THE RELATIONSHIP OF AGING ON THE RATE OF VOCAL FATIGUE BASED ON PERSONAL RATING SCALES AND FUNDAMENTAL FREQUENCY By Olivia Rae Sowa A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Communicative Sciences and Disorders – Master of Arts 2018 ABSTRACT THE RELATIONSHIP OF AGING ON THE RATE OF VOCAL FATIGUE BASED ON PERSONAL RATING SCALES AND FUNDAMENTAL FREQUENCY By Olivia Rae Sowa INTRODUCTION: The relationship between aging and the quality of voice has been researched for the last several decades. Multiple studies have found that females older than 40 consistently experience more instances of voice problems than their male counterparts. RESEARCH QUESTION: to what degree does a vocal loading task fatigue older women? How does the rate of vocal fatigue compare to that found in younger women? METHODS: Eleven female subjects from the age of 55-70 years old were recruited. Participants read aloud for 36 minutes at predetermined dB levels in an attempt to induce vocal fatigue. Subjective and acoustic vocal measures were taken not before and after the vocal loading task and at intervals during the task. RESULTS: Subjective results, acoustical parameters, and comparison data from the younger population were found to not be significant. CONCLUSION: The results of the study showed that vocal loading task did vocally fatigue the participants based on subjective ratings. Acoustically there trending evidence of fatigue from the loading task. The aging female population did not vocally fatigue at any greater rate than their younger counterparts. However, there was evidence that fatigue rate did relate to pulmonary health (estimated lung age) in the older population. ACKNOWLEDGEMENTS I would like to acknowledge the kind cooperation of the participants that made this work possible. I want to acknowledge my family, Edward and friends for their unlimited support and understanding for the busy and crazy schedule that being a thesis candidate allows. Special thanks to Rachel Burtka, Stirling Witthoeft, Courtney Bodrie, and Elizabeth Tompkins for their assistance in making this protocol possible. My deepest thanks to the Ph.D candidates, Russell Banks, Mark Berardi, and David Ford for their hard work, and dedication to my study and countless hours in the lab. I want to thank Lady Catherine Cantor Cutiva for her assistance in the lab as well as for assistance with statistical analysis. I would like to extend my thanks to my thesis committee, Matthew Phillips, Dr. Peter LaPine, and Dr. Jeffrey Searl for guiding my study and asking all of the right questions. Finally, I would like to thank Dr. Eric Hunter for his guidance, teachings, many hours, and input over the past year in making this thesis possible. Thank you for sharing your love of voice and research with me, and for giving me the inspiration to continue to pursue research. I would not be the budding clinician I was without this amazing experience of being a thesis track student under your guidance. Research reported in this thesis document was supported by the College of Communication Arts and Sciences and the National Institute on Deafness and Other Communication Disorders of the National Institutes of Health under Award Number R01DC012315. Content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. iii TABLE OF CONTENTS LIST OF TABLES ............................................................................................................. vi LIST OF FIGURES .......................................................................................................... vii INTRODUCTION ...............................................................................................................1 What is Fatigue? ......................................................................................................1 What is Vocal Fatigue? ............................................................................................2 Attempts to Measure Vocal Fatigue ........................................................................2 Acoustic Measures - Relevant Studies ...................................................3 Aerodynamic Measures .........................................................................3 Visual Perceptual Judgement .................................................................4 Auditory-Perceptual Judgment .............................................................4 Self-Ratings of Vocal Fatigue ................................................................5 Females and Vocal Fatigue .....................................................................................5 Aging and Quality of Voice Function ......................................................................6 Research Question and Hypothesis ..........................................................................7 METHODS ..........................................................................................................................9 Participants ...............................................................................................................9 Pre-Vocal Loading Task ..........................................................................................9 Vocal Loading Task ...............................................................................................10 Post- Vocal Loading Task ......................................................................................12 Measures of Interest ...............................................................................................12 Measurement and Data Analysis ...........................................................................14 RESULTS ..........................................................................................................................16 Participants .............................................................................................................16 Subjective Ratings by Participants.........................................................................17 Acoustical Parameters ............................................................................................19 Steady Vowel Analysis - Perturbation - Jitter and Shimmer ...............19 Running Speech –Rainbow Passage –Fundamental Frequency – Variation of dB ....................................................................................21 Rate of Perceived Vocal Fatigue and Lung Age in Aging Population .... ..............................................................................................................21 Comparison Data – Between Aging and College-Age Females ............................23 DISCUSSION ....................................................................................................................25 Subjective Ratings .................................................................................................25 Acoustical Parameters ............................................................................................26 Rate of Perceived Vocal Fatigue and Lung Age in Aging Population .... ..............................................................................................................27 Comparison Data (Old vs. Young) ........................................................................28 Limitations .............................................................................................................29 iv Future Research .....................................................................................................29 Clinical Applications .............................................................................................30 CONCLUSION ..................................................................................................................31 APPENDICES ..................................................................................................................32 APPENDIX A: Protocol Measures not statistically analyzed for this study .........33 APPENDIX B: Protocol Instructions.....................................................................35 APPENDIX C: Passages used for Vocal Loading Task ........................................42 APPENDIX D: Tables used in previous presentation but not imperative to this study .......................................................................................................................45 APPENDIX E: IRB Approval Form ......................................................................48 APPENDIX F: IRB Approved Recruitment Flyer.................................................50 APPENDIX G: IRB Approved Consent Form ......................................................52 APPENDIX H: Information and Biographical Sheet ............................................57 APPENDIX I: Respiration Form ...........................................................................59 APPENDIX J: Survey Questions ...........................................................................61 APPENDIX K: Specific information used in Methods Section ............................66 BIBLIOGRAPHY ..............................................................................................................73 v LIST OF TABLES Table 1. Participant information and screening data. Average age of participants was 65 years old with a standard deviation of 4.6. ........................................................................16 Table 2. Vocal Fatigue Ratings for each participant with calculated slope. .....................18 Table 3. Pre and Post measures analyzed with regression analysis using a Gamma Distribution with Beta values from post-loading task. ......................................................20 Table 4. Comparison data of aging population for correlational analysis. .......................22 Table 5. Correlational data table of the data from the tables above. A majority of the data that was analyzed was found to have a correlation greater than 50%. ..............................22 Table 6. Slope values compared between age groups. ......................................................24 Table 7. Participant’s lung age and actual age. .................................................................46 Table 8. Participant’s Vocal Fatigue Ratings (over time in minutes). ..............................47 vi LIST OF FIGURES Figure 1. Examples of the screen displayed to participants during the pre-, post- and vocal-loading task. The highlighted lettering “STAY ABOVE THE LINE” indicates the portions of the vocal loading task where participants were required to speak above 72 and 76 dB. Left: Step 7 with each “AAAH” presented individually per predetermined timing. Right: Step 8 indicating the passage to be read. ................................................................11 Figure 2. A visual of the protocol for the present study with specific data collected. ......13 Figure 3. Participants’ subjective ratings of their vocal fatigue over the vocal loading task over time. The lines indicate the slope. ......................................................................17 Figure 4. All participant’s individual vocal fatigue ratings over time. .............................18 Figure 5. Perturbation measures, Jitter and Shimmer average measures taken at pre- vocal, vocal loading and post-vocal loading task with slope lines provided. Note the insignificant change from pre- to post-vocal loading task with an overall decreasing trend line for both jitter and shimmer. ........................................................................................20 Figure 6. Comparison between the aging population and college-age population, vocal fatigue ratings over time. ...................................................................................................23 vii INTRODUCTION Extensive research has shown that females have a higher prevalence of voice disorders compared to men. The reason behind a greater rate of voice problems, particularly in those who use their voice as an occupational tool, has often been related to physiological differences, leading to vocal fatigue (Hunter et al, 2011). For example, the decline of pulmonary function can be a primary cause for vocal complaints and vocal fatigue in the elderly population (Mueller, 1997). Likewise, there have been studies that have established various relationships between the aging population and the quality of voice (Goy et al., 2016). The planned study aims to assess the fatigue ratings of an aging population of females. First, a review of vocal fatigue is presented followed by measures related to vocal fatigue. Next, voice production changes with aging is reviewed. Finally, the research question and hypothesis are presented. What is Fatigue? Before vocal fatigue can be defined, a general definition of fatigue will be presented for context. For this study, fatigue will be defined using the Merriam-Webster Dictionary’s definition, which describes fatigue as, “weariness or exhaustion from labor, exertion, or stress” (Merriam-Webster, 2017). Medical researchers refer to fatigue as “an overwhelming sense of tiredness, lack of energy, and feeling of exhaustion” (Cristian & Griswald, 2009, p. 1). Medical professionals categorize fatigue into two different categories based on symptoms. For example, fatigue with the key symptom as decreased maintenance of focused attention is defined as central fatigue. Fatigue that is characterized as reduced exercise tolerance is defined by medical researchers as peripheral fatigue (Cristian & Griswald, 2009). Specifically in this study, the fatigue of interest is vocal fatigue. Vocal fatigue, which includes some of the same descriptors as 1 in general fatigue, will be defined below in terms of physiology, acoustical and aerodynamic measures. For the remainder of the thesis, use of the word fatigue refers to vocal fatigue. What is Vocal Fatigue? Vocal fatigue is a concept that has been used in a variety of ways in the literature, thus it is without an agreed upon means of quantifying and there are not specific nor agreed upon indicators that it is part of a genetic disposition. It has been used both as a specific symptom and a vocal diagnosis. Vocal fatigue and the potential physiology has been studied and reviewed by Welham & Maclagan (2003), who found that vocal fatigue has both physiological and biomechanical components involved. Components of physiology such as neuromuscular fatigue of the laryngeal musculature, increased vocal fold viscosity, reduced blood circulation to the vocal process and surround structures, and non-muscular tissue strain are all elements shown to be involved in progression of vocal fatigue (Welham & Maclagan, 2003). The definition of vocal fatigue can, additionally, be described by the symptoms one presents with when experiencing vocal fatigue. Kostyk and Rochet (1988) defined symptoms of vocal fatigue including but not limited to: hoarse/husky voice quality, breathy vocal quality, loss of voice, pitch breaks, reduced pitch range, lack of vocal carrying power, the need to use greater vocal effort, unsteady voice, throat fatigue, throat/neck pain, pain on swallowing, increased need to throat clear as well as other symptoms. Additionally, vocal fatigue has been a concept that many can measure by how an individual can perceive in themselves, such as after a long phone conversation or during a long speaking task (Hunter, E.J., 2011; Titze, 2000). Attempts to Measure Vocal Fatigue There have been many ways to quantify vocal fatigue. These include acoustic, aerodynamic and perceptual measures. The more objective measures, being acoustic and 2 aerodynamic measures are collected through instrumental means. Subjective measures are subject to critique and possible misinterpretation because these measures can be based off of personal and psychological ratings; however, subjective ratings can be beneficial for clinical research, where an individual's personal perception of their voice may take precedent. Below is a brief review of these measures. Acoustic Measures - Relevant Studies There have been a multitude of studies that have aimed to acoustically measure vocal fatigue. Acoustic measures have included fundamental frequency, sound pressure level (SPL), phonation type reflecting alpha ratio, jitter, and shimmer. A study conducted by Laukkanen, Ilomaki, Leppanen and Vilkman (2006) used acoustical measures as one component for assessment of vocal fatigue in female teachers. The study found that in the teacher’s working day, there was an increase in fundamental frequency, alpha ratio, jitter and shimmer. For the experimental vocal loading task, the participants had increased jitter and mean fundamental frequency (Laukkanen, Ilomaki, Leppanen and Vilkman, 2006). Another study, used dB SPL as well as electromyography (EMG) for the acoustical analysis during a vocal fatiguing task (Boucher & Ayad, 2008). More recently, in a study of school teachers, a standard deviation of the vocal sound pressure level had relationship to fewer voice complaints of fatigue, thus potentially being a parameter of interest in quantifying vocal fatigue severity (Cantor Cutiva et al, 2017). Aerodynamic Measures For vocal fatigue, aerodynamic measures usually refer to Phonation Threshold Pressure (PTP), or the amount of air needed to initiate and sustain vocal fold oscillation (Chang & Karnell, 2004). PTP was measured in a vocal loading and vocal effort study conducted by Chang 3 and Karnell (2004) who found a strong relationship between PTP and perceived phonatory effort, a subjective index for vocal fatigue (Chang & Karnell, p. 454, 2004); however, a similar study that observed female speakers before and after a vocal loading task did not find a significant relationship between Perceived Phonatory Effort (PPE) and PTP due to variability between subjects (Milbrath & Solomon, 2003). While useful, it should be noted that aerodynamic measures can have a high variability and require a specific and time-consuming protocol. Visual Perceptual Judgment Visual perception of vocal fatigue can be examined through videostroboscopy, involving a strobe light attached to a camera that is able to visualize the motion of the vocal folds. Studies have included visual perception in their studies to show the effects of a vocal loading tasks on the vocal process (Stemple et al., 1995; Linville, 1995). However, these studies require a pre- and post-stroboscopic exam which not all persons tolerate well. For this study, videostroboscopy was not used due to lack of proper observation determined by the Institutional Review Board. Auditory-Perceptual Judgment Listener ratings of vocal fatigue is a subjective measure of vocal fatigue that has been briefly researched in the voice community. Notably, Barsties et al. (2017) has conducted recent work in the areas of voice quality and auditory-perceptual judgement. In that study, the GRBAS (grade, roughness, breathiness, asthenia and strain) was used as the main rating scale. Researchers found that while there were statistically significant results when comparing specific vocal loading tasks to auditory-perceptual ratings and that there is high variability in auditory perceptual judgement, that the amount of training in the ratings can affect consistent ratings of voice quality (Barsties et al., 2017). 4 Self-Ratings of Vocal Fatigue Subjective measures that aim to demonstrate the participant’s feelings both emotionally and physically in regards to the voice have been carefully examined through multiple surveys created for research purposes and clinical voice patients. Research studies have used multiple methods for evaluating subjective ratings during vocal loading tasks. Hunter and Titze (2009) used three rating scales for a study that analyzed the relationship between vocal fatigue recovery after a vocal loading task. The scales included inability to produce soft voice (IPSV), current speaking level (EFFT), and laryngeal discomfort level (DISC). All scales were numbered 1-10 and were taken every 2 hours after the vocal loading exercise. Another study used the Glottal Function Index (GFI) scale as a voice symptom severity index, rated from numbers 1 (no problem) to 5 (severe problem). The GFI was used in this study over an extended period of time, and was rated by participants who were diagnosed with vocal fold paresis (Stager & Beilamowicz, 2014). More recently, Bottalico et al. (2016) illustrated how a simple self-rating of vocal fatigue (1 - no fatigue; 10 - highest fatigue) related to vocal duration and variation of dB SPL. The scale used by Bottalico et al. (2016) will be used for the present study. Females and Vocal Fatigue Based on previous research, it has been illustrated that voice disorders due to prolonged speaking have vocal fatigue as a precursor, and that voice disorders are found more often in women (Smith et al., 1998). Additionally, occupations that are dominantly female are often occupations that experience voice disorders more commonly such as performers, call center workers and teachers (Hunter, Tanner & Smith, 2011). In that report, Hunter and colleagues suggested that women vocally fatigue at a greater rate due in part to several structural and physiological differences. For example, post-pubescence, women have thinner/shorter vocal 5 folds and a smaller larynx overall compared to men (Titze, 1994). These differences result in the vocal folds to oscillate more frequently; given a comparable amount of talk-time the average female vocal folds will vibrate many more times than male vocal folds. The higher rate of vibration cycles these folds experience may be one of several contributing factors for the vocal problems that occurs more often in women. Another physiological distinction that creates differences that may be a contributing factor in the rate of vocal fatigue in women is the endocrine function. After the onset of puberty, changes in hormones affect the laryngeal function of women during menses as well as during menopause. This can be linked to hormone receptors that are located in the tissues of the vocal folds, affecting the voice both acoustically and perceptually. Previous studies have shown increased vocal complaints during periods of an increase in hormones, specifically during menstruation as reported by females (Schneider et al., 2004). Aging and Quality of Voice Function The relationship between aging and quality of voice has been researched with structural and functional changes in mind across multiple studies. For example, several reports (Mueller, 1997; Roy, et al., 2004; and Russell, Oates, & Greenwood, 1998) have discussed how elderly persons report greater vocal difficulties as well as a decrease in vocal function as a result of decreased lung, vocal fold and laryngeal function. Presbyphonia, or vocal difficulties with aging have been related back to physiological changes from aging, genetic factors, and diseases such as COPD or asthma. Without adequate and coordinated breath support from the pulmonary system , the vocal folds will not receive enough airflow for phonation, resulting in the tendency for a talker too overuse, overextend or push their vocal folds to fatigue, potentially leading to unneeded vocal damage. 6 A study relating to vocal fatigue of the aging system stated, “Age related voice problems due to hormonal changes appear in females predominantly between the ages of 45 and 70 years… During menopause the lack of estrogen causes an overbalance in aldosterone production in the suprarenal gland and the average pitch of the speaking voice is lowered” (am Zehnhoff- Dinnesen, A., Angerstein, W., & Deuster, D., 2010, p. 185). Likewise, Gama et al. (2009) found that “acoustic analyses have indicated a decrease in the fundamental frequency of women, as well as increased percentage of jitter and shimmer” (Gama et al., 2009, p. 128-129). Gama et al. (2009) evaluated acoustic measures of voice for 96 women aged between 60 and 103 years and found a significant and gradual reduction in fundamental frequency for aged women (Gama et al., 2009, p. 128-129). The reasoning for why these changes may occur in the aging population could be due to thinning or changes of vocal folds, causing a change in fundamental frequency (Titze, 1994) Based on these studies, it would be a fair hypothesis to state that the current study will find an increase in vocal fatigue in the aging population. Research Question and Hypothesis Due to the aging population and the higher incidence of voice problems among women, this study is designed to quantify vocal fatigue from a vocal loading task among older women. The question for this study was as follows: To what degree does a vocal loading task fatigue older women? A secondary question is, how does the rate of vocal fatigue compare to that found in younger women? To address the primary research question, pre and post-vocal loading measures around a vocal loading task were compared; specifically compared were vocal fatigue ratings and acoustical parameters. This is a descriptive aim without a hypothesis other than the expectation that pre-post differences will be found. The study is designed to expose older women to a vocal 7 loading task, quantify the rate of perceived vocal fatigue, and compare it to previous studies which used a younger population. The hypothesis is that there will be an increase in rate of vocal fatigue in aging females compared to younger females. If the hypothesis is accepted, then using the measures taken, researchers can infer potential underlying reasons for the vocal fatigue due to aging physiology. 8 Participants METHODS Thirteen females between the ages of 55 and 70 were recruited into the study with eleven completing the protocol. The inclusion and exclusion criteria for participants as approved by the IRB is outlined in Appendix A and B. Briefly, the criteria was that participants have normal and age appropriate hearing and speech systems. For this study, the primary researcher will keep in mind the possibility that menopause, hormonal changes, as well as other structural and physiological changes that occur the pharynx of an aging female. Participants reported the following: no smoking within the last 5 years, no previous vocal, speech, pulmonary or hearing impairment or history of vocal problems that required the services of a physician or speech-language pathologist, native speakers of English. Participants signed an IRB approved consent form and underwent a baseline and screening process that included a pulmonary function test, a voice survey, breathing screening and hearing screening with details described in the appendices. After the screening, the participants completed in a vocal loading task (prolonged speaking task) that was preceded by a short voice recording used to assess change due to the loading task. Participants were compensated for their time. Pre-Vocal Loading Task Prior to the vocal loading task, an audio recording of each participant reading a standard passages, including the Marvin Williams Passage, Stella Passage or the Rainbow Passage (See Appendix C for specific passages read; Figure 1 for specific loading task order) was collected while participants were seated in a sound booth. The researcher instructed participants to talk in a normal and comfortable voice with no specified decibel range. The pre-vocal loading task was recorded using a head mounted microphone (Countryman B3, 20-20kHz +-3dB) attached to a 9 Millennia HV-3D 8 channel microphone pre-amplifier (Millennia Media). The pre-amp delivers the signal to an RME ADI-8 DS AD/DA Converter and an RME Multiface II 36 channel digital audio interface (RME, Germany). The digital interface connects to the primary recording computer via a proprietary PCI expansion card on the motherboard. Reaper sound editing software (www.reaper.fm) was used to interface with this hardware to finalize the speech recordings (44,100Hz, 16bit) that were saved for later analysis in WAV format. The head mounted microphone recording setup was used also used for the post loading task. During this pre loading task, participants were asked to rate their vocal fatigue from 1-10, with 1 being no vocal fatigue and 10 being the greatest vocal fatigue. Vocal Loading Task The vocal loading task consisted of a smaller set of predetermined tasks to be completed in 6 minutes, repeated 6 times (36 minutes total). Each time the participant repeated the smaller set of tasks, they were given a radiated dB goal that alternated between 72 dB and 76 dB (ref 30 cm) every 6 minutes. Screen prompts presented the instructions with a predetermined start time keep the participant on time. Computer prompts required the participant to indicate when they were completed with the task (a small wireless clicker was provided). The set of tasks (listed below), which lasted 6 minutes, were repeated 6 times: 1. Drink water (using a small cup provided) 2. Self-rating of vocal fatigue (between 1 and 10), participant clicked to advance 3. Say “ahh” extended productions three times in a comfortable 4. Say “afa” three times in comfortable voice 5. Read aloud the Marvin Williams Passage at the indicated loudness goal of 72 or 76 dB (which stays consistent through Task 9), participant clicked to advance 10 6. Read aloud the Stella Passage, participant clicked to advance 7. Say “Ahh” three times, extended productions, (see Figure 1, left) 8. Read the first paragraph of the Rainbow Passage (See Figure 1, right), participant clicked to advance 9. Continue to read the remainder of the Rainbow Passage, repeating until the 6 minute interval is reached 10. Return to Step 1, Drink water Figure 1. Examples of the screen displayed to participants during the pre-, post- and vocal- loading task. The highlighted lettering “STAY ABOVE THE LINE” indicates the portions of the vocal loading task where participants were required to speak above 72 and 76 dB. Left: Step 7 with each “AAAH” presented individually per predetermined timing. Right: Step 8 indicating the passage to be read. The tasks above were practiced by the participant in a comfortable voice during the pre- vocal loading task to familiarize the participants with the task in order to reduce initial learning affect and related variability across all of the participants. Water drinking time was specified within the visual directions presented on the screen and water was provided in small cups so that the amount was controlled (1 ounce of water per drinking time). Reading at alternating dB levels 11 continued until a total speaking of 36 minutes accrued. The alternating dB levels were controlled for every other cycle of the tasks during the study. For example, for one cycle the participant will read the passages at 72 dB, then for the following cycle they would read at 76 dB. The passages read by the participants include the Rainbow Passage, Stella Passage, Marvin Williams Passage (see Appendix C for full text), as well as a set of sustained set of steady vowels. Between each 2 and a half minute reading segment, the subject stated, “My vocal fatigue level is _____”, and would give their rating of vocal fatigue between 1-10 as stated previously. To help participants maintain the target dB SPL during reading segments, a dB indicator (LingWAVES) was presented to the participant, showing a time history of the vocal level and a large green arrow pointing up which would appear when the target dB was not met. Post-Vocal Loading Task The post-vocal loading task included the same readings from the pre-loading task with the participant reading “/a/” vowels and Part 1 of the Rainbow Passage. For the post-vocal loading task, participants were also allowed to follow through with these tasks at a comfortable voice, or pitch. Measures of Interest Figure 2 depicts the protocol carried out for this study and the measures gathered. As stated previously, acoustical and perceptual measures were taken from the pre- post, and during the vocal loading tasks. Part 1 of the Rainbow Passage as well as the extended /a/ steady vowel were the specific tasks taken from recordings for acoustical analysis. Participants were asked to complete the task; however, if they are unable or become too fatigued, they were allowed to leave at any time and were compensated for their time. 12 Figure 2. A visual of the protocol for the present study with specific data collected. From the above described protocol, three measures were obtained to quantify the effect of the vocal loading task as experienced by the participants. As shown in Figure 3, these measures included: (1) the vocal fatigue self-ratings during the vocal loading task, (2) acoustic perturbation measures from the steady vowels (/a/) from pre-, post- and during the vocal loading task, and (3) variation of dB SPL from a read passage (Rainbow Passage) recorded before and after the loading task. Specifically, the data taken for analysis includes: the steady vowel /a/ taken from the recordings for analysis was from pre-vocal loading task, during the vocal loading task at the 6th repetition of the protocol (around approximately 18-20 minute time stamp in recordings) and at the post-vocal loading task which were completed in a comfortable voice, with the exact procedures for clipping the recordings of the steady vowel were applied when clipping the Rainbow Passage. The vocal fatigue rating would be given to the researcher verbally by the 13 participant. This rating would range from 1 (no fatigue) to 10 (the greatest amount of fatigue experienced). Ratings will be given out verbally by the participant at 2 and a half-minute timed intervals during the screening tasks and vocal loading tasks. In summary, the participants underwent a loading task of a specific vocal amplitude (72-76 dB) and duration of 36 minutes (independent variables), where perceptual and acoustic metrics were obtained (dependent variables) to quantify the effect of vocal loading in the aging female. Measurement and Data Analysis Data was analyzed for multiple parameters including the subjective ratings taken by the participants, and the acoustical data. The clipped recordings were analyzed using custom MATLAB (https://www.mathworks.com/products/matlab.html) scripts which did the bulk file handling for analysis in PRAAT (http://www.fon.hum.uva.nl/praat/); these scripts are custom made and part of standard speech acoustic analysis at the Voice and Biomechanics and Acoustics Laboratory (Michigan State University). The results of these scripts were (among many others) the perturbation measures (jitter, shimmer), as descriptive statistics of the fundamental frequency (in Hz), and descriptive statistics of the estimated dB (including dB standard deviation). To address the primary research question, student t-tests for paired homogeneous data were applied to each of the following pre- and post-vocal loading measures: participant rating of vocal fatigue, jitter, shimmer, and dB SPL standard deviation. Parameter changes were considered statistically significant when p < .05. Additionally, regression analysis were performed using excel and SPSS. Normal distribution of the data used was assessed by means of the Shapiro-Wilk test. Some of the tested variables were normally distributed but others were not. Therefore, ANOVA and Kruskal-Wallis was used to assess differences on acoustic parameters of voice in the middle and at the end of the vocal loading task. In addition, since the 14 dependent variable (repetitions) was a categorical variable and each participant had three repetitions (repeated measures), we used the Generalized Estimating Equations with a Gamma distribution analysis to assess the association between the vocal loading task and the voice acoustic parameters. Subjective vocal fatigue ratings were listed and timed via Excel from researchers analyzing recordings, and then the data was used to create a t-test and p-values. To address the secondary research question comparing the rate of vocal fatigue to a younger pool of participants, the data from a similar study conducted by Burtka (2018) was compared using calculated slope values and t-tests in Excel. 15 RESULTS The questions proposed previously are as follows: To what degree does a vocal loading task fatigue older women? A secondary question is, how does the rate of perceived vocal fatigue compare to that found in younger women? The following sections aim to present the results of this study with use of statistical analysis with both trending and significant data. Subjective ratings of the participant’s vocal fatigue will be outlined and described; additionally, acoustical parameters and correlational analysis will also be included in the results. Participants As stated previously, eleven participants completed the study’s protocol. Table 1 provides individual participant age as well as participant screening results for hearing, breathing and lung function. The average age was 65 years old (standard deviation. 4.6 years). The spirometry software (Carefusion) estimated “lung age” of the participants by using standard spirometry metrics and the following formula females: (0.022 × height (cm)-0.005-FEV1 (L))/0.022” (Toda et al., 2008, p. 514). Overall, the estimated lung age was higher than chronological age. Table 1. Participant information and screening data. Average age of participants was 65 years old with a standard deviation of 4.6. Participant O3 O4 O5 O6 O7 O8 O9 Lung Age 80 80 79 65 70 70 67 Actual Age 70 62 64 67 57 59 64 16 Hearing/Breathing Screening Normal/Normal Normal/Normal Normal/Normal Normal/Normal Normal/Normal Normal/Normal Normal/Normal Table 1. (cont’d) O10 O11 O12 O13 Average: 59 66 70 80 71 61 69 69 70 65 Normal/Normal Normal/Normal Normal/Normal Normal/Normal Subjective Ratings by Participants During the vocal loading task, participants were asked to rate their vocal fatigue from 1- 10, with 1 being the least amount of vocal fatigue, and 10 being the greatest amount of vocal fatigue felt. After two and a half minutes, participants were asked to rate their vocal fatigue during the recorded fatiguing tasks. Results of all participant’s ratings over time are shown in Figure 3 and listed in Table 2 with the calculated slopes also listed. Figure 3. Participants’ subjective ratings of their vocal fatigue over the vocal loading task over time. The lines indicate the slope. 17 Table 2. Vocal Fatigue Ratings for each participant with calculated slope. -5 0 2.5 5 8 Vocal Fatigue Ratings 11 14 17 20 23 Slope 4 4 O3 2 1 O4 6 5 O5 3 2 O6 4 2 O7 2 1 O8 4 O9 3 O10 1 4 1 O11 1 3 O12 1 O13 4 5 Avg 2 2.3 2.4 2.6 2.9 3.5 4 1 5 2 2 1 3 2 1 1 4 4 1 6 2 3 2 3 3 1 2 5 4 1 5 2 3 1 3 3 1 2 4 4 1 5 2 2 1 3 1 1 1 4 5 3 6 3 4 3 4 4 3 4 5 4 37. 5 0.087 7 5 0.246 9 5 0.026 6 6 0.179 9 4 0.140 7 5 0.148 7 3 9 0.186 5 10 0.209 5 9 0.233 4 0.138 6 5 5 6 0.053 4.7 5.1 5.7 6.2 6.8 7.1 7.3 7.7 0.150 28. 5 6 8 6 7 6 5 9 8 8 6 6 31. 5 7 9 6 7 7 6 9 8 8 5 6 25. 5 6 7 6 6 6 4 8 7 7 6 5 34. 5 7 9 6 8 7 6 9 8 9 5 6 6 7 6 5 6 4 7 6 5 6 5 5 7 6 5 5 3 5 6 4 5 5 Each participant’s perceptual ratings indicated an increase in fatigue during the vocal fatiguing task; as would be expected, every participant had a different rate (slope) of fatigue over time. Figure 4 below displays graphs of each participant’s fatigue rating over time. It should be noted that participants O5 and O13 responded with a very different rate of fatigue compared to their peers. Figure 4. All participant’s individual vocal fatigue ratings over time. 18 Figure 4. (cont’d) Acoustical Parameters Steady Vowel Analysis - Perturbation - Jitter and Shimmer The perturbation measures were taken using Jitter and Shimmer analyzed from the clipped recordings during the pre, post, and during the vocal loading task 6th repetition of the steady vowel (/a/). Statistical analysis of a regression analysis using a Gamma Distribution of the perturbation measures compared the pre-vocal loading task to the vocal loading task, and the vocal loading task to the post-vocal loading task during the steady vowels. Table 3 can be used as a reference for the specific data points when comparing pre- and, post-vocal loading task to the exact moment within the experimental protocol. Table 3 below lists the results of the statistical analysis of jitter and shimmer (/a/). The table also displays an additional acoustical parameter harmonic-to-noise ratio (/a/), decibel standard deviation, and fundamental frequency 19 from the pre- and post-vocal loading tasks from running speech. The degrees of freedom were 1 for the Beta analysis. Table 3. Pre and Post measures analyzed with regression analysis using a Gamma Distribution with Beta values from post-loading task. Pre-Vocal Loading Post-Vocal Loading Analysis Acoustic Parameter avg SD avg SD Beta SE Jitter (/a/) 0.004 0 0.003 0.001 -0.14 0.1 Shimmer (/a/) 0.03 0.01 0.023 0.009 -0.12 0.1 H2Noise (/a/) 20.32 4.82 21.66 3.34 0.06 0.1 dB stdev (Rainbow Passage) Fo (Rainbow Passage) 66.88 5.8 65.9 5.05 -0.05 0.1 220.5 37.8 218.5 47.53 -0.05 0.1 Figure 5. Perturbation measures, Jitter and Shimmer average measures taken at pre- vocal, vocal loading and post-vocal loading task with slope lines provided. Note the insignificant change from pre- to post-vocal loading task with an overall decreasing trend line for both jitter and shimmer. 20 Table 3, and Figure 5 display the perturbation measures, taken and analyzed with steady vowels during the pre-, post- and vocal loading tasks. When regression analysis using a Gamma Distribution was completed comparing the pre-vocal loading task and vocal loading task, both jitter and shimmer measurements were found to not significantly change (Beta=0.01; p>0.05). Figure 5 displays the Jitter and Shimmer throughout the pre, post- and during the vocal loading task. While the data was not significant, there is a trending decrease for Jitter throughout the vocal loading task. Running Speech –Rainbow Passage –Fundamental Frequency - Variation of dB The next acoustical measure for comparison is the pre- and post-vocal loading variation of dB SPL (Sound Pressure Level) from a read passage just before and after the loading task (Refer to Figure 2). The data was analyzed using the same methods as the perturbation measures explained above, but using recordings from the Rainbow Passage (see Table 3). While the change in dB stdev did not significantly decrease, a second analysis was done to check if there was a greater difference in another measure of dB variation, that of the inter-quartile range. A paired student t-test analysis of the dB stdev and the dB IQR resulted in p values of 0.106 and 0.048 respectively. This indicates that there was a general trend to decrease variation in dB from pre to post though only the dB IQR resulted in a significant decrease. Rate of Perceived Vocal Fatigue and Lung Age in Aging Population While not part of the research question, estimated lung age (per spirometry) and participant chronological ages were compared to the vocal fatigue ratings via a correlation matrix. Table 4 lists a compilation of lung age, chronological age, and the vocal fatigue ratings from the 11 participants. Using Excel’s built in correlation analysis, Table 5 was generated from 21 the data in Table 4. The comparison revealed a notable trend of a high percentage of the measures including, baseline slope of vocal fatigue ratings, fatigue change and lung age. Table 4. Comparison data of aging population for correlational analysis. Vocal Fatigue Rating Baseline 4 1 5 2 2 1 3 1 1 1 4 Slope of Fatigue Rating 0.087 0.246 0.026 0.179 0.14 0.148 0.186 0.209 0.233 0.138 0.053 Post- Loading Fatigue 7 9 6 9 7 7 9 10 9 6 6 Fatigue Change Chronological (post-pre) Lung Age 3 8 1 7 5 6 6 9 8 5 2 80 80 79 65 70 70 67 59 66 70 80 Age 70 62 64 67 57 59 64 61 69 69 70 As seen in the Table 5, there is a high correlation between the baseline fatigue ratings to the slope of the fatigue ratings. Additionally, there is a modest negative correlation between the slope of fatigue rating and chronological age (-.245) indicating some potential relation between perceived rate of vocal fatigue and aging. When comparing estimated lung age from the pulmonary task and the rate of vocal fatigue, the correlation is even stronger (-.576). Table 5. Correlational data table of the data from the tables above. A majority of the data that was analyzed was found to have a correlation greater than 50%. Slope of Fatigue Rating Post- Loading Fatigue Fatigue Change (post- pre) Lung Age Chrono Age 1 1 Vocal Fatigue Rating Baseline 1 Vocal Fatigue Rating Baseline Slope of Fatigue Rating -0.828 Post-Loading Fatigue -0.504 0.850 22 Table 5. (cont’d) Fatigue Change (post-pre) -0.867 0.968 0.867 1 Lung Age 0.597 -0.576 -0.625 -0.704 1 Chronological Age 0.318 -0.245 -0.201 -0.299 0.258 Comparison Data- Between Aging and College-Age Females 1 The data used to compare the present study to a younger population was from Rachel Burtka’s (2018) study titled, “The Impact of Expiratory Muscle Strength Training on Vocal Fatigue”. The participant’s data was taken from college-age females under the height of 5’2” who had undergone an identical fatiguing task as the aging females in the present study. The data compared was the fundamental frequencies and the subjective ratings given by the young females. This data was then compared over time to the aging population. The Figure 6 below illustrates the subjective ratings over time, while Table 8 lists the comparison of the Vocal Fatigue Ratings between the two different populations. The comparison in Table 8 also has the slope and average slope. Figure 6. Comparison between the aging population and college-age population, vocal fatigue ratings over time. 23 Table 6. Slope values compared between age groups. College-Age Participants P1 P2 P4 P5 P6 P7 P9 P10 Slope 0.182 0.096 0.156 0.204 0.222 0.202 0.060 0.098 Average 0.152 Aging Participants O3 O4 O5 O6 O7 O8 O9 O10 O11 O12 O13 Average Slope 0.087 0.246 0.026 0.179 0.140 0.148 0.186 0.209 0.233 0.138 0.053 0.150 Table 6 distinguishes between the slopes of the data from each participant as well as a comparison between the two age groups. On average, the younger population started and ended on a higher vocal fatigue rating than their older counterparts. The average slopes between the two populations were almost identical and the results of the t-test performed on this data was not significant. 24 DISCUSSION The purpose of this study was to observe vocal fatigue in the aging female and to compare if there is a significant difference in rate of vocal fatigue in the aging females versus college- age females. To accomplish this purpose, fatigue ratings and acoustic measures were obtained prior to vocal loading, during 36 minutes of vocal loading, and immediately after vocal loading in a group of older females. Data from the older females were also compared to the younger college-age female data obtained in a prior study that used the same protocol. The main findings for the primary research question and the secondary question are presented below. Additionally, future research in regards to the aging female and vocal fatigue will be discussed. And finally, clinical applications of these results for the speech-language pathologist, and possibly the voice instructor will be discussed. Subjective Ratings Participants in this study showed an increase in the perception of vocal fatigue over time. Eight of the eleven participants had a rate of vocal fatigue that was much higher overall during the vocal loading task with a majority of the eight participants increasing their fatigue rating by at least 5 points by the conclusion of the vocal loading task. This indicated that the vocal loading task did indeed impact the participant’s perception of their voice. While most of the participants’ rate of fatigue was similar, two participants only reported mild increases in their vocal fatigue levels during the vocal loading task (See Table 2). Possible reasoning behind the differences in each participant could be due to a multitude of factors including age, lung age, and previous voice use during the day, time of day, whether or not the participant had encountered menopause or not, or even the participant’s level of hydration. 25 In terms of age, some participants may have fatigued at a faster rate due to their age being higher in the allowable range (i.e., 66-70 years old). Additionally, time of day and varied amounts of voice usage during the day could have impacted the participant’s specific vocal fatigue. A majority of these participants had participated in the evening, when, more than likely, their voices had been used more than if they had participated in the study in the morning when voice usage was likely to be minimal. Furthermore, correlations between vocal fatigue ratings and lung age shown below had a stronger impact on results of the study than any of the other parameters measured. Acoustical Parameters When comparing the pre- and post-vocal loading tasks, there were no significant results found from statistical analysis; however, there was a trending relationship from the jitter and shimmer from the pre- to post-vocal loading data, indicating that there was a slight change in the perturbation measures during the vocal loading task. The fundamental frequency of the Rainbow Passage was not presented as a metric of interest in the research question, but analysis revealed trending numbers from pre- to the post- vocal loading task. Using a one-tailed t-test, the fundamental frequency mean was statistically significant when comparing the pre- and post-vocal loading tasks. The fundamental frequency also increased over time during the vocal loading task but was not found to be significant. Conclusively, the acoustical data analyzed yielded non-significant results from the specific research question; likewise, the data also indicated trending values and significant values between the pre- to during the vocal loading task as well as the vocal loading task to the post- vocal loading task. There are multiple indications for why the data had presented this way, including the data found for the comparison of lung age verses the slope values of the vocal 26 fatigue ratings. Exact reasoning for the trending and non-significant data would be need to be investigated further to find the cause for the non-significant findings in acoustical parameters seen in this data pool. Laukkanen (2006) had similar findings between pre- and post-vocal loading task that were also not overall significant statistically. This led researchers to compare the pre-vocal loading data to the vocal loading data, and the vocal loading data to the post-loading data. Consequently, the data was statistically significant or trending when comparing the before and after tasks to the middle of the vocal loading tasks, indicating that there may have been a notable difference during the middle of fatiguing tasks, but not as much before and after. Laukkanen’s similar results, has led researchers of this study to wonder if the significance of the vocal loading tasks may be found during the loading task itself, instead of a before and after comparison. The concept of measuring vocal fatigue during the middle of a vocal loading task may be a proposal for future research due to this study’s and multiple experiments inconclusive results from the pre- and post-vocal fatiguing tasks. There appears to be a significant change between the pre-vocal loading task and the vocal loading task, but this change appears to become non- significant by the end of the vocal loading task. Rate of Perceived Vocal Fatigue and Lung Age in Aging Population While there can be many assumptions for why the participant’s fatigued at varied rates, much can be said about the correlational data compared between the slopes of the lung age, chronological age and the participant’s vocal fatigue rating data. If reference is given to the table above comparing the lung age, actual age and vocal fatigue ratings, there are indications that the lung age of the participants was highly correlated as an indicator for rate and severity of vocal fatigue that the participants experienced. 27 A high percentage correlation greater than 50% were found between the participant’s lung and vocal fatigue rating, indicating that there was an association between an individual’s lung age and the rate of vocal fatigue for this study and pool of participants. The participant’s chronological age could also foretell the rate at which these participant’s fatigued over time during the vocal loading task due to the high correlational percentage as seen in Table 6. The results of the correlational analysis completed for this data tells the researchers that the chronological age may only be on factor for determining the severity of vocal fatigue that the participants experienced. Additionally, the correlational analyzation reveals to the lung age of the participants as a notable indicator for the rate at which these participant’s fatigued. The data points the researchers to explore the possibility that aging may not be just the individual’s chronological age, but the aging of the biological systems and the relationships with the rest of the body, as is captured in the estimate of the lung age. Perhaps some of the females fatigued at a greater rate due to their higher lung age, compared to their peers who did not fatigue at a great rate because they had a lower lung age. The points made and found in this research may indicate that more research is still needed to be done in the area of aging and vocal fatigue. Comparison Data (Old vs. Young) One of the questions for this study must be answered through comparison of the data from both the aging population and the college-age population data. The comparison data of the two different participant pools was found to be insignificant, and in fact, rather close in comparison to the slope values of the averaged vocal fatigue ratings. The following data is both a cross comparison of the vocal fatigue ratings over time, during the study. 28 Limitations As with all studies, there are limitations. Such limitations may have impacted the significance and overall impact of the results of the data presented. For example, the low number of recruited subjects for this study is a limitation, with only 11 subjects recruited and fully participated in the study; additional subjects would have allowed for a more generalizable discussion of the results. Additionally, another better selection control of study participants, including stricter inclusion/exclusion criteria for participants, as well as better protocol control would could reduce potential outcome variation. Controlling the subject’s start time specifically for the morning may also have reduced the variability of the participant’s rates of fatigue due to varied voice use of each person’s everyday life. And finally, the literature does not discuss the variability from repeated vocal loading tasks. Therefore, it is possible that a participant’s data were not an average representation of the participant. Future Research Future studies could address these limitations as well as add additional metrics to the analysis as the protocol was complex with opportunity to measure many more voice metrics. For example, focusing on fundamental frequency, or noise-to-harmonic ratio during statistical analysis, may reveal aspects of vocal fatigue not discussed in recent literature. Additionally, phonation aerodynamic measures have previously been shown to be sensitive to vocal loading and could be added in the future. Finally, a more in-depth look at the estimated lung age and vocal loading would be in order. Furthermore, the results of this study may point to future research in the areas related to vocal fatigue and the aging female. Presently, most data and knowledge that exists involving the aging voice is based on physiological or anatomical information or changes that occur within the 29 body as human’s age. A discussed proposal for future research involves comparing more statistical data of the acoustic measures from during the vocal loading task with using the pre- vocal loading task as a reference point and the post-vocal loading task as the comparison. Additionally, controlling the participants for time of day during vocal fatiguing task, smaller age range, and an increase in amount of participants would improve the understanding of what may be happening to participants on a self-perceptual level. Understanding the acoustical changes, and the subjective opinions about this specific population should be observed more closely in future research to fully understand vocal fatigue. Clinical Applications The data from this thesis study will give information to clinicians on what may be leading to an increase in vocal fatigue in the aging female population. The results of this thesis may also indicate that when performing acoustic measures on patients with voice disorders, some participants may have differences in their fundamental frequency over time as they perform vocal warm-up exercises, or vocal fatigue. 30 CONCLUSION The results of this study indicate that the aging female will vocally fatigue over time during a vocal loading task; however, the specific rate of this fatigue will vary across individuals and may be impacted by other factors not controlled for in this study (height, body type, previous medical histories, menopause, etc.). Secondly, the results of this study indicate that females in the aging population of this study fatigued at a rate that was no different than when compared to their younger college-age counterparts, specifically by ratings of vocal fatigue and perturbation measures over time. And lastly, a significant trend was found when comparing the participants’ vocal fatigue ratings to the lung age calculated via spirometry leading researchers to believe that the participant’s vocal fatigue could be directly linked to the lung age rather than the participant’s chronological age overall. 31 APPENDICES 32 APPENDIX A Protocol Measures not statistically analyzed for this study 33 APPENDIX A Protocol Measures not statistically analyzed for this study 1a. Survey The survey for this study will be used for later studies and is located online via Qualitrics. The survey is a combination of the Vocal Handicap Index (VHI), the Vocal Fatigue Index and Voice-Related Quality of Life (VR-QOL). 1b. Pulmonary Function Test The pulmonary function test is used to compare and analyze pulmonary function pre- vocal loading task. A spirometer was used to measure pulmonary function on a computer. The participant will be instructed to blow into the spirometer and are required to pass three consistent breathes as measured by software and indicated as a “pass” on the screen. Once they have three successful trials on the spirometer, the participant will have their results printed, and a copy will be printed for the researcher’s records. 1c. Hearing Screening The hearing screening will be administered in a soundproof booth, via an audiometer. The participant’s hearing will be tested at 25 dB, and at the frequencies of 500, 1000, 2000, and 4000 Hz. If the participant must pass hearing screening to continue with study. 1d. Breathing Screening The breathing screening will use the chart in the protocol below. The participant’s breathing will be assessed by where in their body it takes place. If the patient has inadequate breathing patterns for the vocal loading task, they will be finished with the study and asked to leave. 34 APPENDIX B Protocol Instructions 35 APPENDIX B Protocol Instructions PROTOCOL INSTRUCTIONS SUBJECT NUMBER: _________________ DATE: ____________________________ TIME: ____________________________ “Hello and welcome to our study! --- Do you have any questions before we begin?” SCREENINGS: □ Seat the subject “Please read and sign this consent form. Let me know if you have any questions.” Hand subject consent form & provide a take-home copy as well □ Non-fatiguing tasks “Before we begin, between each screening and survey task, you will be asked to rate your current vocal fatigue. I want you to read the following script “My vocal fatigue level is _____”, with your current vocal fatigue on a scale of 1 being no level of fatigue and 10 being the greatest amount of fatigue. You cannot rate 0 or greater than 10. Other than that during the next tasks I would like you to talk as little as possible, until I redirect you to rate your vocal fatigue level. Please use post it notes or hand gestures to communicate other than that.” Provide accessible post it notes and script reading “My current vocal fatigue level is ____”- when participants rate themselves write the rating on the provided line. □ Ask subject to complete surveys via Qualtrics questionnaire- (which includes: VHI, VFI, BFI- 10, V-RQOL) ”Use this time to completely fill out the questionnaires on this online survey.” Direct subject to online surveys: *PROMPT VOCAL FATIGUE RATING- _____* □ Hearing screening: Make sure earphones are present, turn on audiometer. Mark results on following screening form using 25dB Left Ear Right Ear ● 500 Hz ● 500 Hz 1000 Hz ● 1000 Hz 2000 Hz ● 2000 Hz 4000 Hz ● 4000 Hz □ Pulmonary Function Test: 36 ● Pre-check: turn on computer, having testing tubes nearby, be sure spirometer is plugged into computer and turned on. ● Enter participant’s number into the software and begin exam: o “First I will provide you with instructions to follow the test. I will then model the procedure for you, if you have any questions at this point please ask. I will then place a nose plug on your nose, and you will inhale as much and deeply as you can. Then, exhale as fast as you can pushing all the air out of your lungs. Do not pause in between the inhalation and the exhalation. Feel free to use the rest of your body in order to move as much air as possible.” o Offer the participant a copy of the PFT results. *OFFER A SIP OF WATER* *PROMPT VOCAL FATIGUE RATING- _____* □ Breathing Screening: identifies types of breathing. Mark results on respiration screening form Type of breathing Score Present Absent Diaphragmatic The belly pushes out with an inhale Thoracic During inhale the chest expands to accommodate the air that has been sucked into the lungs Clavicular During inhale the clavicle goes up to accommodate the air that has been sucked into the upper part of the lungs Paradoxical The chest compresses on the inhale rather than expands and vice versa Type of respiration cycle in non-speech: Coordination of respiration-voice: Oral-Oral Nasal-Nasal Oral-Nasal Adequate 37 Not adequate *OFFER A SIP OF WATER* *PROMPT VOCAL FATIGUE RATING- _____* VOCAL LOADING TASKS □ Pre-check: ● Are computers on and logged in? ● Is equipment present and working- microphones in sound booth, recording devices both in and outside the sound booth? ● Are water and water cups full and available next to participant’s seat? o Fill water with about 2 bottles per person. ● Did participant sign the consent form and understand they can leave at any point? □ Equipment check: 1. Turn on equipment inside sound booth- power 1 (power strip), power 2 (sound level meter), and clicker. Make sure the clicker is within easy reach of the participant. 1. Check batteries on power 2 (sound level meter)- there will be a warning for low batteries. 2. Have programs on computer in lab up- (lingWAVES, PowerPoint, Reaper). 1. Reaper 1. Z:\fatigue protocol\fatigue_project_template.RPP 2. Put ID number (m or f to identify sex, with participant number)- in the upper left double click ID# to insert information. 3. Note the record button on the bottom left. 1. Start lingWAVES 1. Z:\fatigue protocol\fatigue template.lwp 1. This will open a template that is ready to go. 2. Note green circle button at the top is the record button. 1. Open PowerPoint 1. Z:\fatigue protocol\Powerpoints 2. Water_fatigue_7.pptx 2. Align PowerPoint and LingWAVES on screen for participant (use screen and duct tape on far upper left for reference). LingWAVES should be open on the bottom on PowerPoint. Open LingWAVES to fit half of the screen and move over to left side until the green record button is in the middle of the tape mark. Lay PowerPoint on top of LingWAVES, open to the height of the screen and over to the tape mark). (tape is here) 1. Screen should look as follows: 38 □ Participant instructions: go through the following ensuring no questions along the way. 1. Make sure participant is seated in an upright posture in booth. 1. “Get into a comfortable seated position. Try to stay in an upright posture and limit movements during tasks to avoid the microphone moving”. 2. Adjust sound-level meter (SLM) to be string length nose distance away from participant. 2. Adjust screen to preferred distance from participant. 1. “Let me know which distance you are comfortable with the screen. You should be able to clearly read the screen with no difficulty.” 2. Make sure PowerPoint is on top of LingWAVES and the arrow is present when LingWAVES turns on. 2. Put microphone around participant’s ears. 3. Show the participant how to get a cup of water. 1. “One big push will fill the cup, watch as I do it. Each time you are prompted to drink on the screen you are to fill the water cup and drink it. I would recommend filling the cup right after you finish it to make it easier next time you are prompted to drink.” 2. Show the participant the click and which button to push. 1. “You click this button to move forward in the PowerPoint to your desired pace. The PowerPoint will indicate whether or not to click when done with a slide”. 2. Begin the PowerPoint and go through the practice tasks (DO NOT RECORD YET): 1. Inform them that there will be an arrow on the right side that they want to disappear 1. “You want to speak loudly enough that the big arrow disappears”. 1. The PowerPoint will then go through the following examples: 1. Vocal fatigue level 1. “As previously done in between tasks, I want you to rate your current vocal fatigue level. With 1 being 1 being no level of fatigue and 10 being the greatest amount of fatigue.” 2. Steady vowel phonation and vowel-consonant-vowel 39 1. “The next two tasks are going to be producing ‘ah’ and ‘afa’. First you will produce ‘ah’ for about 3-5 seconds in a comfortable extended tone for as long as the ‘ah’ is present on the screen. After that you will see ‘afa’ on the screen, you will produce ‘afa’ in a shorter comfortable tone without changing pitch. For example: ‘ahhhhh’ ‘afa’- *provide examples” 3. 2 readings 1. “Read through the next readings in a comfortable pitch.” 2. “Note that on the last slide it says not to click to advance. When you get to this slide repeat the passage until the slide automatically changes. You may or may not have to repeat this slide, it is completely dependent on your pace”. 2. Exit sound booth to begin Reaper recording. 1. Have participant give 3 loud “Hey’s” to test for acoustics. 1. “Pretend you are saying hey to someone in the outdoor lab area, loudly and clearly.” 2. INSERT PARTICIPANT NUMBER IN REAPER. 2. Re-enter sound booth, shut the outer door to begin vocal measures. 1. State date, time and participant number (DO NOT mention actual name) 2. Begin PowerPoint pre-tasks: 1. “When you are ready you are going to complete some initial tasks. The PowerPoint will indicate what you should do, it will be similar to the practice tasks.” 2. Tasks: 1. Count to five drawing out the “I” in five 2. Extended “ah” 3. “Afa” 4. Rainbow passage 2. Ready to begin. 1. “Do you have any questions? I will do the rest of my communicating with you through the intercom. Give me a thumbs up to let me know when you are ready and I will tell you to begin by clicking the clicker which will start the PowerPoint.” 2. When the participant gets to the first slide and is drinking water, begin LingWAVES and click back to the PowerPoint so it is layered on top. 1. NOTE: failure to click back on the PowerPoint will result in the clicker not working when participant clicks it! □ Data collection- participant tasks, equipment: PowerPoint, LingWAVES and Reaper ● Task 1: o Drink water ● Task 2: o Self-rating of vocal fatigue ● Task 3 & 4: o “Aaaah” extended productions x3 in a comfortable voice and “afa” x3 in a comfortable voice ● Task 5: 40 o Marvin Williams Passage- loudness guided by arrows to indicate desired loudness (which will stay consistent through task 9) ● Task 6: o Stella Passage ● Task 7: o “Aaaah” extended productions ● Task 8: o Rainbow Passage ● Task 9: o Rainbow Passage Part 2 *Participants will go through a cycle of these 9 tasks until time limit is complete. ● Researchers during this time will remain outside the sound booth, only entering if signaled by participant. Ensure all recordings are working functionally throughout collection. □ Post data-collection: 1. When PowerPoint is complete, LingWAVES will automatically shut off. 2. Participant will be guided through the following post vocal loading tasks: 1. Count to five drawing out the “I” in five 2. Extended “ah” 3. “Afa” 4. Rainbow passage 2. When they have finished the rainbow passage, instruct them via intercom that you will be re-entering the room. 3. Turn off Reaper and go in sound booth. 4. Thank participant and bring them outside booth for instructions on at home use of respiratory training device. 1. “Thank you for coming today. We are finished with the tasks for today. If you do not have any further questions, I will compensate you for your time, and you may leave” 41 APPENDIX C Passages used for Vocal Loading Task 42 APPENDIX C Passages used for Vocal Loading Task The Rainbow Passage Part 1: When the sunlight strikes raindrops in the air, they act like a prism and form a rainbow. The rainbow is a division of white light into many beautiful colors. These take the shape of a long round arch, with its path high above and its two ends apparently beyond the horizon. There is, according to legend, a boiling pot of gold at one end. People look, but no one ever finds it. When a man looks for something beyond his reach his friends say he is looking for the pot of gold at the end of the rainbow.” First paragraph of the Rainbow Passage (Fairbanks, 1960). Rainbow Passage Part 2 Throughout the centuries people have explained the rainbow in various ways. Some have accepted it as a miracle without physical explanation. To the Hebrews it was a token that there would be no more universal floods. The Greeks used to imagine that it was a sign from the gods to foretell war or heavy rain. The Norsemen considered the rainbow as a bridge over which the gods passed from earth to their home in the sky. Others have tried to explain the phenomenon physically. Aristotle thought that the rainbow was caused by reflection of the sun's rays by the rain. Since then physicists have found that it is not reflection, but refraction by the raindrops which causes the rainbows. Marvin Williams Passage Marvin Williams is only nine. Marvin lives with his mother on Monroe Avenue in Vernon Valley. Marvin loves all movies, even eerie ones with evil villains in them. Whenever a new 43 movie is in the area, Marvin is usually an early arrival. Nearly every evening Marvin is in row one, along the aisle. Stella Passage Please call Stella. Ask her to bring these things with her from the store: Six spoons of fresh snow peas, five thick slabs of blue cheese, and maybe a snack for her brother Bob. We also need a small plastic snake and a big toy frog for the kids. She can scoop these things into three red bags, and we will go meet her Wednesday at the train station. 44 APPENDIX D Tables used in previous presentation but not imperative to this study 45 APPENDIX D Tables used in previous presentation but not imperative to this study Table 7. Participant’s lung age and actual age. Participant ID Lung Age OSF13 Actual Age 70 62 64 67 57 59 64 61 69 69 70 65 OSF03 OSF04 OSF05 OSF06 OSF07 OSF08 OSF09 OSF10 OSF11 OSF12 Average 80 80 79 65 70 70 67 59 66 70 80 71 46 Table 8. Participant’s Vocal Fatigue Ratings (over time in minutes). Time -5 0 2.5 5 8 11 OS03 OS04 OS05 OS06 OS07 OS08 OS09 OS10 OS11 OS12 OS13 4 1 5 2 2 1 3 1 1 1 4 4 1 5 2 2 1 3 1 1 1 4 4 1 5 2 2 1 3 2 1 1 4 4 1 5 2 3 1 3 3 1 2 4 4 1 6 2 3 2 3 3 1 2 5 4 2 6 3 4 2 4 4 1 3 5 Avg 2.3 2.3 2.4 2.6 2.9 3.5 Std Dev 1.4 1.4 1.4 1.3 1.5 1.4 1 4 5 3 6 3 4 3 4 4 3 4 5 4 0. 9 1 7 5 5 6 4 5 3 5 5 4 5 5 4. 7 0. 8 2 0 5 7 6 5 5 3 5 6 4 5 5 5. 1 0. 9 2 3 6 7 6 5 6 4 7 6 5 6 5 5. 7 0. 8 2 5. 5 6 7 6 6 6 4 8 7 7 6 5 6. 2 1. 0 2 8. 5 6 8 6 7 6 5 9 8 8 6 6 6. 8 1. 2 3 1. 5 7 9 6 7 7 6 9 8 8 5 6 7. 1 1. 2 3 4. 5 7 9 6 8 7 6 9 8 9 5 6 7. 3 1. 4 3 7. 5 7 9 6 9 7 7 9 1 0 9 6 6 7. 7 1. 4 47 APPENDIX E IRB Approval Form 48 49 APPENDIX F IRB Approved Recruitment Flyer 50 51 APPENDIX G IRB Approved Consent Form 52 53 54 55 56 APPENDIX H Information and Biographical Sheet 57 58 APPENDIX I Respiration Form 59 60 APPENDIX J Survey Questions 61 62 63 64 65 APPENDIX K Specific information used in Methods Section 66 APPENDIX K Specific information used in Methods Section METHODS Inclusion & Exclusion Criteria In order to control for criteria in the test population, researchers had refine how the aging population was defined. The aging female population in this study was defined as women who were within the age range of 55 to 70 years of age. This age range has been created based on the average starting age range of menopause, with the average starting age being 51 years old (NIA, menopause.org, & ehealthmedicine). Menopause was chosen as an indicator of aging due to the known change in hormone levels of females. Based on the average age range of menopause in women, the menopausal/post- menopausal population included females between the ages of 55-70 years with no smoking history within the last 5 years, and no previous vocal, speech, pulmonary or hearing impairment or history of vocal problems that required the services of a physician or speech-language pathologist. Participants that were vocally or athletically trained within this age range were excluded (i.e., professional singer, voice user or professional athlete). Additionally, non-native speakers of English and any person older than 70 or younger than 55 were excluded from this study. Inclusion and exclusion criteria was assessed during the consent process and the initial screenings of the participants. The screenings conducted included a hearing screening, a breathing and pulmonary screening, and an endoscopic exam of the larynx and vocal processes. Included in the screening were questions about past medical history which includes a history of 67 allergies, medications and any other medical conditions that could impact the results of the study. Patients must not have a history of gastroesophageal reflux disease (GERD), vocal nodules, upper respiratory infection or benign lesions in or on the vocal processes to be included in the study. Further, a participant who can only hear at an intensity greater than 25 dB at the frequencies 500, 1000, 2000 & 4000 Hz will be excluded from the study because this characteristic is linked to hearing loss (WHO, 2017). During the screening, if a participant was found to be within the exclusion criteria, they were compensated and dismissed from the study. Recruiting for this study will continue until 8-10 participants have been selected. Procedures Consent Form Subjects were first consented to participate in the study. A consent form was signed by each participant before they can continue in the study. This form included any risks and the information about the study. Subjects were also not be informed of the hypothesis of the study during any time in order to avoid skewing the data on any subjective measures. Vocal Fatigue Rating The participant were then asked to rate their current vocal fatigue between the numbers of 1 and 10, with 1 being the lowest amount of vocal fatigue and 10 being the highest amount of vocal fatigue. The participants could not rate any lower than 1 and any higher than 10. The participant were to verbally rate their vocal fatigue to the researcher, either during the non-vocal loading tasks or during the vocal loading tasks. Screenings (Non-Vocal Loading Tasks) 68 The screenings included a survey, a pulmonary function test, a hearing screening, and a breathing screening. Completion of the survey, an adequate pulmonary function test as well as a passed hearing and breathing screening in order to continue with participation in the study. Survey A survey including questions from the Vocal Fatigue Index (VFI), vocal handicap index (VHI), Big Five Inventory (BFI-10) and the Voice-Related Quality of Life (V-RQOL) will be administered to the participants before the study begins as a baseline measure, in addition to the self-rating scale after each non-fatiguing and fatiguing task. The surveys were combined and administered using the survey portal, Qualtrics, and the results of the participant’s responses could be analyzed after the study has been conducted. The goal of these surveys are meant to assess the impact of vocal problems in the quality of life, assess patient ratings of vocal function, identify individuals with vocal fatigue based on symptoms of voice problems. Survey questions will be listed in the appendix below. Pulmonary Function Test Spirometry intends assess the pulmonary function and capacity of each participant. Spirometry will be completed using CareFusion, a computer-based program dedicated to pulmonary assessment. Prior to testing, the participant’s age, height, smoking history, gender and race will be recorded in the CareFusion system. The norms for pulmonary measures will be taken from the measures used by the previous study to determine a normal percentage of pulmonary functions. To test pulmonary function, the participant will in an upright position. A demonstration will be performed by the researcher and the following explanation will be given to ensure that the procedure is executed correctly: “Inhale as deeply and quickly as you can. Then, without 69 pausing, exhale as hard as you can, pushing all the air out of your lungs. You may start whenever you feel ready.” The participant will be required to have three successful attempts that are within 5% accuracy of each other as measured by CareFusion. If the participant cannot produce three consistent trials during the screening process, they will be dismissed from the study and compensated for their time. The measures collected will be forced vital capacity (FVC), forced expiratory volume in the first second (FEV1), peak expiratory flow (PEF), and the ratio of forced expiratory volume in the first second of forced vital capacity (FEV1/FVC). After the data has been collected and the pulmonary function tests have been completed, each participant was allowed to view her spirometry results. The data from this screening was recorded into the database and possibly used for data analysis comparison in future studies. Hearing Screening A hearing screening was conducted to rule out if the participant’s hearing is within normal limits to continue with the study. The participant’s hearing will be screened at the frequencies: 500, 1000, 2000 and 4000 Hz; the amplitude of 25 decibels will be tested using these frequencies. These frequencies and the amplitude are based on normal hearing parameters given by ASHA. Breathing Screening The different types of breathing were measured in the participants to assess whether the participant could continue to participate in the study with a short task with the help of Lady Catherine Cantor Cutiva. The breathing types of an individual can include: diaphragmatic (inhale pushes diaphragm outwards), thoracic (chest expands outward to accommodate the air that fills 70 the lungs), clavicular (during inhale the clavicle rises to accommodate for the space filling the upper part of the lungs) and paradoxical (the chest compresses on an inhale rather than expands). Any participants who show clavicular or paradoxical breathing will be excluded from the study because it is considered abnormal (for at rest; clavicular would not be unusual after very heavy exertion) and does not reflect the typical breathing type. Additionally, the area of inhalation and exhalation of breathing was assessed in participants. The patterns, or areas of inhalation and exhalation can be nasal-nasal, oral-nasal or oral-oral, in terms of the upper airway posturing during inhalation and exhalation. Vocal Loading Task The vocal loading task was approximately 35 minutes of speaking at an amplitude range of 72 and 76 decibels. Participants were asked to complete the task; however, if they are unable or become too fatigued, they were allowed to leave at any time and were compensated for their time. The vocal loading tasks are designed to purposefully vocally fatigue the participants. Each participant participated in an approximate 35-minute vocal-loading task. LingWAVES was used to analyze and compare the recordings to the vocal fatigue ratings before and after the vocal loading tasks. The participants were given instruction on this task prior to the recording portion with a practice task. This is the experimental task of the study where each participant was required to complete the tasks as explained below. During the approximate 35-minute vocal loading task, the participants were to alternate vocal intensity between 72.0 dB and 76.0 dB every five minutes, as based on the ISO standard of normal and raised speech level (ISO 1921, 2003). The participants will be seated in a sound- treated booth, with a microphone approximately 30 centimeters from their mouth, and their 71 reading material as well as SPL levels were displayed on a computer screen. During the task, the lingWAVES software monitored the participant’s vocal intensity, and the software Reaper recorded the participant’s audio output. Participants will read from passages commonly used in voice research (i.e., Rainbow Passage, Marvin Williams Passage, etc.) and would rate their vocal fatigue (from 1 through 10) every few minutes. 72 BIBLIOGRAPHY 73 BIBLIOGRAPHY am Zehnhoff-Dinnesen, A., Angerstein, W., & Deuster, D. (2010). Communication disorders in the elderly. Audiological Medicine, 8(4), 184-190. doi:10.3109/1651386X.2010.522859 Barsties, B., Beers, M., Ten Cate, L., Van Ballegooijen, K., Braam, L., De Groot, M., . . . Maryn, Y. (2017). The effect of visual feedback and training in auditory-perceptual judgment of voice quality. Logopedics Phoniatrics Vocology, 42(1), 1. Bottalico P, Graetzer S, Hunter EJ. Effects of speech style, room acoustics, and vocal fatigue on vocal effort. The Journal of the Acoustical Society of America. 2016;139(5):2870–2879. Boucher, V., & Ayad, T. (2008). Physiological attributes of vocal fatigue and their acoustic effects: a synthesis of findings for a criterion-based prevention of acquired voice disorders. Journal of Voice, 24: 3. p. 324-336. Burtka, Rachel E. (2018). The Impact of Expiratory Muscle Strength Training on Vocal Fatigue. Unpublished master’s thesis, Michigan State University, East Lansing, Michigan. Chang, Ann, and Karnell, Michael. (2004). Percieved phonatory effort and phonation threshold pressure across a prolonged voice loading task: a study of vocal fatigue. Journal of Voice, 18;4, p.454-466. Cristian, A., Greenwald, Brian. (2009). Medical literature generally defines fatigue as an overwhelming sense of tiredness. Preface. Physical Medicine and Rehabilitation Clinics of North America, 20(2), P. 11-13. Cutiva, L.C.C., Puglisi, G.E., Astolfi, A., & Carullo, A. (2016). Four-day follow-up study on the self-reported voice condition and noise condition of teachers: relationship between vocal parameters and classroom acoustics. Journal of Voice. Gama, Ana Cristina Côrtes, Alves, Cláudia Fernanda Tolentino, Cerceau, Janaína da Silva Berto, & Teixeira, Letícia Caldas. (2009). Correlação entre dados perceptivo-auditivos e qualidade de vida em voz de idosas. Pró-Fono Revista de Atualização Científica, 21(2), 125-130. https://dx.doi.org/10.1590/S0104-56872009000200007 Gavigan C, and Hunter EJ. (Nov. 2016). “Vocal Fatigue Metrics & Pulmonary Function.” 2016 American Speech-Language-Hearing Association (ASHA) Convention. Philadelphia, PA. Goy H, Pichora-Fuller MK, van Lieshout P. Effects of age on speech and voice quality ratings. The Journal of the Acoustical Society of America. 2016;139(4):1648–1659. 74 Maxfield L, Hunter EJ, Graetzer S. The effect of compromised pulmonary function on speech production among female school teachers. The Journal of the Acoustical Society of America. 2016;139(4):2105-2105. doi:10.1121/1.4950255. Hunter, E. J., & Titze, I.R. (2009). Quantifying vocal fatigue recovery: Dynamic vocal recovery trajectories after a vocal loading exercise. The Annals of Otology, Rhinology, and Laryngology, 118(6), 449. Hunter, E. J., Tanner, K., & Smith, M. E. (2011). Gender differences affecting vocal health of women in vocally demanding careers. Logopedics Phoniatrics Vocology, 36(3), 128-136. doi:10.3109/14015439.2011.587447 Hunter, EJ. NCVS Memo No 11. General Statistics of the NCVS Self-Administered Vocal Rating (SAVRa). 2008. http://www.ncvs.org/ncvs/library/tech Hunter, E., and Titze, I. (2009). Quantifying vocal fatigue recovery: dynamic vocal recovery trajectories after a vocal loading exercise. Annals of Otology, Rhinology and Laryngology, 118(6):449-360. Hunter, E. J., Tanner, K., & Smith, M. E. (2011). Gender differences affecting vocal health of women in vocally demanding careers. Logopedics Phoniatrics Vocology, 36(3), 128-136. doi:10.3109/14015439.2011.587447. Kostyk, Barbara E. & Rochet, A.P. (1988) Laryngeal airway resistance in teachers with vocal fatigue: A preliminary study. Journal of Voice. Volume 12:3, p. 287 – 299. Laukkanen, A., Ilomaki, I., Leppanen, K., and Vilkman, E. (2006). Acoustic measures and self- reports of vocal fatigue by female teachers. Journal of Voice, 22:1. P.283-289. Linville, Sue E. (1995). Changes in glottal configuration in women after loud talking. Journal of Voice, 9: 1. 57-65. Milbrath R., & Solomon, N.P. (2003). Do vocal warm-up exercises alleviate vocal fatigue? Journal of Speech, Language and Hearing Research, 46. p. 422-436. Mueller, Peter. (1997). The aging voice. Seminars in Speech and Language, 18(2), 1-11. Schneider B, van Trotsenburg M, Hanke G, Bigenzahn W, Huber J. Voice impairment and menopause. Menopause. 2004;11:151 – 8. Smith, Elaine, Kirchner, L.H.,Taylor, M., Hoffman, H., Lemke,J.H. (1998). Voice problems among teachers: Differences by gender and teaching characteristics, Journal of Voice, 12: 3, p. 328-334,Retrieved from https://doi.org/10.1016/S0892-1997(98)80022-2. 75 Solomon, N. P, Glaze, L. E., Arnold, R. R, & Mersbergen, M. V. (2003). Effects of a vocally fatiguing task and systemic hydration on men's voices. Journal of Voice - Official Journal of the Voice Foundation, 17(1), 31-46. doi:10.1016/S0892-1997(03)00029-8. Solomon, N. P. (2008). Vocal fatigue and its relation to vocal hyperfunction. International Journal of Speech-Language Pathology, 10(4), 254-266. doi:10.1080/14417040701730990. Stathopoulos, E. T., & Sapienza, C. (1993). Respiratory and laryngeal function of women and men during vocal intensity variation. Journal of Speech and Hearing Research, 36(1), 64- 75. Stager, S., & Beilamowicz, S. (2014). Perceived vocal fatigue and effort in relation to laryngeal functional measures in paresis patients. The Laryngoscope, 124;1631-1636. Stemple, Joseph C.; Stanley, J.; Lee, L. (1995). Objective measures of voice production in normal subjects following prolonged voice use. Journal of Voice, 9: 2. 127-133. Titze IR. Principles of Voice Production. Englewood Cliffs, NJ: Prentice Hall; 1994. Titze IR. Toward occupational safety criteria for vocalization. Log.Phon.Vocol. 1999; 24:49–54. Titze IR. Vocal fatigue: some biomechanical considerations. In: Lawerence VL, ed. Transcripts of the Twelfth Symposium: Care of the Professional Voice. Part One: Scientific Papers. New York: The Voice Foundation; 1984;97-104. Titze, I. R. (1997). Populations in the U.S. workforce who rely on voice as a primary tool of trade: A preliminary report. Journal of Voice - Official Journal of the Voice Foundation, 11(3), 254-59. doi:10.1016/S0892-1997(97)80002-1. Toda, Reiko; Hoshino, Tomoaki; Kawayama, Tomotaka, Haruki Imaoka, Yuki Sakazaki, Toru Tsuda, Shohei Takada, Masaharu Kinoshita, Tomoaki Iwanaga, Hisamichi Aizawa (2008). Validation of "Lung Age" Measured by Spirometry and Handy Electronic FEV1/FEV6 Meter in Pulmonary Diseases. Internal Medicine, 48 (7). Doi: https://doi.org/10.2169/internalmedicine.48.1781. World Health Organization. (February 2017). Deafness and hearing loss. Retrieved from: http://www.who.int/mediacentre/factsheets/fs300/en/. Welham, Nathan V., and Margaret A. Maclagan. "Vocal Fatigue: Current Knowledge and Future Directions." Journal of Voice, vol. 17, no. 1, 2003, p. 21+. Health Reference Center Academic, http://link.galegroup.com/apps/doc/A194405723/HRCA?u=msu_main&sid=HRCA&xid =9eeeaf69. Accessed 6 Feb. 2018 76