Combined AM-and-FM Sensitivities at the Auditory Midbrain

碩士 === 國立成功大學 === 生理學研究所 === 88 === Abstract Speech sounds contain components of amplitude modulation (AM) and frequency modulation (FM) that are detected by many central auditory neurons. Hence AM and FM sounds are considered basic elements of communication signals. But they were oft...

Full description

Bibliographic Details
Main Authors: mei-fang Lee, 李美芳
Other Authors: Poon. P.W.F.
Format: Others
Language:en_US
Published: 2000
Online Access:http://ndltd.ncl.edu.tw/handle/15585425383710875642
id ndltd-TW-088NCKU0116002
record_format oai_dc
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 國立成功大學 === 生理學研究所 === 88 === Abstract Speech sounds contain components of amplitude modulation (AM) and frequency modulation (FM) that are detected by many central auditory neurons. Hence AM and FM sounds are considered basic elements of communication signals. But they were often studied independently and never together. Experimental evidence is not yet available for the existence of a combined sensitivity to a dual AM-and-FM sound, which could be important for speech sound coding. The aim of this study is to determine if there exist cells at the level of the auditory midbrain that are sensitive to both AM and FM tones. And if so, to determine the difference in the response to dual AM-and-FM sounds (i.e., an AM tone plus an FM tone) and to AM or FM sounds given alone. Electrophysiological recordings were performed on urethane-anesthetized rats (n=107) to study single cell responses at the inferior colliculus (IC) using computer-synthesized acoustic signals with exponential modulation envelops. The direction of modulation was chosen to be opposite between AM and FM. About 1/3 of 993 single units recorded in the IC responded to AM and FM stimuli when given separately. For the AM or FM responses, three general types of response were observed (i.e., “insensitive”, “mixed” and “specialized”). With a dual FM-and-AM stimulus, the so-called “combined AM-and-FM sensitivities” could be classified into two main types. One type appeared to be the result of complex interactions between AM and FM sensitivities, as the combined response looked markedly different from either the AM or FM response of the same cell. Whereas the other type resembled either AM or FM response in an intensity-dependent way. To relate the “combined AM-and-FM sensitivities” to the frequency and intensity selectivity of the cell, a slow tone sweep with systematic attenuation in level was used. Four types of response area (RA) were found, viz., simple, asymmetric, multi-band and empty RA. Interestingly some apparent relationships between RAs and AM or FM sensitivities were observed: firstly, similar AM and FM response types tended to co-exist for the same cell (e.g., AM-“insensitive” and FM “insensitive” responses were found together, so were AM-“specialized” and FM-“specialized” responses or AM-“mixed” and FM-“mixed” responses). Secondly, simple RAs associated more often with AM- and FM-“insensitive” cells, and complex RAs (e.g., multi-band and empty RAs) associated more often with AM- and FM-“specialized” cells or with AM- and FM-“mixed” cells. Thirdly, “combined AM-and-FM sensitivities” were more predictable for cells with simple RAs, based on the AM or FM response alone. In contrast, it was less predictable with multi-band RAs and literally unpredictable for cells with empty RAs. The unpredictable responses of “combined AM-and-FM sensitivities” could be due to non-linear interactions between the AM and FM responses. To explore the possible conditions which would allow an optimal non-linear interaction between the AM and FM responses, a number of stimulus parameters were tested (e.g., carrier frequency, intensity, velocity and direction of modulation). Results showed that: (1) interactions between AM and FM responses could be observed more clearly with pseudo-random modulation envelops. (2) By varying carrier frequency of AM at fixed FM stimulation, “combined AM-and-FM sensitivities” were found to be rather consistent across carrier frequencies. (3) Envelop periodicity of stimulation in general would not affect the “combined AM-and-FM sensitivities”. (4) Direction of FM in dual AM-and-FM stimulation would not markedly affect the “combined AM-and-FM sensitivities”. In a limited number of cells, responses to a recorded rat vocalization sound were also studied. The responses were often better than the artificial AM-and-FM stimuli. Even when a cell failed to respond to dual AM-and-FM stimuli, it could be activated by the vocalization sound, suggesting the existence of some untested combinations of AM and FM sound in the complex stimulus. In conclusion, there exist cells at the IC that respond to dual AM-and-FM stimuli. The responses likely involve complex interactions that occurred in a non-linear way. Such interactions could be important in coding multiple stimulus features of the biologically relevant sounds.
author2 Poon. P.W.F.
author_facet Poon. P.W.F.
mei-fang Lee
李美芳
author mei-fang Lee
李美芳
spellingShingle mei-fang Lee
李美芳
Combined AM-and-FM Sensitivities at the Auditory Midbrain
author_sort mei-fang Lee
title Combined AM-and-FM Sensitivities at the Auditory Midbrain
title_short Combined AM-and-FM Sensitivities at the Auditory Midbrain
title_full Combined AM-and-FM Sensitivities at the Auditory Midbrain
title_fullStr Combined AM-and-FM Sensitivities at the Auditory Midbrain
title_full_unstemmed Combined AM-and-FM Sensitivities at the Auditory Midbrain
title_sort combined am-and-fm sensitivities at the auditory midbrain
publishDate 2000
url http://ndltd.ncl.edu.tw/handle/15585425383710875642
work_keys_str_mv AT meifanglee combinedamandfmsensitivitiesattheauditorymidbrain
AT lǐměifāng combinedamandfmsensitivitiesattheauditorymidbrain
AT meifanglee zhōngnǎoduìhùnhédiàofújídiàopínshēngzhīfǎnyīngtèxìng
AT lǐměifāng zhōngnǎoduìhùnhédiàofújídiàopínshēngzhīfǎnyīngtèxìng
_version_ 1716834586321747968
spelling ndltd-TW-088NCKU01160022015-10-13T10:57:06Z http://ndltd.ncl.edu.tw/handle/15585425383710875642 Combined AM-and-FM Sensitivities at the Auditory Midbrain 中腦對混合調幅及調頻聲之反應特性 mei-fang Lee 李美芳 碩士 國立成功大學 生理學研究所 88 Abstract Speech sounds contain components of amplitude modulation (AM) and frequency modulation (FM) that are detected by many central auditory neurons. Hence AM and FM sounds are considered basic elements of communication signals. But they were often studied independently and never together. Experimental evidence is not yet available for the existence of a combined sensitivity to a dual AM-and-FM sound, which could be important for speech sound coding. The aim of this study is to determine if there exist cells at the level of the auditory midbrain that are sensitive to both AM and FM tones. And if so, to determine the difference in the response to dual AM-and-FM sounds (i.e., an AM tone plus an FM tone) and to AM or FM sounds given alone. Electrophysiological recordings were performed on urethane-anesthetized rats (n=107) to study single cell responses at the inferior colliculus (IC) using computer-synthesized acoustic signals with exponential modulation envelops. The direction of modulation was chosen to be opposite between AM and FM. About 1/3 of 993 single units recorded in the IC responded to AM and FM stimuli when given separately. For the AM or FM responses, three general types of response were observed (i.e., “insensitive”, “mixed” and “specialized”). With a dual FM-and-AM stimulus, the so-called “combined AM-and-FM sensitivities” could be classified into two main types. One type appeared to be the result of complex interactions between AM and FM sensitivities, as the combined response looked markedly different from either the AM or FM response of the same cell. Whereas the other type resembled either AM or FM response in an intensity-dependent way. To relate the “combined AM-and-FM sensitivities” to the frequency and intensity selectivity of the cell, a slow tone sweep with systematic attenuation in level was used. Four types of response area (RA) were found, viz., simple, asymmetric, multi-band and empty RA. Interestingly some apparent relationships between RAs and AM or FM sensitivities were observed: firstly, similar AM and FM response types tended to co-exist for the same cell (e.g., AM-“insensitive” and FM “insensitive” responses were found together, so were AM-“specialized” and FM-“specialized” responses or AM-“mixed” and FM-“mixed” responses). Secondly, simple RAs associated more often with AM- and FM-“insensitive” cells, and complex RAs (e.g., multi-band and empty RAs) associated more often with AM- and FM-“specialized” cells or with AM- and FM-“mixed” cells. Thirdly, “combined AM-and-FM sensitivities” were more predictable for cells with simple RAs, based on the AM or FM response alone. In contrast, it was less predictable with multi-band RAs and literally unpredictable for cells with empty RAs. The unpredictable responses of “combined AM-and-FM sensitivities” could be due to non-linear interactions between the AM and FM responses. To explore the possible conditions which would allow an optimal non-linear interaction between the AM and FM responses, a number of stimulus parameters were tested (e.g., carrier frequency, intensity, velocity and direction of modulation). Results showed that: (1) interactions between AM and FM responses could be observed more clearly with pseudo-random modulation envelops. (2) By varying carrier frequency of AM at fixed FM stimulation, “combined AM-and-FM sensitivities” were found to be rather consistent across carrier frequencies. (3) Envelop periodicity of stimulation in general would not affect the “combined AM-and-FM sensitivities”. (4) Direction of FM in dual AM-and-FM stimulation would not markedly affect the “combined AM-and-FM sensitivities”. In a limited number of cells, responses to a recorded rat vocalization sound were also studied. The responses were often better than the artificial AM-and-FM stimuli. Even when a cell failed to respond to dual AM-and-FM stimuli, it could be activated by the vocalization sound, suggesting the existence of some untested combinations of AM and FM sound in the complex stimulus. In conclusion, there exist cells at the IC that respond to dual AM-and-FM stimuli. The responses likely involve complex interactions that occurred in a non-linear way. Such interactions could be important in coding multiple stimulus features of the biologically relevant sounds. Poon. P.W.F. 潘偉豐 2000 學位論文 ; thesis 79 en_US