Psychophysiology, 52 (2015), 782–789. Wiley Periodicals, Inc. Printed in the USA. C 2015 Society for Psychophysiological Research Copyright V DOI: 10.1111/psyp.12407

Hyperactive auditory processing in Williams syndrome: Evidence from auditory evoked potentials

OMER ZARCHI,a,b,c CHEN AVNI,c JOSEF ATTIAS,b,d AMOS FRISCH,c,e MIRI CARMEL,c,e ELENA MICHAELOVSKY,e TAMAR GREEN,c,f ABRAHAM WEIZMAN,c,e,g AND DORON GOTHELFa,c a

Behavioral Neurogenetics Center, The Edmond and Lily Safra Children’s Hospital, Sheba Medical Center, Tel Hashomer, Israel Institute for Clinical Neurophysiology and Audiology, Rabin Medical Center and Schneider Children’s Medical Center, Petah Tikva, Israel c Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel d Department of Communication Sciences & Disorders, Haifa University, Haifa, Israel e Biochemical Genetics Laboratory, Felsenstein Medical Research Center, Petah Tikva, Israel f Nes-Ziyyona-Beer Yaakov Mental Health Center, Nes-Ziyyona, Israel g Research Unit, Geha Mental Health Center, Petah Tikva, Israel b

Abstract The neurophysiologic aberrations underlying the auditory hypersensitivity in Williams syndrome (WS) are not well defined. The P1-N1-P2 obligatory complex and mismatch negativity (MMN) response were investigated in 18 participants with WS, and the results were compared with those of 18 age- and gender-matched typically developing (TD) controls. Results revealed significantly higher amplitudes of both the P1-N1-P2 obligatory complex and the MMN response in the WS participants than in the TD controls. The P1-N1-P2 complex showed an age-dependent reduction in the TD but not in the WS participants. Moreover, high P1-N1-P2 complex was associated with low verbal comprehension scores in WS. This investigation demonstrates that central auditory processing is hyperactive in WS. The increase in auditory brain responses of both the obligatory complex and MMN response suggests aberrant processes of auditory encoding and discrimination in WS. Results also imply that auditory processing may be subjected to a delayed or diverse maturation and may affect the development of high cognitive functioning in WS. Descriptors: Williams syndrome, Auditory processing, Hyperacusis, Evoked response potentials (ERP)

& Rourke, 1999; Lenhoff, Perales, & Hickok, 2001; Lense, Shivers, & Dykens, 2013; Levitin & Bellugi, 1998; Levitin et al., 2004; Martınez-Castilla, Sotilla, & Campos, 2013). Moreover, it was suggested that the apparent strength in the verbal domain relies on spared components of the auditory working memory and, particularly, on the phonological loop more in subjects with WS than in typically developing (TD) individuals (Robinson, Mervis, & Robinson, 2003; Vicari, Carlesimo, Brizzolara, & Pezzini, 1996). While there is some debate as to whether individuals with WS are endowed with special analytical auditory and musical skills, it is clear that they are highly engaged with their auditory environment and perceive it in a unique way. Because WS is etiologically homogeneous, due its resulting from a hemizygous microdeletion of approximately 28 genes on the long arm of chromosome 7 (Meyer-Lindenberg, Mervis, & Berman, 2006), it serves as an excellent model for the study of the physiological processes underlying auditory hypersensitivity. Understanding the pathophysiology of auditory hypersensitivity in WS may also have implications for other psychiatric disorders associated with pathological auditory perception and processing, such as autism, schizophrenia, posttraumatic stress disorder, Down syndrome, and attention-deficit hyperactivity disorder (ADHD; Chemtob, Roitblat,

Williams syndrome (WS) is a neurodevelopmental disorder with a salient auditory phenotype characterized by extremely high rates of debilitating hyperacusis and phonophobia that affect up to 84% and 95% of individuals with the syndrome, respectively (Gothelf, Farber, Raveh, Apter, & Attias, 2006; Klein, Armstrong, Green, & Brown Lii, 1990). Along with those untoward auditory symptoms, many individuals with WS display fascination with auditory stimuli (Levitin, Cole, Lincoln, & Bellugi, 2005). Reports of precise auditory detection and discrimination abilities (e.g., perfect pitch) and musicality are also very common in WS, although they are not consistently evident when tested experimentally (Don, Schellenberg,

This work was supported by the Basil O’Connor Starter Scholar Research Award of the March of Dimes (grant number 5-FY06-590 to DG), the National Alliance for Research on Schizophrenia and Depression Young Investigator Award, and the United States–Israel Binational Science Foundation (grant 2011378). We express our gratitude to the participants and their families for making this study possible. The authors are grateful to Esther Eshkol for her help in the preparation of the manuscript. Address correspondence to: Prof. Doron Gothelf, M.D., The Child Psychiatry Unit, Sheba Medical Center, Tel Hashomer 52621, Israel. E-mail: [email protected] 782

Hyperactive auditory processing in Williams syndrome Hamada, Carlson, & Twentyman, 1988; Cheung & Siu, 2009; Hetrick, Erickson, & Smith, 2012; Levitin et al., 2005). While behavioral studies in WS showed an atypical auditory phenotype and, particularly, hyperacusis and phonophobia to be striking characteristics of WS, little is known about the underlying mechanisms that mediate such auditory hypersensitivity. Previous investigations in WS found evidence for peripheral impairments, including a characteristic mild sensorineural high tone hearing loss (Cherniske et al., 2004; Johnson, Comeau, & Clarke, 2001; Marler, Elfenbein, Ryals, Urban, & Netzloff, 2005; Marler, Sitcovsky, Mervis, Kistler, & Wightman, 2010; Zarchi et al., 2011). An association of hearing loss with acoustic reflex dysfunction in WS as well as a deficit in the efferent auditory system, which modifies cochlear function, as indicated by otoacoustic emissions, was described in several reports (Attias, Raveh, Ben-Naftali, Zarchi, & Gothelf, 2008; Zarchi et al., 2011). Few studies, however, have thus far looked into the central auditory processing in WS, leaving the question of central involvement in the WS auditory phenotype largely unknown. Those neurophysiologic and neuroimaging studies suggest that auditory processing in individuals with WS may be characterized by neural hyperexcitability mediated by neural circuits partly different from those activated in TD individuals (Levitin et al., 2003; Mills et al.,  2013; Neville, Mills, & Bellugi, 1994; Pinheiro, Galdo-Alvarez, Sampaio, Niznikiewicz, & Gonc¸alves, 2010). Auditory evoked response potentials (ERPs) of participants with WS were shown to be less refractory and more excitable than those of control participants, a neural pattern that did not extend to visual modalities (Neville et al., 1994). Moreover, participants with WS demonstrated prominent early positivity in response to auditory presented words, an atypical sensory processing pattern that was interpreted as being related to a diverse maturation trajectory of auditory processing and to the auditory hypersensitivity typical of the syndrome (Mills et al., 2013; Neville, Holcomb, & Mills, 1989; Pinheiro et al., 2010). Taken together, accumulated findings proposed possible corticallevel involvement in auditory hyperactivity, specifically in the stages of auditory sensory encoding. Previous auditory ERP studies among individuals with WS targeted mainly language processing and did not, however, investigate other important central auditory processes, such as those underlying sound discrimination. The current study was designed to delineate the neurophysiologic aberrations underlying the atypical auditory phenotype in WS by investigating the P1-N1-P2 obligatory complex and mismatch negativity (MMN) response in individuals diagnosed as having WS compared with age- and gender-matched TD controls, and associate those findings with behavioral measures in the WS group. We hypothesized that the amplitudes of the P1-N1-P2 obligatory complex in response to tonal stimulation, which is an index of transient encoding of acoustic sound features (Ceponiene et al., 2001), will be larger in the WS participants compared to TD controls. We also hypothesized that the MMN amplitudes, which index preattentive processes of auditory sensory memory and auditory discrimination (N€a€at€anen, 1990), will be larger in the WS participants than those measured in the TD controls. Amplitudes of the P1-N1-P2 complex and MMN were further hypothesized to be associated with hyperacusis and working memory performances, respectively, in the WS group.

783 and 9 females; mean age 5 20.1, SD 5 8.2). The groups were similar in mean age, t(34) 5 -.37; p 5 .707, and gender distribution, v2 5 2.37; p 5 .306. The same TD participants had previously served as controls in an investigation of ours on 22q11.2 deletion syndrome (Zarchi et al., 2013). The WS participants were recruited from the Behavioral Neurogenetics Center at a large tertiary referral center in Israel that coordinates research and treatment of individuals with WS who are referred from departments of genetics and parents’ associations throughout the country. The diagnosis of WS was confirmed by fluorescent in situ hybridization. The TD controls were recruited through advertisements in the local community, and they were all students in mainstream classes and free of any major psychopathology. The study was approved by the Institutional Review Board of Rabin Medical Center. After providing a complete description of the nature of this study, we obtained written informed consent from the participants or their parents. Neurophysiology Prior to undergoing the neurophysiologic evaluation, the participants underwent audiometric testing to determine hearing thresholds and ensuring hearing thresholds of  35 dB HL for frequencies 250–8000 Hz (Zarchi et al., 2011). Auditory stimuli were presented through a GSI-16 audiometer and TDH headphones. Recordings and paradigm. Recordings took place in an electroacoustic shielded room while watching a silent movie. The EEG was recorded from five scalp electrodes (Fz, C3, Cz, C4, and Pz sites), keeping resistance at  10 kX. An electrooculogram (EOG) was recorded from electrodes placed above the right eye and below the right outer canthus. Recordings were referenced to the nose tip, and a ground electrode was placed on the right mastoid. Continuous EEG was acquired via a Ceegraph Bio-Logic system with system band-pass of 0.1–100 Hz and a digital sampling rate of 256 Hz. A paradigm was constructed according to N€a€at€anen et al.’s “optimal” MMN paradigm (N€a€at€anen, Pakarinen, Rinne, & Takegata, 2004), which consists of a standard stimulus and five types of deviants. The standard stimuli were harmonic tones composed of three sinusoidal partials of 500, 1000, and 1500 Hz, and were 75 ms in duration presented at 60 dB above the individualized sensation level of each ear (dB SL). When stimuli evoked an eye blink startle response, their intensity was reduced gradually by 5 dB until the startle response was abolished and a comfortable loudness level was obtained. The deviant tones differed from the standard ones either in frequency (6 10%), duration (25 ms), intensity (6 10 dB), directionality (interaural time difference of 800 ls to the left or right ear), or by the insertion of a gap (7 ms of silence in the middle of the tone). The first 15 tones were standards in each sequence. All deviant tones were presented pseudorandomly in each sequence so that each standard tone was followed by one of the deviant tones. The stimuli were presented in three 5-min sequences with a total of 1,845 stimuli at a stimulus-onset asynchrony of 500 ms.

Method Participants The current study included 18 WS participants (6 males and 12 females; mean age 5 19.1, SD 5 8.2) and 18 TD controls (9 males

ERP waveform and component analysis. A component analysis was done by MATLAB software and the EEGLAB toolbox (Delorme & Makeig, 2004). The MMN responses and the P1N1-P2 ERPs were derived by filtering the EEG offline using a

O. Zarchi et al.

784

Figure 1. A P1-N1-P2 obligatory complex in Williams syndrome (WS) and in typically developing (TD) controls. The WS participants demonstrated a significantly larger P1-N1-P2 complex compared to that of the controls, F(1,32) 5 7.0, p 5 .012, g2p 5 .18.

1–30 Hz band-pass filter. Epochs of the 100-ms prestimulus and 600-ms poststimulus periods were separately averaged for the standard and the five types of deviant stimuli. The baseline value was set according to the mean voltage of the 100-ms prestimulus period. Epochs that included EEG or EOG changes exceeding 6 100 mV and those for the first 15 standards of a sequence were omitted from the calculations. An eye movement correction procedure was performed offline using the eye movement correction algorithm (Orgil Ltd., Israel) described in previous studies (Henkin, Kishon-Rabin, Gadoth, & Pratt, 2002). The procedure identifies and corrects ocular artifacts that contaminate ERP recordings. It includes detection of an epoch in which an eye artifact contaminated a single ERP trial and correction for each epoch and each channel separately, based on the eye artifact potential as recorded at the electrode below the eye. Estimation and correction of the contamination are performed by linear regression between each electrode within the detected epoch and the electrode below the eye. The MMN response waveform was delineated by subtracting standard stimulus ERPs from the corresponding deviant stimulus ERPs. The MMN response amplitudes were calculated as a mean voltage at the 43-ms period centered at the most negative peak in the grandaverage waveform that occurred at 90–250 ms poststimulus (N€a€at€anen et al., 2004; Zarchi et al., 2013). The P1-N1-P2 ERP waveform was constructed by averaging the responses for all standard stimuli, excluding the responses for the 15 consecutive standard tones that initiated each sequence. The acrossparticipants latency distribution of the P1-N1-P2 ERP wave was first determined for the two groups by means of visual inspection. As can be seen in Figure 1, data implied a long-lasting trend covering all the P1-N1-P2 components, rather than a sepa-

rate amplitude modulation per component. Hence, the P1-N1-P2 complex was computed as the area under the curve in time window of 0–250 ms poststimulus.

Psychiatric and Cognitive Assessments The WS participants and their parents were interviewed by senior child psychiatrists using the Hebrew version of the Schedule for Affective Disorders and Schizophrenia for School-Aged Children, Present and Lifetime (K-SADS-PL; Kaufman et al., 1997). Adult participants and their parents (when available) were interviewed with the Structured Clinical Interview for Axis I DSM-IV Disorders (SCID; First, Gibbon, Spitzer, & Williams, 1996). In order to determine the presence or absence of auditory hypersensitivity in the pediatric and adult WS participants, we modified the structured psychiatric interview to include questions on the presence of fears from and aversion to sounds as part of the screening interview of specific phobias. The WS participants completed the age-appropriate Wechsler Intelligence Scale to evaluate global intellectual functioning: the Wechsler Intelligence Scale for Children-Revised (WISC-R, for participants younger than 17 years; Lieblich, Ben-Schachar, & Ninio, 1976) and the Wechsler Adult Intelligence Scale-Third Edition (WAIS-III, for participants 17 years or older; Goldman, Gottlieb, & Stine-Sagi, 2001). Verbal comprehension and working memory indices were computed for each WS participant to assess for potential associations between these domains’ scores and neurophysiological measures. Since WISC-R does not provide IQ norms for those indices, indices were computed as the sum of the standardized scores: verbal comprehension index

Hyperactive auditory processing in Williams syndrome

785

Figure 2. Mismatch negativity (MMN) responses in Williams syndrome (WS) and typically developing (TD) controls. The WS group had significantly larger overall MMN amplitudes compared to those of the controls, F(1,32) 5 6.6, p 5 .015, gp2 5 .18.

(vocabulary, comprehension, and similarities subtests) and working memory index (digit span and arithmetic subtests).

Statistical Analysis The group comparison of the MMN and P1-N1-P2 complex was carried out with a general linear model for repeated measures analysis of variance (ANOVA). Linear stepwise regression models were carried out to assess effects of auditory and psychiatric factors on MMN and P1-N1-P2 complex in the WS group. Statistics were performed using SPSS 17.

Results The groups did not differ significantly in age or gender distribution. Ten of the 18 WS participants met the DSM-IV criteria for ADHD, 7 for anxiety disorder, 1 for autistic spectrum disorder, 1 for psychosis disorder, and none for mood disorder. Auditory hypersensitivity was detected in 14 of the WS participants. P1-N1-P2 and MMN amplitudes differed significantly from the 0 lV baseline at Fz, C3, Cz, and C4 electrodes in the two groups, and thus they were included in the analysis. In order to analyze the P1-N1-P2 complex, we performed repeated measures ANOVA with the study groups as factors, age and gender as covariates, and the area under the curve in 0–250 ms time window of the four frontocentral electrodes (Fz, Cz, C3, and C4) as outcome variables. Confirming our hypothesis, the WS participants demonstrated significantly larger P1-N1-P2 complexes than those of the controls, F(1,32) 5 7.0, p 5 .012, gp2 5 .18 (see Figure 1). There was also a significant age effect on the P1-N1-P2 complex, F(1,32) 5 11.0,

p 5 .002, gp2 5 .26, demonstrating a smaller P1-N1-P2 complex in older participants. There was no significant gender effect. To further assess the age effect detected in the general linear model, we performed an age–amplitude correlation analysis separately for WS and TD participants. To account for multiple testing, we used the Bonferroni correction and considered only those results for which p < .0125 (.05/4) as being significant. We found that, while TD participants had a significant reduction in the P1-N1-P2 complex with age in all but the C3 electrode (Fz: r 5 -.68, p 5 .002; C3: r 5 -.51, p 5 .032; Cz: r 5 -.59, p 5 .010; and C4: r 5 -.63, p 5 .005), the WS participants showed no such significant correlations in any of the electrodes (Fz: r 5 -.36, p 5 .144; C3: r 5 -.48, p 5 .046; Cz: r 5 -.39, p 5 .107; and C4: r 5 -.39, p 5 .182). For analyzing the MMN response, we performed repeated measures ANOVA with the study groups as factors, age and gender as covariates, and the four frontocentral electrodes (Fz, Cz, C3, and C4) as outcome variables. Assessment of the overall group effect on the MMN response was carried out by first comparing the overall mean MMN response to all five deviants. As expected, we found that the WS group had significantly larger overall mean MMN amplitudes than those of the controls, F(1,32) 5 6.6, p 5 .015, gp2 5 .18 (see Figure 2). There were no significant age or gender effects. To identify the variables that accounted for the significant group effect on the MMN response, we further evaluated the five MMN deviant types (i.e., frequency, intensity, directionality, duration, and a silent gap). We used a multivariate ANOVA model with group as factor, age and gender as covariates, and the mean amplitude of the analyzed electrodes (Fz, C3, Cz, and C4) for the five MMN deviant types as outcome variables. The multivariate ANOVA model revealed that the MMN amplitudes were larger in

O. Zarchi et al.

786

Table 1. Comparison of the Five Mismatch Negativity (MMN) Types Between the Williams Syndrome (WS) and Typically Developing (TD) Groups WS mean 6 SD Frequency-MMN* Intensity-MMN Directionality-MMN Duration-MMN* Gap-MMN

23.80 23.73 21.97 25.28 22.48

6 6 6 6 6

2.60 3.12 1.83 3.25 1.62

TD mean 6 SD 22.16 23.71 21.90 22.53 22.17

6 6 6 6 6

3.37 4.61 1.59 2.25 2.08

F

p values

gp2

3.4 0.3 0.3 3.2 0.2

.029 .82 .844 .036 .902

.25 .03 .03 .24 .02

Note. The p values represent the results of a multivariate ANOVA model with group as factor, age and gender as covariates, and the amplitudes of the five MMN deviant types as outcome variables. *p < .05.

the WS group compared to those of the controls for frequency deviant, F(1,32) 5 3.4, p 5 .029, gp2 5 .25, as well as for duration deviant, F(1,32) 5 3.2, p 5 .036, gp2 5 .24. No significant group differences were detected for intensity, directionality, or gap deviants (Table 1). To test whether neurophysiologic abnormalities in WS are associated with psychiatric morbidity as well as with auditory factors, we performed two separate stepwise linear regression models, one for the P1-N1-P2 complex and the other for MMN amplitudes. The mean P1-N1-P2 area under the curve and the mean MMN amplitude of the analyzed electrodes (Fz, C3, Cz, and C4) were entered as the dependent variable in the P1-N1-P2 and MMN regression models, respectively. As independent factors, we entered the demographic variables (age and gender) in a first block, and the psychiatric factors (presence of auditory hypersensitivity, anxiety, and ADHD, as determined by the SADS-PL), cognitive factors (working memory and verbal comprehension Wechsler test indexes), and mean audiometric hearing threshold (mean of frequencies 0.25, 0.5, 1, 2, 4, and 8 KHz) in a second block. Two WS participants were not included in this analysis due to missing data on part of the independent factors.

Figure 3. Scatter plot delineating the relationship between standardized scores on the Wechsler Verbal Comprehension index and mean P1-N1P2 area under the curve in the Williams syndrome group. The P1-N1-P2 complex regression model was significant (adjusted r2 5 .33, p 5 .019). Scores on the verbal comprehension index predicted 33% of the complex area variance, with high verbal comprehension scores predicting a low complex area (b 5 -.615, r2 change 5 .33).

The P1-N1-P2 complex regression was significant (adjusted r2 5 .33; p 5 .019). Scores on the verbal comprehension index predicted 33% of the complex area variance, with high verbal comprehension scores predicting a low complex area (see Figure 3) (b 5 .615, r2 change 5 .33). The MMN regression model was not significant, rejecting all variables from the equation.

Discussion An atypical auditory phenotype with high rates of hyperacusis and phonophobia is well established in WS. Individuals with WS are also attracted to sound and music and may possess special abilities in this domain. In the current study, we attempted to identify the neurophysiologic aberrations underlying the unique auditory phenotype and examine associations between the two. Our present investigation on the neural mechanisms underlying auditory processing in WS is the first to detect abnormal MMN responses in this syndrome. We found the amplitudes of both the P1-N1-P2 obligatory complex and of the MMN response to be larger in the WS group than in the TD controls. High verbal comprehension performances predicted a low P1-N1-P2 response in the WS participants. WS presents a complex auditory profile consisting of common aversive auditory phenomena as well as heightened attraction to sounds and music (Gothelf et al., 2006; Lenhoff, Wang, Greenberg, & Bellugi, 1997; Levitin et al., 2005). Elevated hearing thresholds, low uncomfortable hearing levels, and phonophobia are all prevalent in WS. Paradoxically, high levels of attraction to music and fascination towards auditory stimuli are also common characteristics of the syndrome (Gothelf et al., 2006; Lenhoff et al., 1997; Levitin et al., 2005). While cochlear, acoustic reflex, and efferent auditory impairments in WS may account for some of the auditory intolerance due to deficient protection against excessive auditory input and the presence of pathological loudness recruitment, those impairments apparently cannot fully explain the WS auditory phenotype (Attias et al., 2008; Zarchi et al., 2011). Indeed, similar peripheral impairments in individuals with 22q11.2 deletion syndrome are not associated with such a high prevalence of hyperacusis and phonophobia (Zarchi et al., 2011). Moreover, the extreme fear response to aversive auditory perception (Gothelf et al., 2006), the high level of engagement in music (Levitin et al., 2004), as well as the unique verbal profile in WS (Brock et al., 2007) point to a possible contribution of central mechanisms to the WS auditory phenotype. The hearing thresholds of all the participants in the current study were within normal limits or indicated no more than a mild hearing loss, and the stimuli did not exceed the participant’s

Hyperactive auditory processing in Williams syndrome comfortable level of loudness. Moreover, hearing thresholds in our WS sample did not predict the P1-N1-P2 response or MMN amplitudes. Thus, the possibility that pathological loudness recruitment may account for the increased brain responses in our WS participants seems unlikely. We identified atypical hyperactive processes of tonal encoding and auditory preattentive auditory deviance detection. Similar to our current observations, previous neurophysiologic investigations in WS revealed an atypical morphology of the auditory response for word and tone stimuli (Mills et al., 2013; Neville et al., 1994; Pinheiro et al., 2010). The word stimuli evoked prominent early positivity while the tone stimuli evoked prominent N1-P2 responses at high repetition rates in WS participants compared to controls (Mills et al., 2013; Neville et al., 1994; Pinheiro et al., 2010). Taken together with our findings of robust P1–N1-P2 amplitudes to tonal stimuli, these findings suggest aberrant auditory encoding of both linguistic and nonlinguistic auditory information. We also found that the MMN response of the WS group had higher amplitudes than that of the controls. To the best of our knowledge, this is the first report on MMN responses in WS. Our results may imply that the automatic change detector mechanism, which is thought to underlie the MMN response, is hyperexcited in WS. This finding is of particular interest to the debate on perfect pitch abilities in WS and may suggest that the processes underlying such performances are atypical in WS (Lenhoff et al., 2001; Martınez-Castilla et al., 2013). MMN can be elicited without focused attention and was indeed recorded in inattentive conditions in this study. However, attention can influence processing stages even before deviance detection, such as when attention influences the standard tone formation phase, which is a fundamental part of the MMN-generating process. Such influences of attention occur before deviance detection and can modify MMN (Sussman, 2007). The enhanced MMN response among our WS participants may be interpreted through this framework of an indirect effect of attention on MMN. Indeed, our findings in WS and those of others showed enhanced brain responses in the early stages of sound encoding (e.g., the P1-N1-P2 complex; Mills et al., 2013; Neville et al., 1989; Pinheiro et al., 2010). A general intentional bias toward the auditory environment in WS is evidenced by a wide range of auditory aversion and attraction behaviors, all suggestive of a salient attentive and affective involvement in sound perception in this syndrome (Don et al., 1999; Gothelf et al., 2006; Lenhoff et al., 2001; Lense et al., 2013; Levitin et al., 2004, 2005). Moreover, an atypical contribution of limbic regions in sound processing was also noted on a functional MRI and in postmortem studies (Holinger et al., 2005; Levitin et al., 2003). Thus, our findings of enhanced auditory processing, both in the stages of sound encoding and deviance detection, may be related to the predominance of auditory input in WS, possibly attributed to a global attention-affective bias towards sounds. Interestingly, previous studies found higher MMN responses in individuals with Asperger syndrome and posttraumatic stress disorder (Ge, Wu, Sun, & Zhang, 2011; Kujala et al., 2007; Morgan Iii & Grillon, 1999). It is possible that patients with both disorders as well as those with WS are hypersensitive to auditory alterations in their environment due to a hyperaroused auditory change-detection mechanism. Auditory ERPs are known to be subject to maturation processes. The obligatory auditory response is dominated by enhanced P1 and N2 peaks in early childhood, gradually turning into a W-shaped component governed by the P1 and P2 peaks (Ponton, Eggermont, Kwong, & Don, 2000; Rinne & N€a€at€anen, 2002). The amplitudes

787 of the obligatory complex peaks around the age of 10 years and decreases during the second decade of life (Ponton et al., 2000; Rinne & N€a€at€anen, 2002). One ERP study suggested that the maturation of the N1 component in response to word stimuli is delayed in WS (Mills et al., 2013). Relating these findings to our results of increased P1-N1-P2 response in the WS group, it may be that delayed or atypical maturation processes of the networks serving auditory processing are involved in WS. Indeed, an age-dependent reduction in the P1-N1-P2 complex was detected in our TD controls but not in the WS participants. Moreover, unlike the Wshaped response of the P1-N1-P2 complex on all electrodes observed in our control group, the WS group produced a childlike P1-N2 response in the Fz electrode (Figure 1). This lack of agerelated reduction in the P1-N1-P2 complex and childlike pattern, which was detected at the frontal site, may suggest an immature response of the auditory and/or frontal cortices involved in the obligatory auditory response (Rinne & N€a€at€anen, 2002). Interestingly, concomitant with the immature excessive auditory brain responses seen in TD children, auditory hypersensitivity is prevalent among them and typically vanishes during adolescence (Coelho, Sanchez, & Tyler, 2007). In WS, although there is some attenuation in the severity of hyperacusis during adolescence, hyperacusis may persist after childhood (Gothelf et al., 2006), a possibility that is consistent with the excessive auditory brain responses demonstrated in the present study. We performed regression analysis for the P1-N2-P2 complex and MMN amplitudes in search of neurophysiological-behavioral associations among our WS participants. Demographic variables, such as age and gender, psychiatric diagnosis of ADHD, anxiety disorder and hyperacusis, and hearing level, were ruled out as predictors of the neurophysiological results. When we looked into cognitive functions (e.g., working memory and verbal comprehension), however, we found that a large P1-N1-P2 complex was associated with low verbal comprehension scores in the WS group. Those results suggest that the atypical sound encoding processes in WS may interfere with the development of high-cognitive functions, such as receptive language. Interestingly, hyperacusis was not associated with either the P1-N1-P2 complex or the MMN amplitudes. Thus, it may be that the larger P1-N1-P2 complex and MMN amplitudes found in participants with WS compared to the TD controls are due to other factors, such as group differences in latency jitter across trials, skull thickness, myelination, maturation, or other differences in brain structure or function. Since our hyperacusis measure was a dichotic one, and given that only four of our 18 WS participants did not report hyperacusis, it may be that our sample was underpowered to detect hyperacusis-neurophysiological associations. Our study could benefit from a comprehensive clinical assessment of the severity of hyperacusis and symptoms of phonophobia, which could enable the identification of subtle associations between behavioral and neurophysiologic phenotypes. Furthermore, including an ERP control test of an unrelated system (e.g., early visual processing) could offer this study a more precise view on the specificity of the findings to early auditory processing. Further studies with large samples of both WS and TD participants should assess the contribution of specific genetic variations within the WS-deleted region to the atypical auditory processing in WS. Such studies may also shed light on genetic involvement in the development and regulation of central auditory processing. In conclusion, the present study suggests that central auditory processing is hyperactive in WS. We revealed increased auditory brain responses of both the obligatory complex and MMN responses, implying that the encoding of sounds and

O. Zarchi et al.

788 preattentive change detection are altered in WS. Our findings also imply that auditory processing may be subjected to a delayed or diverse maturation and may affect the development

of high cognitive functioning in WS. These results may contribute to our understanding of the mechanisms underlying the unique auditory phenotype typical to individuals with WS.

References Attias, J., Raveh, E., Ben-Naftali, N. F., Zarchi, O., & Gothelf, D. (2008). Hyperactive auditory efferent system and lack of acoustic reflexes in Williams syndrome. Journal of Basic and Clinical Physiology and Pharmacology, 19, 193–208. doi: 10.1515/JBCPP.2008. 19.3-4.193 Brock, J., Jarrold, C., Brock, J., Farran, E. K., Laws, G., McCormack, T., . . . Bromberg, H. S. (2007). Language abilities in Williams syndrome: A critical review. Development and Psychopathology, 19, 97–127. doi: 10.1017/S095457940707006X Ceponiene, R., Shestakova, A., Balan, P., Alku, P., Yiaguchi, K., & N€a€at€anen, R. (2001). Children’s auditory event-related potentials index sound complexity and “speechness.” International Journal of Neuroscience, 109, 245–260. doi: 10.3109/00207450108986536 Chemtob, C., Roitblat, H. L., Hamada, R. S., Carlson, J. G., & Twentyman, C. T. (1988). A cognitive action theory of posttraumatic stress disorder. Journal of Anxiety Disorders, 2, 253–275. doi: 10.1016/0887-6185(88)90006-0 Cherniske, E. M., Carpenter, T. O., Klaiman, C., Young, E., Bregman, J., Insogna, K., . . . Pober, B. R. (2004). Multisystem study of 20 older adults with Williams syndrome. American Journal of Medical Genetics Part A, 131, 255–264. doi: 10.1002/ajmg.a.30400 Cheung, P. P. P., & Siu, A. M. H. (2009). A comparison of patterns of sensory processing in children with and without developmental disabilities. Research in Developmental Disabilities, 30, 1468–1480. doi: 10.1016/j.ridd.2009.07.009 Coelho, C. B., Sanchez, T. G., & Tyler, R. S. (2007). Hyperacusis, sound annoyance, and loudness hypersensitivity in children. Progress in Brain Research, 166, 169–178. doi: 10.1016/S0079-6123(07) 66015-4 Delorme, A., & Makeig, S. (2004). EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of Neuroscience Methods, 134, 9–21. Don, A. J., Schellenberg, G. E., & Rourke, B. P. (1999). Music and language skills of children with Williams syndrome. Child Neuropsychology, 5, 154–170. doi: 10.1076/chin.5.3.154.7337 First, M. B., Gibbon, M., Spitzer, R. L., & Williams, J. B. W. (1996). User’s guide for the Structured Clinical Interview for DSM-IV Axis I Disorders—Research Version. SCID-I, version 2. Washington, DC: American Psychiatric Association Ge, Y., Wu, J., Sun, X., & Zhang, K. (2011). Enhanced mismatch negativity in adolescents with posttraumatic stress disorder (PTSD). International Journal of Psychophysiology, 79, 231–235. doi: 10.1016/ j.ijpsycho.2010.10.012 Goldman, L., Gottlieb, D., & Stine-Sagi, L. (2001). WAIS-III Hebrew Manual. Jerusalem, Israel: PsychTech Corp. Gothelf, D., Farber, N., Raveh, E., Apter, A., & Attias, J. (2006). Hyperacusis in Williams syndrome: Characteristics and associated neuroaudiologic abnormalities. Neurology, 66, 390–395. doi: 10.1212/01.wnl.0000196643.35395.5f Henkin, Y., Kishon-Rabin, L., Gadoth, N., & Pratt, H. (2002). Auditory event-related potentials during phonetic and semantic processing in children. Audiology & Neuro-otology, 7, 228–239. Hetrick, W. P., Erickson, M. A., & Smith, D. A. (2012). Phenomenological dimensions of sensory gating. Schizophrenia Bulletin, 38, 178– 191. doi: 10.1093/schbul/sbq054 Holinger, D. P., Bellugi, U., Mills, D. L., Korenberg, J. R., Reiss, A. L., Sherman, G. F., & Galaburda, A. M. (2005). Relative sparing of primary auditory cortex in Williams syndrome. Brain Research, 1037, 35–42. doi: 10.1016/j.brainres.2004.11.038 Johnson, L. B., Comeau, M., & Clarke, K. D. (2001). Hyperacusis in Williams syndrome. The Journal of Otolaryngology, 30, 90–92. doi: 10.1212/01.wnl.0000196643.35395.5f Kaufman, J., Birmaher, B., Brent, D., Rao, U. M. A., Flynn, C., Moreci, P., . . . Ryan, N. (1997). Schedule for affective disorders and schizophrenia for school-age children—present and lifetime version (K-SADS-PL): Initial reliability and validity data. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 980–988. doi: 10.1097/00004583-199707000-00021

Klein, A. J., Armstrong, B. L., Greer, M. K., & Brown Iii, F. R. (1990). Hyperacusis and otitis media in individuals with Williams syndrome. Journal of Speech and Hearing Disorders, 55, 339–344. doi: 10.1044/jshd.5502.339 Kujala, T., Aho, E., Lepist€ o, T., Jansson-Verkasalo, E., Nieminen-von Wendt, T., von Wendt, L., & N€a€at€anen, R. (2007). Atypical pattern of discriminating sound features in adults with Asperger syndrome as reflected by the mismatch negativity. Biological Psychology, 75, 109–114. doi: 10.1016/j.biopsycho.2006.12.007 Lenhoff, H. M., Perales, O., & Hickok, G. (2001). Absolute pitch in Williams syndrome. Music Perception, 18, 491–503. doi: 10.1525/ mp.2001.18.4.491 Lenhoff, H. M., Wang, P. P., Greenberg, F., & Bellugi, U. (1997). Williams syndrome and the brain. Scientific American, 277, 68–73. doi: 10.1038/scientificamerican1297-68 Lense, M. D., Shivers, C. M., & Dykens, E. M. (2013). Musicality in Williams syndrome: Examining relationships among auditory perception, musical skill, and emotional responsiveness to music. Frontiers in Psychology, 4. doi: 10.3389/fpsyg.2013.00525 Levitin, D. J., & Bellugi, U. (1998). Musical abilities in individuals with Williams syndrome. Music Perception, 15, 357–389. doi: 10.2307/40300863 Levitin, D. J., Cole, K., Chiles, M., Lai, Z., Lincoln, A., & Bellugi, U. (2004). Characterizing the musical phenotype in individuals with Williams syndrome. Child Neuropsychology, 10, 223–247. doi: 10.1080/09297040490909288 Levitin, D. J., Cole, K., Lincoln, A., & Bellugi, U. (2005). Aversion, awareness, and attraction: Investigating claims of hyperacusis in the Williams syndrome phenotype. Journal of Child Psychology and Psychiatry, 46, 514–523. doi: 10.1111/j.1469-7610. 2004.00376.x Levitin, D. J., Menon, V., Schmitt, J. E., Eliez, S., White, C. D., Glover, G. H., . . . Reiss, A. L. (2003). Neural correlates of auditory perception in Williams syndrome: An fMRI study. NeuroImage, 18, 74–82. doi: 10.1006/nimg.2002.1297 Lieblich, A., Ben-Schachar, N., & Ninio, A. (1976). WISC-R Hebrew Manual. Jerusalem, Jerusalem Psychological Corporation and the Israeli Ministry of Education. Marler, J. A., Elfenbein, J. L., Ryals, B. M., Urban, Z., & Netzloff, M. L. (2005). Sensorineural hearing loss in children and adults with Williams syndrome. American Journal of Medical Genetics Part A, 138A, 318–327. doi: 10.1002/ajmg.a.30970 Marler, J. A., Sitcovsky, J. L., Mervis, C. B., Kistler, D. J., & Wightman, F. L. (2010). Auditory function and hearing loss in children and adults with Williams syndrome: Cochlear impairment in individuals with otherwise normal hearing. American Journal of Medical Genetics Part C: Seminars in Medical Genetics (Vol. 154, pp. 249–265): Wiley Online Library. doi: 10.1002/ajmg.c.30262 Martınez-Castilla, P., Sotillo, M., & Campos, R. (2013). Do individuals with Williams syndrome possess absolute pitch? Child Neuropsychology, 19, 78–96. doi: 10.1080/09297049.2011.639755 Meyer-Lindenberg, A., Mervis, C. B., & Berman, K. F. (2006). Neural mechanisms in Williams syndrome: A unique window to genetic influences on cognition and behaviour. Nature Reviews Neuroscience, 7, 380–393. doi: 10.1038/nrn1906 Mills, D., Dai, L., Fishman, I., Yam, A., Appelbaum, L., St. George, M., . . . Korenberg, J. (2013). Genetic mapping of brain plasticity across development in Williams syndrome: ERP markers of face and language processing. Developmental Neuropsychology, 38, 613–642. doi: 10.1080/87565641.2013.825617 Morgan Iii, C.A., & Grillon, C. (1999). Abnormal mismatch negativity in women with sexual assault-related posttraumatic stress disorder. Biological Psychiatry, 45, 827–832. doi: 10.1016/S0006-3223(98) 00194-2 N€a€at€anen, R. (1990). The role of attention in auditory information processing as revealed by event-related potentials and other brain measures of cognitive function. Behavioral and Brain Sciences, 13, 201–233. doi: 10.1017/S0140525X00078407

Hyperactive auditory processing in Williams syndrome N€a€at€anen, R., Pakarinen, S., Rinne, T., & Takegata, R. (2004). The mismatch negativity (MMN): Towards the optimal paradigm. Clinical Neurophysiology, 115, 140–144. doi: 10.1016/j.clinph.2003.04.001 Neville, H. J., Holcomb, P. J., & Mills, D. L. (1989). Auditory sensory and language processing in Williams syndrome: An ERP study. Journal of Clinical and Experimental Neuropsychology, 11, 52. Neville, H. J., Mills, D. L., & Bellugi, U. (1994). Effects of altered auditory sensitivity and age of language acquisition on the development of language-relevant neural systems: Preliminary studies of Williams syndrome. Hillsdale, NJ: Erlbaum.  Pinheiro, A. P., Galdo-Alvarez, S., Sampaio, A., Niznikiewicz, M., &  F. (2010). Electrophysiological correlates of semantic Gonc¸alves, O. processing in Williams syndrome. Research in Developmental Disabilities, 31, 1412–1425. doi: 10.1016/j.ridd.2010.06.017 Ponton, C. W., Eggermont, J. J., Kwong, B., & Don, M. (2000). Maturation of human central auditory system activity: Evidence from multichannel evoked potentials. Clinical Neurophysiology, 111, 220–236. doi: 10.1016/S1388-2457(99)00236-9 Rinne, T., & N€a€at€anen, R. (2002). Maturation of cortical sound processing as indexed by event-related potentials. Clinical Neurophysiology, 113, 870–882. doi: 10.1016/S1388-2457(02)00078-0 Robinson, B. F., Mervis, C. B., & Robinson, B. W. (2003). The roles of verbal short-term memory and working memory in the

789 acquisition of grammar by children with Williams syndrome. Developmental Neuropsychology, 23, 13–31. doi: 10.1080/87565641.2003. 9651885 Sussman, E. S. (2007). A new view on the MMN and attention debate. Journal of Psychophysiology, 21, 164–175. doi: 10.1027/0269-8803. 21.34.164 Vicari, S., Carlesimo, G., Brizzolara, D., & Pezzini, G. (1996). Shortterm memory in children with Williams syndrome: A reduced contribution of lexical-semantic knowledge to word span. Neuropsychologia, 34, 919–925. doi: 10.1016/0028-3932(96)00007-3 Zarchi, O., Attias, J., Raveh, E., Basel-Vanagaite, L., Saporta, L., & Gothelf, D. (2011). A comparative study of hearing loss in two microdeletion syndromes: Velocardiofacial (22q11. 2 deletion) and Williams (7q11. 23 deletion) syndromes. Journal of Pediatrics, 158, 301–306. doi: 10.1016/j.jpeds.2010.07.056 Zarchi, O., Carmel, M., Avni, C., Attias, J., Frisch, A., Michaelovsky, E., . . . Gothelf, D. (2013). Schizophrenia-like neurophysiological abnormalities in 22q11.2 deletion syndrome and their association to COMT and PRODH genotypes. Journal of Psychiatric Research, 47, 1623–1629. doi: 10.1016/j.jpsychires.2013.07.004 (RECEIVED June 4, 2014; ACCEPTED December 3, 2014)

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material.