Speech-in-noise Perception in Autism and Fragile X
Study Details
Study Description
Brief Summary
The goal of this study is to identify which brain regions are active during speech-in-noise perception, as well as how those regions interact. The investigators are studying brain activation during speech-in-noise in autism and controls as well as individuals with Fragile X Syndrome. The main question[s] it aims to answer are: 1) How does the brain's response to background noise affect a person's ability to understand speech? 2) Can visual cues improve hearing in background noise?
Participants will complete the following:
-
hearing tests
-
cognitive and behavioral measures
-
questionnaires about their symptoms
-
both passive and active hearing tasks while brain activity is recorded with a neuroimaging cap Results will be compared between individuals with autism with and without Fragile X Syndrome as well as individuals without autism.
| Condition or Disease | Intervention/Treatment | Phase |
|---|---|---|
|
N/A |
Detailed Description
Participants will be 15-35 year-olds with autism, Fragile X Syndrome and autism, and a typical developing group, matched on age and sex (n=20 per group). These ages are wide enough to support sufficient recruitment without the potential confound of age-related hearing loss. The investigators will not exclude participants in either autism group based on cognitive ability if they are able to complete study tasks; the investigators expect to determine effects of cognitive and language abilities on neural activation statistically by modeling continuous effects of IQ on findings.
Recruitment Strategy is through email and social media ads, contacting previous study participants who have consented to contact about future studies, and through local and national organizations and foundations, including the National Fragile X Foundation.
Hearing measures will include 1) behavioral audiometry presented via calibrated earphones 2) Otoacoustic Emissions (OAEs) between 2000-8000 Hz. The investigators will use two tasks to measure speech-in-noise perception. The 3) Digits-In-Noise task (DIN) requires participants to repeat single digits spoken in background noise. The loudness of the spoken digits then varies adaptively based on the participant's performance, and a Speech Reception Threshold (SRT) is calculated. The SRT indicates the ratio of speech to noise level in dB at which the participant correctly reports the digit 50% of the time. The investigators will use a variation of the DIN that includes both interaural "in-phase" (Diotic) and "out-of-phase" (Antiphasic) presentation, which allows for characterization of temporal fine structure perception.
The second speech-in-noise task is the 4) Auditory Attention subtest of the Woodcock Johnson
- This task requires no verbal output; instead participants touch or point to the picture of a word that they hear. Background noise is varied, and the individual's ability to correctly identify the word at different speech-to-noise ratios is recorded. The outcome variables for this test include both a raw score of number correct over a range of speech-to-noise ratios and an age-normed standard score. The investigators are using tests that minimize verbal requirements for two reasons. First, this will allow for inclusion of more participants rather than excluding those with cognitive impairment. Second, by minimizing contributions of working memory required for digit strings and use of context clues (as can occur with sentences in noise), the investigators are able to focus on true differences between groups in speech-in-noise perception.
Finally, the investigators will use an audiovisual DIN task where single digit stimuli are paired with the dynamic oval stimulus used in the neuroimaging paradigm to determine behavioral effects of adding a visual stimulus.
Behavioral measures will include standardized assessments of sensory symptoms and experiences (Sensory Profile, Sensory Experiences Questionnaire, Sensory Perception Quotient [33]), social communication symptoms (Autism Spectrum Quotient, Social Responsiveness Scale), restrictive/ repetitive behavior symptoms (Repetitive Behaviors Questionnaire-2 and Adult Repetitive Behaviours Questionnaire-2 [34]), verbal and nonverbal cognitive abilities (Stanford Binet V), and handedness (Purdue Pegboard task, Edinburgh Handedness Inventory). Most of these measures have both a self-report and parent-report option, use of both of which will enable the investigators to get data from participants with a wide range of abilities. They are also normed on both adolescent and adult samples.
Physiological stimuli: Speech stimuli include repeated consonant-vowel syllables presented within an oddball/mismatch negativity (MMN) paradigm that includes frequent, standard syllables and infrequent deviant syllables. Noise stimuli is "babble" noise (speech-like but unintelligible).
Paradigm: The investigators will use two types of neuroimaging within one cap to measure brain activity: 1) functional Near Infrared Spectroscopy (fNIRS) uses light delivered and measured at particular frequencies on the scalp to determine changes in oxygenation of hemoglobin on the cortical surface and 2) Event Related Potentials (ERP) records changes in voltage on the surface of the scalp in relation to changes in stimulus presentation. The investigators will compare brain activity between four conditions: 1) speech alone 2) noise alone 3) speech in noise and 4) speech in noise with a Dynamic Oval. Stimuli will be presented in short trials (ERP) embedded in longer blocks (fNIRS). Specifically, repeated standard and deviant syllables will be presented constantly for 15 seconds followed by 13-17 seconds of jittered rest. This allows the investigators to measure the blood oxygen level dependent (BOLD) response over the 30 second block (fNIRS) while measuring ERPs to deviant stimuli within those blocks. The investigators will also present speech embedded in noise with an animated oval that expands and contracts with the amplitude of the consonant-vowel sounds to determine effects of a non-social visual cue on hearing perception.
Neuroimaging: Integrated 32-channel gel-based EEG and 30 channel fNIRS system with standard 10-20 placement. EEG variables will include MMN amplitude and latency in response to deviant syllables, event-related power during the noise and speech conditions, and resting state oscillations. fNIRS variables will include lateralization and amplitude of neural response to speech and noise in both auditory and executive functioning regions of interest. Similar to speech-in-noise studies with fMRI, the investigators will compare speech alone, noise alone, and speech in noise to map cortical activity during speech-in-noise processing [38].
After completing phone-based consent/assent/permission, participants and their guardians (as needed) will complete an interview over the phone including administration of the Vineland Adaptive Behavior Scales and a brief review of medical and behavioral history. They will also receive a packet in the mail with forms to be completed prior to the in-person visit, which can be mailed back with a pre-stamped and addressed envelope or brought to the study visit and completed there. These forms will include 1) Sensory processing measures (Sensory Profile, Sensory Experiences Questionnaire, Sensory Perception Quotient) 4) Social Communication measures (Autism Spectrum Quotient, Social Responsiveness Scale) and Restrictive/Repetitive behavior assessment (Repetitive3 Behaviors Questionnaire, Adult Repetitive Behaviors Questionaire). All data collected prior to the study visit will be labeled with a subject ID number and will include the participant's date of birth. Study ID's are linked to each participant's name in a single, password protected database that is only accessible by the study PI and other persons approved by the institutional review board for this specific study.
During the in-person study visit, participants and/or parents will review the consent/assent/permission forms prior to beginning study procedures. Participants will then complete hearing measures, including hearing thresholds, otoacoustic emissions, digits-in-noise perception, and the Auditory Attention task. They will also complete a measure of cognitive abilities (Stanford-Binet 5th edition) and may complete the Autism Diagnostic Observation Schedule as needed for confirmation of diagnostic status.
Finally, participants will complete the cortical speech-in-noise perception paradigm. To complete this paradigm, participants will wear an elastic cap that is individually-sized based on head circumference (ActiCap outfitted with NIRX laser sources and detectors in addition to 32 channel EEG). Electrodes are gel-based, and gel will be applied prior to cap application or after cap placement depending on participant comfort. While measuring head circumference, a small amount of electrode gel will be placed on the forearm to ensure that the participant does not display skin sensitivities. Once it is determined that no sensitivities are present, the cap will be placed, with additional gel application as needed to reduce impedance; preapplication of gel reduces cap placement time. Once the cap is placed, sound stimuli will be presented via speakers on either side of a computer monitor, while non-social videos are played to maintain attention. Sound stimuli include short speech sounds (/ba/,/da/) and background babble noise. These will occur in block format, with blocks lasting 15 seconds and jitter of 13-17 seconds between blocks. Within the blocks with speech sounds, the sounds will change between consonants ba and da in order to capture neural responses to sound change. The total paradigm is 25 minutes, but participants will be able to stop at any time prior to that timepoint. Data collected during the neuroimaging protocol includes 1) a video of the cap placement to be used for photometric optode localization 2) video of the participant to be used later to exclude any trials with clear movement and 3) collection of hemodynamic and electrophysiological data from the brain. Videos used for photometric optode localization and trial exclusion will be labelled by subject ID and stored in a password protected folder in a drive that is only accessible to IRB-approved study personnel. Hemodynamic and electrophysiological data will be labelled with the subject ID and stored on the acquisition computer, which is also password protected, prior to data analysis.
Participants will then complete the audiovisual condition while they continue to wear the fNIRS/EEG cap. In this condition, participants will hear sounds as described above but will also see a shape on the computer screen that moves. The shape will be a black filled oval on a white background, with height of the oval varying similarly to sound intensity such that it is temporally aligned with vowel onset. There is no behavioral task for this measure. The investigators will use videos collected during the task to determine if the participant is looking at the screen or away prior to analysis of brain activity data.
Data will only be accessed by study personnel approved by the CCHMC IRB for those specific studies, and will be provided to the PI for this study as per data sharing agreements including participant consent for data sharing. After data is labelled with the study ID for the present study, it will not be linked to the participant's name.
Study Design
Arms and Interventions
| Arm | Intervention/Treatment |
|---|---|
| Experimental: Speech Sounds Participants will hear repeated speech sounds while wearing a neuroimaging cap. |
Behavioral: Mismatch negativity
Event related potentials are compared between frequent and infrequent speech sounds (i.e., "ba" vs "da").
|
Outcome Measures
Primary Outcome Measures
- Speech reception threshold for single digits in noise- auditory only [Baseline]
The SRT is the ratio between the loudness of speech and the loudness of background noise in decibels at which a participant can correctly report the digit that is presented at 50% accuracy.
- Speech reception threshold for single digits in noise- with visual stimuli [Baseline]
The SRT is the ratio between the loudness of speech and the loudness of background noise in decibels at which a participant can correctly report the digit that is presented at 50% accuracy. A higher SRT indicates more difficulty hearing speech in background noise
- Scaled score for auditory attention subtest [Baseline]
This task is from the Woodcock Johnson III. In this task, participants point to a picture after hearing a word presented in background noise, and accuracy is compared to same-age norms. Scores range from 1-19, with higher scores indicating better accuracy.
- correlation between neural activation to speech (fNIRS) and neural differentiation of speech-in-noise (ERP) [Baseline]
Percent change in oxygenation in auditory cortex while hearing speech alone will be correlated with amplitude of the mismatch negativity to deviant sounds. This correlation reflects how neural activation affects neural differentiation for each individual participant
- correlation between neural activation to background noise (fNIRS) and neural differentiation of speech-in-noise (ERP) [Baseline]
Percent change in oxygenation in auditory cortex while hearing background noise will be correlated with amplitude of the mismatch negativity to deviant sounds. This correlation reflects how neural activation affects neural differentiation for each individual participant
- Sensory Profile 2 [Baseline/before baseline]
The Sensory Profile is a self report or parent report measure of responses to sensory stimuli. Total symptoms are categorized as "much less than others", "less than others", "same as others", "more than others", and "much more than others" for each participant.
- Sensory Experiences Questionnaire (SEQ) [Baseline/before baseline]
The Sensory Experiences Questionnaire is a self report or parent report measure of responses to sensory stimuli in the everyday environment
- Sensory Perception Quotient [Baseline/before baseline]
The Sensory Perception Quotient is a self-reported measure of sensory concerns and symptoms
- Repetitive Behavior Questionnaire-2 [Baseline/before baseline]
Self or parent report measure of restrictive and repetitive behaviors
- Adult Repetitive Behaviours Questionnaire-2 [Baseline/before baseline]
Self or parent report measure of restrictive and repetitive behaviors
- Stanford-Binet Intelligence Scales, Fifth edition (SB-5) [Baseline]
IQ is a standardized measure of both verbal and nonverbal abilities, which will be measured with the Stanford Binet V. Standard scores range from 40-160, with a mean of 100 and higher scores indicating increased cognitive ability.
- hand dexterity ratio from the Purdue Pegboard task [Baseline]
The number of pegs placed for the left and right hands in a given amount of time in ratio form, reflects relative dexterity of the right and left hands
- Edinburgh Handedness questionnaire [Baseline/before baseline]
self or parent report measure of which hand (left or right) is used for everyday tasks and how consistently
- Autism Spectrum Quotient [Baseline/before baseline]
self or parent report of an individual's experience of autism symptoms in daily life
- Social Responsiveness Scale, Second Edition [Baseline/before baseline]
self or parent report of symptoms of autism seen in daily life. T scores have a mean of 50 and a range of 32-114, with higher scores indicating more autism symptoms.
Eligibility Criteria
Criteria
Inclusion Criteria:
-
normal audiograms (PTA ≤ 20 dB HL)
-
corrected 20/20 vision (Snellen chart)
-
no history of premature birth (prior to 36 weeks gestation)
-
no medications known to affect EEG signal
-
English as the first language
Inclusion for the Autism group requires the following:
- diagnosis of Autism Spectrum Disorder either based on previous ADOS administration and developmental history, or confirmed via ADOS and developmental history
Inclusion for the Autism + FXS group requires the following:
-
Documented PCR/Southern Blot genetic testing confirming full mutation FXS
-
Diagnosis of Autism Spectrum Disorder, per Autism group.
Inclusion for the Typically Developing group requires the following:
-
no siblings or parents with an Autism Spectrum Disorder or Fragile X Syndrome
-
no current neurological or psychiatric diagnoses
-
IQ over 75
Exclusion Criteria:
-
Hearing loss or uncorrected vision loss
-
history of premature birth (prior to 36 weeks gestation)
Contacts and Locations
Locations
No locations specified.Sponsors and Collaborators
- Children's Hospital Medical Center, Cincinnati
Investigators
- Principal Investigator: Elizabeth Smith, PhD, Children's Hospital Medical Center, Cincinnati
Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- 2023-0532