Auditory-motor rhythm synchronization in children with autism spectrum disorder

Research in Autism Spectrum Disorders 35 (2017) 51–61

Contents lists available at ScienceDirect

Research in Autism Spectrum Disorders journal homepage: http://ees.elsevier.com/RASD/default.asp

Auditory-motor rhythm synchronization in children with autism spectrum disorder Ana Tryfona,b,* , Nicholas E. Fostera,b , Tia Ouimeta,b , Krissy Doyle-Thomasc , Evdokia Anagnostouc, Megha Shardaa , Krista L. Hydea,b , for NeuroDevNet ASD imaging group1 a International Laboratory for Brain Music and Sound Research (BRAMS), Pavillon 1420 Mont-Royal, Department of Psychology, University of Montreal, C.P. 6128, succ. Centre-Ville, Montreal, Quebec, H3C 3J7, Canada b Faculty of Medicine, McIntyre Medical Building, McGill University, 3655 Sir William Osler, Montreal, Quebec H3G 1Y6, Canada c Holland Bloorview Kids Rehabilitation Hospital, University of Toronto, 150 Kilgour Road, Toronto, Ontario, M4G 1R8, Canada

A R T I C L E I N F O

A B S T R A C T

Article history: Received 31 May 2016 Received in revised form 21 December 2016 Accepted 22 December 2016 Number of reviews completed is 3 Available online xxx

Background: Autism spectrum disorder (ASD) is characterized by difficulties in social and communication skills as well as atypical sensory perception and motor skills. Sensorimotor abilities such as auditory-motor integration are essential for social interaction and communication. The goal of this research was to investigate the development of auditorymotor rhythm synchronization for the first time in ASD versus typically-developing (TD) children. Methods: Participants were 31 boys with ASD and 23 TD boys that were matched in age and IQ. Participants were tested on an auditory-motor rhythm synchronization task in which they tapped in synchrony with rhythms of varying metrical complexity. Results: Both children with ASD and TD performed similarly on this task and both groups performed better with age. Conclusions: This work demonstrates that non-verbal rhythm synchronization is intact in ASD over the course of childhood development. This research serves to better understand sensorimotor interactions in ASD and to better define sensory phenotypes in ASD. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Autism Auditory Motor Rhythm Synchronization Development

1. Introduction The coordination and regulation of sensory and movement information is essential for social interaction and communication (Donnellan, Hill, & Leary, 2012). In particular, auditory-motor integration is critical for acquiring complex skills such as speech, language and musical proficiency (Zatorre, Chen, & Penhune, 2007). However, many individuals with autism spectrum disorder (ASD) have atypical sensory (Leekam, Nieto, Libby, Wing, & Gould, 2007) and motor skills (Gowen & Hamilton, 2013; Green et al., 2009). Studies on auditory-motor integration in ASD are limited, but some work has shown impairments in the context of complex and speech-related auditory-motor tasks in ASD. However, it is unclear how individuals with ASD perform on more basic and non-verbal auditory-motor rhythm synchronization tasks. Moreover, the

* Corresponding author at: International Laboratory for Brain Music and Sound Research (BRAMS), Pavillon 1420 Mont-Royal, Department of Psychology, University of Montreal, C.P. 6128, succ. Centre-Ville, Montreal, Quebec, H3C 3J7, Canada. E-mail address: [email protected] (A. Tryfon). 1 http://www.neurodevnet.ca/research/asd http://dx.doi.org/10.1016/j.rasd.2016.12.004 1750-9467/© 2016 Elsevier Ltd. All rights reserved.

52

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

developmental course of auditory-motor rhythm synchronization is unclear in ASD. To these aims, in the present study we examined basic auditory-motor rhythm synchronization in children with ASD versus TD for the first time with a focus on age-related changes in performance. 1.1. Auditory-motor skills in ASD versus TD In the auditory domain, individuals with ASD tend to have impaired high-level and complex sensory processing, but intact or enhanced detail-oriented processing of low-level stimuli (Mottron, Dawson, Soulières, Hubert, & Burack, 2006). For example, while the perception of complex linguistic and socially-relevant speech stimuli is often impaired in ASD compared to TD, certain aspects of musical information such as pitch discrimination may be intact or enhanced in ASD (for a review, see Ouimet, Foster, Tryfon, & Hyde, 2012; O’Connor, 2012). In the motor domain, studies have found both fine and gross motor differences in ASD (Freitag, Kleser, Schneider, & von Gontard, 2007; Green et al., 2009). For instance, fine motor abilities necessary for daily living skills have been shown to be delayed in children with ASD, as well as gross motor skills such as running and throwing (Provost, Lopez, & Heimerl, 2007; Lane, Harpster, & Heathcock, 2012). Motor skill performance largely relies on intact sensory abilities (Tseng, Diedrichsen, Krakauer, Shadmehr, & Bastian, 2007; Shadmehr, Smith, & Krakauer, 2010; Liu, 2013). However, there has been limited study of auditory-motor integration in ASD, and most work has been conducted in the context of gross motor and complex speech tasks. For example, Moran, Foley, Parker, and Weiss (2013) examined two-legged hopping in sync with an auditory cue in young ASD adults with impaired expressive language versus TD. While TD performed well on this task, individuals with ASD were unable to regulate their motor production to match an external auditory cue. Other studies have investigated auditory-motor integration in the context of speech-related tasks. Lin et al. (2015) examined auditory-motor control in adults with ASD versus TD by delaying their speech feedback or adding loud noise to their auditory feedback. They found that performance in ASD was less affected by noise and more affected by the delayed speech feedback. These findings indicate that, in contrast to TD, individuals with ASD rely more on feedback control than on feedforward control in speech production. In another study, Russo, Larson, & Kraus (2008) examined audio-vocal regulation in children with ASD by presenting pitch-shifted voice auditory feedback to vocalizing participants. A subset of children with ASD who showed larger responses to perturbed auditory feedback also had lower receptive language scores relative to TD, signalling a possible dysfunction in the audio-vocal system for voice pitch regulation in ASD. Taken together, the studies reviewed above indicate that auditory-motor integration in ASD is altered in the context of gross motor and speech-related tasks. Some work has begun to examine the potential of auditory-motor based therapies in ASD. For example, Wan et al. (2011) effectively used Auditory-Motor Mapping Training (AMMT) to promote speech production in non-verbal children with ASD. In another study, Srinivasan et al., 2015 showed that socially embedded movement-based contexts are valuable in promoting imitation/praxis, interpersonal synchrony, and motor performance in children with ASD. In addition, Hardy and Lagasse (2013) proposed the use of auditory rhythmic cueing to improve motor functioning in ASD. However, the authors signal the important need for more research to be conducted on auditory-motor rhythm synchronization in ASD before such clinical applications might be applied. The present study addresses this important gap between fundamental science and clinical application by examining auditory rhythm synchronization in ASD with a novel focus on the effect of development. 1.2. The effect of age on auditory-motor skills in ASD versus TD Recent reports indicate that the interaction between movement and sound is present from birth and plays a critical role in communication at the beginning of life (for a review, see Provasi, Anderson, & Barbu-Roth, 2014). For example, children between 18 months to 4 years old are able to synchronize finger tapping to an external rhythm if the external rhythm is close to their own internal tapping tempo (Provasi & Bobin-Begue, 2003). Studies in TD have shown that rhythm synchronization and tempo discrimination improves with age (Drake, Jones, & Baruch, 2000). Studies of auditory processing in ASD and TD have reported developmental differences in performance. In TD, pitch discrimination ability increases with age whereas in ASD, it is enhanced in childhood but then stabilizes across development (Mayer, Hannent, & Heaton, 2014). Similarly, our laboratory found enhanced abilities in auditory global-local pitch processing in ASD at younger ages (Foster et al., 2016). Moreover, while auditory echoic memory in TD takes time to develop fully into adulthood, in ASD this development stops prematurely (Erviti et al., 2014). In terms of motor development, young children with ASD have both gross and fine motor differences that can become more pronounced with age (Lloyd, MacDonald, & Lord, 2013), but the results are mixed. In a recent longitudinal study, Travers et al. (2016) examined the developmental course of fine motor ability in ASD versus TD using a finger tapping task in which participants were asked to tap the index finger as many times as possible within 10 s. No group differences were found in childhood, but the ASD group showed slower tapping versus TD in adolescence and adulthood. In another longitudinal study, development of gross and fine motor skills, visual reception, and receptive and expressive language was investigated in infants at high and low risk for ASD (Landa & Garrett-Mayer, 2006). By as early as 14 months of age, the high risk ASD group performed worse overall, and showed a slower developmental trajectory relative to TD. The developmental course of auditory-motor integration in ASD is unclear. As reported above, audio-vocal regulation differs in children with ASD (Russo et al., 2008). In TD, audio-vocal regulation generally improves with age (Liu, Chen, Jones, Huang, & Liu, 2011); however, it is unclear how it evolves with age in ASD. In a study by Liu (2013), the authors reported

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

53

correlated sensory (auditory and visual) and motor changes in young children with ASD, but again, it is not clear how these differences might evolve throughout childhood and beyond. Taken together, previous studies indicate that there are developmental differences in auditory and motor skills in ASD versus TD. However, the results vary and no studies have specifically examined the effect of age on auditory-motor rhythm synchronization in ASD as done in the present study. 1.3. Study objectives The overall goal of this research was to investigate auditory-motor rhythm synchronization in children with ASD versus TD. Specifically, the aims were to test: 1) whether there are group differences in a non-verbal auditory-motor rhythm synchronization task in children with ASD versus TD, and 2) how age affects performance on this task in both groups. Based on previous findings of atypical sensory and motor skills in ASD, it was expected that children with ASD would have decreased auditory-motor rhythm synchronization performance compared to TD, and also show differences in development. Findings from the present study promise to further our understanding of sensorimotor processing and developmental differences in ASD. 2. Methods 2.1. Participants Two groups of children participated in the present study: 1) 31 boys with ASD and 2) 23 TD boys that were matched on a group level for age (ASD: mean 11.5 years, SD 2.8, range 6.9–15.6; TD: mean 11.1 years, SD 3.1 range 6.8–15.7; p = [12_TD$IF]0.60) and intelligence quotient (IQ; ASD [n = 30]: mean 108.5, SD 13.7, range 84–146; TD: mean 114.2, SD 10.4, range 88–133; p = [13_TD$IF]0.10). Participants were recruited as part of the “NeuroDevNet Autism Demonstration Project”, a multi-site initiative to study brain structural and behavioral development in children with ASD (Zwaigenbaum et al., 2011). Participants were recruited and tested at two sites: 1) in Montreal, Canada at the Montreal Neurological Institute (14 ASD, 15 TD), and, 2) in Toronto, Canada at the Holland Bloorview Kids Rehabilitation Hospital and Toronto Sick Kids Hospital (17 ASD, 8 TD). Individuals with ASD were diagnosed by expert opinion (American Psychiatric Association, 2000) and diagnoses were supported by standard diagnostic measures including the Autism Diagnostic Interview – Revised (ADI-R) (Lord, Rutter, & Le Couteur, 1994) and the Autism Diagnostic Observation Schedule (ADOS) (Lord et al., 2000). The TD participants had no history of neurological or psychiatric illness and all participants had normal hearing based on parent report. Exclusion criteria included IQ less than 70, gestational age of 35 weeks or less, any primary psychiatric diagnoses (with the exception of ASD for the clinical group), a medical history of neurological disease, family history of ASD (for TD group), or hearing impairment. While any primary psychiatric diagnoses were an exclusion criteria for TD, participants with ASD who had comorbidity with disorders such as ADHD or anxiety (and consequently were on medication) were not excluded from our sample due to the high incidence of comorbidity with such disorders in ASD (Leyfer et al., 2006). Six participants with ASD had a comorbid diagnosis of ADHD/anxiety or were on medication at the time of testing (including Concerta, Risperdal, Abilify, Zoloft, Effexor, Biphentin, and Prozac). IQ was assessed using the full-scale score on the Wechsler Abbreviated Scale of Intelligence (WASI) or WASI-II (Wechsler, 1999; Wechsler, 2011). The full-two scale score was used where full-four scale score was not available. One participant with ASD did not have a WASI completed at time of testing. The Edinburgh handedness questionnaire was available for 11 ASD (7 right-handed, and 4 ambidextrous) and 14 TD participants (8 righthanded, 3 left-handed, 2 ambidextrous) (Dragovic, 2004; Oldfield, 1971). Participant characteristics are presented in Table 1. The present study was approved by local ethics committees at each site. All guardians provided written informed consent and participants above the age of 14 provided assent. All participants were compensated for their time. 2.2. Auditory-motor rhythm synchronization task Participants were tested on an auditory-motor rhythm synchronization task previously used in TD adults (Chen, Penhune, & Zatorre, 2008). This task was chosen for several reasons. First, this particular rhythm synchronization task’s sensitivity to fine-grained tapping timing has been demonstrated across a wide range of ability (non-musicians and musicians) (e.g., Table 1 Characteristics of participants. Mean

ASD (n = 31)

TD (n = 23)

Significance

Age in years (SD/range) Full Scale IQ (SD/range) ADOS social interaction (n = 28; SD/range) ADOS communication (n = 28; SD/range) ADI-R social (n = 28; SD/range) ADI-R communication (n = 28; SD/range) ADI-R restricted and repetitive behaviors (n = 28; SD/range)

11.5 (2.8/6.9–15.6) n = 30; 108.5 (13.7/84–146) 7.9 (2.5/3–13) 2.9 (1.9/0–8) 17.7 (6.0/9–30) 15.3 (4.1/7–24) 6.5 (2.6/2–12)

11.1 (3.1/6.8–15.7) 114.2 (10.4/88–133) – – – – –

p = [97_TD$IF]0.60 p = [98_TD$IF]0.10 – – – – –

54

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

Bailey & Penhune, 2010; Chen et al., 2008). Moreover, given that many individuals with ASD have some form of atypical language profile, the non-verbal aspect of this task is ideal to use in the ASD population. Finally, there is some indication that activities based around low-level auditory rhythm synchronization may have therapeutic impact in ASD (Hardy & Lagasse, 2013), but as reviewed above, the performance characteristics of individuals with ASD on such tasks has not yet been well defined. Participants were instructed to listen carefully to a woodblock rhythm and then tap back in synchrony with it on the button of a computer mouse. Participants were free to position the mouse in a comfortable position and complete the task with their preferred hand. The rhythms were randomly presented at three different levels of increasing metrical complexity: ‘strongly metric’ (easiest level), ‘medium metric’, or ‘weakly metric’ (hardest level). Each complexity group had 2 rhythms, and each rhythm had 11 notes and lasted 4.5–5.75 s in duration. Here we adapted the task used in Chen et al. (2008) to be more suited for children. Specifically, we included an easier (more metrical) rhythm level, removed the most difficult/ complex non-metrical rhythm used in Chen et al. (2008), and provided more breaks during the testing session. Five practice trials were included to ensure task comprehension. A cartoon giraffe wearing headphones was present on screen during both practice trials and testing blocks. Participants were instructed to listen to the rhythm the first time it was played, and then to tap along with it when it repeated the second time. When the participant was supposed to be listening to the rhythm, the headphones were highlighted on the cartoon giraffe. When the participant was supposed to be tapping along with the rhythm, the giraffe’s foot was highlighted. This also allowed testers in the room with the child to observe the child’s behavior and redirect if necessary (i.e. child is tapping while supposed to be listening only). Presentation software only recorded mouse clicks during tapping trials. Practice trials consisted of 5 trials of strongly metric rhythm patterns not used in the testing blocks. No participants were excluded due to task demands or comprehension issues. A schematic of the stimuli and a timeline of the listen and tap in synchrony trials is shown in Fig. 1. Within each block, the order of rhythm patterns was randomized, but the repetitions of a given rhythm were presented consecutively (e.g. 5-5-5 11-1 4-4-4 6-6-6 2-2-2 3-3-3). This was done to facilitate learning of each pattern’s structure. In total, participants completed 3 blocks, with 18 trials per block. A metronome-based tapping task was also administered to examine tapping ability. In this task, there were six trials where each trial consisted of 15 s tapping in synchrony with the metronome, followed by 15 s of

[(Fig._1)TD$IG] a)

Strongly Metric #2

Strongly Metric #1

0 sec

1.0

2.0

3.0

4.0

5.0

6.0

1.0

2.0

3.0

4.0

5.0

6.0

1.0

2.0

3.0

4.0

5.0

6.0

0 sec

1.0

2.0

3.0

4.0

5.0

6.0

2.0

3.0

4.0

5.0

6.0

Weakly Metric #6

Weakly Metric #5

0 sec

1.0

Medium Metric #4

Medium Metric #3

0 sec

0 sec

2.0

3.0

4.0

5.0

6.0

0 sec

1.0

end of listen trial, start of tap trial

b)

0 sec

1.0

visual cue to listen

2.0

auditory cue

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

visual cue to tap

11.0

12.0

auditory cue

+

13.0

14.0

15.0

16.0

17.0

18.0

Fig. 1. a) Examples of stimuli used in the auditory-motor rhythm synchronization task. ‘Strongly Metric’ (easiest level), ‘Medium Metric’, and ‘Weakly Metric’ (hardest level) refer to temporal organization of events within the sequence. Intervals between woodblock sounds were multiples of 250 ms.[96_TD$IF] b) Example of stimuli presentation: ‘Listen’ trial followed by a ‘Listen and tap in synchrony’ trial.

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

55

free-tapping wherein subject maintains the tempo of the metronome without ceasing tapping. [14_TD$IF]Inter-tap interval (ITI) [15_TD$IF] synchrony was measured during the free-tapping phase of this task. The stimuli were presented binaurally through Sennheiser HD 25-1 II headphones at a consistent and comfortable volume. The experiment was administered on a laptop computer using Presentation software (Neurobehavioral Systems, http://www.neurobs.com) which also recorded the performance for each trial. The task took 30 min to complete. 2.3. Data analyses Task performance was evaluated using four different measures. Three of these (ITI synchrony, onset asynchrony, and global accuracy) were previously described in Chen et al. (2008). The first measure, [16_TD$IF]ITI synchrony, was based on the accuracy with which subjects reproduced each time interval between events in a sequence. Specifically this score was calculated from the stimulus time interval (t) and the actual response time interval (r) using the formula Score = 1 abs(r t)/t. Higher values indicate better performance. The second measure, onset asynchrony, was measured by determining the absolute value of the time difference between the stimulus onset and the response onset for each sound in the pattern. Lower values for onset asynchrony indicate better performance. A third measure, the non-absolute value for onset asynchrony, was included to measure negative (early) and positive (late) time difference between the stimulus onset and the response onset for each sound in the pattern. Lastly, a fourth measure called global accuracy was calculated, which was a correct/incorrect score reflecting whether a tap response onset fell within each stimulus’ scoring window, [17_TD$IF]was defined as half the onset-to-onset interval before or after each stimulus onset. For all measures, if more than one tap response fell within the same scoring window, the response closest to the stimulus was taken. In order to test for group differences in performance, repeated measures linear fixed-effects models were used to analyze ITI synchrony, onset asynchrony, global accuracy, and non-absolute value on onset asynchrony. These models included group as a between-subjects fixed factor, rhythm complexity as the repeated measure, and age as a covariate using a compound symmetry covariance structure. The full model used for each performance metric was 1 + Group + Complexity + Age + Group  Complexity + Group  Age. Inclusion of covariates for site and IQ did not affect findings so they were not included in the final model. Effect sizes were calculated as partial eta squared using F values (Lakens, 2013). Lastly, performance on the metronome tapping task was analyzed using a linear fixed-effects model 1 + Age + Group + Group  Age.

Table 2 Summary of results for all performance measurements. Effect size was calculated using partial eta squared with 90% confidence interval. F

P

h2p

90% Confidence Interval (h2p )

ITI Synchrony Complexity Age Group Group [4_TD$IF] Age Complexity [9_TD$IF] Group

(2, 104) = 106.18 (1, 50) = 34.92 (1, 50) = 0.18 (1, 50) = 3.34 (2, 104) = 0.90

<0.001 <0.001 0.67 0.07 0.41

0.67 0.41 0.004 0.06 0.02

0.58–0.73 0.23–0.54 0.00–0.07 0.00–0.19 0.00–0.06

Onset Asynchrony Complexity Age Group Group [4_TD$IF] Age Complexity [9_TD$IF] Group

(2, 104) = 175.72 (1, 50) = 30.95 (1, 50) = 0.59 (1, 50) = 0.56 (2, 104) = 0.74

[10_TD$IF]<0.001 <0.001 0.45 0.46 0.48

0.77 0.38 0.01 0.01 0.01

0.71–0.81 0.20–0.51 0.00–0.10 0.00–0.10 0.00–0.06

Non-absolute Asynchrony Complexity Age Group Group [4_TD$IF] Age Complexity [9_TD$IF] Group

(2, 104) = 6.40 (1, 50) = 0.05 (1, 50) = 0.18 (1, 50) = 0.33 (2, 104) = 0.17

0.002 0.82 0.67 0.57 0.84

0.11 0.001 0.004 [10_TD$IF]0.01 0.003

0.03–0.20 0.00–0.04 0.00–0.07 0.00–0.08 0.00–0.02

Global Accuracy Complexity Age Group Group [4_TD$IF] Age Complexity [9_TD$IF] Group

(2, 104) = 68.0 (1, 50) = 8.94 (1, 50) = 0.49 (1, 50) = 0.02 (2, 104) = 2.43

<0.001 0.004 0.49 0.88 0.09

0.57 0.15 0.01 0.0004 0.04

0.46–0.64 0.03–0.30 0.00–0.09 0.00–0.02 0.00–0.11

Metronome Tapping ITI Synchrony Age Group Group [4_TD$IF] Age

(1, 47) = [102_TD$IF]23.81 (1, 47) = 0.74 (1, 47) = 0.31

<0.001 0.39 0.58

0.34 0.02 [10_TD$IF]0.01

0.16–0.48 0.00–0.11 0.00–0.09

56

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

3. Results A full table of F, P and h2p values for all dependent measures is shown in Table 2. Analyses of performance on the auditorymotor rhythm synchronization task revealed a significant main effect of rhythm complexity for each of the four measures: ITI synchrony [F(2, 104) = 106.18, p < 0.001, h2p = 0.67] onset asynchrony [F(2, 104) = 175.72, p < 0.001, h2p = 0.77], non-absolute value of onset asynchrony [F(2, 104) = 6.40, p = 0.002, h2p = 0.11] and global accuracy [F(2, 104) = 68.0, p < 0.001, h2p = 0.57] (Fig. 2). Unstandardized effect sizes and confidence intervals for complexity levels are shown in Table 3. No main effect of group was found for any of the four measures:[18_TD$IF] ITI synchrony, onset asynchrony, non-absolute value of onset asynchrony, or global accuracy (all p > 0.4, h2p  0.01). There was also no interaction for group by complexity for any of the four performance measurements (all p > 0.09, h2p  0.04). A significant main effect of age was found for ITI synchrony, [F(1, 50) = 34.92,

[(Fig._2)TD$IG] **

**

**

150

**

**

Group ASD TD

Onset Asynchrony (msec)

ITI Synchrony (%)

80

75

70

130

110

90

65

60

70 Strongly Metric

Medium Metric

Weakly Metric

Strongly Metric

Rhythm Complexity

**

**

Medium Metric

Weakly Metric

Rhythm Complexity

*

*

*

Non-Absolute Onset Asynchrony (msec)

20

Global Accuracy (%)

90

85

80

10

0

-10

-20

75 Strongly Metric

Medium Metric

Rhythm Complexity

Weakly Metric

Strongly Metric

Medium Metric

Weakly Metric

Rhythm Complexity

Fig. 2. Performance on the auditory-motor rhythm synchronization task. Panels from top left clockwise: Mean Inter-tap interval (ITI) synchrony scores, mean onset asynchrony (msec), mean non-absolute value of onset asynchrony (msec), and mean global accuracy. Values are adjusted for age. Data are reported as mean  95% CI. Performance differences between complexity levels shown by brackets: *indicates p < 0.05, **indicates p < 0.001.

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

57

Table 3 Effect sizes and confidence intervals for each performance measurement between complexity levels.

ITI Synchrony Strong–Medium Strong–Weak Medium–Weak Onset Asynchrony Strong–Medium Strong–Weak Medium–Weak

Difference in score

95% Confidence Interval of difference (Bonferroni corrected)

P of difference (Bonferroni corrected)

0.13 0.15 0.02

0.10 [103_TD$IF]to 0.16 0.12 [104_TD$IF]to 0.18 0.005 to 0.05

<0.001 <0.001 0.14

29.48 56.65 27.17

36.84 to 64.01 to 34.53 to

22.12 49.30 19.82

<0.001 <0.001 <0.001

Non-Absolute Asynchrony Strong–Medium 10.56 Strong–Weak 9.79 Medium–Weak 0.77

2.56 [105_TD$IF]to 18.57 1.78 [106_TD$IF]to 17.80 8.78 to 7.24

0.005 0.01 1.0

Global Accuracy Strong–Medium Strong–Weak Medium–Weak

4.62 [107_TD$IF]to 8.65 7.38 [108_TD$IF]to 11.41 0.75 [109_TD$IF]to 4.78

<0.001 <0.001 0.004

6.64 9.40 2.76

p < 0.001, h2p = 0.41], onset asynchrony [F(1, 50) = 30.95, p < 0.001, h2p = 0.38], and global accuracy [F(1, 50) = 8.94, p = 0.004,

h2p = 0.15] but not for the non-absolute value for onset asynchrony [F(1, 50) = 0.05, p = 0.82, h2p = 0.001]. The interaction of group by age was trending but not significant for ITI synchrony [F(1, 50) = 3.34, p = 0.07, h2p = 0.06] and non-significant for onset asynchrony, non-absolute value of onset asynchrony, or global accuracy (all p > 0.4, h2p  0.01) (Fig. 3). The adjusted R2 goodness of fit values for each of these models are as follows: ITI synchrony R2 = 0.52, onset asynchrony R2 = 0.57, nonabsolute value of onset asynchrony R2 = 0.01, and global accuracy R2 = 0.31. The metronome tapping task yielded only a significant main effect of age [F(1, 47) = [19_TD$IF]23.81, p < 0.001, h2p = 0.34]. No significant effects of group or group by age interaction were found (all p > 0.3, h2p  0.02). 4. Discussion The goal of the present study was to examine auditory-motor rhythm synchronization in children with ASD versus TD children with a special focus on the effect of age on task performance. Children with ASD performed similarly to TD children on the auditory-motor rhythm synchronization task, and both groups performed better with age. These results offer new insight into the development of auditory-motor synchronization in ASD and TD. 4.1. Intact basic and non-verbal auditory-motor synchronization in children with ASD As expected, TD children performed well on the auditory rhythm synchronization task but performed worse as a function of increasing rhythmic complexity. These results are consistent with previous findings of similar performance profiles in TD adults on this same task (Chen et al., 2008; Bailey & Penhune, 2010). Individuals with ASD also performed similar to TD on the auditory rhythm synchronization task overall. Specifically, the ASD group showed no differences on measures of precise temporal interval reproduction, tap synchrony, bias toward early or late responses, or coarse-level reproduction of the rhythmic patterns. Thus, the present results indicate that auditory-motor rhythmic synchronization is intact in children with ASD, at least in the context of the low-level and non-verbal type of task used here. However, the present findings of intact rhythm synchronization are in contrast with previous auditory-motor deficits found in ASD. These differing results may be related to differences in participant characteristics and the type of tasks used. In the present study, we tested individuals with ASD without significant language or intellectual impairments on a low-level and non-verbal auditory rhythm synchronization task that relies on fine motor synchronization skills. However, previous work has tested individuals with ASD with important language impairments and found deficits on more complex and gross motor tasks involving two-legged hopping to an auditory cue (Moran et al., 2013). Auditory-motor differences in ASD have also been found in the context of more complex speech tasks. For example, Lin et al. (2015) found that individuals with ASD had an atypical delayed auditory feedback effect on speech production. In another study, Russo et al. (2008) found that ASD had an atypical audio-vocal system regulation. Taken together, the present results indicate that auditory-motor integration is intact at least in the context of basic and non-verbal rhythm synchronization, but can be impaired in the context of more complex and gross motor tasks as well as speech-related tasks. The variation in results across studies signal the need to carefully control and describe participant samples and type of tasks used in ASD research.

58

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

[(Fig._3)TD$IG]

ASD TD Line of fit

90

Overall Onset Asynchrony (msec)

Overall ITI Synchrony (%)

175

80

70

60

150

125

100

75

50

50

p < 0.001

p < 0.001

40

6

8

10

12

14

16

6

8

Age (years)

12

14

16

Overall Non-Absolute Onset Asynchrony (msec)

Age (years)

95

90

Overall Global Accuracy (%)

10

85

80

75

70

40

20

0

-20

-40

p = 0.8

p = 0.004 -60

65 6

8

10

12

14

16

Age (years)

6

8

10

12

14

16

Age (years)

Fig. 3. The effect of age on performance on the auditory motor synchronization task. Panels from top left clockwise: ITI synchrony, onset asynchrony, nonabsolute value of onset asynchrony, and global accuracy. All four measurements are averaged over all rhythm complexities. P values indicate main effect of age.

4.2. Auditory-motor synchronization in development For the first time, here we examined the developmental course of low-level non-verbal rhythm synchronization in ASD versus TD. In TD, performance on the rhythm synchronization task improved as a function of age. These results are consistent with previous findings that showed that rhythm synchronization and tempo discrimination improves with age (Drake et al., 2000). The ASD group showed similar performance to TD on the auditory-motor rhythm synchronization task and performance also improved with age just as in TD. These results differ from previous reports of developmental differences in both auditory and motor abilities in ASD. Previous work has shown developmental differences in auditory pitch processing in early childhood in ASD (Foster et al., 2016; Mayer et al., 2014), and motor differences can become more pronounced with age in ASD (Lloyd et al., 2013; Travers et al., 2016). There is also some evidence for developmental differences in auditory-motor integration. For example, auditory-vocal regulation differs in children with ASD (Russo et al., 2008), and Liu (2013) reported correlated sensory (auditory and visual) and motor changes in young children with ASD. However, this is the first study to specifically examine basic non-verbal auditory-motor rhythm synchronization in ASD as a function of development. Taken together with previous work, the results indicate that the developmental course of basic rhythm synchronization is stable across childhood in ASD, whereas developmental differences may occur in the context of more complex and speech-related

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

59

auditory-motor integration tasks. This work signals the need for future research to examine the development of auditorymotor integration at different levels of processing in the same participants. 4.3. Implications The present findings have important fundamental implications to better understand auditory-motor integration in ASD and TD children overall and specifically with respect to development. These results also have implications for clinical practice. While some work has proposed the use of auditory rhythmic cueing to improve auditory-motor integration in ASD behavior such as speech (Wan et al., 2011) or motor functioning (Hardy & Lagasse, 2013) intervention-based studies still lack the relevant basic empirical work. The present findings of intact auditory-motor synchronization in ASD help to provide some of the needed empirical support to be able to conduct better-targeted interventions in ASD. For example, based on the present work, clinicians might capitalize on intact auditory rhythm synchronization skills to facilitate higher-level auditorymotor function in ASD such as speech or gross motor function. Finally, the present findings serve to guide future directions in ASD research related to auditory-motor integration that will have important impact on clinical practice, such as in developing neurobiological markers of ASD that can guide clinical practice. 4.4. Future directions The present results motivate various new areas of study in ASD. For example, our laboratory is currently examining brainbehavioral studies in a subsample of the participants tested here to link auditory-motor rhythm synchronization with measures of brain function and structure. Such brain imaging studies may help to better understand sensorimotor differences in ASD, and lead to new neurobiological markers of ASD. Another important extension of the present research would be to examine auditory-motor rhythm synchronization in a broader sample of individuals with ASD, including children with ASD with significant language or intellectual impairments, to see if our results generalize to a wider autism spectrum. In particular, it will be important to examine the connection between auditory-motor rhythm synchronization and language ability in ASD, based on previous research suggesting that auditory-motor based therapies can facilitate speech production in non-verbal ASD cases (Wan et al., 2011). It will also be important to investigate whether auditory-motor synchronization remains intact from adolescence into adulthood, as previous research has shown motor differences can be affected during this period (Travers et al., 2016). Lastly, a future longitudinal study of auditory-motor rhythm synchronization would contribute to a better understanding of its developmental trajectory in ASD by measuring change within individuals rather than the present cross-sectional design. 4.5. Conclusions In the present study, auditory-motor rhythm synchronization was investigated in ASD versus TD children. Rhythm synchronization was intact overall in children with ASD, and both TD and ASD showed improved task performance with age. These results serve to better characterize sensorimotor processing and auditory-motor integration in ASD and TD children. The present findings provide a foundation for future work examining the brain basis of perceptual differences in ASD. Such combined behavioral and brain-based studies may guide future sensory-based interventions in ASD. Conflict of interest EA has received consultation fees from NOVARTIS and Seaside Therapeutics and has an unrestricted grant from Sanofi Canada. Acknowledgements The authors thank the families who participated in this study. We thank C. D’Aguiar and E. Kossivas for their assistance with data collection. We thank V. Penhune for access to the auditory-motor task used in the present study as well as valuable feedback on this study. We thank the Data Coordinating Centre at the Montreal Neurological Institute (A.C. Evans, S. Das, C. Rogers, P. Kostopoulos, V. Fonov, I. Leppert) and NeuroDevNet Informatics Core for data management infrastructure. We acknowledge the principal investigators and members of the Pathways in ASD Study Team (www.asdpathways.ca) and Simons Simplex Collection (www.sfari.org) projects, under which certain clinical and diagnostic measures were collected. This work was sponsored by NeuroDevNet (www.neurodevnet.ca) and the Canadian Institutes of Health Research to KH. References American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders, 4th ed.,[120_TD$IF] text revision ed. Washington, DC: American Psychiatric Association.

60

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

Bailey, J. A., & Penhune, V. B. (2010). Rhythm synchronization performance and auditory working memory in early- and late-trained musicians. Experimental Brain Research, 204(1), 91–101. http://dx.doi.org/10.1007/s00221-010-2299-y. Chen, J. L., Penhune, V. B., & Zatorre, R. J. (2008). Moving on time: Brain network for auditory-motor synchronization is modulated by rhythm complexity and musical training. Journal of Cognitive Neuroscience, 20(2), 226–239. http://dx.doi.org/10.1162/jocn.2008.20018. Donnellan, A. M., Hill, D. A., & Leary, M. R. (2012). Rethinking autism: Implications of sensory and movement differences for understanding and support. Frontiers in Integrative Neuroscience, 6, 124. http://dx.doi.org/10.3389/fnint.2012.00124. Dragovic, M. (2004). Categorization and validation of handedness using latent class analysis. Acta Neuropsychiatry, 16(4), 212–218. http://dx.doi.org/10.1111/ j.0924-2708.2004.00087.x. Drake, C., Jones, M. R., & Baruch, C. (2000). The development of rhythmic attending in auditory sequences: Attunement, referent period, focal attending. Cognition, 77(3), 251–288. Erviti, M., Semal, C., Wright, B. A., Amestoy, A., Bouvard, M. P., & Demany, L. (2014). A late-emerging auditory deficit in autism. Neuropsychology, [12_TD$IF]29(3), 454– 462. http://dx.doi.org/10.1037/neu0000162. Foster, N. E., Ouimet, T., Tryfon, A., Doyle-Thomas, K., Anagnostou, E., & Hyde, K. L. (2016). Effects of age and attention on auditory global-local processing in children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 46(4), 1415–1428. http://dx.doi.org/10.1007/s10803-015-2684-2. Freitag, C. M., Kleser, C., Schneider, M., & von Gontard, A. (2007). Quantitative assessment of neuromotor function in adolescents with high functioning autism and Asperger Syndrome. Journal of Autism and Developmental Disorders, 37(5), 948–959. http://dx.doi.org/10.1007/s10803-006-0235-6. Gowen, E., & Hamilton, A. (2013). Motor abilities in autism: A review using a computational context. Journal of Autism and Developmental Disorders, 43(2), 323–344. http://dx.doi.org/10.1007/s10803-012-1574-0. Green, D., Charman, T., Pickles, A., Chandler, S., Loucas, T., Simonoff, E., et al. (2009). Impairment in movement skills of children with autistic spectrum disorders. Developmental Medicine and Child Neurology, 51(4), 311–316. http://dx.doi.org/10.1111/j.1469-8749.2008.03242.x. Hardy, M. W., & Lagasse, A. B. (2013). Rhythm, movement, and autism: Using rhythmic rehabilitation research as a model for autism. Frontiers in Integrative Neuroscience, 7, 19. http://dx.doi.org/10.3389/fnint.2013.00019. Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs. Frontiers in Psychology, 4, 863. http://dx.doi.org/10.3389/fpsyg.2013.00863. Landa, R., & Garrett-Mayer, E. (2006). Development in infants with autism spectrum disorders: A prospective study. Journal of Child Psychology and Psychiatry, 47(6), 629–638. http://dx.doi.org/10.1111/j.1469-7610.2006.01531.x. Lane, A., Harpster, K., & Heathcock, J. (2012). Motor characteristics of young children referred for possible autism spectrum disorder. Pediatric Physical Therapy, 24(1), 21–29. http://dx.doi.org/10.1097/PEP.0b013e31823e071a. Leekam, S. R., Nieto, C., Libby, S. J., Wing, L., & Gould, J. (2007). Describing the sensory abnormalities of children and adults with autism. Journal of Autism and Developmental Disorders, 37(5), 894–910. http://dx.doi.org/10.1007/s10803-006-0218-7. Leyfer, O. T., Folstein, S. E., Bacalman, S., Davis, N. O., Dinh, E., Morgan, J., . . . Lainhart, J. E. (2006). Comorbid psychiatric disorders in children with autism: Interview development and rates of disorders. Journal of Autism and Developmental Disorders, 36(7), 849–861. http://dx.doi.org/10.1007/s10803-0060123-0. Lin, I. F., Mochida, T., Asada, K., Ayaya, S., Kumagaya, S., & Kato, M. (2015). Atypical delayed auditory feedback effect and Lombard effect on speech production in high-functioning adults with autism spectrum disorder. Frontiers in Human Neuroscience, 9, 510. http://dx.doi.org/10.3389/fnhum.2015.00510. Liu, P., Chen, Z., Jones, J. A., Huang, D., & Liu, H. (2011). Auditory feedback control of vocal pitch during sustained vocalization: A cross-sectional study of adult aging. PLoS One, 6(7), e22791. http://dx.doi.org/10.1371/journal.pone.0022791. Liu, T. (2013). Sensory processing and motor skill performance in elementary school children with autism spectrum disorder. Perceptual and Motor Skills, 116 (1), 197–209. http://dx.doi.org/10.2466/10.25.PMS.116.1.197-209. Lloyd, M., MacDonald, M., & Lord, C. (2013). Motor skills of toddlers with autism spectrum disorders. Autism, 17(2), 133–146. http://dx.doi.org/10.1177/ 1362361311402230. 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. Journal of Autism and Developmental Disorders, 24(5), 659–685. Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Leventhal, B. L., DiLavore, P. C., . . . Rutter, M. (2000). The autism diagnostic observation schedule-generic: A standard measure of social and communication deficits associated with the spectrum of autism. Journal of Autism and Developmental Disorders, 30(3), 205–223. Mayer, J. L., Hannent, I., & Heaton, P. F. (2014). Mapping the developmental trajectory and correlates of enhanced pitch perception on speech processing in adults with ASD. Journal of Autism and Developmental Disorders, [123_TD$IF]46(5), 1562–1573. http://dx.doi.org/10.1007/s10803-014-2207-6. Moran, M. F., Foley, J. T., Parker, M. E., & Weiss, M. J. (2013). Two-legged hopping in autism spectrum disorders. Frontiers in Integrative Neuroscience, 7, 14. http://dx.doi.org/10.3389/fnint.2013.00014. Mottron, L., Dawson, M., Soulières, I., Hubert, B., & Burack, J. (2006). Enhanced perceptual functioning in autism: An update, and eight principles of autistic perception. Journal of Autism and Developmental Disorders, 36(1), 27–43. http://dx.doi.org/10.1007/s10803-005-0040-7. O'Connor, K. (2012). Auditory processing in autism spectrum disorder: A review. Neuroscience and Biobehavioral Reviews, 36(2), 836–854. http://dx.doi.org/ 10.1016/j.neubiorev.2011.11.008. Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia, 9(1), 97–113. Ouimet, T., Foster, N. E., Tryfon, A., & Hyde, K. L. (2012). Auditory-musical processing in autism spectrum disorders: A review of behavioral and brain imaging studies. Annals of the New York Academy of Sciences, 1252, 325–331. http://dx.doi.org/10.1111/j.1749-6632.2012.06453.x. Provasi, J., & Bobin-Begue, A. (2003). Spontaneous motor tempo and rhythmical synchronisation in 2-1/2 and 4-year-old children. International Journal of Behavioral Development, 27(3), 220–231. http://dx.doi.org/10.1080/01650250244000290. Provasi, J., Anderson, D. I., & Barbu-Roth, M. (2014). Rhythm perception, production, and synchronization during the perinatal period. Frontiers in Psychology, 5, 1048. http://dx.doi.org/10.3389/fpsyg.2014.01048. Provost, B., Lopez, B. R., & Heimerl, S. (2007). A comparison of motor delays in young children: Autism spectrum disorder, developmental delay, and developmental concerns. Journal of Autism and Developmental Disorders, 37(2), 321–328. http://dx.doi.org/10.1007/s10803-006-0170-6. Russo, N., Larson, C., & Kraus, N. (2008). Audio-vocal system regulation in children with autism spectrum disorders. Experimental Brain Research, 188(1), 111– 124. http://dx.doi.org/10.1007/s00221-008-1348-2. Shadmehr, R., Smith, M. A., & Krakauer, J. W. (2010). Error correction, sensory prediction, and adaptation in motor control. Annual Review of Neuroscience, 33, 89–108. http://dx.doi.org/10.1146/annurev-neuro-060909-153135. Srinivasan, S. M., Kaur, M., Park, I. K., Gifford, T. D., Marsh, K. L., & Bhat, A. N. (2015). The effects of rhythm and robotic interventions on the imitation/praxis, interpersonal synchrony, and motor performance of children with Autism Spectrum Disorder (ASD): A pilot randomized controlled trial. Autism Research and Treatment736516. http://dx.doi.org/10.1155/2015/736516. Travers, B. G., Bigler, E. D., Duffield, T. C., Prigge, M. D., Froehlich, A. L., Lange, N., . . . Lainhart, J. E. (2016). Longitudinal development of manual motor ability in autism spectrum disorder from childhood to mid-adulthood relates to adaptive daily living skills. Developmental Science. http://dx.doi.org/10.1111/ desc.12401 [124_TD$IF]9[Epub ahead of print]. Tseng, Y. W., Diedrichsen, J., Krakauer, J. W., Shadmehr, R., & Bastian, A. J. (2007). Sensory prediction errors drive cerebellum-dependent adaptation of reaching. Journal of Neurophysiology, 98(1), 54–62. http://dx.doi.org/10.1152/jn.00266.2007. Wan, C. Y., Bazen, L., Baars, R., Libenson, A., Zipse, L., Zuk, J., . . . Schlaug, G. (2011). Auditory-motor mapping training as an intervention to facilitate speech output in non-verbal children with autism: A proof of concept study. PLoS One, 6(9), e25505. http://dx.doi.org/10.1371/journal.pone.0025505. Wechsler, D. (1999). Wechsler abbreviated scale of intelligence. San Antonio, TX: The Psychological Corporation. Wechsler, D. (2011). Wechsler abbreviated scale of intelligence—Second edition (WASI-II). San Antonio, TX: The Psychological Corporation.

A. Tryfon et al. / Research in Autism Spectrum Disorders 35 (2017) 51–61

61

Zatorre, R. J., Chen, J. L., & Penhune, V. B. (2007). When the brain plays music: Auditory-motor interactions in music perception and production. Nature Reviews Neuroscience, 8(7), 547–558. http://dx.doi.org/10.1038/nrn2152. Zwaigenbaum, L., Scherer, S., Szatmari, P., Fombonne, E., Bryson, S. E., Hyde, K., . . . Volden, J. (2011). The NeuroDevNet autism spectrum disorders demonstration project. Seminars in Pediatric Neurology, 18(1), 40–48. http://dx.doi.org/10.1016/j.spen.2011.02.003.