Studies On The Auditory Response Properties To Different Sound Stimulus Patterns In Inferior Collicular Neurons

Under free field stimulation condition, we studied the auditory response properties of inferior collicular (IC) neurons of leaf-nosed bat (Hipposideros armiger), Miniopterus magnater, and mouse (Mus musculus) to different sound stimulus patterns using in vivo neuroelectrophysiological recording, and obtained the results as following.1. The recorded echolocation calls of the leaf-nosed bats were a constant frequency-frequency modulation (CF-FM) signal with 2-3 harmonics. The recoded CF components of three harmonics from 18 leaf-nosed bats (n=18) were 33.3±0.2 kHz,66.5±0.3 kHz, and 99.4±0.5 kHz, respectively, and the second one among these three harmonics was its dominant frequency. The best frequencies (BFs) had linear correlation with recording depth of all the recorded IC neurons (n=169, r=0.716, P<0.001). The neurons tuned at DF had lower minimal threshold (MT) and larger Qn values than other neurons. All the neurons could generate a single onset response (single-on response, SO) to both CF and FM sound stimuli. When using CF-FM sound stimulus, most (76%) neurons were still to generate a SO response to CF-FM stimulus, which were called SO responders, but the remaining (24%) neurons could generate severally an onset response (double-on (DO) response) to both CF and FM components of CF-FM sound, which were called DO responders. On the other hand, the DO responders have higher minimum threshold (MT) and longer latency to the alone FM stimulus than alone CF. The results showed that there was certain relationship between echolocation call and frequency tuning of IC neurons in the leaf-nosed bats, and the FM component of the CF-FM sound took part in shapping discharge pattern, latency and firing rate of IC neurons. The data also suggested that these two types of responders might play respectively different roles in echo analysis during hunting.2. Under CF and CF-FM stimulation conditions, rate-amplitude functions (RAF) of the leaf-nosed bats'IC neurons could be classified into three types, i.e monotonic, saturated, and non-monotonic. CF-FM stimulus induced higher proportions of the nonmonotonic and saturated than CF stimulus. Comparison of best amplitude (BA), dynamic range (DR), and slope of the RAEs obtained by CF and CF-FM stimuli, CF-FM stimulus decreased BA from 93.0±9.8 to 84.4±11.3 dB SPL (P<0.01), DR from 27.5±9.4 to 22.5±8.5 dB (P<0.05), and increased slope from 3.1±1.4 to 4.1±1.7%/dB (P<0.05). On the other hand, the result showed that FM component could increase firing rates of the IC neurons at less than 30 dB above MT while could decrease firing rates at more than 30 dB above MT (P<0.001). It is suggested that the mechanism underlying the firing rate increasing at lower amplitude could be a summational effect induced by partial frequencies of FM near CF, while the firing rate decreasing at higher amplitude was perhaps because of lateral inhibition activated by those frequencies of FM away from the CF at higher amplitude. Therefore, FM of the echolocation call of CF-FM bat could increase the sensitivity of IC neurons to sound amplitude.3. Generally, the frequency and amplitude of echo would change during echolocation, but the echo duration could keep no change, as such the echo duration was served as a "tag" to recognize a bat's own call. By presenting single sound stimulus of different presentation rate (PR) and pulse-echo pair stimulus of different duration to the Miniopterus magnater, we examined the duration tuning and forward masking of IC neurons. The duration tuning of most IC neurons were changed from all-pass or long-pass toward band-pass or short-pass and their critical bandwidths were decreased (P<0.001) with PR increasing. There was no any relationship between types of duration tuning and forward masking, the R value of forward masking increased with the masker's duration increasing for all types of duration tuning (P<0.001). These results indicated that the higher PR could induce a large proportion of short-pass and band-pass neurons and enhance the ability of the IC neurons to follow high PR. So, it was matched better with the echolocation of bats which increase their emitted pulse repletion rate with approaching target during hunting.4. The forward masking of IC neurongs in Miniopterus magnater decreased gradually with off-BF of masker frequency toward low and high frequency sides under pulse-echo pair stimulation condition. The neurons affected by maskers could be categorized into three types, i.e. low frequency side long-masking, high frequency side long-masking, and both side equal-masking neurons according to their half-band widths of low (half-band widthlow) and high (half-band widthhigh) frequency sides and the forward masking index "R" at 50% forward masking. There was linear correlated between half-band widths of low and high frequency sides in all neurons (n=24, r=0.47, P<0.05). Among theses neurons, half-band widthlow in 50% of the neurons was remarkable more than half-band widthhigh (P<0.001), half-band widthhigh, in 25% of the neurons was remarkable more than half-band widthlow (P<0.05), and half-band widths of low and high frequency sides in the remaining 25% of neurons were equal (P>0.05). These results suggested that forward masking caused by off-BF sound could be the base of depression induced by bat vocalization and precedence effect in the hearing.5. Under paired sound stimulation condition, the changes in membrane potentials of mouse IC neurons were studied. The types of action potentials (AP) elicited by single sound could be classified into the single (n=11), double (n=8) and train (n=5) while inhibitory postsynaptic potential (IPSP) evoked by sound stimuation could occur at pre- or/and post-AP firing. The AP firing to probe was completely recovery when the paired sound gap was longer than the duration of masker-evoked IPSP at post-AP in 58.3% of IC neurons. The AP firing of the remaining (41.7%) IC neuronms could be completely recovery at paired sound gap less than or equal to the duration of masker-evoked IPSP at post-AP. However, the response latency to probe could be completely recovery when paired sound gap was longer than the duration of masker-evoked IPSP at post-AP in almost all the IC neurons (23/24,95.8%). By statistical analysis, only latency recovery need paired sound gap longer than the duration of the masker-evoked IPSP at post-AP (P<0.05). These results suggested that the postsynaptic inhibition and membrane potential level of the IC neurons might participate in the forward masking.