Fuh-Cherng Jeng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Fuh-Cherng Jeng has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
Chia-Der Lin
Department of Otolaryngology-HNS, Medical University Hospital, Taichung City
12 shared publicationsChing-Yuan Wang
Department of Otolaryngology-HNS, China Medical University Hospital, Taichung, Taiwan
5 shared publicationsJiong Hu
Communication Sciences and Disorders, Ohio University
5 shared publicationsAmanda E. Carriero
Ohio University
3 shared publicationsBrenda Dickman
3 shared publicationsGrant R. Hollister
Commun. Sci. and Disord., Ohio Univ., Athens, OH
3 shared publications
Latest Publications
- 29 May 2025
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 133 (1) , 25-40
Detecting Human Frequency-Following Responses Using an Artificial Neural Network
Frequency-following responses (FFRs) are neural signals that reflect the brain’s encoding of acoustic characteristics, such as speech intonation. While traditional machine learning models have been used to classify FFRs elicited under various conditions, the potential of deep learning models in FFR research remains underexplored. This study investigated the efficacy of a three-layer artificial neural network (ANN) in detecting the presence or absence of FFRs elicited by a rising intonation of the English vowel /i/. The ANN was trained and tested on FFR recordings, using F0 estimates derived from the spectral domain as input data. Model performance was evaluated by systematically varying the number of inputs, hidden neurons, and the number of sweeps included in the recordings. The prediction accuracy of the ANN was significantly influenced by the number of inputs, hidden neurons, and sweeps. Optimal configurations included 6–8 inputs and 4–6 hidden neurons, achieving a prediction accuracy of approximately 84% when the signal-to-noise ratio was enhanced by including 100 or more sweeps. These results provide a foundation for future applications in auditory processing assessments and clinical diagnostics.
- 29 May 2025
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 133 (1) , 25-40
Detecting Human Frequency-Following Responses Using an Artificial Neural Network
Frequency-following responses (FFRs) are neural signals that reflect the brain’s encoding of acoustic characteristics, such as speech intonation. While traditional machine learning models have been used to classify FFRs elicited under various conditions, the potential of deep learning models in FFR research remains underexplored. This study investigated the efficacy of a three-layer artificial neural network (ANN) in detecting the presence or absence of FFRs elicited by a rising intonation of the English vowel /i/. The ANN was trained and tested on FFR recordings, using F0 estimates derived from the spectral domain as input data. Model performance was evaluated by systematically varying the number of inputs, hidden neurons, and the number of sweeps included in the recordings. The prediction accuracy of the ANN was significantly influenced by the number of inputs, hidden neurons, and sweeps. Optimal configurations included 6–8 inputs and 4–6 hidden neurons, achieving a prediction accuracy of approximately 84% when the signal-to-noise ratio was enhanced by including 100 or more sweeps. These results provide a foundation for future applications in auditory processing assessments and clinical diagnostics.
- 16 August 2024
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 131 (5) , 1584-1602
Machine Learning Recognizes Frequency-Following Responses in American Adults: Effects of Reference Spectrogram and Stimulus Token
Electrophysiological research has been widely utilized to study brain responses to acoustic stimuli. The frequency-following response (FFR), a non-invasive reflection of how the brain encodes acoustic stimuli, is a particularly propitious electrophysiologic measure. While the FFR has been studied extensively, there are limitations in obtaining and analyzing FFR recordings that recent machine learning algorithms may address. In this study, we aimed to investigate whether FFRs can be enhanced using an “improved” source-separation machine learning algorithm. For this study, we recruited 28 native speakers of American English with normal hearing. We obtained two separate FFRs from each participant while they listened to two stimulus tokens /i/ and /da/. Electroencephalographic signals were pre-processed and analyzed using a source-separation non-negative matrix factorization (SSNMF) machine learning algorithm. The algorithm was trained using individual, grand-averaged, or stimulus token spectrograms as a reference. A repeated measures analysis of variance revealed that FFRs were significantly enhanced ( p < .001) when the “improved” SSNMF algorithm was trained using both individual and grand-averaged spectrograms, but not when utilizing the stimulus token spectrogram. Similar results were observed when extracting FFRs elicited by using either stimulus token, /i/ or /da/. This demonstration shows how the SSNMF machine learning algorithm, using individual and grand-averaged spectrograms as references in training the algorithm, significantly enhanced FFRs. This improvement has important implications for the obtainment and analytical processes of FFR, which may lead to advancements in clinical applications of FFR testing.
- 16 August 2024
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 131 (5) , 1584-1602
Machine Learning Recognizes Frequency-Following Responses in American Adults: Effects of Reference Spectrogram and Stimulus Token
Electrophysiological research has been widely utilized to study brain responses to acoustic stimuli. The frequency-following response (FFR), a non-invasive reflection of how the brain encodes acoustic stimuli, is a particularly propitious electrophysiologic measure. While the FFR has been studied extensively, there are limitations in obtaining and analyzing FFR recordings that recent machine learning algorithms may address. In this study, we aimed to investigate whether FFRs can be enhanced using an “improved” source-separation machine learning algorithm. For this study, we recruited 28 native speakers of American English with normal hearing. We obtained two separate FFRs from each participant while they listened to two stimulus tokens /i/ and /da/. Electroencephalographic signals were pre-processed and analyzed using a source-separation non-negative matrix factorization (SSNMF) machine learning algorithm. The algorithm was trained using individual, grand-averaged, or stimulus token spectrograms as a reference. A repeated measures analysis of variance revealed that FFRs were significantly enhanced ( p < .001) when the “improved” SSNMF algorithm was trained using both individual and grand-averaged spectrograms, but not when utilizing the stimulus token spectrogram. Similar results were observed when extracting FFRs elicited by using either stimulus token, /i/ or /da/. This demonstration shows how the SSNMF machine learning algorithm, using individual and grand-averaged spectrograms as references in training the algorithm, significantly enhanced FFRs. This improvement has important implications for the obtainment and analytical processes of FFR, which may lead to advancements in clinical applications of FFR testing.
- 28 December 2023
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 131 (2) , 417-431
Advancing Auditory Processing by Detecting Frequency-Following Responses Through a Specialized Machine Learning Model
In this study, we explore the feasibility and performance of detecting scalp-recorded frequency-following responses (FFRs) with a specialized machine learning (ML) model. By leveraging the strengths of feature extraction of the source separation non-negative matrix factorization (SSNMF) algorithm and its adeptness in handling limited training data, we adapted the SSNMF algorithm into a specialized ML model with a hybrid architecture to enhance FFR detection amidst background noise. We recruited 40 adults with normal hearing and evoked their scalp recorded FFRs using the English vowel/i/with a rising pitch contour. The model was trained on FFR-present and FFR-absent conditions, and its performance was evaluated using sensitivity, specificity, efficiency, false-positive rate, and false-negative rate metrics. This study revealed that the specialized SSNMF model achieved heightened sensitivity, specificity, and efficiency in detecting FFRs as the number of recording sweeps increased. Sensitivity exceeded 80% at 500 sweeps and maintained over 89% from 1000 sweeps onwards. Similarly, specificity and efficiency also improved rapidly with increasing sweeps. The progressively enhanced sensitivity, specificity, and efficiency of this specialized ML model underscore its practicality and potential for broader applications. These findings have immediate implications for FFR research and clinical use, while paving the way for further advancements in the assessment of auditory processing.
- 28 December 2023
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 131 (2) , 417-431
Advancing Auditory Processing by Detecting Frequency-Following Responses Through a Specialized Machine Learning Model
In this study, we explore the feasibility and performance of detecting scalp-recorded frequency-following responses (FFRs) with a specialized machine learning (ML) model. By leveraging the strengths of feature extraction of the source separation non-negative matrix factorization (SSNMF) algorithm and its adeptness in handling limited training data, we adapted the SSNMF algorithm into a specialized ML model with a hybrid architecture to enhance FFR detection amidst background noise. We recruited 40 adults with normal hearing and evoked their scalp recorded FFRs using the English vowel/i/with a rising pitch contour. The model was trained on FFR-present and FFR-absent conditions, and its performance was evaluated using sensitivity, specificity, efficiency, false-positive rate, and false-negative rate metrics. This study revealed that the specialized SSNMF model achieved heightened sensitivity, specificity, and efficiency in detecting FFRs as the number of recording sweeps increased. Sensitivity exceeded 80% at 500 sweeps and maintained over 89% from 1000 sweeps onwards. Similarly, specificity and efficiency also improved rapidly with increasing sweeps. The progressively enhanced sensitivity, specificity, and efficiency of this specialized ML model underscore its practicality and potential for broader applications. These findings have immediate implications for FFR research and clinical use, while paving the way for further advancements in the assessment of auditory processing.
- 3 August 2023
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 130 (5) , 1834-1851
Effects of Silent Intervals on the Extraction of Human Frequency-Following Responses Using Non-Negative Matrix Factorization
Source-Separation Non-Negative Matrix Factorization (SSNMF) is a mathematical algorithm recently developed to extract scalp-recorded frequency-following responses (FFRs) from noise. Despite its initial success, the effects of silent intervals on algorithm performance remain undetermined. Our purpose in this study was to determine the effects of silent intervals on the extraction of FFRs, which are electrophysiological responses that are commonly used to evaluate auditory processing and neuroplasticity in the human brain. We used an English vowel /i/ with a rising frequency contour to evoke FFRs in 23 normal-hearing adults. The stimulus had a duration of 150 ms, while the silent interval between the onset of one stimulus and the offset of the next one was also 150 ms. We computed FFR Enhancement and Noise Residue to estimate algorithm performance, while silent intervals were either included (i.e., the WithSI condition) or excluded (i.e., the WithoutSI condition) in our analysis. The FFR Enhancements and Noise Residues obtained in the WithoutSI condition were significantly better ( p < .05) than those obtained in the WithSI condition. On average, the exclusion of silent intervals produced a 11.78% increment in FFR Enhancement and a 20.69% decrement in Noise Residue. These results not only quantify the effects of silent intervals on the extraction of human FFRs, but also provide recommendations for designing and improving the SSNMF algorithm in future research.
- 3 August 2023
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 130 (5) , 1834-1851
Effects of Silent Intervals on the Extraction of Human Frequency-Following Responses Using Non-Negative Matrix Factorization
Source-Separation Non-Negative Matrix Factorization (SSNMF) is a mathematical algorithm recently developed to extract scalp-recorded frequency-following responses (FFRs) from noise. Despite its initial success, the effects of silent intervals on algorithm performance remain undetermined. Our purpose in this study was to determine the effects of silent intervals on the extraction of FFRs, which are electrophysiological responses that are commonly used to evaluate auditory processing and neuroplasticity in the human brain. We used an English vowel /i/ with a rising frequency contour to evoke FFRs in 23 normal-hearing adults. The stimulus had a duration of 150 ms, while the silent interval between the onset of one stimulus and the offset of the next one was also 150 ms. We computed FFR Enhancement and Noise Residue to estimate algorithm performance, while silent intervals were either included (i.e., the WithSI condition) or excluded (i.e., the WithoutSI condition) in our analysis. The FFR Enhancements and Noise Residues obtained in the WithoutSI condition were significantly better ( p < .05) than those obtained in the WithSI condition. On average, the exclusion of silent intervals produced a 11.78% increment in FFR Enhancement and a 20.69% decrement in Noise Residue. These results not only quantify the effects of silent intervals on the extraction of human FFRs, but also provide recommendations for designing and improving the SSNMF algorithm in future research.
- 1 August 2022
- Published by Georg Thieme Verlag KG in Seminars in Hearing
- Vol. 43 (03) , 251-274
Implementation of Machine Learning on Human Frequency-Following Responses: A Tutorial
The frequency-following response (FFR) provides enriched information on how acoustic stimuli are processed in the human brain. Based on recent studies, machine learning techniques have demonstrated great utility in modeling human FFRs. This tutorial focuses on the fundamental principles, algorithmic designs, and custom implementations of several supervised models (linear regression, logistic regression, k-nearest neighbors, support vector machines) and an unsupervised model (k-means clustering). Other useful machine learning tools (Markov chains, dimensionality reduction, principal components analysis, nonnegative matrix factorization, and neural networks) are discussed as well. Each model's applicability and its pros and cons are explained. The choice of a suitable model is highly dependent on the research question, FFR recordings, target variables, extracted features, and their data types. To promote understanding, an example project implemented in Python is provided, which demonstrates practical usage of several of the discussed models on a sample dataset of six FFR features and a target response label.
- 1 August 2022
- Published by Georg Thieme Verlag KG in Seminars in Hearing
- Vol. 43 (03) , 251-274
Implementation of Machine Learning on Human Frequency-Following Responses: A Tutorial
The frequency-following response (FFR) provides enriched information on how acoustic stimuli are processed in the human brain. Based on recent studies, machine learning techniques have demonstrated great utility in modeling human FFRs. This tutorial focuses on the fundamental principles, algorithmic designs, and custom implementations of several supervised models (linear regression, logistic regression, k-nearest neighbors, support vector machines) and an unsupervised model (k-means clustering). Other useful machine learning tools (Markov chains, dimensionality reduction, principal components analysis, nonnegative matrix factorization, and neural networks) are discussed as well. Each model's applicability and its pros and cons are explained. The choice of a suitable model is highly dependent on the research question, FFR recordings, target variables, extracted features, and their data types. To promote understanding, an example project implemented in Python is provided, which demonstrates practical usage of several of the discussed models on a sample dataset of six FFR features and a target response label.
- Open Access
- 6 May 2022
- Published by Taylor & Francis in International Journal of Audiology
- Vol. 62 (7) , 688-698
Non-negative matrix factorization improves the efficiency of recording frequency-following responses in normal-hearing adults and neonates
One challenge in extracting the scalp-recorded frequency-following response (FFR) is related to its inherently small amplitude, which means that the response cannot be identified with confidence wh...
- 6 May 2022
- Published by Taylor & Francis in International Journal of Audiology
- Vol. 62 (7) , 688-698
Open AccessNon-negative matrix factorization improves the efficiency of recording frequency-following responses in normal-hearing adults and neonates
One challenge in extracting the scalp-recorded frequency-following response (FFR) is related to its inherently small amplitude, which means that the response cannot be identified with confidence wh...
- 29 September 2021
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 128 (6) , 2527-2543
Separating the Novel Speech Sound Perception of Lexical Tone Chimeras From Their Auditory Signal Manipulations: Behavioral and Electroencephalographic Evidence
Previous research has shown the novelty of lexical-tone chimeras (artificially constructed speech sounds created by combining normal speech sounds of a given language) to native speakers of the language from which the chimera components were drawn. However, the source of such novelty remains unclear. Our goal in this study was to separate the effects of chimeric tonal novelty in Mandarin speech from the effects of auditory signal manipulations. We recruited 20 native speakers of Mandarin and constructed two sets of lexical-tone chimeras by interchanging the envelopes and fine structures of both a falling/yi4/and a rising/yi2/Mandarin tone through 1, 2, 3, 4, 6, 8, 16, 32, and 64 auditory filter banks. We conducted pitch-perception ability tasks via a two-alternative, forced-choice paradigm to produce behavioral (versus physiological) pitch perception data. We also obtained electroencephalographic measurements through the scalp-recorded frequency-following response (FFR). Analyses of variances and post hoc Greenhouse-Geisser procedures revealed that the differences observed in the participants’ reaction times and FFR measurements were attributable primarily to chimeric novelty rather than signal manipulation effects. These findings can be useful in assessing neuroplasticity and developing speech-processing strategies.
- 29 September 2021
- Published by SAGE Publications in Perceptual and Motor Skills
- Vol. 128 (6) , 2527-2543
Separating the Novel Speech Sound Perception of Lexical Tone Chimeras From Their Auditory Signal Manipulations: Behavioral and Electroencephalographic Evidence
Previous research has shown the novelty of lexical-tone chimeras (artificially constructed speech sounds created by combining normal speech sounds of a given language) to native speakers of the language from which the chimera components were drawn. However, the source of such novelty remains unclear. Our goal in this study was to separate the effects of chimeric tonal novelty in Mandarin speech from the effects of auditory signal manipulations. We recruited 20 native speakers of Mandarin and constructed two sets of lexical-tone chimeras by interchanging the envelopes and fine structures of both a falling/yi4/and a rising/yi2/Mandarin tone through 1, 2, 3, 4, 6, 8, 16, 32, and 64 auditory filter banks. We conducted pitch-perception ability tasks via a two-alternative, forced-choice paradigm to produce behavioral (versus physiological) pitch perception data. We also obtained electroencephalographic measurements through the scalp-recorded frequency-following response (FFR). Analyses of variances and post hoc Greenhouse-Geisser procedures revealed that the differences observed in the participants’ reaction times and FFR measurements were attributable primarily to chimeric novelty rather than signal manipulation effects. These findings can be useful in assessing neuroplasticity and developing speech-processing strategies.
- 1 September 2018
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 144 (3_Suppleme) , 1837
Frequency-following responses elicited by a consonant-vowel with intonation
To date, two major types of acoustic stimuli have been used to evoke frequency-following responses (FFRs)—either by using a consonant-vowel or by using a vowel with intonation, but not both. The goals of this study were to (1) determine the feasibility of recording FFRs by forging a CV and intonation into one acoustic stimulus, and (2) examine the characteristics of such responses. Twelve Chinese adults (8 females, 4 males, age = 25.58 ± 4.23 years old) were recruited. FFRs were elicited by using the consonant-vowel /da/ combined with a rising intonation pattern (fundamental frequencies 85-93 Hz) for a total of 8000 sweeps from each participant. Six objective indices (consonant amplitude, consonant latency, consonant latency, frequency error, tracking accuracy, and pitch strength) were derived from all recordings. Results demonstrated that it was feasible to record the consonant, vowel, and intonation responses simultaneously. Results also demonstrated distinctive FFR trends with increasing number of sweeps for the consonant, vowel, and intonation responses. Taken together, these findings may have important implications for basic research and clinical applications.
- 1 September 2018
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 144 (3_Suppleme) , 1837
Frequency-following responses elicited by a consonant-vowel with intonation
To date, two major types of acoustic stimuli have been used to evoke frequency-following responses (FFRs)—either by using a consonant-vowel or by using a vowel with intonation, but not both. The goals of this study were to (1) determine the feasibility of recording FFRs by forging a CV and intonation into one acoustic stimulus, and (2) examine the characteristics of such responses. Twelve Chinese adults (8 females, 4 males, age = 25.58 ± 4.23 years old) were recruited. FFRs were elicited by using the consonant-vowel /da/ combined with a rising intonation pattern (fundamental frequencies 85-93 Hz) for a total of 8000 sweeps from each participant. Six objective indices (consonant amplitude, consonant latency, consonant latency, frequency error, tracking accuracy, and pitch strength) were derived from all recordings. Results demonstrated that it was feasible to record the consonant, vowel, and intonation responses simultaneously. Results also demonstrated distinctive FFR trends with increasing number of sweeps for the consonant, vowel, and intonation responses. Taken together, these findings may have important implications for basic research and clinical applications.
- 1 September 2018
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 144 (3) , 1837-1838
Machine learning in detecting frequency-following responses
To improve the efficiency and timeliness in frequency-following response (FFR) testing, the purpose of this study was to investigate the capabilities of machine learning in the detection of an FFR. Continuous brain waves were recorded from 25 Chinese adults in response to a pre-recorded Mandarin monosyllable \yi2\ with a rising frequency contour. A total of 8000 artifact-free sweeps were recorded from each participant. Continuous brain waves accumulated (from the first sweep) up to the first 500 sweeps were considered FFR absent; brain waves accumulated (from the first sweep) up to the last 1000 sweeps (i.e., from 7001 to 8000 sweeps) were considered FFR present. Six response features (frequency error, slope error, tracking accuracy, spectral amplitude, pitch strength, and root-mean-square amplitude) were extracted from each recording and served as key predictors in the identification of a response. Twenty-three supervised machine-learning algorithms, with a 10-fold cross-validation procedure, were implemented via a Classification Learner App in MATLAB. Two algorithms yielded 100% efficiency (i.e., 100% sensitivity and 100% specificity) and 14 others produced efficiency ≥99%. Results indicated that a majority of the machine-learning algorithms provided efficient and accurate predictions in whether an FFR was present or absent in a recording.
- 1 September 2018
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 144 (3) , 1837-1838
Machine learning in detecting frequency-following responses
To improve the efficiency and timeliness in frequency-following response (FFR) testing, the purpose of this study was to investigate the capabilities of machine learning in the detection of an FFR. Continuous brain waves were recorded from 25 Chinese adults in response to a pre-recorded Mandarin monosyllable \yi2\ with a rising frequency contour. A total of 8000 artifact-free sweeps were recorded from each participant. Continuous brain waves accumulated (from the first sweep) up to the first 500 sweeps were considered FFR absent; brain waves accumulated (from the first sweep) up to the last 1000 sweeps (i.e., from 7001 to 8000 sweeps) were considered FFR present. Six response features (frequency error, slope error, tracking accuracy, spectral amplitude, pitch strength, and root-mean-square amplitude) were extracted from each recording and served as key predictors in the identification of a response. Twenty-three supervised machine-learning algorithms, with a 10-fold cross-validation procedure, were implemented via a Classification Learner App in MATLAB. Two algorithms yielded 100% efficiency (i.e., 100% sensitivity and 100% specificity) and 14 others produced efficiency ≥99%. Results indicated that a majority of the machine-learning algorithms provided efficient and accurate predictions in whether an FFR was present or absent in a recording.
- 1 February 2018
- Published by American Academy of Audiology in Journal of the American Academy of Audiology
- Vol. 29 (2) , 125-134
Exponential Modeling of Frequency-Following Responses in American Neonates and Adults
Background:The scalp-recorded frequency-following response (FFR) has been widely accepted in assessing the brain’s processing of speech stimuli for people who speak tonal and nontonal languages. Characteristics of scalp-recorded FFRs with increasing number of sweeps have been delineated through the use of an exponential curve-fitting model in Chinese adults; however, characteristics of speech processing for people who speak a nontonal language remain unclear.Purpose:This study had two specific aims. The first was to examine the characteristics of speech processing in neonates and adults who speak a nontonal language, to evaluate the goodness of fit of an exponential model on neonatal and adult FFRs, and to determine the differences, if any, between the two groups of participants. The second aim was to assess effective recording parameters for American neonates and adults.Research Design:This investigation employed a prospective between-subject study design.Study Sample:A total of 12 American neonates (1‐3 days old) and 12 American adults (24.1 ± 2.5 yr old) were recruited. Each neonate passed an automated hearing screening at birth and all adult participants had normal hearing and were native English speakers.Data Collection and Analysis:The English vowel /i/ with a rising pitch contour (117‐166 Hz) was used to elicit the FFR. A total of 8,000 accepted sweeps were recorded from each participant. Three objective indices (Frequency Error, Tracking Accuracy, and Pitch Strength) were computed to estimate the frequency-tracking acuity and neural phase-locking magnitude when progressively more sweeps were included in the averaged waveform. For each objective index, the FFR trends were fit to an exponential curve-fitting model that included estimates of asymptotic amplitude, noise amplitude, and a time constant.Results:Significant differences were observed between groups for Frequency Error, Tracking Accuracy, and Pitch Strength of the FFR trends. The adult participants had significantly smaller Frequency Error (p < 0.001), better Tracking Accuracy (p = 0.001), and larger Pitch Strength (p = 0.003) values than the neonate participants. The adult participants also demonstrated a faster rate of improvement (i.e., a smaller time constant) in all three objective indices compared to the neonate participants. The smaller time constants observed in adults indicate that a larger number of sweeps will be needed to adequately assess the FFR for neonates. Furthermore, the exponential curve-fitting model provided a good fit to the FFR trends with increasing number of sweeps for American neonates (mean r2 = 0.89) and adults (mean r2 = 0.96).Conclusions:Significant differences were noted between the neonatal and adult participants for Frequency Error, Tracking Accuracy, and Pitch Strength. These differences have important clinical implications in determining when to stop a recording and the number of sweeps needed to adequately assess the frequency-encoding acuity and neural phase-locking magnitude in neonates and adults. These findings lay an important foundation for establishing a normative database for American neonates and adults, and may prove to be useful in the development of diagnostic and therapeutic paradigms for neonates and adults who speak a nontonal language.
- 1 February 2018
- Published by American Academy of Audiology in Journal of the American Academy of Audiology
- Vol. 29 (2) , 125-134
Exponential Modeling of Frequency-Following Responses in American Neonates and Adults
Background:The scalp-recorded frequency-following response (FFR) has been widely accepted in assessing the brain’s processing of speech stimuli for people who speak tonal and nontonal languages. Characteristics of scalp-recorded FFRs with increasing number of sweeps have been delineated through the use of an exponential curve-fitting model in Chinese adults; however, characteristics of speech processing for people who speak a nontonal language remain unclear.Purpose:This study had two specific aims. The first was to examine the characteristics of speech processing in neonates and adults who speak a nontonal language, to evaluate the goodness of fit of an exponential model on neonatal and adult FFRs, and to determine the differences, if any, between the two groups of participants. The second aim was to assess effective recording parameters for American neonates and adults.Research Design:This investigation employed a prospective between-subject study design.Study Sample:A total of 12 American neonates (1‐3 days old) and 12 American adults (24.1 ± 2.5 yr old) were recruited. Each neonate passed an automated hearing screening at birth and all adult participants had normal hearing and were native English speakers.Data Collection and Analysis:The English vowel /i/ with a rising pitch contour (117‐166 Hz) was used to elicit the FFR. A total of 8,000 accepted sweeps were recorded from each participant. Three objective indices (Frequency Error, Tracking Accuracy, and Pitch Strength) were computed to estimate the frequency-tracking acuity and neural phase-locking magnitude when progressively more sweeps were included in the averaged waveform. For each objective index, the FFR trends were fit to an exponential curve-fitting model that included estimates of asymptotic amplitude, noise amplitude, and a time constant.Results:Significant differences were observed between groups for Frequency Error, Tracking Accuracy, and Pitch Strength of the FFR trends. The adult participants had significantly smaller Frequency Error (p < 0.001), better Tracking Accuracy (p = 0.001), and larger Pitch Strength (p = 0.003) values than the neonate participants. The adult participants also demonstrated a faster rate of improvement (i.e., a smaller time constant) in all three objective indices compared to the neonate participants. The smaller time constants observed in adults indicate that a larger number of sweeps will be needed to adequately assess the FFR for neonates. Furthermore, the exponential curve-fitting model provided a good fit to the FFR trends with increasing number of sweeps for American neonates (mean r2 = 0.89) and adults (mean r2 = 0.96).Conclusions:Significant differences were noted between the neonatal and adult participants for Frequency Error, Tracking Accuracy, and Pitch Strength. These differences have important clinical implications in determining when to stop a recording and the number of sweeps needed to adequately assess the frequency-encoding acuity and neural phase-locking magnitude in neonates and adults. These findings lay an important foundation for establishing a normative database for American neonates and adults, and may prove to be useful in the development of diagnostic and therapeutic paradigms for neonates and adults who speak a nontonal language.