[MM ' N 103, 229 THS LIBRARY xylem?! Michigan St?” University This is to certify that the thesis entitled ACOUSTIC ANALYSIS OF ARTICULATION PRODUCED UNDER THE INFLUENCE OF ORAL ANES'I‘HESIA presented by Gerald J. Brochu has been accepted towards fulfillment of the requirements for M.A. Audiology and Speech Sciences degree in Paul Alan Cooke, M.S. Major professor Date November 22, 1978 0-7639 ACOUSTIC ANALYSIS OF ARIICULATION PRODUCED UNDER THE INFLUENCE OF ORAL ANESTHESIA ‘By Gerald J. Brochu A THESIS Submitted to ' Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Audiology and Speech Sciences 1978 ABSTRACT ACOUSTIC ANALYSIS OF ARTUCULATION PRODUCED UNDER THE INFLUENCE OF ORAL ANESTHESIA BY Gerald J. Brochu The purpose of this study was to determine the effects of repeated oral sensory deprivation (via nerve block anesthesia) on speech output. The subjects were five normal speaking and hearing adults. They were required to read a list of words under four conditions: Injection Set 1, Pre and Post Anesthesia, and Injection Set 2, Pre and Post Anesthesia. The tape recorded words were analyzed using wide-band spectrograms. The vowel duration and fundamental frequency of words produced in the control conditions were compared to those produced in the test condi- tions. The results of the analysis revealed that there was little dif- ference between the vowel duration and fundamental frequency of words produced under the influence of oral sensory deprivation and words pro- duced under normal conditions. Also, there was no significant dif- ference between the speech produced in the test condition of Injection Set 1 as compared to Injection Set 2. The results of this study were interpreted to indicate that (l) Vowels are relatively feedback free and are more open-loop in nature. Gerald J. Brochu (2) Consonants, which require more precise articulatory configura- tions, are feedback dependent and are controlled by closed loop programming. (3) No compensatory adjustments were learned by the subjects between Injection Sets 1 and 2. This Thesis is dedicated to my parents, Mr. Gerald D. Brochu and Mrs. LaVerle J. Brochu ii ACKNOWLEDGEMENTS The author would like to express appreciation to Dr. Leo V. Deal for helping me throughout my undergraduate and graduate programs. Appreciation is also expressed to the members of my thesis committee (Paul Cooke, Dr. Leo V. Deal, Dr. Oscar T081, and Dr. Y. Pal Kapur). Special appreciation is expressed to Paul Cooke for his direction in the completion of this thesis. Special appreciation is also expressed to Dr. Daniel S. Beasley and Dr. John M. Hutchinson for their guidance in the initial stages of this study and to Dr. Edmund Hagen for volunteering his time and expertise in the administration of the anesthetic injections. Further appreciation is expressed to the subjects who participated in this study and to those who helped operate the study. 111 TABLE OF CONTENTS LIST OF TABLES . . . . . . LIST OF FIGURES. . . . . . . . . . . . . . . . MRODUCT ION O O O O O O O O O O O O O O O O 0 Open Loop System . . . . . . . . . . . . Closed Loop System . . . . . . . . . . . Simultaneous Open and Closed Loop System. Measuring The Motor Control of Speech Via Criticism Of The Use Of Oral Anesthesia - Statement Of The Problem. . . . . . . . EXPERIMENTAL PROCEDURES. . . . . . . . . . . . Subjects. . . . . . . . . . . . . . . . . Speech Stimuli. . . . . . . . . . . . . . Oral Anesthetization. . . . . . . . . . . Methods . . . . . . . . . . Recording Procedures. . . . . . . . . . Analysis Procedure. . . . . . . . . . Date Comparisons. . . . . . . . . . . . . RESULTS 0 O O O O O O O 0 O O O O O O O O 0 Reliability Of Measurements . . . . . . . Fundamental Frequency Of Vowels . . . . vowel Duration. . . . . . . . . . . . . . Summary Of Results. . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . Recommendations . . . . . . . . . . APPENDICES Appendix A. Human Consent Form. . . B. Medical History Information Form. . . C. Stimulus Word List. . . . . . . . . . iv Oral Anesthe sia Page . vi .vii H OOU‘l-l-‘NH . 13 .. 14 . 15 . 15 . 17 . 18 . 20 . 21 . 21 . 29 TABLE OF CONTENTS (cont.) Application Of Oral Anesthetic . Anesthetic Dosage Per Subject . Pre-dismissal Instructions. . . Speech Sound Development Chart. HEG‘J'TJNU LIST OF REFERENCES . Directions Prior To Anesthetization . Pre-stimulus Word Production Instructions 43 . 44 . 49 50 52 53 . 54 Table l. 2. LIST OF TABLES Summary of analysis of variance for fundamental frequency . Summary of analysis of variance for vowel duration . . . vi LIST OF FIGURES Figure 1, Mysak's model of a closed loop system . . . . . . . . 2. Average fundamental frequency across all vowels and subjects . . . . . . . . . . . . . . . . . . . . 3. Average change in fundamental frequency between pre- and post anesthesia conditions in injection sets 1 and 2 for all vowels . . . . . . . . . . . . . 4. Average vowel duration across all vowels and subjects 5. Average change in vowel duration between pre- and post anesthesia conditions in injection sets 1 and 2 for all vowels and subjects . . . . . . . vii INTRODUCTION In recent years there has been increasing interest in and contro- versy over the processes which underlie the motor control of speech production. Some investigators have theorized that speech production is controlled by a predominantly open loop system (Lashley, 1951; MacNeilage, 1970), whereas others have supported the concept of a predominantly closed loop speech production system.(Fairbanks, 1954; Sussman, 1972). Further, several investigators have supported the idea that there is a simultaneous operation of both closed and open loop speech production systems (Scott and Ringel, 1971; Putnam and Ringel, 1972; Putnam, 1973; Hutchinson and Putnam, 1974). Open Loop System Lashley (1951) theorized that motor movements, including those involved in the production of speech, are a result of both specific patterning and a general facilitation of the central nervous system. He contended that sensory factors (sensory feedback) play only a minor role in regulating the intensity and duration of the neural discharge responsible for motor movements. He also maintained that the initiation and timing of the contraction of muscles in a complex movement are also independent of sensory control. The open loop speech production system as described by MacNeilage (1970) depicts the motor movements of speech as pre-programmed. That is, a series of motor movements for a specific phoneme are stored in the higher motor-speech areas of the brain. When.a specific phoneme or series of phonemes (such as a syllable, word, phrase, or larger unit) is required for spech production, the motor area of the brain transmits a motor command to the speech production apparatus. The muscles involved in the speech production process then carry out this command, resulting in the performance of the desired speech move- ments with miminal reference to peripheral sensory data. Closed Loop System i The original theory of a closed loop speech production system was presented by Fairbanks (1954). He described a "closed cycle system or servosystem" in which the control mechanism employed feedback information from the output signal. The feedback information was compared to the output signal by the central mechanism so that the speech production mechanisms could produce speech which has the same form as the input. An extension of the model developed by Fairbanks was given by Mysak (1966), and is presented below in Figure 1. fig? '““ . m rzeLi-g] F mew-m um l![_ meflew;__l ' reusa- «2' l l . ""$:‘.'::"' Iii was.“ I" “nu“, «cum I i mun-non [l' i nan-sun" “i g 8 l «5M I I o , m r ' . ' 'i g : 3 i —-. - . :11. '":"-—..‘:.'€.':.—-l'. "1" "ta—ma“ 2 1;: L01. . I L_,_J%3 I" g j" fig—“J; / ' Mu- I I i ' Man i - i i L ' : :tf-a‘l' % :.....' ' - E {an I I E but; : E: if : “no. can» .i 3 E v» -------- -‘ ”ease? I i- :2 s. ......... (Duh-A10. i 'z , . ""9- I sense» and bin}; .3 I :-.-._ 0.- l l L. “0". A!" . ~ ' : '"-L_ji_;!3 n__T_J: . h I l l I v I L _ I“. | l l I l | L r l L Figure 1. Mysak's model of a closed loop system. According to Mysak, the receptor unit is the first part of the internal loop. This section processes radiant energy via the eye, sound pressure energy via the ear, and mechanical energy via the touch receptors. The second section of the system is the integrator which is composed of three basic units. The Phase 1 and Phase 2 integrators store incoming speech sounds or other percepts. Phase 1 integration involves the recognizing and the attaching of significance to incoming stimuli. Phase 2 integration involves the interpretation of the incoming stimuli. The speech content error signal mechanism compares the content of the actual speech with a prescribed speech content and determines whether there is any error in the speech signal. The next section, the transmitter, is composed of three parts. The Phase 2 transmitter component receives ideas of speech intentions from the integrator unit which automatically excites word patterns in the Phase 2 transmitter. The Phase 2 transmitter then activates the appropriate signals in the Phase 1 transmitter. The third part of the transmitter operates as a corrector device which can send error signals to the speech product corrector if there is an error in the speech signal or, if the signal is correct, can trigger the next command. The corrected signal is then sent to the effector unit which is responsible for the production of speech events. It is composed of three units: the motor unit, which is responsible for the column of air necessary for speech; the generator, which is responsible for the vibrating of the air column or voicing; and the modulator, which is responsible for breaking up the voiced or unvoiced air stream into articulatory units. The sensor unit is the last section of the internal loop system and is responsible for feedback of the speech product and speech content data. Sensor 1 is responsible for auditory feedback, sensory 2 for tactile feedback, and sensor 3 for kinesthetic feedback. This completes the entire cycle of the speech mechanism. When all of the components of the system are operating, the resulting speech is normal. However, a breakdown in any one of the units or a part of one of the units may cause a breakdown in the production of speech. The closed loop speech production system has also been described by Sussman (1972). He describes the closed loop system as a mechanism which continuously samples neural information on the spatial position of the articulators - their direction of movement, their rate of movement -- and sends this information to higher neural centers. This neural information is used to enable the speech production musculature to make the transitions from one phoneme to another. These transitions are made with prior knowledge of which phonemes are to follow one another, several phonemes in advance (coarticula- tion). In this sytem speech movements are not pre-programmed as in the open loop system but are made from moment to moment using positional, tactile, auditory, and other feedback information. Simultgneous Qpeg_Andelosgd Lobp_§ystem One of the descriptions of the simultaneous operation of an open and closed loop speech production system.was given by Scott and Ringel (1971). They explained that the open loop system is responsible for the initiation of the motor speech commands for all phonemes. The closed loop system operates as a servo-system which samples the speech Output and simultaneously makes refinements in the motor behavior of phonemes, especially those which require a more precise articula- tory placement, such as silibants and the phoneme /r/. Putnam and Ringel (1972) interpreted their data to support the hypothesis that open loop programming controls the initial portion of a word, including the vowel and the final portion of a single word utterance. The closed loop system then operated utilizing oral sensory feedback to refine the articulation in an utterance. Putnam (1973) interpreted her data to indicate that a speaker's ability to maintain intelligibility while under the influence of oral sensory anesthesia was qualitative evidence of the operation of both open and closed loop speech controls. She added "...that trigeminal feedback contributed to the unconscious closed-loop functions operating to refine articulation and to the conscious perception of tongue and lip positions." Measuring_the Motor Control of Speech Via Oral Anesthesia One of the tools employed by past investigators in forming hypotheses concerning the motor control of speech has been the use of topical and/or local anesthetization of the oral cavity. McCroskey (1958) used a perceptual analysis to measure the effects of oral anesthesia and delayed side-tone on the rate, articulatory accuracy, and intelligibility of speech. He found that in conditions involving the anesthetization of the articulators the largest adverse influence was on articulation. The mean number of correctly articulated words was reduced, which reduced intelligibility. The rate of speech was not significantly affected by the anesthesia condition but was affected adversely by the air-born delayed side-tone. McCroskey concluded that oral sensory information is an important factor in articulatory accuracy. Ringel and Steer (1963) used both topical and local anesthesia to alter the oral sensory information of speakers. They assessed articulation accuracy, duration, average peak intensity level, and fundamental frequency during conditions of oral sensory deprivation. Their findings indicated that both forms of oral anesthesia signifi- cantly affected speech output. Average peak intensity levels increased, along with articulation duration. The fundamental frequency, however, was not significantly affected. Errors of articulation occurred more frequently in the anesthesia condition, resulting from the injection of anesthetics into the oral region, than in the control (non-anesthe- tized) condition. Gammon et a1. (1971) measured the articulation ability and stress/ juncture characteristics during speech, i.e., the ability to produce correctly words which require the use of an intricate, highly structured set of stress rules, using speakers who were under the influence of "extensive" oral anesthetization. Their findings, based upon a per- ceptual analysis of the oral sensory deprived speech, indicated that the anesthesia did not reduce the speaker's ability to produce correct stress/juncture patterns in speech. However, they found that artic- ulatory production was disrupted by oral sensory deprivation. They concluded that stress/juncture rules affecting the articulatory mechanisms take place at a high decoding level and are largely feed- back free and that tactile feedback was more important for the produc- tion of consonants than it was for vowels. They contended that tongue position and configuration do not have to be as precise for correct vowel production as they do for the production of consonants. To date, this is the only study which found changes in the manner of articulation as opposed to the accuracy of articulation. Scott and Ringel (1971) used an acoustic and perceptual analysis of twenty-four bisyllabic words produced by speakers under the influence of oral sensory deprivation. They reported that the articulatory changes which occurred as a result of oral sensory deprivation were non-phonemic in nature. That is, the articulatory errors were only mildly inaccurate and occurred at the level of the articulatory mechanisms rather than at higher speech production centers. They concluded that the motor control of speech is accomplished by the simultaneous operation of both open and closed loop feedback systems. Putnam (1973) investigated the articulatory behavior of the lips and tongue under the influence of oral sensory anesthesia using cinera- diographic techniques. Her findings supported those of Scott and Ringel (1971), namely, that open loop and closed loop feedback mecha- nisms for articulatory control operate simultaneously. Horii et a1. (1973) employed digital speech processing procedures to compare vowel-to-consonant ratios (the amount of time which is spent in producing the vowels in a word over the amount of time which is spent producing the consonants), long and short time spectra, funda- mental frequency distributions, phonation time ratios, and rate of utterance of normal and sensory deprived speakers. During sensory deprived speech they found: "(1) a reduction in high frequency spectral components, (2) some temporal disorganization (primarily manifested as a decreased rate of utterance and prolongation of voiced syllabic nuclei), and (3) a higher and more variable fundamental frequency." They con- cluded that "Changes in spectral characteristics of some speech sounds are certainly correlated with changes in articulatory gestures" (p. 76). Hutchinson and Putnam (1974) measured the intraoral pressure, intraoral pressure durations, and air flow rate of speakers whose oral cavities had been anesthetized. They reported "very few" qualitative changes in aerodynamic patterns in oral sensory deprived speakers. Their conclusions based upon aerodynamic data were similar to those given by Scott and Ringel (1971), Putnam.and Ringel (1972), and Putnam (1973). That is, the temporal sequencing of speech is feedback free or open loop, whereas the closed loop information is required to provide precision in articulatory targets. Leanderson and Persson (1972) measured the effects of trigeminal nerve block anesthesia on the EMG activity of the facial muscles. "...assess the extent to which The purpose of their study was to somesthetic feedback mechanisms participate in the control of motor commands to lip muscles and labial speech gestures during coarticula- tion" (p. 2). They found that there was a general increase in the .amount of prespeech, background, and articulatory activity. They con- cluded that the EMG changes observed after the application of the anesthetic could not be explained by a disturbed peripheral closed loop feedback system. However, they added that their findings could be accounted for by a disturbed positional sense which could cause a normally unconscious articulatory control mechanism.to become conscious and cause the affected muscles to over-react.. Criticism of the Use of Oral Anesthesia Scott and Ringel (1971), Putnam and Ringel (1972), Bordon (1972), Leanderson and Persson (1972), Abbs (1973), Putnam (1973), and most recently, Hutchinson and Putnam (1974) have presented some of the criticisms of the use of oral anesthesia to produce oral sensory deprivation. These investigators have realized and discussed the complexities of the sensory-motor nerve communications within the oral mechanisms related to speech productions. The major criticism of the use of anesthesia to produce oral sensory deprivation has been that one cannot anesthetize the sensory nerves (the trigeminal nerve and its branches) without some motor nerve (the facial nerve and its branches) contamination. Furthermore, as Baumel (1974) reported, the exact communications between the motor and sensory nerves of the orc- facial areas cannot be accounted for in the literature available at the present time. Therefore, the extent of contamination of the orc- motor nerves as a result of oral sensory anesthesia is not presently known. Another variable which may exist in the motor-sensory contamina- tion controversy is the hypothesis presented by Putnam (1973). She suggests the possibility of a compensatory ore-facial motor adjustment which may result from the anesthetization of the arc-sensory nerves. This "over-reaction" of the oro-motor musculature as a result of the application of an oral anesthetic was also hypothesized by Leanderson and Persson (1972). Bordon et al. (1972) and Abbs (1973) contended that interpretation of data compiled in investigations utilizing methods of oral sensory 10 deprivation by the use of an anesthetic agent should be discontinued. They felt that more information must be compiled and analyzed concerning the extent of sensory-motor nerve contamination before hypotheses con- cerning speech production mechanisms could be formulated. This investigation is concerned with the compensatory movements of the speech production mechanisms while under the influence of oral sensory anesthesia. Statement of the Problem Several investigators (Scott and Ringel, 1971; Putnam and Ringel, 1972; Putnam, 1973; and Hutchinson and Puthan, 1974) have suggested that the oro—facial mechanisms responsible for speech production operate with feedback systems comprised of both open loop and closed loop operations. The reason for the acceptance of this hypotheses is evident. If only an open loop feedback system were in operation during speech production, application of oral sensory anesthesia would have no effect on the production of speech, assuming no motor involve- ment occurred as a result of the anesthesia. On the other hand, if- only a closed loop feedback system were in operation during speech production, the application of oral sensory anesthetics would cause a complete breakdown in speech production. For the purpose of this study, a feedback system composed of both open and closed loop opera- tions to control the speech production mechanisms is accepted. Once the motor commands for speech production, such as articula- tion, intensity, frequency, duration, and voicing of phonemes, are learned, they are stored as motor commands or target specifications in the speech production areas of the brain. The open system then 11 relays these invariant motor commands or target specifications to the mechanisms responsible for the production of speech. The closed loop system is then necessary for the refinement of the specified tar- get movements, especially articulatory movements. Investigations have yet to give a satisfactory explanation of the motor control of speech. However, the research completed thus far has provided insights into this complex process. It is important that future investigations of the motor control of speech, utilizing oral anesthesia techniques, employ an acoustic analysis as indicated by the results of Ringel and Steer, (1963), Scott and Ringel, (1971), and Horii et a1. (1973). The purpose of this study was to determine whether compensatory adjustments can be learned and successfully carried out by speakers under the influence of repeated oral sensory anesthesia. More specifi— cally, the following areas were investigated: 1. A comparison was made of the fundamental frequencies pro- duced under normal conditions to the fundamental frequencies produced under the influence of oral sensory anesthesia. 2. A comparison was made of vowel durations produced under normal conditions to vowel durations produced under the influence of oral sensory anesthesia. 3. A comparison was made of the fundamental frequencies pro- duced under the influence of oral sensory anesthesia to the fundamental frequencies of speech produced under the influence of a second application of oral sensory anesthesia. 4. A comparison was made of vowel durations produced under the influence or oral sensory anesthesia to vowel durations 12 produced under the influence of a second application of oral sensory anesthesia. EXPERIMENTAL PROCEDURES Subjects The subjects used in this study were five adult speakers, three males and two females, who spoke general American dialect as determined by a trained speech and language pathologist. The average age of the subjects was 22.2 years; four of the subjects were twenty-two years of age and one subject was twenty-three years of age. The subjects were all volunteers and signed a Human Consent Form (Appendix A). In addition, potential subjects were required to fill out a Medical History Information Form (Appendix B). Only individuals who met the following criteria, as reported on the above form, were allowed to participate in this study: 1. The subjects reported that they did not have any upper respiratory infections prior to the experimental procedures. 2. The subjects reported that they did not have a history of speech problems and also displayed no speech and/or language problems as determined by an informal evaluation performed by a trained speech and language pathologist. 3. The subjects had hearing within normal limits as determined by an audiological evaluation performed by a trained audio- logist. 4. The subjects reported that they did not have any extensive ore-facial surgery performed on them excluding traditional 13. l4 dental extractions. 5. The subjects had normal oral structures as determined by an oral examination performed by an oral surgeon. 6. The subjects reported that they had never had a history of unusual reactions to oral anesthetics, especially Xylocaine. 7. The subjects reported that they had never participated in an investigation involving the anesthetization of the oral cavity. 8. The subjects reported that they had no history of rheumatic fever, diabetes, cardiac disease, prolonged bleeding ten- dencies, allergies to drugs, pregnancy, or were under a physician's care for depression or high blood pressure. 9. The subjects reported that they would not consume any alcoholic beverage or medication at least 24 hours prior to the experiment. Speech Stimuli The subjects were required to read a list of words (Appendix C) under a normal (control) condition and an experimental (oral anesthesia) condition on two separate occasions. The speech stimuli consisted of a list of 24 bisyllabic words from the spondee word list (C.I.D. Auditory Test Wel). The words were a representative sample of most of the consonant sounds of the English language exclusive of /v/, lbl, /z/, /3/. and In/, and all vowels were represented at least once. In addition, this word list was chosen because information including both percep- tual judgments (close phonetic transcription) and acoustic data w‘-—'-:1-5 15 . (wide band spectrograms) has previously been published by Scott and Ringel (1971). To control for inconsistencies in the production of the words contained in the word list, the subjects were instructed to refrain from practicing the words contained on the stimulus word list outside of the experimental conditions. Oral Anesthetization In the experimental conditions each subject's oral sensitivity was altered by means of dental injections administered by an oral surgeon. The physician was advised to use caution during the injec- tion procedures to avoid accidental contamination of the motor nerves prior to the onset of the experimental procedures. The application of the anesthetic was performed according to the same procedures described by Hutchinson (1973) (Appendix D). The test procedures were not started until the subjects had reported that they were insensitive to probing with a sterile hypo- dermic needle in the gingiva, lips, and tongue. The average anesthe- tic dosage for the subjects are shown in Appendix E. The subject's sensitivity was checked before and after the test procedures. If at any time after the initial administration of the anesthetic the subjects reported sensitivity in any part of the anesthetized area, the testing was discontinued and the subject received further anesthetization until he again reported insensitivity to the hypodermic probing. At no time were the subjects given injections of the anesthetic above 25 cc as recommended by the oral surgeon. Methods Prior to the administration of the anesthetic to the subjects in each of the two injection sets, an indepth explanation of the procedure 16 was read to each subject (Appendix F). Administration of the anesthe- tic and subsequent testing took place in the same building, however, the administration of the anesthetic took place in an otolaryngologist's office and the stimulus word recordings took place in an audiology test suite. The anesthetic was administered with the subjects seated in an otolaryngology examination chair. After the administration of the anesthetic, the tests of oral sensitivity were carried out. After it was determined that the subject's oral cavity and related structures had been effectively desensitized, the subjects were transferred from the examination chair to a wheelchair and taken directly to an I.A.C. audiology test suite. The subjects remained seated in the wheelchair and a stationary boom microphone was placed approximately 10 centimeters away from the subject's mouth. The subjects received verbal instructions to speak in a "normal" manner, as they would had they been talking under normal conditions and to speak each word, one at a time, only when a red light which was placed directly in front of them flashed (Appendix G). The subjects were then handed a typed list of stimulus words containing the twenty-four bisyllabic words. First, all twenty-four words appeared on the list three times. After each word had been listed three times, eight of these words appeared ten times in succession. The subjects were required to read the list of stimulus words in both the control and experimental conditions. The order in which these words appeared on the lists were re-ordered in each condition and injection set to prevent the effects of repeated intonational and/or stress characteristics. 17 Three injection sets were performed on each subject. Prior to each injection set the subjects read the stimulus words. This normal condition was used as a control procedure. Thus, the subjects read the stimulus words under six conditions: Injection set 1 -- pre and post-anesthesia, Injection set 2 -— pre and post-anesthesia, and Injection set 3 -~ pre and post-anesthesia. There was a 5 day interval between Injection set 1 and 2 and between 2 and 3. Because of a recording equipment failure and subsequent loss of data recorded during Injection set 1, this injection set was termed a practice trial. Injection sets 2 and 3 were then renumbered Injection set 1 and 2 for the purposes of this study. Following each of the three test sessions involving the use of anesthetics, the subjects were escorted to a "recovery room." Here the subjects were required to lie down and rest for no less than one hour. At no time were the subjects left unattended. Before the sub- jects were driven to his or her home from the test sight, a detailed explanation of the procedures they were to follow were read to them (Appendix H). Repordipg_Procedures The subjects were seated in an I.A.C. sound treated audiology test suite with a Bruel and Kjar 4145 microphone (characterized by a flat frequency response from 20 to 20K Hz) placed approximately 10 centimeters away at mouth level and 0 degrees azimuth. All speech stimuli produced by the subjects were recorded on an Ampex A G 600 tape recorder (characterized by a flat frequency response from 30 to 3K Hz) at 7 1/2 inches per second. The entire recording system was 18 calibrated after each injection set in the following manner: 1. A constant 60 dB tone sweeping all frequencies between 20 and 20K Hz was patched into a Bruel and Kjar micro- phone and recorded on an Ampex A G 600 tape recorder. 2. The signal was played back through a Bruel and Kjar 2112 Spectrometer into a Bruel and Kjar 2305 sound level recorder which served to record the frequency response of the system, followed by a graphic display on specially designed paper. Wide-band spectrograms were made directly from the original stimulus tapes on a Voiceprint Laboratories 469l-A spectrograph. All spectrograms were made in one session, and the spectrograph machine was calibrated prior to use in this study. Also, the recording/ playback heads of the spectrograph were cleaned after every tenth spectrogram to ensure accurate results. Analysis Procedure Wide-band spectrograms provided the means by which the acoustic variables of vowel duration and fundamental frequency were recorded for analysis. Spectrograms were made of the first, fifth, and tenth productions of the eight stimulus words which were produced ten times in succession in both the control and experimental (oral sensory anes- thesia) conditions of Injection sets 1 and 2. Vowel duration was measured on the spectrograms in the following manner: A vertical line was drawn from the baseline of the spectro- gram at the first appearance of darkness on the vowel formant which was considered to be the darkest or most defined for each vowel. Another vertical line was drawn for the baseline at the point where it appeared that the darkness of the formant was completed. These vertical lines 19 were considered to be the boundaries of the vowel for the purposes of this sutdy. The distance between the vertical lines was measured in millimeters and was considered to be the length of the vowel in milli- meters. If the vertical lines representing the length of the vowel fell between the millimeter markings on the measuring scale, the value of the next lowest full millimeter marking was considered to be the length of the vowel in millimeters. The length of the vowels were converted from millimeters to milli- seconds by using the following mathematical formula: 1. The length of the vowel formant in millimeters was represented - by N. 2. The total length of the spectrogram was 320 millimeters. 3. The total elapsed time of the spectrogram was 2.4 seconds, represented by T. 4. 320 8 N 2.4 T 5. N x 2.4 _?ER1_—_ - Length of vowel in seconds (T). Fundamental frequency was measured using the same vowel boundaries as were used for determining vowel duration. The following formula was used to arrive at the fundamental frequency of the vowels in Hz. 1. The number representing the length of the vowel in seconds (T sec. taken from the formula for deriving vowel duration) was changed to milliseconds by moving the decimal point three digits to the right (represented by T m sec.). 2. Count the number of vertical striations on the spectrogram that lay between the vertical boundaries of the vowel. The number of vertical striations was represented by (S striations). 20 3. S x 1000 T = Fundamental frequency (in Hz). A grid containing all possible values of striations in the vowel formants and all possible values for the length of the vowels in milli- meters was constructed to determine the fundamental frequency and durations of the vowels. This was done to enable the examiner to deter- mine the fundamental frequency and duration of the vowels without having to use the formula and thus minimizing the chances of mathematic and transcription errors. Dataygomparisons After all of the data were compiled and analyzed, the following comparisons were made for both fundamental frequency and vowel duration: 1. Pre-anesthesia to Post-anesthesia, Injection set 1. 2. Pre-anesthesia to Post-anesthesia, Injection set 2. 3. Post-anesthesia Injection set 1 to Post-anesthesia Injection set 2. 4. Individual subject variations for all subjects in all conditions in Injection set 1 and Injection set 2. RESULTS Reliability of Measurement A random sample of 20 spectrograms of stimulus words was selected for reliability purposes, representing 4% of the total number of 480 spectrograms. This sample included spectrograms of words produced by each subject in the Pre-anesthesia and Post- anesthesia conditions of Injection set 1 and Injection set 2. Photo-copies of these spectrograms were made utilizing a Xerox photo-copying machine. The examiner then measured, in the same manner as the original spectrograms were measured, the fundamen- tal frequency and vowel duration of each photo-copied stimulus word. The data obtained from the re-measured photo-copies of the original spectrograms were recorded along with the data obtained from the original spectrograms. A statistical analysis of the data obtained from the original and re-measured spectrograms was made using the Pearson r correlational formula. The Pearson r analysis revealed that the reliability of measurement for vowel duration was 0.78 and 0.94 for fundamental frequency of vowels. The figures derived from the Pearson r analysis were interpreted to indicate that the measurements for both vowel duration and fundamental frequency of vowels were reliable. 21 22 Fundamental Frequency of Vowels Figure 2 shows the mean fundamental frequency of vowels during Pre-anesthesia conditions 1 and 2 and Post-anesthesia conditions 1 and 2. This figure represents values across all eight vowels and all five subjects. As shown, the mean fundamental frequency remained essentially unchanged from Pre-anesthesia condition Injection set 1 (144Hz) to Pre-anesthesia condition Injection set 2 (l43Hz). Figure 1 also shows the relationship in mean fundamental fre- quency of vowels from Pre-anesthesia conditions 1 and 2 to Post- anesthesia conditions 1 and 2. During Injection set 1 the subjects averaged an increase of 4Hz (144 Hz to 148 Hz) from Pre- to Post- conditions. In Injections set 2 the mean fundamental frequency of vowels in the Post-anesthesia condition was 8 Hz higher than the fundamental frequency in the Pre-anesthesia condition (143Hz to 151 Hz). 151 X 150 149 ‘1... x 147 146 145 Fundamental Frequency (in Hz) 144 0 143 42 Injection Set 1 Injection set 2 0 = Pre-anesthesia X = Post-anesthesia Figure 2. Average fundamental frequency across all vowels and subjects 23 The data displayed in figure 2 was supported by an analysis of variance (ANOVA) whose summary appears in Table 1. As illustrated in that table, there was no significant difference between Injection set 1 and Injection set 2 at the 0.5 level of confidence. In addition, there was no significant difference between the Pre-anesthesia and Post-anesthesia conditions in either Injection set 1 or Injection set 2. Table 1. Summary of analysis of variance of fundamental frequency. _S_o_1_1£_c_e_ _D_l~_‘ _S_§ _M_S Fobs FOL I3.05 Injections ‘(A) l 6.05 6.05 1 7.71 Time (B) 1 186.05 186.05 2.2 7.71 Subjects \(8) 4 33376.28 334.05 32.2 Injections x Time (AxB) l 18.05 18.05 7.71 Injections x Subjects (AxS) 4 47.2 11.8 Time x subjects (BxS) 4 337.2 84.3 Inj. x sub. x Time (AxBxS) 4 2.25 0.56 TOTALS 19 33973 1788 Intersubject differences in fundamental frequency of vowels between Pre-anesthesia and Post-anesthesia conditions are depicted in Figure 3. The Y axis shows the amount of change that occurred as a result of the nerve block anesthetic to the subjects (Post- anesthesia condition). An examination of the values found in Figure 3 shows that changes greater than 10 Hz were found in subjects P. F. and B. M. The greatest changes between Pre-anesthesia and Post- anesthesia conditions were 20 Hz for Subject B. M. and 15 Hz for 24 X = Injection set 1 0 I Injection set 2 20 l 19 18 X 17 >~. l6 2 15 i) m 5:" 14 o :2 13 '3' 12 x U 8 11 5 10 g . k. 9 E. 8 3 7 c 3 5 0 U .5 5 4 3 2 1 0 1 J c: H - 1 0 -a a u >» - 2 a a g a as ' 3 x u'u a- u c: m - 4 X m s u cak.k. - 5 Subjects G. 0. P. F. B. M. D. M. M. M. Figure 3. Average change in fundamental frequency between pre- and post- anesthesia conditions in injection sets 1 and 2 for all vowels. 25 subject P. F. in Injection set 2. Hence, for subjects B. M. and P. F. there was an increase in fundamental frequency after nerve block anesthetic injections were administered (Post-anesthesia condition). Subjects G. 0., D. M., and M. M. were minimally affected by the application of nerve block anesthesia (Post-anesthesia condi- tion) as indicated by fundamental frequency difference values clustered between +6 Hz and -4 Hz. Vowel Duration Figure 4 illustrates the mean duration of vowels produced by speakers in the experiment. As shown in Figure 4, the two Pre- anesthesia conditions differed on an average of 4 milliseconds, 93 milliseconds for Injection.set 1 and compared to 97 milliseconds for Injection set 2. Figure 4 also shows the changes that occurred during the Post- anesthesia condition, i.e., after the nerve block anesthetic was applied. In Injection set 1 the mean Post-anesthesia vowel duration of 109 milliseconds was 16 milliseconds longer than the Pre-anesthesia mean vowel duration of 93 milliseconds. Also, in Injection set 2 the mean Post-anesthesia vowel duration of 113 milliseconds was 16 milliseconds longer than the Pre-anesthesia mean vowel duration of 97 milliseconds. A statistical analysis was accomplished on these data, and the analysis of variance (ANOVA) is summarized in Table 2. Although the main effects between Injection set 1 and Injection set 2 were non- significant, there was a significant effect between Pre-anesthesia and Post-anesthesia conditions at the 0.5 level of confidence. After 26 114 113 X 112 111 110 109 X 108 107 106 105 104 103 102 101 99 98 Vowel Duration (Milliseconds) 96 95 94 93 0 92 Injection set 1 Injection set 2 0 a Pre-anesthesia X a Post-anesthesia Figure 4. Average vowel duration across all vowels and subjects. 27 the application of nerve block anesthesia (Post-anesthesia condition), the vowels, on the average, were significantly longer than those vowels produced under normal circumstances (Pre-anesthesia condition). Individual differences appeared to vary across subjects. Figure 5 represents the average change in vowel durations between Pre- anesthesia and Post-anesthesia conditions in both Injections set 1 and Injection set 2. Four out of five subjects demonstrated a greater change during Injection set 2 in relationship to Injections set 1. Ira-l! M‘l’u—hfl- 0‘ 3 Among these four subjects the greatest change was 20 milliseconds for subjects B. M. and P. F. Subject D. M. displayed a reversal of this pattern. This subject displayed an average change of 50 milliseconds greater vowel duration change during Injection set 1. This was the greatest change from Pre-anesthesia to Post-anesthesia condition for all subjects. Table 2. Summary of analysis of variance for vowel duration. F F Source DE. SS. M§_ obs a 8.05 Injections (A) l 72 72 1 7.71 Time (B) 1 119.5 118.5 8.5 7.71 subjects (S) 4 11569 392.25 Injections x Time (AxB) l 1 1 0.01 Injections x subjects (AxS) 4 351 87.85 Time x Subjects (BxS) 4 556 139.0 Inj. x Subs. x Time (AxBxS) 4 392 85.5 TOTALS 19 4126 217.1 28 X 8 Injection set 1 0 = Injection set 2 50 X h‘ P‘ P‘ h‘ h) ax ~4 co so <3 #4 {a .43 ta ta s~ u: ee—— F‘ P“ n: u: Increase in Vowel Duration P‘ F‘ c: ha AA 7T f’""*_”'l CI—‘NUJJ-‘UIO‘NCDQ Decrease in Vowel Duration subjects Go 0. P. F. B. M. D. M. M. M. Figure 5. Average change in vowel duration between Pre- and Post- anesthesia conditions in infection sets 1 and 2 for all vowels and all subjects. 29 Summagy of Results The five subjects employed in this investigation were injected with a nerve block anesthetic on two separate occasions. On the average, before the anesthetic injections were applied, the subjects' fundamental frequency and vowel durations were essentially the same. After the anesthetic injections were applied, there were slight alteration in the subjects vowel productions. Fundamental frequency and vowel durations were greater during the Post-anesthesia conditions. While the average increase of 6 Hz was not significant for fundamental ”- rang-turning?“ frequency, the average increase of 16 milliseconds for vowel duration was barely significant. In addition, there was a wider variability within subjects on the vowel duration parameter than on the funda- mental frequency parameter. DISCUSSION A review of the data obtained from this investigation revealed the following: (1) No significant difference between the fundamental fre- quency of speech produced under normal conditions and speech produced under the influence of oral sensory anesthesia. (2) No significant difference between the vowel durations of speech produced under normal conditions and speech produced under the influence of oral sensory anesthesia. (3) No significant difference between the fundamental fre- quency of speech produced under the influence of oral sensory anesthesia and the fundamental frequency of speech produced under the influence of a second application of oral sensory anesthesia. ' (4) No significant difference between the vowel durations of speech produced under the influence or oral sensory anesthesia and the vowel durations of speech produced under the influence of a second application of oral sensory anesthesia. As illustrated in Figures 2 and 4, the values for fundamental frequency and vowel duration Pre-anesthesia (control condition) in 30 31 injections sets 1 and 2 were essentially the same. These figures revealed that the average fundamental frequency for all subjects during the Pre—anesthesia condition in Injection set 1 was 144 Hz, as compared to 143 Hz in the Pre-anesthesia conditions resulted in values of 93 milliseconds and 97 milliseconds for the two Pre- anesthesia conditions. The fact that the values in the two trials were very similar allows for comparisons between the Pre-anesthesia conditions and the Post-anesthesia conditions in the two trials. That is, one was able to determine whether the subjects were able to learn compensatory behaviors between the two injection sets. As shown in Figures 2 and 4, both fundamental frequency and vowel duration increased after the application of the oral sensory anesthetic, although statistical significance only occurred with the vowel duration values. In addition, the statistically significant difference between pre- and post-anesthesia conditions for vowel duration was enchanced by subject D. M. who exhibited an average increase in duration of 50 milliseconds in Injection set 1. Nine out of ten data points displayed increased vowel durations as a result of the application of oral sensory anesthesia in both Injection sets 1 and 2. This overall increase of vowel duration values during oral sensory deprivation supports the overall increase in physiological variables as reported by other investigators. Ringel and Steer (1963) reported increases in average peak level of intensity and articulation duration as the result of the applica- tion of both topical and local anesthetization of the oral cavity. 32 Horii egnal, (1973) reported that oral sensory deprived speakers displayed prolongations of voiced syllabic nuclei and higher funda- mental frequency. Leanderson and Persson (1972) found that during oral sensory deprivation there was a general increase in the amount of pre-speech background and articulatory activity. Hutchinson and Putnam (1974) found that during oral sensory deprivation speakers displayed consistently higher intraoral air pressure, intraoral air pressure duration, and increased oral air flow rates. Hutchinson and Putnam also discussed the possibility of "A subglottal compen- sation (an elevated respiratory driving force) for a reduced supra- glottal feedback load (oral sensory anesthesia)." Leanderson and Persson accounted for these increases caused by oral sensory anesthesia as a disturbed positional sense which caused the normal unconscious control mechanism to become conscious and cause the affected muscles to over-react. Putnam (1973) also describes the effects of oral sensory anesthesia as an over compensation of the ore-facial muscula- ture . The overall increases in psychological activity during oral sensory deprived speech is similar to the Lombard Effect. The Lombard Effect is a phenomenon whereby an individual will increase vocal intensity when their auditory feedback is reduced. Likewise, when there is a reduction in the amount of sensory information that the ore-facial muscles are receiving, there is an overall increase in their activity. As revealed in this study, there were slightly greater changes in duration and frequency during the application of 33 the oral sensory anesthetic when compared to normal speaking condi- tions. This may be interpreted to indicate that not only were the subjects not able to learn compensatory adjustments in reaction to the reduced oral sensory information but that the overcompensation as previously described was not enhanced by the second application of the anesthetic. It should be noted that the general increase in physiological variables found in other studies were substantially larger than those demonstrated in this investigation. Several explanations may account for these differences. VMany of the previously mentioned studies examined consonants, whereas this study focused specifically on vowels. If consonants are controlled via closed loop programming, then a reduction in oral sensation should have the greatest effect on consonants. Indeed, Scott and Ringel (1971) have demonstrated that sibilants and the phoneme /r/, which require a precise blade configuration, are most affected by oral sensory deprivation. Mbst consonants rely on lingua-alveolar, lingua-palatal, labial, labial dental, lingua-dental, or bilabial placement for correct production. It is possible that such tactile-dependent productions are goverened by a closed loop servo-system. Conversely, v0wels do not rely on accurate placement of the articulators, are not as affected by oral sensory deprivation, and may be more open loop in nature. Scott and Ringel (1972) provided information which may be used to support this possibility when they reported that close high front and back vowels were more affected than the less precise vowels. 34 In addition, Gammon et a1. (1971) concluded that oral sensory feed- back is necessary for consonants because they are more precise and that vowels are less reliant upon oral sensory information because of the less precise tongue position and configuration required. It may be assumed that the more accurate the placement needed for the production of a phoneme, the greater the need for closed loop information. In considering the order of emergence of speech sounds in child— ren, it is generally accepted that children develop the vowel sounds first followed by the consonants in a prescribed developmental order (Appendix I). The consonants lpl, /b/, and /m/ are the consonant sounds that are first learned by'the average child. These sounds are relatively simple and do not require precise control of the oro- facial musculature. However, as the child develops, he begins to make more and more complex sounds until he has the ability to produce the more complex sounds such as /s/, /z/, and /r/. These consonant sounds are relatively complex and may require more closed loop infor- mation to be produced correctly. In addition, it is possible that the very early learned vowel sounds become open loop in nature because they are learned at a very early age; and by the time a child is seven or eight years of age they become "pre-programmed." In comparison, more precise consonant sounds are learned at a later time (seven to eight years of age) and are more reliant upon closed loop feedback information to achieve accurate production. This closed loop feedback information continues 35 to be necessary for correct consonant production even in adulthood because of the increasing complexity of our utterances and because of the affects of coarticulation. Recommendations As the data acquired in this study were analyzed it became apparent that there may have been some measurable differences between the post anesthesia variables (fundamental frequency and vowel :3: duration) of the original Injection set 1 (data lost due to equipment F breakdown) and the post anesthesia variables of Injection set 2. It i— is possible that compensatory adjustments may have occurred between the first and second Injection sets and that no further compensation occurred between Injection sets 2 and 3 (present Injection sets 1 and 2). Future investigations utilizing repeated applications of an oral sensory anesthetic may reveal such adjustments. It may be possible that a perceptual analysis of the same stimulus words recorded in this study would reveal more information regarding the effects of oral sensory deprivation. Perhaps a percep- tual analysis would include a qualitative judgment of the articulatory accuracy and/or intelligibility of the speech produced under the influence of oral sensory anesthesia. Although in this study, via an acoustic analysis, it was found that the subjects did not learn compensatory adjustments between Injection set 1 and Injection set 2, perhaps some learned adjustments may be detected perceptually. Another means by which we may gain more information regarding the effects of oral sensory deprivation on speech production would be to duplicate this study increasing the number of injection sets. 36 This would increase the amount of practice each subject would have while under the influence or oral sensory anesthesia. This increased "practice time" may allow the subjects to learn compensatory adjust- ments that were not present over two applications of oral sensory anesthesia. Future investigators may also want to utilize more accurate means of measuring fundamental frequency. This may include narrow band spectrograms, oscilloscopic photograph tracings, photophonophenello- grams or any other process which may provide more accurate fundamental frequency values. Highly accurate fundamental frequency values may permit future investigators to detect small changes in fundamental frequency values which were undetected in this study. APPENDICES APPENDIX A HUMAN CONSENT FORM Subject's Name: Subject's Address: Investigators: Gerald J. Brochu (Principal Investigator) Y. Pal Kapur, M.D. Edmund H. Hagan, D.D.s. Daniel S. Beasley, Ph.D. John M. Hutchinson, Ph.D. 1. I, . being of sound mind and (Name of Subject) and years of age, do hereby consent to, authorize and request each of the investigators named above (and the agents, employees and fellow employees of each of them) to undertake and perform on the proposed procedure, treatment, research, or investigation (herein called "Procedure") identified and explained in the document entitled "ACOUSTIC ANALYSIS OF ARTICULATION PRODUCED UNDER THE INFLUENCE OF ORAL ANESTHESIA." 2. I have read the document referred to above, and I have been fully advised of the nature of the Procedure and the possible risks and complications involved in it, all of which risks and complications I hereby assume voluntarily. 3. I have not been coerced in any way to participate in this pro- cedure by Michigan State University, the Trustees of Michigan State‘ University, the Michigan State University Speech and Hearing Clinic, the investigators listed above, or anyone acting on their behalf. 4. I understand that this Procedure will not necessarily advantage me in any way. 37 38 APPENDIX A (continued) 5. I hereby understand that all personal information will remain confidential. 6. I understand that I am free to withdraw from this study at any- time without prejudice. Signed at Michigan State University this day of , 1975, in the presence of the witnesses whose {1 signatures appear below, opposite my signature. ‘ E L (Subject) Witnessed by: APPENDIX B MEDICAL HISTORY INFORMATION FORM Subject's Name: Subject's Birthdate: Subject's Sex: This information is required to ensure your safety during the test procedures and to enable to examiners to compile data relevant to this study. It is extremely important that you answer the follow- ing questions as accurately as possible. fr—m-m‘fntq (1) Have you ever had any speech and/or language problems? If you have, explain as accurately as possible. (2) Have you ever had any extensive ore-facial surgery or any come plications as a result of a dental extraction? If you have, please explain as accurately as possible. (3) Have you ever participated in an investigation involving the use of oral anesthesia? (4) Have you ever had any unusual reactions to local anesthetics. especially Xylocaine? If you have explain as accurately as possible. (5) Are you presently taking any medication? If you are, give the kind of medication and the frequency with which you take this medication. 39 40 Appendix B (continued) (6) Do you currently suffer from, or have you ever suffered from, rheumatic fever, diabetes, cardiac disease, prolonged bleeding tendencies, or allergic drug reactions? If you have, explain as accurately as possible. (7) Are you presently pregnant or under a physician's care for depression or high blood pressure? If you are, explain as accurately as possible. (8) Do you presently have a cold or other upper respiratory in- fection? "a _.__ [Pa-1r. u .1“. ii. nutmeg backbone oatmeal headlight hotdog toothbrush shipwreck hedgehog northwest birthday duckpond sundown APPENDIX C Stimulus Word List schoolboy airplane eggplant scarecrow earthquake whitewash footstool hardware woodwork drawbridge mushroom woodchuck Also, the eight words listed below were chosen to be repeated ten times in succession by each subject in each of conditions in both Injection sets because Scott and Ringel (1971) stated that the sounds most affected by oral sensory anesthesia were sibilants and the phoneme /r/ and the least affected was the phoneme /l/. words listed below contain these phonemes. mushroom whitewash shipwreck northwest 41 The eight oatmeal headlight footstool schoolboy ‘El SHIP. fi Mi” ' | 42 APPENDIX C (continued) Each subject was required to read each of the words listed above. These words were reordered so as not to be influenced by rhythmic or intonational variables in each of the conditions in Injection set 1 and Injection set 2. APPENDIX D Application of Oral Anesthetic The application of the anesthetic was performed according to the same procedure described by Hutchinson (1973). Sensory deprivation was achieved through a series of nerve block injections of Xylocaine (with Epin- ephrin l/lO0,000). The series consisted of (10 bi- laterial mandibular injections involving the lingual and inferior alveolar branches of the mandibular division of the trigeminal nerve (at the point distal to the ramification of the mylohyoid nerve) to elim- inate sensation of the lower lip, lower teeth, anter- ior two-thirds of the tongue, the skin of the cheek, and the mucous membrane of the mouth and lower gingiva; (2) bilateral posterior palatine injections involving the middle and posterior branches of the nasopalatine nerve to eliminate sensation of the mucous membrame of the velum and uvula; (3) medial nasopalatine in- jections to eliminate sensation in the mucous membrane of the alveolus and anterior hard palate; and (4) bi- lateral infraorbital injections to eliminate sensation of the upper lip and upper teeth. 43 APPENDIX E Anesthetic Dosage Per Subject Subject's Initials D.M. Injection set 1 Nerve Branch Right Side Left Side Mandibular 3.0 cc 2.0 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc Nasopalatine 0.3 cc Infraborbital 1.0 cc 1.0 cc TOTAL 4.9 cc 3.6 cc GRAND TOTAL 8.5 cc Injection set 2 Nerve Branch Right Side Left Side Mandibular 2.7 cc 2.7 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc Nasopalatine 0.3 cc Infraorbital 1.0 cc 1.0 cc TOTAL , b 4.6 cc “ 4.3 cc GRAND TOTAL 8.9 cc 44 45 APPENDIX E (continued) Anesthetic Dosage Per Subject Subject's Initials M.M. Injection set 1 Nerve Branch Right Side Left Side Mandibular 2.0 cc 2.0 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc Nasopalatine 0.3 cc Infraorbital 1.0 cc 1.0 cc TOTAL 3.9 cc 3.6 cc GRAND TOTAL 7.5 cc Injection set 2 Nerve Branch. Right Side Left Side Mandibular 2.2 cc 2.7 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc Nasopalatine 0.3 cc Infraorbital 0.8 cc 0.8 cc TOTAL 3.9 cc 4.1 cc GRAND TOTAL 8.0.cc 46 APPENDIX E (continued) Anesthetic Dosage Per Subject Subject's Initials P.F. Injection set 1 Nerve Branch Right Side Left Side Mandibular 3.0 cc 2.0 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc ‘ Nasopalatine 0.3 cc Infraorbital 1.0 cc 1.0 cc TOTAL 4.9 cc 3.6 cc GRAND TOTAL 8.5 cc Injection set 2 Nerve Branch Right Side Left Side Mandibular 1.5 cc 1.2 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc Nasopalatine 0.3 cc Infraorbital 0.8 cc 0.8 cc TOTAL 3.2 cc 2.6 cc GRAND TOTAL 5.8 cc Anesthetic Dosage Per Subject Subject's Initials APPENDIX E (continued) B.M. 47 Injection set 1 Nerve Branch Right Side Left Side Mandibular 2.0 cc 2.0 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc NasoPalatine 0.3 cc Infraorbital 1.0 cc 1.0 cc TOTAL 3.9 cc 3.6 cc GRAND TOTAL 7.5 cc Injection set 2 Nerve Branch Right Side Left Side Mandibular 1.2 cc 1.2 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.8 cc 0.8 cc Nasopalatine 0.3 cc Infraorbital 0.3 cc .3 cc TOTAL 2.9 cc .6 cc GRAND TOTAL 5.5 cc Subject's Initials Anesthetic Dosage Per Subject APPENDIX E (continued) G.O. 48 Injection set 1 Nerve Branch Right Side Left Side Mandibular 2.0 cc 2.0 cc Posterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc Nasopalatine 0.3 cc Infraorbital 1.0 cc 1.0 cc TOTAL 3.9 cc 3.6 cc GRAND TOTAL 7.5 cc Injection set 2 Nerve Branch Right side Left Side Mandibular 1.2 cc 1.2 cc APosterior Palatine 0.3 cc 0.3 cc Anterior Palatine 0.3 cc 0.3 cc Nasopalatine 0.3 cc Infraorbital 0.8 cc 0.8 cc TOTAL 2.9 cc 2.6 cc GRAND TOTAL 5.5 cc APPENDIX F Directions Prior to Anesthetization A series of dental injections, similar to those administered by your dentist prior to filling a cavity or pulling a tooth, will be administered to you by an oral surgeon. You may feel some pain and/or discomfort with the initial hypodermic insertion and sub- sequent testing for sensitivity. After the oral surgeon feels the anesthesia has taken full effect, he will give you a series of "tests" to ensure that your oral structures are completely anesthetized. One of these "tests" will involve the pricking of your oral cavity with a sterilized hypodermic needle. This will not cause pain unless the anesthetic has not reached its full potency or if its effects are wearing off. In either case, the pain will be minimal. You may feel somewhat self-conscious about your appearance and/or your speech after the anesthesia has taken its full effect. This will be due to the total de—sensitization of your oral cavity. When this occurs, remember that the persons involved in this study have, as their first priority, your safety. The completion of this study is of secondary importance. An oral surgedn will be present at all times during test procedures involving anesthetics to handle any difficulties which might arise. Again, let me stress that your safety is of prime importance. 49 APPENDIX G Pre-stimulus Word Production Instructions During the reading of the stimulus word lists try to remain still; do not turn your head to either side or look up or down. Speak directly into the microphone. There is a red light directly in front of you. Can you see it? After one of the examiners signals for you to begin you will watch the red light. Each time the red light flashes you will read one of the stimulus words on the list in front of you out loud. You are to read only one word at a time and only when the red light flashes. The light will flash at approximately three second intervals. This will give you ample time to read the word out loud and to place your finger on the next word which appears on the stimulus word list. We advise you to use your finger to mark the word you are to read next so you will not loose your place or get confused. You are to read the words on the stimulus word list beginning with the first word which appears at the top of column one at the left hand side of your stimulus word sheet. Read the words going down the column one until you reach the last word in that column. Then go on the column two and read the words in exactly the same manner. When you are finished reading the words in column two, go on to column three and read the words in the same manner as columns one and two. 50 51 APPENDIX G (continued) After you have completed reading all of the stimulus words on page one, you will go on and read the stimulus words on pages two and three in the same manner. You will be given time to turn the page. The red light will not signal you to read the stimulus words until you show the examiner, by raising your hand, that you are ready to resume reading the stimulus words. Nod your head "yes" if you understand the directions that have just been read to you. Nod your head "no" if you do not understand the directions that have just been read to you, and I will read them to you again. Again, let me remind you if you start to feel uncomfortable, stop reading immediately and tell the examiner who will be monitoring your activities at all times through the window directly in front of you. APPENDIX E Pre-dismissal Instructions One of the attendants will drive you home and escort you into your home. After you are in your home, you should not leave your home for at least two hours. It is especially important that you do not attempt to drive a vehicle for at least two hours after you have arrived at home. After you are in your home, you should not attempt to eat any solid foods or drink hot liquids for at least three hours. The reason for this is that your oral cavity sensitivity has been significantly reduced. If you were to eat solid food or drink a hot liquid, it would be possible for you to bite the insides of your mouth, especially your tongue, or to burn your mouth or throat because of the decreased sensitivity of your oral cavity. It is also possible that you may inhale food or drink into your trachea causing you to choke and/or perhaps suffocate. Again, I would like to stress that we are only concerned with your safety. We will call you at your home every hour, if necessary, until you report that all of the sensitivity of your oral cavity has been regained. 52 APPENDIX I Speech Sound Development Chart Speech Sound Approximate Age Sound Develops /m/ 3 years /n/ 3 years /p/ 3 years /h/ 3 years /W/ 3 years /b/ 4 years /k/ 4 years /g/ 4 years /f/ 4 years /y/ 5 years /n/ 5 years /d/ 5 years /1/ 6 years /t/ 6 years /I / 6 years /tI/ 6 years /dr/ /cl/ /b1/ /gl/ 6 years /r/ 7 years /V/ 7 years /d3/ 7 years [9/ 7 years /tr/ /st/ /sl/ /sw/ /sp/ 7 years /8/ 8 years /2/ 8 years /°/ 8 years 53 LIST OF REFERENCES LIST OF REFERENCES Abbs, James, H. "The Influence of the Gamma Motor System On Jaw Movements During Speech: A Theoretical Framework and Some Preliminary Observations," J. Speech Hear. Res., 1973, 16, 175-200. Baumel, Julian, J. "Trijeminal-Facial Nerve Communications," Arch. Otolaryngology, Jan. 1974, 99, 34-44. Borden, G. J. "Some Effects of Oral Anesthesia On Speech: An Electromyographic Investigation," Status Rep. Speech Res., 1972, SR 29/30 (Haskins Laboritories, New Haven Conn.). Borden, G. J., Harris, K. S. and Catena, L. "Electromyographic Study of Speech Musculature During Lingual Nerve Block," paper presented at Ann. Conv. Am. Speech Hear. Assoc., San Francisco, 1972. Fairbanks, Grant. "Systematic Research In Experimental Phonetics. *1. A Theory of the Speech Mechanism as a Servo-System," J. Speech Hear. Dis., 1954, 133-139. Gammon, s. A., Smith, P. J., Daniloff, R. G., and Kim, C. W. "Articulation and Stress/Juncture Production Under Oral Anesthetization and Masking," J. Speech Hear. Res., 1971, 14, 271-281. Horii, Y., House, S. S., Li, D. P., and Ringel, R. "Acoustic Characteristics of Speech Produced Without Oral Sensation," Hutchinson, J. M. "Effect of Oral Sensory Deprivation on Stuttering Behavior," Ph.D. Dissertation, Purdue University, 1973. Hutchinson, J. M., and Putnam, A. H. B. "Aerodynamic Aspects of Sensory Deprived Speech," J. Acoust. Soc. Am., 1974, Vol. 56, No. 5, Nove., 1612-1617. Lashley, K. S. "The Problem of Serial Order in Behavior." reprinted from Cerebral Mechanisms in Behavior, 1951, L. A. Jeffress, (ed.) Wiley, 112-136. 54 55 Leanderson & Persson. "The Effect of Trigeminal Nerve Block on the Articulatory Activity of Facial Muscles," in On the Functional Organization of Facial Muscles of Speech, R. Leanderson, Ed. 1972, (Karolinska Suukhuset, Stockholm, Sweden). MacNeilage, P. F. "The Motor Control of Serial Ordering of Speech," Psychological Review, 1970, 77, 182-196. McCroskey, E. D. "The Relative Contributions of Auditory and Tactile Cues to Certain Aspects of Speech," South Speech J., 1958, 24, 84-90. Mysak, E. D. "Speech Pathology and Feedback Theory," Charles C. Thomas, Publisher, Bannerstone House, East Lawrence Ave., Springfield, 111., 1966, 301-327. Putnam, A. H. B. "Articulation with Reduced Oral Sensory Control: A Cineradiographic Study," Ph.D. Dissertation, Purdue University, 1973. Putnam, A. H. B., and Ringel, R. "Some Observations of Articulation During Labial Sensory Deprivation," J. Speech Hear. Res., 1972, 14, 529-542. Ringel, R. "Oral Sensation and Perception: A Selective Review," Amer. Speech Hear. Ass. Reports, 1970, 5, 188-207. Ringel, R. and Steer, M. D. "Some Effects of Tactile and Auditory Alterations on Speech Output," J. Speech Hear. Res., 1963, 6, 368-378. Scott, C. M., and Ringel, R. "Articulation Without Oral Sensory Control," J. Speech Hear. Res., 1971, 14, 804-818. Scott, C. M., and Ringel, R. "The Effects of Motor and Sensory Disruptions on Speech: A Description of Articulation," J. Speech Hear. Res., 1971, 14, 819-828. Sussman, H. M. "What the Tongue Tells the Brain," Psycol. Bul., 1972, V01. 77, No. 4, 262-272. 08 0322 03 RI M" B"- Hi" V." mi “I” '- Mi U" l l jji llllllfllllll