Accepted Manuscript

Abnormal auditory synchronization in stuttering: A magnetoencephalographic study

Yoshikazu Kikuchi, MD, PhD, Tsuyoshi Okamoto, Katsuya Ogata, Koichi Hagiwara,
Toshiro Umezaki, Masamutsu Kenjo, Takashi Nakagawa, Shozo Tobimatsu

PII: S0378-5955(16)30168-X
DOI: 10.1016/j.heares.2016.10.027
Reference: HEARES 7262

To appear in: Hearing Research

Received Date: 29 April 2016

Accepted Date: 31 October 2016

Please cite this article as: Kikuchi, Y., Okamoto, T., Ogata, K., Hagiwara, K., Umezaki, T.,
Kenjo, M., Nakagawa, T., Tobimatsu, S., Abnormal auditory synchronization in stuttering: A
magnetoencephalographic study, Hearing Research (2016), doi: 10.1016/j.heares.2016.10.027.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 Kikuchi et al. 1
ACCEPTED MANUSCRIPT
1 Abnormal auditory synchronization in stuttering: A magnetoencephalographic

2 study

4 Running title: Increased auditory synchrony in stuttering

PT
6 Yoshikazu Kikuchi,a Tsuyoshi Okamoto,b,c Katsuya Ogata,d Koichi Hagiwara,d Toshiro

Umezaki,e,f Masamutsu Kenjo,g Takashi Nakagawa,a Shozo Tobimatsud

RI
7

SC
a
9 Department of Otorhinolaryngology, Graduate School of Medical Sciences, Kyushu

10 University, Fukuoka, Japan

11
b

U
Faculty of Arts and Science, Kyushu University, Fukuoka, Japan
AN
c
12 Graduate School of Systems Life Sciences, Kyushu University, Fukuoka, Japan
d
M

13 Department of Clinical Neurophysiology, Faculty of Medical Sciences, Kyushu

14 University, Fukuoka, Japan

D

e
15 Voice and Swallowing Center, Fukuoka Sanno Hospital, Fukuoka, Japan
TE

f
16 International University of Health and Welfare, Fukuoka, Japan
g
17 Department of Special Needs Education, University of Teacher Education Fukuoka,
EP

18 Fukuoka, Japan
C

19
AC

20 Corresponding author: Yoshikazu Kikuchi, MD, PhD, Department of

21 Otorhinolaryngology, Graduate School of Medical Sciences, Kyushu University, 3-1-1

22 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan.

23 Tel.: +81-92-642-5668; Fax: +81-92-642-5685;

24 E-mail: kikuci@med.kyushu-u.ac.jp
 Kikuchi et al. 2
ACCEPTED MANUSCRIPT
25 Abstract

26 In a previous magnetoencephalographic study, we showed both functional and structural

27 reorganization of the right auditory cortex and impaired left auditory cortex function in

28 people who stutter (PWS). In the present work, we reevaluated the same dataset to

29 further investigate how the right and left auditory cortices interact to compensate for

PT
30 stuttering. We evaluated bilateral N100m latencies as well as indices of local and

RI
31 inter-hemispheric phase synchronization of the auditory cortices. The left N100m

32 latency was significantly prolonged relative to the right N100m latency in PWS, while

SC
33 healthy control participants did not show any inter-hemispheric differences in latency. A

34 phase-locking factor (PLF) analysis, which indicates the degree of local phase

35
U
synchronization, demonstrated enhanced alpha-band synchrony in the right auditory
AN
36 area of PWS. A phase-locking value (PLV) analysis of inter-hemispheric
M

37 synchronization demonstrated significant elevations in the beta band between the right

38 and left auditory cortices in PWS. In addition, right PLF and PLVs were positively
D

39 correlated with stuttering frequency in PWS. Taken together, our data suggest that
TE

40 increased right hemispheric local phase synchronization and increased inter-hemispheric

41 phase synchronization are electrophysiological correlates of a compensatory mechanism

EP

42 for impaired left auditory processing in PWS.

C

43
AC

44 Keywords: stuttering; magnetoencephalography; phase-locking factors; phase-locking

45 values; synchronization

46

47 Abbreviations:

48 AEF, auditory event-related field; DAF, delayed auditory feedback; ECD, equivalent
 Kikuchi et al. 3
ACCEPTED MANUSCRIPT
49 current dipole; fMRI, functional magnetic resonance imaging;

50 HCs, healthy controls; HPI, head-position indicator; MC-tSSS, movement

51 compensation-temporal extension signal−space separation; MEG,

52 magnetoencephalography; PLF, phase-locking factor; PLV, phase-locking value; PWS,

53 people who stutter; RMS, root mean square.

PT
54

RI
U SC
AN
M
D
TE
C EP
AC
 Kikuchi et al. 4
ACCEPTED MANUSCRIPT
55 1. Introduction

56 Stuttering is a developmental disorder that affects speech fluency. People who

57 stutter (PWS) can temporarily decrease their rate of stuttering by using a metronome,

58 masking noise, or altering auditory feedback (Altrows and Bryden, 1977; Andrews et al.,

59 1983; Hampton and Weber-Fox, 2008; Kalinowski et al., 1993; Lincoln et al., 2006).

PT
60 Recent advances in the pathophysiology of stutter have suggested that multiple neural

RI
61 systems in both hemispheres are involved in stutter generation. PWS typically show

62 impaired auditory and speech production and processing in the left hemisphere;

SC
63 cortico-cortical connectivity is also impaired among areas related to the left superior

64 longitudinal fasciculus, which connects auditory areas with frontal speech

65
U
motor-planning areas (including Broca’s area) and motor regions (Chang et al., 2008,
AN
66 2011; Cykowski et al., 2010; Neef et al., 2015; Sommer et al., 2002; Watkins et al.,
M

67 2008). Stutter is also associated with decreased structural and functional connectivity in

68 the basal ganglia-thalamocortical loops (Chang and Zhu, 2013; Lu et al., 2010).
D

69 Interestingly, several studies have demonstrated enhanced right-hemispheric

TE

70 function in PWS; PWS show broad activation in the frontal operculum,

71 temporo-parietal junction, dorsolateral prefrontal cortex (Kell et al., 2009; Neumann et

EP

72 al., 2005), and superior temporal gyrus (STG) (Chang et al., 2008; Kell et al., 2009;

Kikuchi et al., 2011). Fluency-shaping therapies reduce activity in the right primary
C

73
AC

74 auditory cortex (BA 41) and lead to reactivation of the left primary auditory cortex in

75 both overt and covert reading paradigms (Kell et al., 2009). Accordingly, it has been

76 argued that altered right hemispheric function could reflect compensation for, rather

77 than a contribution to, the emergence of stutter (Chang et al., 2011; Kell et al., 2009).

78 Yet, the electrophysiological characteristics of this potential compensatory mechanism

79 are poorly understood.

 Kikuchi et al. 5
ACCEPTED MANUSCRIPT
80 In a recent magnetoencephalography (MEG) study (Kikuchi et al., 2011), we

81 demonstrated that PWS demonstrate disturbed auditory sensory gating in the left

82 hemisphere as an electrophysiological signature of stutter. We also identified increased

83 gray matter volume in the right auditory cortex in association with the expansion of

84 tonotopic organization, which supported the idea that dysfunction of the left auditory

PT
85 cortex produces compensatory reorganization in the right auditory cortex. However,

RI
86 several issues with this theory remain to be addressed. First, it is necessary to quantify

87 the functional inter-relationship between impaired left auditory processing and

SC
88 reorganization of the right auditory cortex. Second, although we were previously unable

89 to detect alterations in auditory event-related fields (AEFs) in the right hemisphere of

90
U
PWS, altered activity in the right auditory cortex should be examined using other
AN
91 indices such as phase synchrony. For example, hypersynchrony in the right hemisphere
M

92 of PWS would support the idea of hyperactivity and expanded tonotopic organization as

93 a compensatory mechanism for stutter.

D

94 On the above basis, we re-analyzed our previous MEG data using two indices
TE

95 of phase synchrony: the phase-locking factor (PLF) and phase-locking value (PLV)

96 indices. PLF is an index of local phase synchrony (Palva et al., 2005; Sinkkonen et al.,
EP

97 1995) whereas PLV is an index of phase synchrony between two signals of different

spatial origin (Lachaux et al., 1999), and has accordingly been utilized for evaluating
C

98
AC

99 functional connectivity between cortical areas (Hagiwara et al., 2010, 2014; Simões et

100 al., 2003). We hypothesized that the compensatory mechanism in PWS would be

101 expressed as altered synchronization in auditory processing in the right hemisphere. We

102 also analyzed latency differences at N100m between the two hemispheres in response to

103 monaural stimulation in order to assess trans-callosal conduction. It was anticipated that

104 the present study would provide evidence for alterations in basic auditory processing in
 Kikuchi et al. 6
ACCEPTED MANUSCRIPT
105 PWS and inform the nature and utility of compensatory mechanisms in PWS.

106

107 2. Materials and Methods

108 2.1 Participants

109 Sixteen male PWS (mean age, 30 ± 5.8 years; range, 21–41 years) and 16 male

PT
110 healthy controls (HCs) (mean age, 30.8 ± 6.4 years; range, 22–43 years) participated in

RI
111 this study. All participants provided written informed consent for participation and the

112 study was approved by the Ethics Committee of Kyushu University. None of the

SC
113 participants had any history of otological or neurological disorders, and all participants

114 were right-handed according to the Edinburgh Handedness Inventory (Oldfield, 1971).

115
U
The frequency of stuttering was assessed by a speech-language-hearing
AN
116 therapist and quantified as a percentage of stuttered syllables among at least 300
M

117 analyzable syllables, averaged across three different speaking contexts (reading a short

118 story, describing pictures, and asking questions). Only instances of unambiguous
D

119 stuttering were counted (Jones et al., 2000) and stuttering was defined as either syllable
TE

120 repetition or audible/inaudible sound prolongation (Conture, 2001); interjections,

121 polysyllabic whole-word repetitions, monosyllabic whole-word repetitions atypical of

EP

122 stuttering, revisions, and phrase-repetitions were not counted. The mean stuttering

frequency was 8.4% (range, 0.8%–36.7%) for PWS and 0.2% (range: 0–0.5 %) for
C

123
AC

124 control participants. One PWS participant was excluded due to low dysfluency (less

125 than 1%) (Jones et al., 2005; Kell et al., 2009).

126

127 2.2 Stimulus setting

128 Auditory tone-burst stimuli (250, 1000, and 4000 Hz) were each monaurally

129 presented for 300 ms (10-ms and 20-ms rise and fall times, respectively) more than 128
 Kikuchi et al. 7
ACCEPTED MANUSCRIPT
130 times in a randomized order. The interstimulus interval was 2.5–3.5 s. For each

131 participant, the hearing threshold was determined and stimuli were delivered at an

132 intensity of 30 dB above this threshold so as not to induce cross-hearing. Stimuli were

133 delivered using a Tone Burst Generator (Kyushu Keisokuki, Fukuoka, Japan) and

134 passed through a plastic tube (length, 6 m; inner diameter, 8 mm) into sponge earpieces

PT
135 fitted into the participants’ ears. Participants attended to auditory stimuli while in the

RI
136 supine position.

137

SC
138 2.3 MEG recording

139 AEFs were measured using a whole-head 306-channel biomagnetometer

140
U
system (Elekta-Neuromag, Helsinki, Finland) in a quiet magnetically shielded room.
AN
141 The detector array included 102 identical triple-sensor elements, with each sensor
M

142 element composed of two orthogonally oriented planar-type gradiometers and one

143 magnetometer. In the present study, MEG data was analyzed from 204 planar-type
D

144 gradiometers, as gradiometers reduce signal from external artifacts such as geomagnetic
TE

145 signal and other environmental influences. Additionally, the amplitude of the root mean

146 square (RMS) from a pair of gradiometers indicates a maximum value when the pair is
EP

147 located immediately above the activated brain region, such that gradiometer pairs are

useful for the approximation of the activated brain region (Hämäläinen et al., 1993).
C

148
AC

149 Prior to recording, four head-position indicator (HPI) coils were attached to the scalp

150 and a three-dimensional digitizer was used to measure anatomical landmarks of the

151 head with respect to the HPI coils. During data acquisition, the HPI coils were

152 continuously active and the head position was continuously monitored. Magnetic

153 responses were digitally sampled at a rate of 1000 Hz. After the completion of

154 recording, movement compensation was performed by temporal extension signal−space

 Kikuchi et al. 8
ACCEPTED MANUSCRIPT
155 separation (MC-tSSS) with Maxfilter 2.0 software (Neuromag®; Elekta, Stockholm,

156 Sweden) and applied off-line to the recorded raw data in order to reduce artifactual

157 signals arising from outside of the sensor array and to correct for changes in head

158 position and associated movement-related artifacts (Medvedovsky et al., 2007; Taulu,

159 2004, 2005). Off-line averaging of AEFs was performed using the

PT
160 MC-tSSS-reconstructed raw data, and a total of 128 responses were averaged for each

RI
161 ear. Finally, a −100 to −10-ms baseline adjustment and a band pass filter of 1–30 Hz

162 were applied to AEFs.

SC
163 Of 204 recording channels, 46 were centered on a gradiometer with a maximal

164 amplitude of the N100m signal and selected for AEF analysis. In our previous study, we

165
U
only examined contralateral N100m responses because the primary objective was to
AN
166 assess tonotopic organization (Kikuchi et al., 2011). Here, we calculated the difference
M

167 between ipsilateral and contralateral N100m latencies in order to evaluate hemispheric

168 interactions. Contralateral N100m responses to high-intensity auditory stimulation are

D

169 transmitted via the corpus callosum to the ipsilateral temporal cortex (Oe et al., 2002).
TE

170 AEF peak latencies and amplitudes were determined from root mean square (RMS)

171 waveforms reconstructed using gradiometer data. We confirmed that activity at RMS
EP

172 peaks originated from the auditory area by estimating the equivalent current dipole

(ECD). To calculate the ECD of the N100m component, 10-ms data across the time
C

173
AC

174 point showing maximal amplitude were selected from RMS waveforms, and the ECDs

175 with the largest dipole moments over this period were further selected. The calculated

176 ECD strengths and locations were accepted for further analysis when they satisfied the

177 following criteria: (1) an N100m latency within 80–160 ms of the RMS peak; (2) a

178 goodness-of-fit value of the ECD greater than 85%; and (3) a confidence volume less

179 than 100 mm3 (Elbert et al., 2002; Rojas et al., 2002; Kikuchi et al., 2011).
 Kikuchi et al. 9
ACCEPTED MANUSCRIPT
180

181 2.4 Assessment of PLFs and PLVs

182 To analyze the time−frequency domain of neural oscillations within auditory

183 areas, we performed continuous wavelet transformation of the single-trial MEG signals

184 using the complex Morlet wavelet as a mother function. We then calculated two

PT
185 measures of neural phases (but not powers). First, we calculated the PLF, which is an

RI
186 index of phase synchronization with respect to a given stimulus. This parameter was

187 calculated according to a previously described method (Palva et al., 2005; Sinkkonen et

SC
188 al., 1995; Hagiwara et al., 2014) using the following equation:

189
U
AN
190 where Φk (t, f0) represents the amplitude as well as the phase of a specific frequency f0

at time point t, and N denotes the total epoch number. PLFs were evaluated for each
M

191

192 sensor representing right and left auditory cortical activity (Palva et al., 2005).
D

193 The second synchronization index was the PLV, which is an index of phase
TE

194 synchronization between signals representing two separate cortical areas. Details of the

195 PLV analysis have previously been described (Hagiwara et al., 2010; Lachaux et al.,
EP

196 1999, Simões et al., 2003). Briefly, PLV was calculated using the following equation:
C
AC

197

198 where m and n denote channels representing right and left auditory activities,

199 respectively. Both PLFs and PLVs were calculated from the two gradiometers of the

200 highest amplitudes, each of which included two orthogonal sensors (Fig. 1).

201

202 2.5 Statistical analysis

 Kikuchi et al. 10
ACCEPTED MANUSCRIPT
203 To analyze the N100m latency and PLFs, we performed a repeated-measures

204 analysis of variance (ANOVA) with group (PWS vs. control) as a between-participant

205 factor and hemisphere (left hemisphere vs. right hemisphere) and stimulation side (left

206 ear vs. right ear) as within-participant factors. We focused on differences between the

207 two groups, two hemispheres, and two stimulation sides; thus, we treated the results of

PT
208 different stimulation frequencies together.

RI
209 To analyze differences between the ipsilateral and contralateral N100m

210 latencies and PLVs, we performed a repeated-measures ANOVA with group (PWS vs.

SC
211 control) as a between-participant factor and stimulation side (left ear vs. right ear) as a

212 within-participant factor. Post-hoc analyses were conducted using the Bonferroni

213
U
correction. Relationships among N100m latency differences, PLFs, PLVs, and stuttering
AN
214 rates were evaluated using Pearson’s correlation coefficient. The time window of
M

215 interest for PLFs and PLVs was set between 80 and 160 ms (Kikuchi et al., 2011). We

216 also analyzed various frequency bands, including the theta (4−7 Hz), alpha (8−13 Hz),
D

217 beta (14−30 Hz), and gamma (31−50 Hz) bands. In all analyses, the significance level
TE

218 was set at 0.05. Statistical analysis and computation was performed using SPSS 11.5

219 (SPSS Inc., Chicago, IL, USA).

EP

220
C

221 3. Results
AC

222 The N100m latency data of both groups are summarized in Table 1. Analysis of

223 the N100m latency revealed a significant main effect of stimulation side (F1, 88 =14, p <

224 0.001). The mean N100m latency for left ear stimulation was significantly longer than

225 that for right ear stimulation. Although the main effects of group and hemisphere were

226 not significant, there was a significant interaction between group and hemisphere (F1, 88

227 = 4.5, p = 0.036). A post-hoc analysis revealed that the left hemispheric latency was
 Kikuchi et al. 11
ACCEPTED MANUSCRIPT
228 significantly longer than the right hemispheric latency in PWS (p = 0.019; Table 1).

229 Observed latency differences between ipsilateral and contralateral N100m

230 responses are summarized in Table 2. There was a significant main effect of stimulation

231 side on the difference between ipsilateral and contralateral N100m latency (F1, 88 =5.6, p

232 = 0.020). Although the main effect of group was not significant, there was a significant

PT
233 interaction between group and stimulation side (F1, 88 = 8.4, p = 0.005). A post-hoc

RI
234 analysis revealed that the N100 latency difference for the left stimulation side in PWS

235 was significantly longer than that for the right stimulation side in PWS (p < 0.01) and

SC
236 that for left stimulation sides in HCs (p = 0.040).

237 The PLF data are shown in Figure 2. Although the main effects of group,

238
U
hemisphere, and stimulation side were not significant in the alpha band, there was a
AN
239 significant interaction between group and hemisphere (F1, 88 = 6.9, p = 0.011). A
M

240 post-hoc analysis revealed that right PLFs were significantly higher than left PLFs in

241 PWS (p = 0.002), and also significantly higher than right PLFs in HCs (p = 0.002).
D

242 There were no significant differences between PWS and HCs in the theta (F1, 88 = 0.63,
TE

243 p = 0.42), beta (F1, 88 = 0.12, p = 0.91), or gamma (F1, 88 = 0.16, p = 0.87) bands.

244 Figure 3 summarizes the PLV data. In the beta band, the analysis showed a
EP

245 significant main effect of group (F1, 88 = 4.6, p = 0.035), but no significant main effects

of stimulation side, or any interaction between group and stimulation side.

C

246
AC

247 Synchronization between the right and left auditory cortices in the beta band was also

248 higher in PWS than in HCs. There were no significant differences between PWS and

249 HCs in the theta (F1, 88 = 0.04, p = 0.08), alpha (F1, 88 = 1.1, p = 0.3), or gamma (F1, 88 =

250 0.88, p = 0.35) bands.

251 Correlation analyses in the alpha band indicated that stuttering frequency was

252 positively correlated with PLF in the right hemisphere (p < 0.05). In addition,
 Kikuchi et al. 12
ACCEPTED MANUSCRIPT
253 correlation analyses in the beta band indicated that stuttering frequency was positively

254 correlated with PLV in PWS (p < 0.05). No correlations were found between stuttering

255 frequency and N100m latency or the difference between ipsilateral and contralateral

256 N100m latency.

257

PT
258 4. Discussion

RI
259 In this study, we found that alpha-band PLFs in the right auditory area were

260 significantly increased in PWS versus HCs. Moreover, PLF in the right auditory area

SC
261 was positively correlated with stuttering frequency in PWS; this is in contrast with

262 previous MEG studies on the N100 amplitude that did not show any correlation between

263
U
right STG activation and stuttering frequency (Beal et al., 2011; Etchell et al., 2016;
AN
264 Salmelin et al., 1998). A lack of observed change in the amplitude appears to be
M

265 contradictory with the results of previous functional magnetic resonance imaging

266 (fMRI) studies demonstrating increased activity in the right STG in PWS (Chang et al.,
D

267 2008; Kell et al., 2009). However, we previously demonstrated that PLF analysis is
TE

268 useful for evaluating the magnitude of cortical hyperactivity, even in the absence of a

269 change in amplitude (Hagiwara et al., 2014). Thus, it is likely that the PLF analysis in
EP

270 the current study detected excitability changes in the right auditory cortex in PWS.

Unlike fMRI studies, our analysis of neural oscillation indices was temporally specific
C

271
AC

272 for the early stage of auditory processing. Taken together, it can be argued based on our

273 data that basic auditory processing in the right hemisphere was significantly altered in

274 PWS. Consistent with our result, an fMRI study of fluency-shaping therapies

275 demonstrated that compensatory mechanisms likely underlie increased activity in the

276 right primary auditory cortex of PWS (Kell et al., 2009). Therefore, we propose that
 Kikuchi et al. 13
ACCEPTED MANUSCRIPT
277 enhanced local alpha-band synchrony in the right auditory area could be a signature

278 compensatory change in PWS.

279 We also found that beta-band PLVs were much higher in PWS than in HCs. In

280 addition, the beta-band PLV was correlated with stuttering frequency. Recently, it has

281 been proposed that a possible cause of stuttering is a deficit in the brain’s timing

PT
282 networks (Alm, 2004; Etchell et al., 2014). On the premise that internal timing derives

RI
283 from beta oscillation activity (Etchell et al., 2015), a MEG study demonstrated phase

284 alternations of beta oscillations in children with stutter versus those with typical

SC
285 development (Etchell et al., 2016). These authors claimed that changes in beta band

286 oscillations could predict auditory function and were ultimately related to stutter.

287
U
Alternatively, it has been reported that the strength of beta band oscillations predicts the
AN
288 magnitude of auditory sensory gating (Hong et al., 2008). Thus, increased beta band
M

289 synchrony (in association with increased stuttering frequency) may have been the result

290 of impaired auditory gating (Kikuchi et al., 2011). It is also known that stuttering is
D

291 improved by delayed auditory feedback (DAF) (Soderberg, 1968). Interestingly, an

TE

292 EEG study demonstrated that beta band hyperactivity in the right temporo-parietal

293 region of PWS was markedly reduced by DAF (Rastatter et al., 1998). In consideration
EP

294 of the above information, we suggest that enhanced beta band synchrony in our study

represented another compensatory change in inter-hemispheric auditory processing in

C

295
AC

296 PWS. While Etchell et al. (2015) proposed a similar hypothesis in the basal ganglia of

297 PWS, and indeed it has been shown that both functional and structural connectivity

298 among the putamen, auditory areas, and motor areas are altered in the left hemisphere of

299 children with stutter (Chang et al., 2013), our study is the first to assess

300 inter-hemispheric beta band synchrony between bilateral auditory areas in PWS.

301 It is interesting to note that enhanced beta band synchrony was observed in the
 Kikuchi et al. 14
ACCEPTED MANUSCRIPT
302 PLV analysis but not in the PLF analysis. This is likely due to the nature of the

303 differences between evoked and induced oscillations. Owing to their close temporal

304 relationship with incoming stimuli, evoked oscillations reflect bottom-up sensory

305 transmission. In contrast, induced oscillations represent the internal dynamics of cortical

306 processing networks and have been related to higher cognitive functions (Uhlhaas and

PT
307 Singer, 2010).

RI
308 The current study also incorporated analyses of ipsilateral N100m latencies as

309 well as ipsilateral-contralateral differences in N100m latencies. As a result, prolonged

SC
310 N100m latencies were found in the left hemisphere of PWS irrespective of stimulation

311 side. Significant inter-hemispheric latency differences in response to left ear stimulation

312
U
were also identified. The former result is consistent with impaired left auditory
AN
313 processing, though the complete results require careful interpretation. Monaural
M

314 auditory stimulation separates contralateral and ipsilateral signals between the brainstem

315 and primary auditory cortex (Eggermont, 2001; Langers et al., 2005). PWS exhibit
D

316 normal auditory latencies in the brainstem (Decker et al., 1982; Newman et al., 1985;
TE

317 Stager, 1990). Thus, delayed left hemispheric latencies in the present study could

318 indicate impaired signaling from the brainstem to the left hemisphere/primary auditory
EP

319 cortex. It is likely that the left auditory cortex itself is impaired in PWS, as suggested by

our previous study (Kikuchi et al., 2011).

C

320
AC

321 A recent tractography study showed that the right and left auditory areas were

322 reciprocally connected via a posterior portion of the corpus callosum (Westerhausen et

323 al., 2009). If PWS have dysfunction of the posterior portion of corpus callosum, an

324 inter-hemispheric latency difference might be suspected in response to auditory

325 stimulation. However, only left ear stimulation produced significantly longer latency

326 differences compared to HCs. There are two possible explanations for this finding. First,
 Kikuchi et al. 15
ACCEPTED MANUSCRIPT
327 dysfunction of the corpus callosum in PWS may have only been present in the anterior

328 corpus callosum, which is associated with speech production rather than auditory

329 processing (Civier et al., 2015). In this scenario, our result would be consistent with

330 preservation of the posterior-dominant trans-callosal auditory processing network.

331 Enhanced beta-band synchrony between the two auditory areas could be mediated by

PT
332 this connection. The other possibility is the presence of ascending pathway dominance

RI
333 over trans-callosal information transfer. In response to monaural auditory stimulation,

334 ascending information crosses to the contralateral side in the lower brainstem, followed

SC
335 by ipsilateral signal transfer towards the auditory cortex (Eggermont, 2001; Langers et

336 al., 2005). In fact, it has been shown that neither unilateral lesion of the contralateral

337
U
auditory cortex nor absence of a corpus callosum affect the ipsilateral response
AN
338 (Gutschalk et al., 2015). Overall, we propose that a delayed N100m response in the left
M

339 hemisphere of PWS resulted from impaired signaling to the left auditory cortex.

340 In conclusion, we suggest that increased local alpha band synchronization in

D

341 the right auditory area and increased inter-hemispheric beta band synchronization are
TE

342 electrophysiological correlates of a compensatory mechanism for impaired left auditory

343 processing in PWS.

EP

344

Funding Source
C

345
AC

346 This work was supported by Grants-in-Aid for Young Scientists (B) 25861566 (Y. K.)

347 and 21791620 (K. O. and Y. K.), and a Grant-in-Aid for Scientific Research on

348 Innovative Areas MEXT KAKENHI 15H05875 (S.T.)

349

350 Conflict of Interest

351 The authors declare no competing financial interests

 Kikuchi et al. 16
ACCEPTED MANUSCRIPT
352

353 Acknowledgements

354 We would like to thank Editage (www.editage.jp) for English language editing.

PT
RI
U SC
AN
M
D
TE
C EP
AC
 Kikuchi et al. 17
ACCEPTED MANUSCRIPT
355 References

356 Alm, P.A., 2004. Stuttering and the basal ganglia circuits: a critical review of possible

357 relations. J. Commun. Disord. 37, 325-369.

358 Altrows, I.F., Bryden, M.P., 1977. Temporal factors in the effects of masking noise on

359 fluency of stutterers. J. Commun. Disord. 10, 315-329.

PT
360 Andrews, G., Craig, A., Feyer, A.M., Hoddinott, S., Howie, P., Neilson, M., 1983.

RI
361 Stuttering: a review of research findings and theories circa 1982. J. Speech Hear.

362 Disord. 48, 226-246.

SC
363 Chang, S.E., Zhu, D.C., 2013. Neural network connectivity differences in children who

364 stutter. Brain 136, 3709-3726.

365
U
Chang, S.E., Horwitz, B., Ostuni, J., Reynolds, R., Ludlow, C.L., 2011. Evidence of left
AN
366 inferior frontal-premotor structural and functional connectivity deficits in adults
M

367 who stutter. Cereb. Cortex. 21, 2507-2518.

368 Chang, S.E., Erickson, K.I., Ambrose, N.G., Hasegawa-Johnson, M.A., Ludlow, C.L.,
D

369 2008. Brain anatomy differences in childhood stuttering. Neuroimage 39,

TE

370 1333-1344.

371 Civier, O., Kronfeld-Duenias, V., Amir, O., Ezrati-Vinacour, R., Ben-Shachar, M., 2015.
EP

372 Reduced fractional anisotropy in the anterior corpus callosum is associated with

reduced speech fluency in persistent developmental stuttering. Brain Lang. 143,

C

373
AC

374 20-31.

375 Conture, E., 2001. Stuttering: Its nature, diagnosis and treatment. Needham Heights,

376 MA: Allyn & Bacon.

377 Cykowski, M.D., Fox, P.T., Ingham, R.J., Ingham, J.C., Robin, D.A., 2010. A study of

378 the reproducibility and etiology of diffusion anisotropy differences in

379 developmental stuttering: a potential role for impaired myelination. Neuroimage 52,
 Kikuchi et al. 18
ACCEPTED MANUSCRIPT
380 1495-1504.

381 Eggermont, J.J., 2001. Between sound and perception: reviewing the search for a neural

382 code. Hear. Res. 157, 1-42.

383 Etchell, A.C., Johnson, B.W., Sowman, P.F., 2014.Behavioral and multimodal

384 neuroimaging evidence for a deficit in brain timing networks in stuttering: a

PT
385 hypothesis and theory. Front. Hum. Neurosci. 8, 467.

RI
386 Etchell, A.C., Johnson, B.W., Sowman, P.F., 2015. Beta oscillations, timing, and

387 stuttering. Front. Hum. Neurosci. 8, 1036.

SC
388 Etchell, A.C., Ryan, M., Martin, E., Johnson, B.W., Sowman, P.F., 2016. Abnormal time

389 course of low beta modulation in non-fluent preschool children: A

390
U
magnetoencephalographic study of rhythm tracking. Neuroimage 25, 953-963.
AN
391 Elbert, T., Sterr, A., Rockstroh, B., Pantev, C., Muller, M.M., Taub, E., 2002. Expansion
M

392 of the tonotopic area in the auditory cortex of the blind. J. Neurosci. 22, 9941-9944.

393 Gutschalk, A., Steinmann, I., 2015. Stimulus dependence of contralateral dominance in
D

394 human auditory cortex. Hum. Brain Mapp. 36, 883-896.

TE

395 Hagiwara, K., Okamoto, T., Shigeto, H., Ogata, K., Somehara, Y., Matsushita, T., Kira,

396 J., Tobimatsu, S., 2010. Oscillatory gamma synchronization binds the primary and
EP

397 secondary somatosensory areas in humans. Neuroimage 51, 412-420.

Hagiwara, K., Ogata, K., Okamoto, T., Uehara, T., Hironaga, N., Shigeto, H., Kira, J.,
C

398
AC

399 Tobimatsu, S., 2014. Age-related changes across the primary and secondary

400 somatosensory areas: an analysis of neuromagnetic oscillatory activities. Clin.

401 Neurophysiol. 125, 1021-1029.

402 Hämäläinen, M., Hari, R., Ilmoniemi, R.J., Knuutila, J., Lounasmaa, O.V., 1993.

403 Magnetoencephalography—theory, instrumentation, and applications to

404 noninvasive studies of the working human brain. Rev. Mod. Phys. 65, 413-497.
 Kikuchi et al. 19
ACCEPTED MANUSCRIPT
405 Hampton, A., Weber-Fox, C., 2008. Non-linguistic auditory processing in stuttering:

406 evidence from behavior and event-related brain potentials. J. Fluency. Disord. 33,

407 253-273.

408 Hong, L.E., Buchanan, R.W., Thaker, G.K., Shepard, P.D., Summerfelt, A., 2008. Beta

409 (~16 Hz) frequency neural oscillations mediate auditory sensory gating in humans.

PT
410 Psychophysiology 45, 197-204.

RI
411 Jones, M., Onslow, M., Harrison, E., Packman, A., 2000. Treating stuttering in young

412 children: predicting treatment time in the Lidcombe Program. J. Speech. Lang. Hear.

SC
413 Res. 43, 1440-1450.

414 Jones, M., Onslow, M., Packman, A., Williams, S., Ormond, T., Schwarz, I., Gebski, V.,

415
U
2005. Randomised controlled trial of the Lidcombe programme of early stuttering
AN
416 intervention. BMJ 331, 659-661.
M

417 Kalinowski, J., Armson, J., Roland-Mieszkowski, M., Stuart, A., Gracco, V.L., 1993.

418 Effects of alterations in auditory feedback and speech rate on stuttering frequency.
D

419 Lang. Speech 36, 1-16.

TE

420 Kell, C.A., Neumann, K., von Kriegstein, K., Posenenske, C., von Gudenberg, A.W.,

421 Euler, H., Giraud, A.L., 2009. How the brain repairs stuttering. Brain 132,
EP

422 2747-2760.

Kikuchi, Y., Ogata, K., Umesaki, T., Yoshiura, T., Kenjo, M., Hirano, Y., Okamoto, T.,
C

423
AC

424 Komune, S., Tobimatsu, S., 2011. Spatiotemporal signatures of an abnormal

425 auditory system in stuttering. Neuroimage 55, 891-899.

426 Lachaux, J.P., Rodriguez, E., Martinerie, J., Varela, F.J., 1999. Measuring phase

427 synchrony in brain signals. Hum. Brain Mapp. 8, 194-208.

428 Langers, D.R., van Dijk, P., Backes, W.H., 2005. Lateralization, connectivity and

429 plasticity in the human central auditory system. Neuroimage 28, 490-499.
 Kikuchi et al. 20
ACCEPTED MANUSCRIPT
430 Lincoln, M., Packman, A., Onslow, M., 2006. Altered auditory feedback and the

431 treatment of stuttering: a review. J. Fluency Disord. 31, 71-89.

432 Lu, C., Peng, D., Chen, C., Ning, N., Ding, G., Li, K., Yang, Y., Lin, C., 2010. Altered

433 effective connectivity and anomalous anatomy in the basal ganglia-thalamocortical

434 circuit of stuttering speakers. Cortex. 46, 49-67.

PT
435 Medvedovsky, M., Taulu, S., Bikmullina, R., Paetau, R., 2007. Artifact and head

RI
436 movement compensation in MEG. Neurol. Neurophysiol. Neurosci. 29, 4.

437 Neef, N.E., Anwander, A., Friederici, A.D., 2015. The neurobiological grounding of

SC
438 persistent stuttering: from structure to function. Curr. Neurol. Neurosci. Rep. 15,

439 63.

440
U
Neumann, K., Preibisch, C., Euler, H.A., von Gudenberg, A.W., Lanfermann, H., Gall,
AN
441 V., Giraud, A.L., 2005. Cortical plasticity associated with stuttering therapy. J.
M

442 Fluency Disord. 30, 23-39.

443 Oe, H., Kandori, A., Yamada, N., Miyashita, T., Tsukada, K., Naritomi, H., 2002.
D

444 Interhemispheric connection of auditory neural pathways assessed by auditory

TE

445 evoked magnetic fields in patients with fronto-temporal lobe infarction. Neurosci.

446 Res. 44, 483-488.

EP

447 Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh

inventory. Neuropsychologia 9, 97-113.

C

448
AC

449 Palva, S., Linkenkaer-Hansen, K., Naatanen, R., Palva, J.M., 2005. Early neural

450 correlates of conscious somatosensory perception. J. Neurosci. 25, 5248-5258.

451 Rastatter, M.P., Stuart, A., Kalinowski, J., 1998. Quantitative electroencephalogram of

452 posterior cortical areas of fluent and stuttering participants during reading with

453 normal and altered auditory feedback. Percept. Mot. Skills. 87, 623-633.

454 Rojas, D.C., Bawn, S.D., Carlson, J.P., Arciniegas, D.B., Teale, P.D., Reite, M.L., 2002.
 Kikuchi et al. 21
ACCEPTED MANUSCRIPT
455 Alterations in tonotopy and auditory cerebral asymmetry in schizophrenia. Biol.

456 Psychiatry 52, 32-39.

457 Simões, C., Jensen, O., Parkkonen, L., Hari, R., 2003. Phase locking between human

458 primary and secondary somatosensory cortices. Proc. Natl. Acad. Sci. U. S. A. 100,

459 2691-2694.

PT
460 Sinkkonen, J., Tiitinen, H., Naatanen, R., 1995. Gabor filters: an informative way for

RI
461 analysing event-related brain activity. J. Neurosci. Methods 56, 99-104.

462 Soderberg, G.A., 1968. Delayed auditory feedback and stuttering. J. Speech Hear.

SC
463 Disord. 33, 260-267.

464 Sommer, M., Koch, M.A., Paulus, W., Weiller, C., Buchel, C., 2002. Disconnection of

465
U
speech-relevant brain areas in persistent developmental stuttering. Lancet 360,
AN
466 380-383.
M

467 Taulu, S., Simola, J., Kajola, M., 2004. MEG recordings of DC fields using the signal

468 space separation method (SSS). Neurol. Clin. Neurophysiol. 35, 1-4.
D

469 Taulu, S., Simola, J., Kajola, M., 2005. Applications of the signal space separation
TE

470 method. IEEE. Trans. Signal. Process. 53, 3359-3372.

471 Uhlhaas, P.J., Singer, W., 2010. Abnormal neural oscillations and synchrony in
EP

472 schizophrenia. Nat. Rev. Neurosci. 11, 100-113.

Watkins, K.E., Smith, S.M., Davis, S., Howell, P., 2008. Structural and functional
C

473
AC

474 abnormalities of the motor system in developmental stuttering. Brain 131:50-59.

475 Westerhausen, R., Grüner, R., Specht, K., Hugdahl, K., 2009. Functional relevance of

476 interindividual differences in temporal lobe callosal pathways: a DTI tractography

477 study. Cereb. Cortex 19, 1322-1329.

478
 Kikuchi et al. 22
ACCEPTED MANUSCRIPT
479 Figure legends

480 Figure 1. Analyses of phase-locking factor (PLF) and phase-locking value (PLV) in a

481 normal control participant. (A) Representative MEG waveforms from 204 planar

482 gradiometers in response to left 1000 Hz stimulation (left). Left and right hemispheric

483 responses in the rectangles (left) are enlarged in the middle panel. These responses were

PT
484 subjected to a time−frequency analysis to obtain the PLF (right). The peak time of PLF

RI
485 was about 100 ms. Higher PLFs are shown in red and lower PLFs are shown in blue.

486 (B) A representative PLV analysis between signals from the two distinct auditory

SC
487 cortices. Similar to the PLF analysis, left and right responses in the rectangles were

488 subjected to time−frequency analysis to obtain PLVs (right). The peak time of PLV was

489
U
about 100 ms. Higher PLVs are shown in red and lower PLVs are shown in blue.
AN
490
M

491 Figure 2. Phase-locking factors in healthy control participants (HCs) versus people who

492 stutter (PWS) in both hemispheres. Right local phase synchronization was significantly
D

493 higher than left local phase synchronization in PWS, and was also higher than that of
TE

494 right local phase synchronization in HCs. Black rectangles indicate alpha band (8–13

495 Hz) synchronization between 80 and 160 ms after stimulus onset.

EP

496
C

497 Figure 3. Phase-locking values in healthy control participants (HCs) and people who
AC

498 stutter (PWS). Synchronization between the right and left auditory cortices in PWS was

499 significantly higher than that in HCs. White rectangles indicate beta band (14–30 Hz)

500 synchronization between 80 and 160 ms after stimulus onset.

501
 ACCEPTED MANUSCRIPT

Kikuchi et al. 23

502 Table 1. N100m latencies in the left and right hemispheres after monaural stimulation with three tonal frequencies.

Participants Stimulation side Left hemisphere Right hemisphere

PT
250 Hz 1000 Hz 4000 Hz 250 Hz 1000 Hz 4000 Hz

RI
HC Right 114.1 ± 3.0 95.5 ± 2.1 102.5 ± 2.1 123.5 ± 10.8 105.1 ± 9.0 118.1 ± 15.6

SC
Left 132.6 ± 8.9 104.6 ± 7.6 116.0 ± 8.1 124.4 ± 2.9 96.2 ± 1.8 101.8 ± 3.5

122.9 ± 3.0a 100.3 ± 2.1a 101.7 ± 2.1a

U
PWS Right 126.0 ± 7.0 105.8 ± 7.8 117.2 ± 18.1

AN
Left 144.9 ± 19.2a 110.4 ± 7.6a 118.5 ± 17a 131.7 ± 2.6 98.7 ± 1.7 102.1 ± 3.3

M
503 Values are expressed as the mean ± SD.
a
504 Significant inter-hemispheric difference at the level of p < 0.05.

D
TE
505 Abbreviations: HC, healthy controls; PWS, people with stuttering.

506
EP
507
C

508
AC

509

510
 ACCEPTED MANUSCRIPT

Kikuchi et al. 24

511

512 Table 2. Latency differences between ipsilateral and contralateral N100m responses.

PT
Right ear stimulation Left ear stimulation 513

RI
250 Hz 1000 Hz 4000 Hz 250 Hz 1000 Hz 4000 Hz514

SC
HC 7.6 ± 1.4 9.4 ± 2.3 14 ± 3.4 7.2 ± 1.5 8.5 ± 1.4 13 ± 2.9 515

10.5 ± 2.1b,c 13 ± 1.4b,c 18 ± 4.3b,c

U
PWS 1.2 ± 1.5 3.8 ± 1.3 14 ± 3.5 516

AN
517

518 Values are expressed as the mean ± SD.

M
b
519 Significant difference between right and left ear stimulations at the level of p < 0.01.

D
c
520 Significant difference between HC and PWS for left ear stimulation at the level of p < 0.05.

521
TE
Abbreviations: HC, healthy controls; PWS, people with stuttering.
EP
522
C

523
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Highlights

• People who stutter (PWS) have significantly delayed left-only N100m latency

• PWS have increased local phase synchronization in the alpha band

• PWS show increased inter-hemispheric synchronization in the beta band

PT
• Right hemisphere hyperactivity offsets impaired left hemispheric function in PWS

RI
U SC
AN
M
D
TE
C EP
AC