of the sensorimotor cortex of guinea-pig that the majority of responsive neurons which generate "thin" spikes are stellate neurons. ~ The authors considered these cells as the most likely candidates for GABAergic inhibitory cortical

319

neurons. However, it should be noted that properties such as neuronal lability and duration of action potential cannot serve as reliable criteria for electrophysiological identification of inhibitory cortical neurons, since a definite portion of stellate cells are known to be excitatory neurons in cortical input networks but none are known to be inhibitory. 19M'61'63 Besides, according to the data available, "thin" spikes can be generated not only by stellate neurons but also by some pyramidal cortical neurons) Cross-correlation analysis of the interaction between tonic neurons and other auditory neurons has shown that this population is functionally heterogeneous. ~ The excitation of tonic neurons has been shown to be accompanied either by excitation of the neurons located in the same or in the neighbouring column, or by their monosynaptic inhibition. Therefore, in the former case the tonic neuron could act as the excitatory unit and in the latter case as the inhibitory one. More precise determination of the inputs to these neurons from subcortical structures, their location in the intracortical transduction and processing chains of auditory information, and the study of morphofunctional organization of their synaptic connections with other auditory cortical neurons will make possible a more precise assessment of their function and role in the analysis of acoustic signals. Acknowledgements--The authors would like to thank Prof. Ph.N. Serkov and Drs M. Ja. Voloshin and D. A. Vasilenko for their support and helpful criticism of the manuscript.

RIEFERENCF~

I. Abcles M. and Goldstein M. N. (1970) Functional architecture in cat primary auditory cortex: columnar organization according to depth. J. Neurophysiol. 39, 172-187. 2. Andersen P., Andersson S. A. and Lomo T. (1967) Nature of thalamocortical relations during spontaneous barbiturate spindle activity. J. Physiol. 192, 283-307. 3. Antonova A. M. (1973) Neuronal organization of the auditory cortex in the cat. Archs Anat. 65, 21-32. 4. Clark G. M. and Dunlop C. W. (1969) Poststimulns-time response patterns in the nuclei of the cat superior olivary complex. Expl Neurol. 23, 266-290. 5. Dykes R. W., Lamour I., Diudori P., Landry P. and Dutar P. (1988) Somatosensory cortical neurons with an identifiable elcctrophysiologlcal signature. Brain Res. 441, 48-58. 6. Eccles J. (1971) The Inhibitory Pathways of the Central Nervous System. Mir, Moscow [in Rusman]. 7. Edelman G. M. and Mountcastle V. B. (1981) The Mindful Brain. Cortical Organization and Group-selective Theory of Higher Brain Fanction. Mir, Moscow [in Russian]. 8. Erulkar S. D., Ro~ J. E. and Davies P. W. (1956) Single unit activity in the auditory cortex of the cat. Bull. Johns Hopkina Hosp. 99, 55-86. 9. Evans E. F. and Whitficld J. C. (1964) Classification of unit l'~-sponsesin unanaesthetized and unrestrained cat. J. Physiol. 171, 476-493. I0. Genis E. D., Scrkov F. N. and Malsky V. A. (1973) On morphology of the cat auditory cortex. Neurophysiology, Kiev 5, 519-524 [in Russ/an]. 1I. Gcrshuni G. V. (I 965) Organization of afferent inflow and the process of discrimination of signals of various duration. Zh. vys~h, nerv. De)at. 15, 260-273 [in Rns~an]. 12. Goldstcin M. N., Hall J. L. and Butterfield B. O. (1968) Single unit activity in primary auditory cortex of unanacsthetized cats. J. acoust. Soc. Am. 43, 444-455. 13. Greenwood D. D. and Maruyama N. (1965) Excitatory and inhibitory response areas of auditory neurons in the nuckus. J. Neurophysiol. 28, 863-892. 14. Harthne H. K. and Rathff F. (1957) Inhibitory interaction of receptor units in the eye ofLimulus. J. gen. Physiol. 40, 357-376. 15. Hellweg F.-C., Schnitz W. and CrcutzfeldtO. D. (1977) Extraccllularand intraccllularrecordings from cat'scortical whisker projection area: thalamocortical response transformation. J. Neurophysiol. 40, 463~$79. 16. Hind .I.E. (1960) Unit activity in the auditory cortex. In Neural Mechanisms of the Auditory and Vestibular System,

pp. 201-210. Thomas, Springfield, IL.

320

I.O. VOL~OV and A. V. GALAZJLK

17. Hub¢l D. H. and Wiesel T. N. (1977) Functional architecture of macaque monkey cortex. Proc. R. Soc. Lond. BI98, 1-59.

18. Imig T. J. and Adrian H. O. (1977) Binaural columns in the primary field (AI) cat auditory cortex. Brain Res. 138, 241-257. 19. Jones E. G. (1975) Lamination and differential distribution of thalamic afferents within the sensorimotor cortex of the squirrel monkey. J. comp. Neurol. 160, 167-204. 20. Katsuki J., Murata K., Suga N. and Takenako T. (1959) Electrical activity of cortex auditory neurons of unanaesthetized and unrestrained cat. Proc. lmig. Japan Acad. 35, 571-574. 21. Katsuki J., Watanabe T. and Maruyama N. (1959) Activity of auditory neurons in upper levels of cat brain J. Neurophysiol. 22, 343-359. 22. Kelly J. P. and Wong D. (1981) Laminar connections of the cat's auditory cortex. Brain Res. 212, 1- 15. 23. Kiang N. J., Pfeiffer R. R., Warr W. B. and Backus A. S. (1965) Stimulus coding in the cochlear nucleus. Anat. Otol. Rhinol. Laryngol. 74, 463--485. 24. Kiang N. J., Watanabe T., Thomas E. C. and Clark L. F. (1962) Stimulus coding in the cat's auditory nerve. Anat. OtoL Rhinol. Laryngol. 71, 1009-1026. 25. Knight P. L. (1977) Representation of the cochlea within the anterior auditory field (AAF) of the cat. Brain Res. 130, 447-467. 26. Krnjcvi6 K., Randic M. and Straughau D. W. (1966) Nature of a cortical inhibition process. J. Physiol. 184, 49-77. 27. Land J. S. and Boethe R. G. (1975) Interlaminar connections and pyramidal neuron organization m the visual cortex area 17 of the macaque monkey. J. comp. Neurol. 159, 305-334. 28. MeCornick D. A., Connors B. W. and Lighthall J. W. (1985) Comparative electrophysiology of pyramidal and sparsely spiny stellat¢ neurons of the neocortex. J. Neurophysiol. 54, 782--806. 29. Merzenich M. M. and Brugge J. F. (1973) Representation of the cochlear portion on the superior temporal plane of the macaque monkey. Brain Res. 50, 275-296. 30. Merzenich M. M., Knight J. L. and Roth G. L. (1975) Representation of cochlear within primary auditory cortex in the cat. J. Neurophysiol. 38, 231-247. 31. Mitani A., Shimokoushi M., lton K., Nomurs S., Kudo M. and Mizuno N. (1985) Morphology and laminar organization of electrophysiologically identified neurons in the primary auditory cortex in the cat. J. comp. Neurol. 235, 430-447. 32. Mountcastle V. B. (1957) Modality and topographic properties of single neurons of cat's somatic sensory cortex. J. Neurophysiol. 20, 408-434. 33. Mountcastle V. B. (1964) Some functional properties of somatic afferent system. In Theory of Communication in Sensory Systems, pp. 185-213. Mir, Moskow [in Russian]. 34. Mountcastle V. B., Talbot W. N., Sakata H. and Hyv/iriven J. (1969) Cortical neuronal mechanisms studied in unanaesthetized monkeys. Neuronal periodicity and frequency discrimination. J. Neurophysiol. 32, 454-484. 35. Parker D. E. (I965) Vertical organization in the auditory cortex of the cat. J. audit. Res. 2, 99 t24. 36. Purpura D. P. and Shofer R. J. (1963) lntraeellular recording from thalamic neurons during reticulo-cortical activation. J. Neurophysiol. 26, 494--505. 37. Radionova E. A. (1971) Functional CharacterL+tic.~ of the Cochlear Nucleus Neurons and Audition. Nauka. Leningrad [in Russian]. 38. Ribaupierre F., Goldstein M. and Komshian G. {1972) Intracellular study of the cat's primary auditory cortex. Brain Res. "/8, 185-204. 39. Rose J. E., Galambos R. and Hughes J. R. (1959) Microelectrode studies of the cochlear nuclei ol the cat. Bull. Johns Hopkins Hosp. 104, 211-251. 40. Rose J. E., Greenwood D. D., Golberg J. and Hind J. E. (1963) Some discharge characteristics of single neurons in the inferior colliculus of the cat. I. Tonotopic organization, relation of spike-count to intensity and firing patterns of single element. J. Neurophy~iol. 26, 294-320. 41. Serkov F. N. (1977) Electrophysiology of Higher Subdivisions of the Acoustic System. Naukova Dumka. Kiev [in Russian]. 42. Serkov F. N. (1985) Nature and synaptic mechanisms of inhibition in cerebral cortex neurons. Phyziol. Zh. 31,513- 525 [in Russianl. 43. Serkov F. N. (1986) Cortical Inhibition. Naukova Dumka, Kiev [in Russianl. 44. Serkov F. N. and Kazakov V. N. (1980) Neurophysioiogy o f the Thalamus. Naukova Dumka, Kiev [in Russian]. 45. Serkov F. N. and Volkov I. O. (1983) Reactions of auditory cortical neurons to tones of different frequencies and electrical stimulation of the corresponding parts of the cochlea in cat. Neurophysiology, Kiev 15, 527. 534 [in Russian]. 46. Serkov F. N. and Volkov I. O. (1984) lntracellular responses of primary auditory cortex neurons in cat to tones of different frequencies and electrical stimulation of spiral ganglion fibres. Neurophysiology, Kiev 16, 123-131 [in Russian]. 47. Serkov F. N. and Volkov I. O. (1984) Characteristic of poststimulus and lateral inhibition in neurons of cat primary auditory cortex. Neurophysiology, Kiev 16, 193-201 [in Russian]. 48. Serkov F. N. and Volkov 1. O. (1985) Role of cortical inhibition in processing acoustic signals. Fizml. Zh. 31,569--578 [in Russian]. 49. Serkov F. N. and Yanovsky E. Sh. (1972) Responses of the auditory cortical neurons to stimulation of geniculocortical fibres. Neurophysiology, Kiev 4, 227-235 [in Russian]. 50. Simons D. S. (1978) Response properties of vibrissa units in rat SI somatosensory neocortex. J Neurophysiol. 41, 798-820. 51. Skrebitsky V. G. (1977) Regulation of Sensory Transmission in the Visual System. Meditsina, Moscow [in Russian]. 52. Sousa-Pinto A. (1973) Structure of the first auditory cortex (AI) in the cat. Archs ital. Biol. 11, 112-137. 53. Volkov I. O. (1982) Reactions of the cat auditory cortex neurons to electrical stimulation of nerve fibres innervating receptor tells of different parts of cochlear organ. Neurophysiology, Kiev 14, 417-425 [in Russian]. 54. Volkov I. O., Dembnovetzky O. F. and Galazjuk A. V. (1985) Patterns of responses of the cat auditory cortex neurons to tonal stimulation under nembutal anaesthesia and in the post,anaesthetized period. Neurophysiologv, Kiet' 17, 728 737 [in Russian].

Formation of spike response in auditory cortex neurons

321

55. Volkov I. O. and Galazjuk A. V. (1985) Responses of auditory cortex neurons to tones of the best frequency in non-anaesthetized cats. Neurophy~iology, Kiev 17, 500-508 [in Russian]. 56. Volkov I. O. and Galazjuk A. V. (1986) Electrical responses of neurons from the second acoustic field ofthe cat cerebral cortex to tones of the characteristic frequency. Phyzioi. Zh. 32, 733-741 [in Ru~an]. 57. Volkov I. O. and Galazjuk A. V. (1989) Response patterns of "tonic-type" neurons of acoustic cortex to tones of different frequencies and intensities. Neurophysiology, Kiev 21, 498-506 [in Rudder]. 58. Volkov I. O., Galazjuk A. V. and Dembnovetzky O. F. (1989) Interaction of neurons characterized by tonic response to sound with other neurons in auditory cortex of the cat. Neurophysiology, Kiev 21, 613-620 [in Rnui~n]. 59. Voronin L. L. and Skrebitaky V. G. (1967) Extra- and intracellular investigation of neuron responses in the motor cortex of alert rabbits to acoustic and photic slJmurt. Zh. vyssh, herr. Deyat. 17, 523-533 [in Russian]. 60. Whitfield I. C. (1967) The Auditory Pathway. Arnold, London. 61. Winer J. A. (1984) The pyramidal neurons in layer III of cat primary auditory cortex (AI). J. comp. Neurol. 229, 476-496. 62. Winer J. A. (1984) The non-pyramidal cells in layer III of cat primary auditory cortex (AI)..J. comp. Neurol. 229, 512-530. 63. Winer J. A. (1984) Anatomy of layer IV in cat primary cortex (AI). J. comp. Neuroi. 229, 535-567. 64. Yanovsky E. Sh. (1985) Inhibitory neurons in structural organi~tion of cortical inhibition. Phyziol. Zh. 31, 554-561 [in Russian].

(Accepted 10 October 1990)