ABR: auditory brainstem response
ASD: autism spectrum disorder
DSM-III: Diagnostic and Statistical Manual of Mental Disorders, 3rd edition
DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders, 3rd edition, revised
DSM-IV: Diagnostic and Statistical Manual of Mental Disorders, 4th edition
DSM-IV-TR: Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision
FFR: frequency-following response
IPL: interpeak latency
IQ: intelligence quotient
MRI: magnetic resonance imaging
PRISMA: Preferred Reporting Items for Systematic reviews and Meta-Analyses
SOC: superior olivary complex
TD: typically developing
Multiple brain regions have been studied for clues to the pathophysiology of autism spectrum disorder (ASD). The auditory brainstem has long been considered a candidate brain region for such pathology (Ornitz et al, 1985 ; Rodier, 2002). Individuals with ASD commonly demonstrate auditory disorders consistent with brainstem abnormalities, including atypical sensitivity to sound (Crane et al, 2009), poor speech recognition in noise (Alcantara et al, 2004), impaired sound localization (Visser et al, 2013), and difficulty with auditory temporal processing, despite otherwise normal hearing (Beers et al, 2014 ; Gravel et al, 2006 ; Tharpe et al, 2006).
Despite considerable research on the structure and function of the auditory brainstem in ASD, there have been no systematic reviews of the literature to date. Assessing the evidence for brainstem abnormalities in ASD has important implications for improving our understanding of the underlying pathophysiology, potentially leading to identification of novel biomarkers and development of targeted clinical interventions and treatments. The goal of this review is to examine existing evidence for auditory brainstem abnormalities in ASD and to identify gaps in the literature and possible directions for future research.
Literature Search Strategies
We conducted a systematic online review of peer-reviewed articles published between January 1975 and August 2016 and available in English. We concurrently searched the PubMed, Embase, Cochrane, CINAHL, PsycINFO, and Web of Science online databases for studies matching any combination of the term autism with brainstem or auditory. A controlled vocabulary (Medical Subject Headings and Emtree) was used, where appropriate, in combination with additional search terms (Supplemental Digital Content 1, http://links.lww.com/CBN/A69).
Figure 1 shows our PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) flow diagram (Moher et al, 2009). Initial search results yielded 687 matches after duplicates were removed. A number of articles identified were not suitable for inclusion in the review because they were not experimental studies, did not directly investigate brainstem structure or function, or did not specify the clinical criteria used to diagnose ASD. To further refine our search, we culled articles if they did not meet three selection criteria: (1) the article was an experimental study, eg, not a review or an editorial article; (2) study participants had a clinical diagnosis of ASD; and (3) the study included at least one measure of auditory brainstem structure and/or function such as imaging or electrophysiology. Based on the second inclusion criterion, articles were excluded if they focused primarily on Rett syndrome, animal models, drug or other therapeutic interventions, neurodegenerative diseases, speech and language studies, and genetic studies. Studies of individuals diagnosed with ASD who carried an additional diagnosis of Asperger syndrome or pervasive developmental delay–not otherwise specified were included. (Hereafter, we use the term ASD to encompass these other two categories, unless noted otherwise.) Based on the third criterion, we also excluded articles that focused exclusively on the peripheral auditory system (eg, not the brainstem), including studies of peripheral hearing loss (ie, conductive, sensorineural, or mixed hearing loss). It has been shown that peripheral hearing loss is not a consistent finding (Gravel et al, 2006 ; Tharpe et al, 2006) and does not occur with increased prevalence in ASD (Beers et al, 2014) and, therefore, cannot account for the high rate of other (eg, nonperipheral) auditory abnormalities observed.
A total of 72 published articles met selection criteria for inclusion in our review. All are listed with summary information in Tables 1 through 3, where they are grouped by primary methodology (eg, anatomical, imaging, electrophysiology) to facilitate access to primary sources and for comparison across studies.
This review is organized into four sections. The first two sections focus on structural brainstem and functional brainstem studies, respectively. The third section discusses limitations of current studies, and the fourth section provides a summary and directions for future research. In addition to consistency of findings across studies, we used the following six methodological criteria to evaluate the quality of studies reviewed:
- ASD diagnosis based on standardized clinical criteria, eg, Diagnostic and Statistical Manual of Mental Disorders, 3rd or 4th edition (DSM-III/IV) (American Psychiatric Association [APA], 1985, 2000).
- Inclusion of a control group versus use of preexisting normative (eg, clinical) data.
- Adequate matching of control participants with ASD participants, eg, by age and sex.
- Audiometric screening or testing performed to exclude individuals with hearing loss.
- Study results based on adequate sample size.
- ASD participants matched by intellectual level, language ability, and chronological age
As indicated in Tables 1 through 3, the majority of studies reviewed failed to meet one or more methodological quality criteria. Methodological limitations of the studies reviewed are discussed, and studies meeting the quality criteria were given greater weight in conclusions drawn from the review.
STRUCTURAL BRAINSTEM ABNORMALITIES IN ASD
We begin with a brief overview of auditory brainstem anatomy to provide a framework for reviewing structural studies. The human auditory brainstem is a complex system of ascending, descending, and crossing pathways that process, integrate, and transmit information from the ear to the brain. The main ascending auditory nuclei are the cochlear nucleus, superior olivary complex (SOC), and inferior colliculus (Figure 2); they are located in pairs on either side of the brainstem and connected by afferent relay pathways, including the lateral lemniscus. The cochlear nucleus receives input from the peripheral auditory system (inner ear) and is located at the junction between the medulla and the pons in the brainstem. Its primary role is to maintain the fine-grained frequency (tonotopic) information extracted by the cochlea in the inner ear. The cochlear nucleus projects bilaterally to the SOC, where binaural sound cues are processed for sound localization. The SOC is also the origin of the efferent brainstem system, known as the olivocochlear bundle, which serves to protect the inner ear from loud sounds and high levels of background noise (Møller, 1983). The inferior colliculus is in the rostral portion of the brainstem and performs multiple functions, including processing frequency information and temporal and spatial sound cues (Ehret, 1997), with output projections to higher auditory structures including the medial geniculate (auditory) body of the thalamus.
Evidence for structural brainstem abnormalities in ASD derives from two main sources: postmortem studies and structural imaging studies.
A number of brain autopsy studies have reported morphological abnormalities in the brainstems of individuals with ASD (Kulesza et al, 2011 ; Kulesza and Mangunay, 2008 ; Rodier et al, 1997b). Brain specimens for these postmortem studies were obtained primarily from the Autism Tissue Program (Pickett, 2001); clinical diagnoses were confirmed retrospectively by medical records.
A number of postmortem studies have reported brainstem abnormalities at the level of the pons where the SOC is located (Kulesza et al, 2011 ; Lukose et al, 2011, 2015). One of the main SOC nuclei, the medial superior olivary nucleus, was found to be consistently abnormal in individuals with ASD as compared to non-ASD individuals (Kulesza et al, 2011). In one study, medial superior olivary nucleus abnormalities were found in all nine participants with ASD, but none of the controls, and included a reduction (77%) in the number of medial superior olivary nucleus neurons as well as differences in the orientation and type of neurons (Kulesza et al, 2011). In a larger follow-up study of 16 individuals with ASD aged 2 to 56 years, a 45% reduction in the number of SOC neurons was observed, and the medial superior olivary nucleus was significantly smaller as compared with 10 typically developing (TD) controls aged 3 to 32 years (Lukose et al, 2015). Similar brainstem structural abnormalities have also been observed in animal models of ASD (Lukose et al, 2011). Evidence for structural abnormalities at the level of the SOC is of interest because a number of associated auditory functions, including sound localization, sound sensitivity, and suppression of background noise, are often impaired in ASD.
Despite reported reduction of SOC nuclei in ASD, no consistent differences in overall brainstem volumes between individuals with ASD and matched controls have emerged from postmortem studies. Indeed, a recent postmortem study using immunocytochemical staining of 10 brains of individuals with ASD aged 1.8 to 28 years reported increased neuronal development, or neurogenesis, in the brainstem, including pons and midbrain as well as cerebellum and primary auditory cortex, that was prolonged compared with individuals without ASD (Azmitia et al, 2016).
Magnetic Resonance Imaging Studies
Magnetic resonance imaging (MRI) studies of brainstem structure and volumes in individuals with ASD have yielded equivocal findings. A number of early MRI studies using area measurements derived from midsagittal images reported reduction of the entire brainstem and/or the pons in individuals with ASD as compared with controls (Ciesielski et al, 1997 ; Gaffney et al, 1988 ; Hashimoto et al, 1992a, 1992b, 1995). However, not all studies that used this approach found differences (Elia et al, 2000 ; Herbert et al, 2003 ; Hsu et al, 1991). Moreover, several MRI studies that used volumetric measurements found no differences in overall brainstem volume between individuals with ASD and controls (Hardan et al, 2001 ; Herbert et al, 2003). However, another study that examined gray and white matter brainstem differences in 22 individuals with ASD reported reduction of gray matter volume but not white matter volume when compared to 22 age- and sex-matched controls (Jou et al, 2009). Interestingly, no correlations were observed between gray matter volume and intelligence quotient (IQ), although oral sensory sensitivity was found to be significantly correlated (Jou et al, 2009). A subsequent longitudinal volumetric MRI study examined brainstem development over a 2-year period in 23 boys with ASD and 23 age- and sex-matched controls (Jou et al, 2013). Participants with ASD experienced more rapid growth, primarily in bilateral gray matter volume, during the 2-year period. However, by age 15, whole brainstem volumes were similar between groups.
In summary, both postmortem and structural MRI studies suggest that brainstem anomalies are present at the level of the pons (SOC) in individuals with ASD (Table 1). However, only 4 of the 16 (25%) studies reviewed met methodological quality criteria. The main methodological limitations identified were small sample sizes, deficiencies in matching participant characteristics, and failure to specify the clinical criteria used in the diagnosis of ASD. Involvement of the SOC is consistent with a cluster of auditory abnormalities often observed in individuals with ASD that includes hypersensitivity to sound, poor sound localization, and impaired listening in background noise. There is some indication that these brainstem anomalies occur early in gestation (<6 weeks), result in increased and prolonged neurogenesis, and may differentially affect gray matter volume. However, with regard to an overall reduction of brainstem size in individuals with ASD, the results remain equivocal.
The majority of postmortem studies reviewed had relatively small samples, ranging from 1 (Rodier et al, 1997b) to a maximum of 16 individuals with ASD (Lukose et al, 2015). The IQ levels of ASD participants were not specified in many of the postmortem studies (Azmitia et al, 2016 ; Kulesza et al, 2011 ; Kulesza and Mangunay, 2008 ; Lukose et al, 2015 ; Rodier et al, 1997b), and one study included participants with the diagnosis of pervasive developmental delay–not otherwise specified (Kulesza et al, 2011). The imaging studies had larger sample sizes overall, ranging from 13 to 102 participants with ASD. A number of factors may have contributed to the variability in findings across studies, including heterogeneity in intellectual levels and ages, small sample sizes, inclusion of participants with non-ASD diagnoses such as pervasive developmental delay, and use of different diagnostic criteria.
FUNCTIONAL BRAINSTEM ABNORMALITIES IN ASD
A number of studies have examined the functional integrity of auditory brainstem structures and pathways in ASD using objective electrophysiological measures such as the auditory brainstem response (ABR) test, frequency-following response (FFR), acoustic reflex testing, reflex modulation, and otoacoustic emissions. The ABR test measures sound-evoked, synchronous neural activity from the auditory nerve up through the inferior colliculus in the brainstem and has been widely used in ASD studies.
The ABR is composed of five major components or waveforms (I-V) that normally occur within 10 to 12 milliseconds after stimulus onset (clicks, tone bursts) and are generated by the auditory nerve (waves I-II), the cochlear nucleus (wave III), the SOC (wave IV), and the lateral lemniscus and inferior colliculus (wave V) (Møller et al, 1981, 1982). ABR studies performed in children younger than 5 years of age are often performed under sedation. The three main ABR waveform measurements used in ASD studies are individual peak (also called absolute) latencies and amplitudes, interpeak latencies (IPLs) (I-III, III-V, I-V), and interaural latencies. Although a few studies have reported abnormal ABR waveform amplitudes (see, eg, Garreau et al, 1984), the majority of studies have focused on waveform latency measurements (peak, interpeak, and interaural) of the three most stable components or waveforms: waves I, III, and V. ABR findings for absolute peak latencies, IPLs, and interaural latencies in ASD studies are discussed separately.
ABR Absolute (Peak) Latencies
ABR studies have yielded equivocal findings regarding wave I latencies in normal-hearing individuals with ASD. A number of studies have reported normal wave I latencies, consistent with the location of the main wave I generator in the auditory nerve proximal to the inner ear (Courchesne et al, 1985 ; Tharpe et al, 2006). However, other studies have reported abnormal absolute wave I latencies that were either prolonged (Tanguay et al, 1982) or early (Dabbous, 2012) despite normal hearing. In contrast to the equivocal wave I findings, numerous ABR studies have reported abnormal (prolonged) absolute latencies for waves III and/or V in infants (Miron et al, 2016) and children (Cohen et al, 2013 ; Dabbous, 2012 ; Rosenblum et al, 1980 ; Roth et al, 2012) with ASD.
In a study of young normal-hearing children with ASD aged 24 to 45 months undergoing sedated ABR (Roth et al, 2012), wave I latencies were normal, but waves III and V were prolonged in 50% of participants, as compared with 8% of age- and sex-matched children with language delays. The authors suggested that ABR abnormalities may be more common in younger children with ASD. This view is supported by a recent study of 118 infants aged 0 to 3 months (n=30) and toddlers aged 1.5 to 3.5 years (n=40) who were later diagnosed with ASD (Miron et al, 2016). A number of participants (n=48) were excluded for a variety of reasons (eg, elevated ABR thresholds, genetic anomalies, or age at time of testing), leaving only 70 participants for examination. For the infant group, wave V latencies were prolonged as compared with matched controls, and absolute wave V latencies for the right ear accurately predicted 21 of 30 individuals with ASD and 24 of 30 individuals in the normative group using a classification algorithm. For the toddler group, absolute waveform latencies were significantly prolonged for both ears. Moreover, abnormal wave III and V latencies have been reported for young children undergoing ABR testing with sedation (Sersen et al, 1990) or without sedation (Magliaro et al, 2010).
In contrast to studies of young children with ASD, absolute ABR peak latencies are not consistently abnormal in studies of older children. A study of ASD children aged 7 to 13 years showed prolonged peak latencies for only the right ear (Fujikawa-Brooks et al, 2010), while a recent study of 60 older children aged 5 to 18 years found that only 7% had any ABR abnormalities when compared with age-matched controls (Demopoulos and Lewine, 2016). A study of children with ASD aged 3 to 10 years reported normal wave I, III, and V peak latencies (Tharpe et al, 2006), as did a study of 14- to 28-year-old participants with ASD (Courchesne et al, 1985), a study of eight ASD participants with a mean age of 23 years (Grillon et al, 1989), and a study of 5- to 40-year-old participants with pervasive developmental delay–not otherwise specified that included autism (Rumsey et al, 1984).
Taken together, these ABR studies suggest that absolute latencies for waves III and V are frequently prolonged in younger but not older children with ASD. However, it is important to note that several studies of older ASD children (Courchesne et al, 1985 ; Grillon et al, 1989) had relatively small sample sizes (≤15 subjects) or did not include a control group, relying instead on normative data (Demopoulos and Lewine, 2016).
ABR Interpeak Latencies
Multiple studies have reported abnormal (eg, prolonged) ABR IPLs in individuals with ASD (Magliaro et al, 2010 ; Maziade et al, 2000 ; Roth et al, 2012 ; Taylor et al, 1982 ; Wong and Wong, 1991). In one study, all IPLs were prolonged in children with ASD and to a lesser extent in children with language delays who did not meet criteria for ASD (Roth et al, 2012). Several ASD studies have reported prolonged IPLs for the I-III and I-V intervals (Magliaro et al, 2010 ; Maziade et al, 2000 ; Taylor et al, 1982 ; Wong and Wong, 1991). In a comparatively large sample of 73 individuals who met DSM-III-R criteria for ASD (mean age=7.3 years), significant prolongation of the IPL for waves I-III and I-V was observed relative to controls matched for age and sex (Maziade et al, 2000). First-degree but not second- or third-degree relatives also had significantly prolonged wave I-III IPLs relative to controls (Maziade et al, 2000). In 48% of families, either the individual with ASD or a relative had a prolonged I-III IPL interval (Maziade et al, 2000). The authors note, however, that approximately 50% of individuals with ASD and their relatives showed no IPL abnormalities.
In addition to a prolonged I-III IPL, studies have also reported prolonged III-V IPLs (Fein et al, 1981 ; Magliaro et al, 2010 ; Rosenhall et al, 2003 ; Sersen et al, 1990 ; Tas et al, 2007 ; Wong and Wong, 1991) and/or I-V IPL interval (Magliaro et al, 2010 ; Rosenhall et al, 2003 ; Skoff et al, 1980 ; Wong and Wong, 1991). In a study involving 71 children who met DSM-IV-TR criteria for ASD, significant prolongation of the I-V and III-V IPLs was observed in comparison to IPLs obtained for a TD control group (Kwon et al, 2007). In another study, 80% of participants with ASD had prolongation of one or more IPLs (Thivierge et al, 1990). A recent study of 30 children aged 3 to 7 years, who met DSM-IV criteria for ASD and represented a wide range of intellectual disabilities, reported significant delays in the I-V and III-V IPLs (Azouz et al, 2014). Based on a sensory checklist, 23 of 30 ASD participants showed some evidence of auditory abnormalities, including hypersensitivity to sound. Atypically short IPL intervals (ie, I-III, III-V, and I-V) have also been reported (Ververi et al, 2015).
ABR testing was performed in a longitudinal study of infants in neonatal intensive care units (Cohen et al, 2013). All children received an ABR assessment while in the neonatal intensive care unit as well as an arousal-modulated attention procedure at age 4 months (Gardner and Karmel, 1995). The arousal-modulated attention procedure purports to be a predictor of later neurodevelopmental outcomes and reflects a homeostatic effect, which is believed to be mediated at brainstem levels (Gardner and Karmel, 1995). Infants with initially abnormal ABRs (n=46) and initially normal ABRs (n=28) were identified. Abnormalities consisted of prolonged IPLs with normal wave I latencies. ABR abnormalities were normalized by 1-month post-term age for all but one of the cases. Follow-up evaluations were undertaken at 24 to 36 months, at which time the mothers completed the Pervasive Developmental Disorder Behavior Inventory (Cohen et al, 2010). At follow-up, 14 cases of ASD were identified; 13 of the 14 cases were in the early ABR abnormality group (Cohen et al, 2013). Early ABR abnormalities and atypical visual preference, as assessed by arousal-modulated attention, were predictive of later competencies in language and in social domains as well as later ASD diagnoses (Cohen et al, 2013).
Twenty-eight of the ABR studies reviewed measured all three IPLs (I-III, III-V, I-V). Of these, 20 studies reported abnormalities of one IPL or more in participants with ASD (Table 2). Four studies that reported normal IPLs had fewer than 20 subjects, raising the possibility that they may not have been adequately powered to detect abnormalities (Courchesne et al, 1985 ; Garreau et al, 1984 ; Grillon et al, 1989 ; Thabet and Zaghloul, 2013). Similarly, several early studies that reported normal IPLs did not include ABR results from control subjects (Novick et al, 1980), included disproportionate numbers of female control participants (Shivashankar and Satishchandra, 1989 ; Student and Sohmer, 1978), used outdated ASD diagnostic criteria (McClelland et al, 1992), or provided no description of the ASD diagnostic criteria used (Takesada et al, 1992).
ABR Interaural Latencies
In addition to measuring IPLs between ABR waveform peaks for a given ear, the presence of asymmetries between ears (ie, interaural latency differences) has also been examined. One study reported that the I-V interaural time difference was significantly prolonged in 18% of the participants with ASD (Rosenhall et al, 2003). In a group of children with ASD and hypersensitivity, shorter wave III-V and I-V IPLs were obtained in comparison to an age- and sex-matched control group; however, only children with ASD who had issues with hypersensitivity were included in the study (Thabet and Zaghloul, 2013). The effect of an increased rate of click presentation on ABR absolute and IPLs was examined in a group of 20 children with ASD and in 20 age- and sex-matched TD children (Fujikawa-Brooks et al, 2010). The latency of wave V for the left ear was significantly prolonged in the ASD group as compared to the TD group.
ABR loudness growth functions have also been studied in ASD to investigate possible neural correlates of the hyperacusis or atypical sensitivity to sound, which is common in ASD. A study of 25 children with ASD aged 1.5 to 3.33 years and 25 TD controls was conducted to objectively estimate uncomfortable loudness levels (Dabbous, 2012). While ABR abnormalities were found in the group of children with ASD, the mean slope of the ABR latency intensity function did not differ between ASD and TD groups, suggesting that hyperacusis is not associated with abnormal loudness growth functions in ASD (Dabbous, 2012).
Another application of the ABR has involved use of a forward masking procedure to assess processing of more complex stimuli at brainstem levels (Källstrand et al, 2010). It was shown that wave III amplitudes were significantly lower in the participants with ASD than in any other groups (ie, controls, patients with schizophrenia, and patients with attention deficit hyperactivity disorder) examined in the forward masking condition (Källstrand et al, 2010). The authors suggested that measures of forward masking may be related to mechanisms underlying echo suppression and the ability to process speech in noisy environments.
Summary of ABR Findings
A majority of the studies reviewed (29 of 36) report atypical ABR findings in individuals with ASD (Table 2). Two largely converging findings emerge from these studies: (1) younger children at risk for ASD appear more likely to have ABR abnormalities than older children; and (2) the most common abnormalities observed in young children with ASD were prolonged waves III and V absolute latencies and the corresponding waves I-III, III-V, and I-V IPLs. These findings are consistent with abnormal neural transmission between lower and higher levels of the brainstem. Evidence for abnormal IPLs involving wave V are also consistent with previously discussed structural brainstem findings in ASD, suggesting involvement at the level of the SOC that could affect transmission of auditory information to the inferior colliculus, the main generators of wave V. In contrast, evidence for ABR abnormalities in older children and abnormal wave I latencies in ASD children of all ages remain equivocal.
Several factors may have contributed to these discrepant findings, including differences in the selection and matching parameters (eg, sex, age) of the control groups, small sample sizes, and inclusion of participants with peripheral hearing loss. For example, inclusion of female participants in a control group could increase the likelihood of reporting ABR abnormalities in predominantly male ASD participants due to overall shorter IPLs of females (Stockard et al, 1979) that may be present from an early age (Stuart and Yang, 2001). Other demographic differences within and between subject groups include differences in ASD diagnostic criteria, functional level, and age range. While some studies focused on a specific age range, eg, young children aged 1 to 3 years or 24 to 45 months (Dabbous, 2012 ; Roth et al, 2012), other studies included wider age ranges, eg, ages 5 to 40 or 23 to 58 years (Källstrand et al, 2010 ; Rumsey et al, 1984). Eight of the studies reviewed did not report participants’ IQs or functional levels. Similarly, while some studies excluded (or analyzed separately) participants with known neurologic disorders such as seizures (Rumsey et al, 1984 ; Sersen et al, 1990), this exclusion criterion does not appear to have been applied routinely across studies. Likewise, some studies included participants with known hearing loss (Taylor et al, 1982), while others excluded participants with hearing loss or history of middle-ear dysfunction (Rumsey et al, 1984 ; Tharpe et al, 2006). Overall, only 2 of the 36 studies reviewed met all six methodological quality criteria.
In addition to the ABR, four other objective measures have been used to assess auditory brainstem function in individuals with ASD: the FFR, the acoustic reflex, otoacoustic emissions, and reflex modulation audiometry. Findings from studies using each of these measures are discussed below (see also Table 3).
The FFR is a sustained electrophysiological response that measures how well neuronal populations phase-lock to the envelope of a low- to mid-frequency periodic stimulus. The FFR is generated by the brainstem, but unlike the ABR, which is elicited using a transient auditory stimulus (eg, click, tone burst), the FFR is elicited using a continuous stimulus that is modulated in amplitude or frequency at a regular (periodic) rate. The FFR can be elicited with a variety of periodic stimuli, including amplitude-modulated tones and vowel sounds.
A small number of studies have used the FFR to measure phase-locking to amplitude- and frequency-modulated tones and speech sounds (eg, vowels) in individuals with ASD. One FFR study reported reduced phase-locking to speech sounds in children with ASD compared to TD controls (Russo et al, 2009), suggesting that impairments in auditory brainstem function may contribute to language deficits in children with ASD. An earlier FFR study from the same group reported abnormal FFR for rising and falling pitch contours in individuals with ASD, suggesting brainstem-level deficits in the encoding and tracking of prosodic elements of speech (Russo et al, 2008). It is of interest that both of these studies excluded ASD participants with prolonged absolute wave V latencies or abnormal click-evoked ABR findings. Thus, the observed deficits in phase-locking and processing of pitch contours in participants with ASD were independent of abnormalities in the click-evoked ABR summarized earlier in this review.
Acoustic Reflex Studies
The acoustic reflex is a standard component of the clinical hearing evaluation and is used in conjunction with tympanometry to assess middle-ear function and the integrity of auditory brainstem pathways. The acoustic reflex is elicited bilaterally when a reflex-eliciting stimulus is presented to either ear and is measured by monitoring changes in acoustic immittance. The neural connections underlying the acoustic reflex form a circuit, or arc, that has both afferent and efferent components (Pillion, 2012).
Studies measuring the acoustic reflex in individuals with ASD are listed in Table 3. An early study reported absent acoustic reflexes in 17 children with ASD aged 3 to 9 years (Suriá and Serra-Raventós, 1975). None of the children had a history of middle-ear pathology or hearing loss. Criteria used to diagnose ASD were not reported. A more recent study reported no differences in ipsilateral or contralateral acoustic reflexes between 40 children with ASD and 40 TD children (Gravel et al, 2006). Three participants with ASD were excluded due to the presence of sensorineural hearing loss, middle-ear dysfunction, or the presence of pressure-equalization tubes. Diagnosis of ASD was based on the Autism Diagnostic Observation Schedule (Lord et al, 1989) and Autism Diagnostic Interview (Lord et al, 1994). Acoustic reflex measurements were completed for all but one child in the ASD group. In that case, a hermetic seal could not be maintained (Gravel et al, 2006).
In a related study, ipsilateral (right ear) acoustic reflexes were measured in 22 children with ASD who ranged in age from approximately 3 to 10 years and in 22 age- and gender-matched TD children (Tharpe et al, 2006). The children with ASD met DSM-IV criteria for ASD. Measurements were completed on 15 of the children with ASD and 21 TD children. No group differences in acoustic reflex thresholds were found between the two groups of children (Tharpe et al, 2006). In another study, no significant differences in ipsilateral acoustic reflexes were found in children and teenagers with ASD who met DSM-IV criteria for ASD and were reported by caregivers to have hypersensitivity as compared to children and teenagers with ASD without hypersensitivity (Gomes et al, 2004). In summary, none of the three studies found acoustic reflex abnormalities in their ASD participants.
In contrast, a study that purported to use a more sensitive measure of the acoustic reflex found lower acoustic reflex thresholds and significantly prolonged ipsilateral reflex latencies in individuals aged 4 to 23 years with ASD relative to TD individuals (Lukose et al, 2013). Interestingly, longer acoustic reflex latencies for ipsilateral stimulation were found only for the left ear (Lukose et al, 2013). While all participants in the group with ASD previously had been diagnosed with ASD, the precise diagnostic criteria were not specified. Participants with ASD had been referred to the facility for auditory processing evaluations because of issues with “auditory sensitivities” or “auditory inattention.” The authors suggested that the acoustic reflex abnormalities observed in their participants with ASD reflect dysfunction at the level of the pons, particularly the SOC, in the brainstem where the medial olivocochlear efferent system originates (Lukose et al, 2013). An earlier study by Khalfa et al (2001b) also found intra-aural asymmetries in efferent activity of the medial olivocochlear system in children and adolescents with ASD.
The finding of lowered acoustic reflex thresholds has been associated with “hyperacute” pure tone thresholds in children with learning disorders, but not hypersensitivity to loud auditory stimuli (Gordon, 1986), which would appear to be at odds with the more commonly observed hypersensitivity to sound in individuals with ASD.
In a recent study, abnormal acoustic reflex findings occurred with greater frequency in participants with ASD (37%) than controls (6%) (Demopoulos and Lewine, 2016). Participants with ASD were also found to experience atypical sound sensitivity with significantly greater frequency (37%) than the control group (0%). Participants with ASD were 60 individuals meeting DSM-IV-TR criteria for ASD. There was a wide range of IQs (46 to 136) represented in the study, and the authors suggest that the discrepancies between their results and those of Gravel et al (2006) reflect inclusion of only individuals with higher-functioning ASD. However, the IQs represented in the two studies appear quite similar, and it seems more likely that the discrepancy in results reflects the more stringent audiometric exclusion criteria used by Gravel et al (2006).
Summary of Acoustic Reflex Studies
Approximately half of the studies reviewed reported abnormal acoustic reflex findings in individuals with ASD (Table 3). Abnormal findings included acoustic reflex thresholds at abnormally low levels, absent acoustic reflexes, and prolonged latencies. However, not all of the studies excluded participants with hearing loss. Moreover, the one study that failed to elicit measurable acoustic reflexes in any ASD participants did not include a control group (Suriá and Serra-Raventós, 1975). The finding of abnormally low acoustic reflex thresholds is consistent with the presence of hyperacusis, but not necessarily hypersensitivity, for moderate to loud auditory signals (Gordon, 1986). Across the six studies reviewed, participants’ ages varied considerably, from 3 to 40 years, with three studies focused exclusively on pediatric participants. Similarly, participants’ functional levels ranged from profound intellectual disability to above-average intelligence. Most of the acoustic reflex studies reviewed did not include control groups and were limited by relatively small sample sizes.
Otoacoustic Emissions Studies
Measurements of otoacoustic emissions have been used to study efferent auditory brainstem pathways in ASD (Collet et al, 1993 ; Danesh and Kaf, 2012 ; Khalfa et al, 2001b). The primary function of the efferent brainstem system is to protect the inner ear from loud sounds by damping the movement of the outer hair cells in the cochlea and to improve signal detection in background noise. The origin of the brainstem efferent response is the medial olivocochlear bundle at the level of the SOC and is assessed using contralateral masking. Collet et al (1993) examined contralateral suppression of transient-evoked otoacoustic emissions in 11 participants who met DSM-III (American Psychiatric Association, 1985) criteria for ASD and ranged in age from 6 to 30 and in controls matched by sex and age. Participants with ASD were found to have reduced contralateral suppression in comparison with controls. Overall, transient-evoked otoacoustic emission amplitudes did not differ between groups. The authors concluded that deficient functioning in the medial olivocochlear bundle may be related to the presence of hypersensitivity and may also be an indication of brainstem dysfunction in ASD (Collet et al, 1993). In a subsequent study, Collet and colleagues found no differences in the amount of contralateral suppression measured between 22 participants, ranging in age from 4 to 18, who met DSM-IV criteria for a diagnosis of ASD and in controls (Khalfa et al, 2001b). Significantly greater suppression was found for the right ear than the left in the group with ASD; no ear difference was found for the control group (Khalfa et al, 2001b). In addition, transient-evoked otoacoustic emission amplitudes decreased as a function of increased age in the ASD group but not in the control group (Khalfa et al, 2001b), suggesting the presence of decreasing peripheral auditory sensitivity with increasing age in the ASD group.
A more recent study showed no differences in the amount of binaural suppression in transient-evoked otoacoustic emissions between 35 participants with ASD and 42 TD controls who ranged in age from 6 through 17 (Bennetto et al, 2017). ASD diagnosis was confirmed by administering the Autism Diagnostic Interview–Revised and the Autism Diagnostic Observation Schedule. A diagnosis of ASD was also ruled out in the TD group by administration of the Autism Diagnostic Observation Schedule and the Social Responsiveness Scale (Constantino and Gruber, 2005). A possible explanation for the differences between the findings of Bennetto et al (2017) and Khalfa et al (2001b) may be that the former focused on individuals with relatively high-functioning ASD and used masking noise delivered both ipsilateral and contralateral to the test ear (Bennetto et al, 2017).
Another type of otoacoustic emission, distortion product otoacoustic emissions, has also been used in examining functioning of the medial olivocochlear bundle in 14 children with ASD aged 6 to 14 years (Danesh and Kaf, 2012). Overall amplitude of distortion product otoacoustic emissions was lower in the group with ASD than in the control group. In addition, the amount of contralateral suppression was greater in the control group than in the group with ASD (Danesh and Kaf, 2012). The authors suggested that hypersensitivity in ASD is related to both cochlear dysfunction and less efficient auditory efferent pathway activity (Danesh and Kaf, 2012). On the other hand, the same authors, in a more recent study, showed no differences between controls and a group diagnosed with higher-functioning Asperger syndrome in contralateral suppression of distortion product otoacoustic emissions (Kaf and Danesh, 2013).
Taken together, these two studies highlight how differences in severity of involvement, subject selection criteria, and/or level of intellectual functioning may contribute to discrepant findings. The 2012 Danesh and Kaf study excluded participants with Asperger syndrome but did not test for normal hearing abilities. In contrast, the 2013 Kaf and Danesh study included only participants with Asperger syndrome, and all had audiometric testing to confirm normal hearing.
Summary of Otoacoustic Emissions Studies
Only five studies met the criteria for review; of these, the severity of ASD, which ranged from mild to moderate intellectual disability, was specified in only three studies (Table 3). All five studies specified the criteria used in participant selection, although the criteria were not uniform across studies. All but one study reviewed above found abnormalities in the nature of contralateral suppression measured in participants with ASD, although the nature of the abnormalities varied across studies.
Reduced contralateral suppression in ASD compared with controls was found in two of five studies (Collet et al, 1993 ; Danesh and Kaf, 2012). One study used binaural suppression and found no differences between participants with ASD and TD controls (Bennetto et al, 2017). One study found a right-left asymmetry in the amount of contralateral suppression only in the group of participants with ASD (Khalfa et al, 2001b). Reduced contralateral suppression is consistent with the presence of a brainstem-level deficit in efferent auditory function mediated by the medial olivocochlear bundle. Outcomes in this area should be viewed with caution because of the small effect size of contralateral noise on otoacoustic emission amplitude, which is on the order of 1 to 2 decibels (de Boer and Thornton, 2008 ; Sun, 2008). Measurement of suppression effects is highly dependent on the signal-to-noise ratio, and the magnitude of the effect can vary considerably across participants (Goodman et al, 2013). Moreover, the majority of studies on contralateral suppression of otoacoustic emissions were based on relatively small numbers of participants.
Although structural abnormalities and comorbid neurologic issues were ruled out in some cases, this was not done uniformly across studies. Additional issues include heterogeneity of the ASD population and differences in diagnostic criteria. It should also be noted that the brainstem receives efferent input from both the thalamus and the cortex. Therefore, abnormalities in either structure could affect brainstem function (Khalfa et al, 2001a).
Reflex Modulation Audiometry
Reflex modulation audiometry has been used more rarely to examine the integrity of brainstem auditory pathways in individuals with ASD, as the procedure requires equipment that is not generally used in clinical practice. The paradigm involves presenting a startle-eliciting stimulus that is preceded by a lower-level auditory signal, termed pre-pulse, which serves to modulate or inhibit the magnitude of the startle response by as much as 80% to 90% (Fendt et al, 2001). Acoustic pre-pulses are processed by the ascending auditory pathway and require an intact inferior colliculus for pre-pulse inhibition to occur (Fendt et al, 2001). One of the first studies examined pre-pulse inhibition in 54 participants who met DSM-III criteria for ASD and ranged in age from 2.8 to 33 years (Ornitz et al, 1993). Electromyogram measurements of the orbicularis oculi were recorded as an index of the magnitude of the startle response. Findings indicated no consistent differences in the impact of pre-pulse stimulation on the startle response between the participants with ASD and age-matched controls.
A subsequent study was undertaken with 21 adult participants with Asperger syndrome (diagnosis based on 10th revision of the International Statistical Classification of Diseases and Related Health Problems criteria) and normal intelligence (McAlonan et al, 2002). The study also included MRI measurements of brain anatomy. Measurements of startle amplitude and pre-pulse inhibition completed with 12 of the participants with Asperger syndrome showed significantly reduced pre-pulse inhibition of the startle response in comparison with controls (McAlonan et al, 2002). MRI findings revealed abnormalities in the frontal-striatal region, a brain region associated with sensory-motor gating (McAlonan et al, 2002). Another study involving 14 adults with ASD (DSM-IV criteria) also measured the startle response by measuring electromyogram signals from the orbicularis oculi muscle (Perry et al, 2007). The authors reported significantly reduced pre-pulse inhibition. In addition, the percent of pre-pulse inhibition correlated negatively with the restrictive and repetitive behaviors subscale on the Autism Diagnostic Interview–Revised. Specifically, reduced startle inhibition was associated with increased restricted or repetitive behaviors (Perry et al, 2007).
However, more recent studies have failed to show differences in the impact of pre-pulse stimulation on startle amplitude between children with ASD (Oranje et al, 2013) or adults with ASD (Kohl et al, 2014) and controls. In adults with ASD, significantly greater overall amplitudes of startle were reported (Kohl et al, 2014). In contrast, some recent studies have shown reduced pre-pulse inhibition for individuals dually diagnosed with ASD and fragile X syndrome (Yuhas et al, 2011) and in children with ASD (Madsen et al, 2014). The later study included 35 participants aged 8 to 12 years who met DSM-IV-TR criteria for ASD.
In another study, acoustic stimuli were used to elicit startle, which was measured by eye blinks (Takahashi et al, 2016). Participants were 17 children aged 8 to 16 years who met DSM-IV-TR criteria for ASD. The children with ASD were found to have significantly prolonged peak startle latencies and greater startle magnitudes at intensities of 95-decibel sound pressure level and lower in comparison to TD controls. No significant differences between participants with ASD and controls were seen for a measure of pre-pulse inhibition or for startle amplitude at stimulus intensities greater than 95-decibel sound pressure level (Takahashi et al, 2016). At 1-year follow-up, the measures were replicated, and there were no significant test-retest differences on any of the measures (Takahashi et al, 2017). Using the same methodology as summarized above, an earlier study had shown a relationship between behavioral traits on the Social Responsiveness Scale (Constantino and Gruber, 2005) and startle amplitude as well as startle latency (Takahashi et al, 2014) in a relatively small sample of 12 children ranging in age from 6 to 17 who met DSM-IV-TR criteria for ASD.
Summary of Reflex Modulation Studies
While 8 of the 10 reflex modulation studies reviewed reported abnormal findings, reduced pre-pulse inhibition was reported in only three of the studies (Table 3). However, one research group consistently found greater startle amplitudes in ASD participants for relatively low-level, startle-eliciting stimuli (Takahashi et al, 2014, 2017). In addition, a relationship was found between behavioral traits on the Social Responsiveness Scale and greater startle amplitude as well as prolongation of startle latency (Takahashi et al, 2014, 2016, 2017). The level of startle-eliciting stimuli appears to be a significant confounding factor in this area. The criteria used in participant selection were reported in all studies, although three different criteria were used in the 10 studies reviewed and 6 of the studies were based on fewer than 20 participants. With the exception of one early study, the functional level in most of the reflex modulation participants appears to be higher than in other studies reviewed (Ornitz et al, 1993).
LIMITATIONS OF EXISTING RESEARCH
Although the majority of studies reviewed provide converging evidence for the presence of auditory brainstem pathology in individuals with ASD, a number of methodological limitations warrant consideration before drawing definitive conclusions.
General Methodological Limitations
A number of methodological differences may have contributed to discrepant findings across ASD studies, including differences in ASD diagnostic criteria, severity of ASD, participants’ ages, small sample sizes, and the presence of comorbid neurologic disorders.
Inclusion criteria may be a significant confounding factor in interpreting studies on auditory function in ASD and other areas as well (Pickett et al, 2009). The criteria for diagnosing autism have evolved over time, and only in recent years have studies included more objective measures such as the Autism Diagnostic Observation Schedule in participant selection (Lord et al, 1989). Early studies often did not provide the clinical criteria used to diagnose participants with ASD as described in the DSM and/or included individuals with other diagnoses such as neurologic disorders or psychosis (Gillberg et al, 1983). Studies have also varied in the level of severity and developmental level in the participants with ASD. A number of studies have focused on participants with higher-functioning ASD, perhaps because those individuals were more readily assessed with the measures included in the studies. However, clinical observations suggest that individuals with lower-functioning ASD may be more likely to experience issues with auditory hypersensitivity (Patten et al, 2013).
The age of participants included in the reviewed studies varies widely, from less than 1 year to 58 years of age. While several studies focused on a relatively limited age range (eg, 24 to 45 months in Roth et al, 2012), other studies have included participants in a wide age range (eg, 35 years) (Källstrand et al, 2010 ; Rumsey et al, 1984). This is a potential concern for several reasons: (1) age is commonly used as a surrogate index of development; and (2) auditory brainstem pathways may be more impaired in younger than older children with ASD (Roth et al, 2012). Another general limitation of existing ASD brainstem studies is the reliance on small sample sizes in a clinical population known to be highly heterogeneous, limiting the statistical power and generalizability of the findings (Ioannidis, 2014 ; Ioannidis and Trikalinos, 2005).
It is increasingly recognized that ASD is more of a syndrome than a specific etiologic entity. Sources of the heterogeneity include genetic (Parikshak et al, 2016), neuroanatomical, and behavioral factors. To date, few if any studies have attempted to take such heterogeneity into account, and many do not even provide the data that would allow reconstruction of the specific characteristics of their participants in light of current knowledge.
Finally, an initial review of the literature showed that almost all the relevant literature was in English or, at the very least, duplicated in English. Although our review was limited to articles in English, this did not appear to be a significant limitation.
Limitations of Anatomical Studies
Several of the early anatomical studies reviewed were based on a single participant with ASD (Rodier et al, 1996, 1997b). Subsequent anatomical studies reporting morphological brainstem anomalies in ASD have been based on small sample sizes and/or have not included a control group. An additional limitation of postmortem anatomical studies is that it is not possible to relate the claimed structural anomalies to functional performance.
Limitations of Functional Studies
Abnormalities in the ABR have been reported across multiple studies, although the specific nature of the abnormalities can vary. It seems likely that participant and methodological variables may account for the disparate results reported in ABR studies and other measures of brainstem function in ASD as noted above. The presence of peripheral hearing loss could also be a significant factor in some studies, in particular studies showing a shortening of the interpeak interval for waves I-V (Keith and Greville, 1987).
Several acoustic reflex studies have reported anomalies suggestive of dysfunction of the efferent brainstem system in ASD. However, this finding remains equivocal, as other studies have reported normal acoustic reflexes. Inclusion of individuals with a history of middle-ear dysfunction, a counter-indication for testing acoustic reflexes, may account, in part, for the discrepant findings.
Although studies using contralateral suppression of otoacoustic emissions have also suggested less efficient efferent auditory pathways in ASD, these findings have not been uniform across studies. Reduced contralateral suppression of otoacoustic emissions would potentially be a finding of some significance, as individuals with ASD have been shown to experience difficulty processing speech in certain types of noise backgrounds (Alcantara et al, 2012). Measurement of contralateral suppression is not widely used clinically, and findings suggesting a relationship between reduced contralateral suppression and speech processing in noise have not always been replicable in TD individuals (Stuart and Butler, 2012). Further research is needed in this area to confirm the presence of reduced contralateral suppression in individuals with ASD.
Reflex modulation appears to be well suited for the investigation of neural mechanisms underlying auditory processing issues in individuals with ASD, since the neural processes mediating pre-pulse inhibition and startle are well understood. Reflex modulation studies have had conflicting results, with several studies showing abnormalities in processes underlying sensory-motor gating. However, not all reflex modulation studies have reported differences between individuals with ASD and matched normal controls. The intensity level of startle-eliciting stimuli appears to be a significant confounding factor in research on reflex modulation in individuals with ASD. One significant advantage for the study of reflex modulation is that an association has been demonstrated between prolonged startle latency, startle magnitude, and scores on the Social Responsiveness Scale (eg, social awareness, social cognition, and autistic mannerisms) (Takahashi et al, 2014). Reflex modulation has been used to study brainstem mechanisms in a wide variety of diagnostic groups and is viewed as a very promising measure for translational research (Takahashi et al, 2011). However, neural mechanisms mediating startle and pre-pulse inhibition are not fully mature until about 8 years of age (Takahashi et al, 2011), which limits the utility of reflex modulation during the early years when the symptomatology of ASD initially appears.
CONCLUSIONS AND DIRECTIONS FOR FUTURE RESEARCH
The majority of the studies reviewed suggest that brainstem pathology is present in individuals with ASD, consistent with the auditory abnormalities commonly observed, including atypical sound sensitivity, poor sound localization, and difficulty ignoring background noise. The anatomical studies reviewed revealed anomalies in ASD brainstem morphology. Postmortem studies have documented anomalies in brainstem morphology in individuals with ASD in brainstem structures, which mediate aspects of binaural processing and the efferent auditory pathways (Kulesza et al, 2011 ; Kulesza and Mangunay, 2008 ; Lukose et al, 2015 ; Rodier et al, 1997b). However, as previously noted, a number of the anatomical and imaging studies reviewed were limited by relatively small sample sizes, inadequate clinical diagnostic criteria, and large intersubject variability.
Some of these anomalies suggest the possibility of early brainstem injury during neural tube formation (Rodier, 2002). The increased prevalence in ASD of morphological anomalies of the external face, neck, and ear (Dawson et al, 2009) is also consistent with injury or other abnormalities early in gestation (Rodier et al, 1996). Prenatal exposure to thalidomide, which has been associated with increased prevalence of ASD and gestational brainstem injury (Rodier, 2002), as well as the increased risk of ASD in other neurodevelopmental disorders known to involve brainstem abnormalities (eg, Moebius syndrome and Joubert syndrome), are also suggestive of a role for early brainstem injury in ASD (Rodier, 2002). The presence of minor external ear malformations in children with ASD has been considered a marker of prenatal insult during the first month of gestation (Rodier et al, 1997a) when the brainstem is developing (Hashimoto et al, 1995).
MRI studies have also supported the presence of brainstem structural anomalies in ASD. MRI studies have shown a reduction in brainstem gray matter volume in individuals with ASD (Jou et al, 2009). Developmental studies have shown more rapid growth specifically in gray matter in participants with ASD (Jou et al, 2013). There is also some evidence that angiogenesis at the level of the pons/midbrain, cerebellum, and primary auditory cortex is prolonged in ASD (Azmitia et al, 2016).
Future studies of the auditory brainstem in ASD would benefit from combining structural and functional methodologies and including measures of speech processing in quiet and background noise as well as loudness processing. These measures could help determine the extent to which brainstem deficits are related to the communication deficits and abnormal auditory responses commonly exhibited by individuals with ASD. Based on our review, the following recommendations are made to help guide future auditory brainstem studies in ASD:
- Careful consideration and attention to the research design is important for ensuring that future studies yield valid, robust, and reliable findings.
- By including both structural and functional auditory brainstem measures from the same ASD individuals, it will be possible to directly correlate observed structural abnormalities on imaging (eg, MRI) with auditory function.
- Characterize ASD participants not only by chronological age, but also by their developmental level and expressive and receptive language abilities.
- Study participants should be screened for hearing loss using standard behavioral and/or objective audiometric measures.
- Use speech materials appropriate for the receptive level of participants with ASD and include objective and, whenever possible, behavioral measures of speech processing in background noise, including temporally modulated noise (Alcantara et al, 2012).
- Include measures of loudness perception. For individuals able to perform behavioral testing, adapted psychophysical methods appropriate for participants with developmental delays are recommended (Serpanos and Gravel, 2000). Objective measures of loudness can be used for participants unable to perform psychophysical measures of loudness (Serpanos, 2004).
- One or more measures of auditory brainstem function used in the studies reviewed here should be included for comparison with prior findings.
- Include documentation and characterization of sensory processing issues, such as hypersensitivity, based on parental reports, questionnaires, or other measures.
Future studies would also benefit from inclusion of well-defined age groups, the inclusion of children under 3 years of age for improved early identification, and use of the most current ASD diagnostic criteria for subject inclusion. Defining phenotypes based on the presence or absence of abnormalities in brainstem structure and function and measures of auditory processing would be one potential outcome of this proposed approach.
The authors are grateful to Carrie Price, MLS, Clinical Informationist at the Welch Library at the Johns Hopkins University School of Medicine, who performed the database literature search. The authors thank Paulo Belli for the diagram of auditory pathways, Johanna Veader for bibliographic assistance, and Hanna Pillion, Johanna Veader, and Nancy Grund for editorial assistance. The authors gratefully acknowledge the thoughtful and insightful comments of the two anonymous reviewers.
Alcantara JI, Cope TE, Cope W, et al. 2012. Auditory temporal-envelope processing in high-functioning children with Autism Spectrum Disorder
. Neuropsychologia. 50:1235–1251.
Alcantara JI, Weisblatt EJ, Moore BC, et al. 2004. Speech-in-noise perception in high-functioning individuals with autism or Asperger’s syndrome. J Child Psychol Psychiatry. 45:1107–1114.
American Psychiatric Association. 1985. Diagnostic and Statistical Manual of Mental Disorders, 3rd ed. Washington, DC: American Psychiatric Association.
American Psychiatric Association. 2000. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Association.
Azmitia EC, Saccomano ZT, Alzoobaee MF, et al. 2016. Persistent angiogenesis in the autism brain: an immunocytochemical study of postmortem cortex, brainstem and cerebellum. J Autism Dev Disord. 46:1307–1318.
Azouz HG, Kozou H, Khalil M, et al. 2014. The correlation between central auditory processing in autistic children and their language processing abilities. Int J Pediatr Otorhinolaryngol. 78:2297–2300.
Beers AN, McBoyle M, Kakande E, et al. 2014. Autism and peripheral hearing loss: a systematic review. Int J Pediatr Otorhinolaryngol. 78:96–101.
Bennetto L, Keith JM, Allen PD, et al. 2017. Children with autism spectrum disorder
have reduced otoacoustic emissions at the 1 kHz mid-frequency region. Autism Res. 10:337–345.
Ciesielski KT, Harris RJ, Hart BL, et al. 1997. Cerebellar hypoplasia and frontal lobe cognitive deficits in disorders of early childhood. Neuropsychologia. 35:643–655.
Cohen IL, Gardner JM, Karmel BZ, et al. 2013. Neonatal brainstem function and 4-month arousal-modulated attention are jointly associated with autism. Autism Res. 6:11–22.
Cohen IL, Gomez TR, Gonzalez MG, et al. 2010. Parent PDD behavior inventory profiles of young children classified according to autism diagnostic observation schedule-generic and autism diagnostic interview-revised criteria. J Autism Dev Disord. 40:246–254.
Collet L, Roge B, Descouens D, et al. 1993. Objective auditory dysfunction in infantile autism. Lancet. 342:923–924.
Constantino JN, Gruber CP. 2005. Social Responsiveness Scale (SRS). Los Angeles, California: Western Psychological Services.
Courchesne E, Courchesne RY, Hicks G, et al. 1985. Functioning of the brain-stem auditory pathway in non-retarded autistic individuals. Electroencephalogr Clin Neurophysiol. 61:491–501.
Crane L, Goddard L, Pring L. 2009. Sensory processing in adults with autism spectrum disorders. Autism. 13:215–228.
Creak EM. 1963. Childhood psychosis: a review of 100 cases. Br J Psychiatry. 109:84–89.
Dabbous AO. 2012. Characteristics of auditory brainstem
response latencies in children with autism spectrum disorders. Audiol Med. 10:122–131.
Danesh AA, Kaf WA. 2012. DPOAEs and contralateral acoustic stimulation and their link to sound hypersensitivity in children with autism. Int J Audiol. 51:345–352.
Dawson S, Glasson EJ, Dixon G, et al. 2009. Birth defects in children with autism spectrum disorders: a population-based, nested case-control study. Am J Epidemiol. 169:1296–1303.
de Boer J, Thornton AR. 2008. Neural correlates of perceptual learning in the auditory brainstem
: efferent activity predicts and reflects improvement at a speech-in-noise discrimination task. J Neurosci. 28:4929–4937.
Demopoulos C, Lewine JD. 2016. Audiometric profiles in autism spectrum disorders: Does subclinical hearing loss impact communication? Autism Res. 9:107–120.
Dunn LM, Dunn LM. 1997. Peabody Picture Vocabulary Test, 3rd ed. Circle Pines, Minnesota: American Guidance Service.
Ehret GEhret G, Romand R. 1997. The auditory midbrain, a “shunting yard” of acoustical information processing. The Central Auditory System. New York, New York: Oxford University Press; 259–316.
Elia M, Ferri R, Musumeci SA, et al. 2000. Clinical correlates of brain morphometric features of subjects with low-functioning autistic disorder. J Child Neurol. 15:504–508.
Fein D, Skoff B, Mirsky AF. 1981. Clinical correlates of brainstem dysfunction in autistic children. J Autism Dev Disord. 11:303–315.
Fendt M, Li L, Yeomans JS. 2001. Brain stem circuits mediating prepulse inhibition of the startle reflex. Psychopharmacology (Berl). 156:216–224.
Fujikawa-Brooks S, Isenberg AL, Osann K, et al. 2010. The effect of rate stress on the auditory brainstem
response in autism: a preliminary report. Int J Audiol. 49:129–140.
Gaffney GR, Kuperman S, Tsai LY, et al. 1988. Morphological evidence for brainstem involvement in infantile autism. Biol Psychiatry. 24:578–586.
Gardner JM, Karmel BZ. 1995. Development of arousal-modulated visual preferences in early infancy. Dev Psychol. 31:473–482.
Garreau B, Tanguay P, Roux S, et al. 1984. Brain stem auditory evoked potentials in the normal and autistic child [in French]. Rev Electroencephalogr Neurophysiol Clin. 14:25–31.
Gillberg C, Rosenhall U, Johansson E. 1983. Auditory brainstem
responses in childhood psychosis. J Autism Dev Disord. 13:181–195.
Gilliam JE. 2013. Autism Rating Scale, 3rd ed. Lutz, Florida: PAR Inc.
Gomes E, Rotta NT, Pedroso FS, et al. 2004. Auditory hypersensitivity in children and teenagers with autistic spectrum disorder. Arq Neuropsiquiatr. 62:797–801.
Goodman SS, Mertes IB, Lewis JD, et al. 2013. Medial olivocochlear-induced transient-evoked otoacoustic emission amplitude shifts in individual subjects. J Assoc Res Otolaryngol. 14:829–842.
Gordon AG. 1986. Abnormal middle ear muscle reflexes and audiosensitivity. Br J Audiol. 20:95–99.
Gravel JS, Dunn M, Lee WW, et al. 2006. Peripheral audition of children on the autistic spectrum. Ear Hear. 27:299–312.
Griffiths R. 1984. The Abilities of Young Children: A Comprehensive System of Mental Development for the First Eight Years of Life. Rev ed. Bucks, United Kingdom: The Test Agency Limited.
Grillon C, Courchesne E, Akshoomoff N. 1989. Brainstem and middle latency auditory evoked potentials in autism and developmental language disorder. J Autism Dev Disord. 19:255–269.
Hardan AY, Minshew NJ, Harenski K, et al. 2001. Posterior fossa magnetic resonance imaging in autism. J Am Acad Child Adolesc Psychiatry. 40:666–672.
Hashimoto T, Murakawa K, Miyazaki M, et al. 1992a. Magnetic resonance imaging of the brain structures in the posterior fossa in retarded autistic children. Acta Paediatr Scand. 81:1030–1034.
Hashimoto T, Tayama M, Miyazaki M, et al. 1992b. Reduced brainstem size in children with autism. Brain Dev. 14:94–97.
Hashimoto T, Tayama M, Murakawa K, et al. 1995. Development of the brainstem and cerebellum in autistic patients. J Autism Dev Disord. 25:1–18.
Herbert MR, Ziegler DA, Deutsch CK, et al. 2003. Dissociations of cerebral cortex, subcortical and cerebral white matter volumes in autistic boys. Brain. 126:1182–1192.
Hsu M, Yeung-Courchesne R, Courchesne E, et al. 1991. Absence of magnetic resonance imaging evidence of pontine abnormality in infantile autism. Arch Neurol. 48:1160–1163.
Ioannidis JP. 2014. How to make more published research true. PLoS Med. 11:e1001747.
Ioannidis JP, Trikalinos TA. 2005. Early extreme contradictory estimates may appear in published research: the Proteus phenomenon in molecular genetics research and randomized trials. J Clin Epidemiol. 58:543–549.
Jou RJ, Frazier TW, Keshavan MS, et al. 2013. A two-year longitudinal pilot MRI study of the brainstem in autism. Behav Brain Res. 251:163–167.
Jou RJ, Minshew NJ, Melhem NM, et al. 2009. Brainstem volumetric alterations in children with autism. Psychol Med. 39:1347–1354.
Kaf WA, Danesh AA. 2013. Distortion-product otoacoustic emissions and contralateral suppression findings in children with Asperger’s syndrome. Int J Pediatr Otorhinolaryngol. 77:947–954.
Källstrand J, Olsson O, Nehlstedt SF, et al. 2010. Abnormal auditory forward masking pattern in the brainstem response of individuals with Asperger syndrome. Neuropsychiatr Dis Treat. 6:289–296.
Keith WJ, Greville KA. 1987. Effects of audiometric configuration on the auditory brain stem response. Ear Hear. 8:49–55.
Khalfa S, Bougeard R, Morand N, et al. 2001a. Evidence of peripheral auditory activity modulation by the auditory cortex in humans. Neuroscience. 104:347–358.
Khalfa S, Bruneau N, Roge B, et al. 2001b. Peripheral auditory asymmetry in infantile autism. Eur J Neurosci. 13:628–632.
Kohl S, Wolters C, Gruendler TO, et al. 2014. Prepulse inhibition of the acoustic startle reflex in high functioning autism. PLoS One. 9:e92372.
Kulesza RJ, Lukose R, Stevens LV. 2011. Malformation of the human superior olive in autistic spectrum disorders. Brain Res. 1367:360–371.
Kulesza RJ, Mangunay K. 2008. Morphological features of the medial superior olive in autism. Brain Res. 1200:132–137.
Kwon S, Kim J, Choe BH, et al. 2007. Electrophysiologic assessment of central auditory processing by auditory brainstem
responses in children with autism spectrum disorders. J Korean Med Sci. 22:656–659.
Lord C, Risi S, Lambrecht L, et al. 2000. The autism diagnostic observation schedule-generic: a standard measure of social and communication deficits associated with the spectrum of autism. J Autism Dev Disord. 30:205–223.
Lord C, Rutter M, DiLavore PC, et al. 2002. Autism Diagnostic Observation Schedule. Los Angeles, California: Western Psychological Services.
Lord C, Rutter M, Goode S, et al. 1989. Autism diagnostic observation schedule: a standardized observation of communicative and social behavior. J Autism Dev Disord. 19:185–212.
Lord C, Rutter M, Le Couteur A. 1994. Autism Diagnostic Interview-Revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. J Autism Dev Disord. 24:659–685.
Lukose R, Beebe K, Kulesza RJ Jr. 2015. Organization of the human superior olivary complex in 15q duplication syndromes and autism spectrum disorders. Neuroscience. 286:216–230.
Lukose R, Brown K, Barber CM, et al. 2013. Quantification of the stapedial reflex reveals delayed responses in autism. Autism Res. 6:344–353.
Lukose R, Schmidt E, Wolski TP Jr, et al. 2011. Malformation of the superior olivary complex in an animal model of autism. Brain Res. 1398:102–112.
Madsen GF, Bilenberg N, Cantio C, et al. 2014. Increased prepulse inhibition and sensitization of the startle reflex in autistic children. Autism Res. 7:94–103.
Magliaro FC, Scheuer CI, Assumpcao Junior FB, et al. 2010. Study of auditory evoked potentials in autism. Pro Fono. 22:31–36.
Maziade M, Merette C, Cayer M, et al. 2000. Prolongation of brainstem auditory-evoked responses in autistic probands and their unaffected relatives. Arch Gen Psychiatry. 57:1077–1083.
McAlonan GM, Daly E, Kumari V, et al. 2002. Brain anatomy and sensorimotor gating in Asperger’s syndrome. Brain. 125:1594–1606.
McClelland RJ, Eyre DG, Watson D, et al. 1992. Central conduction time in childhood autism. Br J Psychiatry. 160:659–663.
Miron O, Ari-Even Roth D, Gabis LV, et al. 2016. Prolonged auditory brainstem
responses in infants with autism. Autism Res. 9:689–695.
Moher D, Liberati A, Tetzlaff J, et al. 2009. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 6:e1000097.
Møller AR. 1983. Auditory Physiology. New York, New York: Academic Press.
Møller AR, Jannetta PJ, Møller MB. 1981. Neural generators of brainstem evoked potentials: results from human intracranial recordings. Ann Otol Rhinol Laryngol. 90:591–596.
Møller AR, Jannetta P, Møller MB. 1982. Intracranially recorded auditory nerve response in man: new interpretations of BSER. Arch Otolaryngol. 108:77–82.
Novick B, Vaughan HG Jr, Kurtzberg D, et al. 1980. An electrophysiologic indication of auditory processing defects in autism. Psychiatry Res. 3:107–114.
Oranje B, Lahuis B, van Engeland H, et al. 2013. Sensory and sensorimotor gating in children with multiple complex developmental disorders (MCDD) and autism. Psychiatry Res. 206:287–292.
Ornitz EM, Atwell CW, Kaplan AR, et al. 1985. Brain-stem dysfunction in autism: results of vestibular stimulation. Arch Gen Psychiatry. 42:1018–1025.
Ornitz EM, Lane SJ, Sugiyama T, et al. 1993. Startle modulation studies in autism. J Autism Dev Disord. 23:619–637.
Parikshak NN, Swarup V, Belgard TG, et al. 2016. Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism. Nature. 540:423–427.
Patten E, Ausderau KK, Watson LR, et al. 2013. Sensory response patterns in nonverbal children with ASD. Autism Res Treat. 2013:436286.
Perry W, Minassian A, Lopez B, et al. 2007. Sensorimotor gating deficits in adults with autism. Biol Psychiatry. 61:482–486.
Pickett E, Pullara O, O’Grady J, et al. 2009. Speech acquisition in older nonverbal individuals with autism: a review of features, methods, and prognosis. Cogn Behav Neurol. 22:1–21.
Pickett J. 2001. Current investigations in autism brain tissue research. J Autism Dev Disord. 31:521–527.
Pillion JP. 2012. Speech processing disorder in neural hearing loss. Case Rep Med. 2012:206716.
Ritvo ER. 1977. National Society for Autistic Children: Definition of the Syndrome of Autism.
Rodier PM. 2002. Converging evidence for brain stem injury in autism. Dev Psychopathol. 14:537–557.
Rodier PM, Bryson SE, Welch JP. 1997a. Minor malformations and physical measurements in autism: data from Nova Scotia. Teratology. 55:319–325.
Rodier PM, Ingram JL, Tisdale B, et al. 1996. Embryological origin for autism: developmental anomalies of the cranial nerve motor nuclei. J Comp Neurol. 370:247–261.
Rodier PM, Ingram JL, Tisdale B, et al. 1997b. Linking etiologies in humans and animal models: studies of autism. Reprod Toxicol. 11:417–422.
Rosenblum SM, Arick JR, Krug DA, et al. 1980. Auditory brainstem
evoked responses in autistic children. J Autism Dev Disord. 10:215–225.
Rosenhall U, Nordin V, Brantberg K, et al. 2003. Autism and auditory brain stem responses. Ear Hear. 24:206–214.
Roth DA, Muchnik C, Shabtai E, et al. 2012. Evidence for atypical auditory brainstem
responses in young children with suspected autism spectrum disorders. Dev Med Child Neurol. 54:23–29.
Rumsey JM, Grimes AM, Pikus AM, et al. 1984. Auditory brainstem
responses in pervasive developmental disorders. Biol Psychiatry. 19:1403–1418.
Russo N, Nicol T, Trommer B, et al. 2009. Brainstem transcription of speech is disrupted in children with autism spectrum disorders. Dev Sci. 12:557–567.
Russo N, Skoe E, Trommer B, et al. 2008. Deficient brainstem encoding of pitch in children with Autism Spectrum Disorders. Clin Neurophysiol. 119:1720–1731.
Rutter MRutter M, Schopler E. 1978. Diagnosis and definition. Autism: A Reappraisal of Concepts and Treatments. New York, New York: Plenum Press.
Rutter M, Bailey A, Berument SK, et al. 2003. Social Communication Questionnaire. Los Angeles, California: Western Psychological Services.
Schopler E, Reichler RJ, Bashford A, et al. 1990. Individualized Assessment and Treatment for Autistic and Developmentally Disabled Children: Vol 1, Psychoeducational Profile Revised (PEP-R). Austin, Texas: Pro-Ed.
Schopler E, Reichler RJ, DeVellis RF, et al. 1980. Toward objective classification of childhood autism: Childhood Autism Rating Scale (CARS). J Autism Dev Disord. 10:91–103.
Semel E, Wiig E, Secord W. 2006. Clinical Evaluation of Language Fundamentals, 4th ed. Sydney, Australia: Pearson.
Serpanos YC. 2004. ABR and DPOAE indices of normal loudness in children and adults. J Am Acad Audiol. 15:555–565.
Serpanos YC, Gravel JS. 2000. Assessing growth of loudness in children by cross-modality matching. J Am Acad Audiol. 11:190–202.
Sersen EA, Heaney G, Clausen J, et al. 1990. Brainstem auditory-evoked responses with and without sedation in autism and Down’s syndrome. Biol Psychiatry. 27:834–840.
Shivashankar N, Satishchandra P. 1989. Auditory brainstem
responses in autistic children. NIMHANS Journal. 7:159–162.
Skoff BF, Mirsky AF, Turner D. 1980. Prolonged brainstem transmission time in autism. Psychiatry Res. 2:157–166.
Sohmer H, Student M. 1978. Auditory nerve and brain-stem evoked responses in normal, autistic, minimal brain dysfunction and psychomotor retarded children. Electroencephalogr Clin Neurophysiol. 44:380–388.
Sparrow S, Balla D, Cicchetti D. 1984. Vineland Adaptive Behavior Scales (Survey Form). Circle Pines, Minnesota: American Guidance Services.
Stockard JE, Stockard JJ, Westmoreland BF, et al. 1979. Brainstem auditory-evoked responses: normal variation as a function of stimulus and subject characteristics. Arch Neurol. 36:823–831.
Stuart A, Butler AK. 2012. Contralateral suppression of transient otoacoustic emissions and sentence recognition in noise in young adults. J Am Acad Audiol. 23:686–696.
Stuart A, Yang EY. 2001. Gender effects in auditory brainstem
responses to air- and bone-conducted clicks in neonates. J Commun Disord. 34:229–239.
Student M, Sohmer H. 1978. Evidence from auditory nerve and brainstem evoked responses for an organic brain lesion in children with autistic traits. J Autism Child Schizophr. 8:13–20.
Sun XM. 2008. Contralateral suppression of distortion product otoacoustic emissions and the middle-ear muscle reflex in human ears. Hear Res. 237:66–75.
Suriá D, Serra-Raventós M. 1975. Acoustic impedance measurement and autistic children. Folia Phoniatr (Basel). 27:387–388.
Takahashi H, Hashimoto R, Iwase M, et al. 2011. Prepulse inhibition of startle response: recent advances in human studies of psychiatric disease. Clin Psychopharmacol Neurosci. 9:102–110.
Takahashi H, Komatsu S, Nakahachi T, et al. 2016. Relationship of the acoustic startle response and its modulation to emotional and behavioral problems in typical development children and those with autism spectrum disorders. J Autism Dev Disord. 46:534–543.
Takahashi H, Nakahachi T, Komatsu S, et al. 2014. Hyperreactivity to weak acoustic stimuli and prolonged acoustic startle latency in children with autism spectrum disorders. Mol Autism. 5:23.
Takahashi H, Nakahachi T, Stickley A, et al. 2017. Stability of the acoustic startle response and its modulation in children with typical development and those with autism spectrum disorders: a one-year follow-up. Autism Res. 10:673–679.
Takesada M, Nakajima A, Matsubayashi T. 1992. Study of auditory brain-stem responses in autistic-children. Neurobiol Infantile Autism. 965:209–212.
Tanguay PE, Edwards RM, Buchwald J, et al. 1982. Auditory brainstem
evoked responses in autistic children. Arch Gen Psychiatry. 39:174–180.
Tas A, Yagiz R, Tas M, et al. 2007. Evaluation of hearing in children with autism by using TEOAE and ABR. Autism. 11:73–79.
Taylor MJ, Rosenblatt B, Linschoten L. 1982. Auditory brainstem
response abnormalities in autistic children. Can J Neurol Sci. 9:429–433.
Thabet EM, Zaghloul HS. 2013. Auditory profile and high resolution CT scan in autism spectrum disorders children with auditory hypersensitivity. Eur Arch Otorhinolaryngol. 270:2353–2358.
Tharpe AM, Bess FH, Sladen DP, et al. 2006. Auditory characteristics of children with autism. Ear Hear. 27:430–441.
Thivierge J, Bedard C, Cote R, et al. 1990. Brainstem auditory evoked response and subcortical abnormalities in autism. Am J Psychiatry. 147:1609–1613.
Ververi A, Vargiami E, Papadopoulou V, et al. 2015. Brainstem auditory evoked potentials in boys with autism: still searching for the hidden truth. Iran J Child Neurol. 9:21–28.
Visser E, Zwiers MP, Kan CC, et al. 2013. Atypical vertical sound localization and sound-onset sensitivity in people with autism spectrum disorders. J Psychiatry Neurosci. 38:398–406.
Wechsler D. 1951. Equivalent test and mental ages for WISC. J Consult Psychol. 15:381–384.
Wechsler D. 1955. Manual for the Wechsler Adult Intelligence Scale. New York, New York: Psychological Corporation.
Wechsler D. 1967. Wechsler Preschool and Primary Scale of Intelligence. New York, New York: Psychological Corporation.
Wechsler D. 1974. Wechsler Intelligence Scale for Children–Revised. New York, New York: The Psychological Corporation.
Wechsler D. 1981. Wechsler Adult Intelligence Scale. New York, New York: Psychological Corporation.
Wechsler D. 1991. Wechsler Intelligence Scale for Children, 3rd ed. New York, New York: Psychological Corporation.
Wechsler D. 1999. Wechsler Abbreviated Scale of Intelligence. Bloomington, Minnesota: PsycCorp.
Wechsler D. 2004. Wechsler Intelligence Scale for Children, 4th ed. San Antonio, Texas: Pearson.
Wechsler D. 2008. Wechsler Adult Intelligence Scale, 4th ed. San Antonio, Texas: Pearson.
Wechsler D. 2012. Wechsler Preschool and Primary Scale of Intelligence. Sydney, Australia: Pearson.
Wong V, Wong SN. 1991. Brainstem auditory evoked potential study in children with autistic disorder. J Autism Dev Disord. 21:329–340.
World Health Organization. 1992. ICD-10 Classification of Mental and Behavioral Disorders: Clinical Descriptions and Diagnostic Guidelines. Geneva, Switzerland: World Health Organization.
Yuhas J, Cordeiro L, Tassone F, et al. 2011. Sensorimotor gating in idiopathic autism and autism associated with fragile X syndrome. J Autism Dev Disord. 41:248–253.