SEARCH

SEARCH BY CITATION

Keywords:

  • Autism Spectrum Disorders;
  • Speech Perception;
  • Auditory Processing;
  • Aging

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

While temporal and perceptual processing abnormalities, identified in a number of electrophysiological and brain imaging studies of individuals with (ASD), are likely to impact on speech perception, surprisingly little is known about the behavioral outcomes of such abnormalities. It has been hypothesized that rapid temporal processing deficits may be linked to impaired language development through interference with acoustic information during speech perception. The present study aimed to investigate the impact of temporal changes on encoding and recall of speech, and the associated cognitive, clinical, and behavioral correlates in adults with ASD. Research carried out with typically developing (TD) adults has shown that word recall diminishes as the speed of speech increases, and it was predicted that the magnitude of this effect would be far greater in those with ASD because of a preexisting rapid temporal processing deficit. Nineteen high-functioning adults with ASD, and age- and intelligence-matched TD controls performed verbatim recall of temporally manipulated sentences. Reduced levels of word recall in response to increases in presentation speed were observed, and this effect was greater in the older participants in the ASD group than in the control group. This is the first study to show that both sensory abnormalities and aging impact on speech encoding in ASD. Auditory processing deficits in ASD may be indicative of an association with the sensory abnormalities and social and communication impairments characterizing the disorder. Autism Res 2014, 7: 40–49. © 2013 International Society for Autism Research, Wiley Periodicals, Inc.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Autism spectrum disorders (ASDs) are neurodevelopmental disorders characterized by impairments in social interaction, deficits in cognitive flexibility, and deficits in communication and language abilities. The Diagnostic and Statistical Manual of Mental Disorders 4th Ed. (DSM-IV) further defines the communication cluster in ASD, specifying that a significant delay in language or a clear deficit in the ability to carry on a conversation with others be present to elicit an ASD diagnosis [American Psychiatric Association, 2000]. While researchers have estimated that between 25% and 50% of individuals with ASD never acquire functional language [Gillberg & Coleman, 2000; Klinger, Dawson, & Renner, 2002], the presentation of language impairments within this population is extremely diverse, and although intellectually able, adults with ASD may experience difficulties carrying on a conversation; their scores on formal tests of verbal intelligence may well be in the normal range or higher.

While temporal and perceptual processing abnormalities, identified in a number of electrophysiological and brain imaging studies of individuals with ASD [Fujikawa-Brooks, Isenberg, Osann, Spence, & Gage, 2010; Gervais et al., 2004; Källstrand, Olsson, Nehlstedt, Skold, & Nielzen, 2010; Lepistö et al., 2009; Whitehouse & Bishop, 2008], are likely to impact on speech perception, surprisingly little is known about the behavioral outcomes of such abnormalities. However, the importance of addressing this question is highlighted by research showing that language processing difficulties limit the psychosocial and vocational opportunities of intellectually able adults with ASD [Howlin, Alcock, & Burkin, 2005]. Research aiming to elucidate interrelationships between information processing abnormalities and core communication deficits in individuals with ASD will make an important contribution to the theoretical and empirical base that informs the development of intervention services for these individuals.

Temporal aspects of auditory stimuli carry important information. For example, Rosen [1992] proposed that temporal cues are the primary component upon which speech perception is based, and studies have demonstrated that deficiencies in speech perception are often associated with deficits discriminating temporal auditory features [Kujala et al., 2000]. Furthermore, changes to the temporal parameters of speech affect other perceptual components of the speech signal; increasing the rate of natural speech also involves changes at the syllable, word, and sentence levels that disrupt the relationship between their timing and that of speech units [Janse, 2004]. Several studies have examined temporal perception of speech through rapid speed processing in typically developing (TD) individuals and have observed significant declines in rates of performance with increasing speech rate [Stine, Wingfield, & Poon, 1986; Tun, 1998; Tun, Wingfield, Stine, & Mecsas, 1992; Wingfield, 1975; Wingfield, Poon, Lombardi, & Lowe, 1985].

Impairments in temporal processing are often noted in individuals with ASD. A recent study by Kwakye, Foss-Feig, Cascio, Stone, and Wallace [2011] using temporal order judgment tasks provided evidence for impairments in both multisensory and auditory temporal processing in children with ASD. Their findings are consistent with other behavioral studies and electrophysiological findings of reduced mismatch negativity in response to duration changes in nonspeech sounds [Lepistö et al., 2005, 2006]. Taken together, these results indicate atypical responses to temporal aspects of auditory information in ASD. Kwakye et al. [2011] suggested that these impairments could be due in part to an extended temporal processing window in individuals with ASD, which affects the rapid processing of sensory information.

Gepner and Féron [2009] put forth a tempo-spatial processing hypothesis to explain various degrees of disability in ASD. Within the auditory domain, the authors noted evidence of impairments in speech flow perception and segmentation in children with ASD [Gepner & Massion, 2002] and increased phoneme categorization performance when phonemes were produced at reduced speeds [Tardif, Thomas, Gepner, & Rey, 2002]. Their hypothesis suggests that rapid changes in the environment, acting on one or more sensory modalities, are implicated in processing impairments in children and adolescents with ASD. However, the extent to which this hypothesis generalizes to adults with ASD has not been investigated. Rapid processing impairments in the auditory domain may well result in difficulties with verbal comprehension and impairments in language abilities [OramCardy, Flagg, Roberts, Brian, & Roberts, 2005]. This is particularly concerning as the ability to integrate temporal information is hypothesized to be vital in the development of those social functions that are impaired in individuals with ASD [Gepner & Tardif, 2006].

Speech processing involves the rapid decoding of a constantly changing signal that must occur in real time. Thus, it is not surprising that individuals who experience temporal processing difficulties overall would have more difficulty with rapid speech. Previous research [Stine et al., 1986; Tun, 1998; Tun et al., 1992; Wingfield et al., 1985] found that elderly adults demonstrated steeper declines in rates of performance with increasing speech rate in comparison with younger individuals. Speech rate is normally under the control of the speaker rather than the listener, and impairments in rapid speech processing could therefore have a direct impact on one's social communication abilities. Studies by Laine, Tardif, and Gepner [2008] and Laine, Rauzy, Gepner, and Tardif [2009] attempted to alleviate the effects of rapid processing impairments in children with ASD by slowing the auditory presentation of sentences. Verbal comprehension was enhanced during slow speech rates, especially in children with low-functioning autism. Thus, temporal manipulations to speech through increasing the rate of presentation may well uncover increased speech and information processing abnormalities in adults with ASD.

One reason we might expect temporal speech manipulations to result in increased processing abnormalities in adults with ASD is the presence of sensory atypicalities, which are known to be pervasive and continue into adulthood [Crane et al., 2009; Leekam, Nieto, Libby, Wing, & Gould, 2007]. Based on Gepner and Féron's [2009] hypothesis and Kwakye et al.'s [2011] findings, it is plausible that increased sensitivity to and awareness of sensory information may make it more difficult for adults with ASD to process rapid speech changes. The recent line of research into aging in ASD also raises interesting questions given the previously cited findings regarding aging and rapid speech processing in TD individuals. Neurologial studies have highlighted correlations between atypical cortical thickness and age in individuals with ASD that are not present in TD controls [Doyle-Thomas & Duerden, 2013; Wallace, Dankner, Kenworthy, Giedd, & Martin, 2010]. However, little research exists regarding the possible behavioral effects of age-related neurological atypicalities in adults with ASD.

The following experiment aims to investigate the impact of temporal changes on encoding and recall of speech in ASD and TD adults. Research carried out with TD adults has shown that word recall diminishes as the speed of speech increases, and it is predicted that the magnitude of this effect will be far greater in those with ASD because of a preexisting rapid temporal processing deficit. In addition to analyzing accuracy scores, the present study also incorporates the use of recall times in order to examine some of the more subtle processing differences. The present study also aims to explore the relationship between cognitive, behavioral, and clinical correlates (i.e. sensory processing abnormalities, autistic traits, chronological age, etc.) and memory and recall for rapidly presented sentences using regression analyses.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Participants

Nineteen adults (4 females/15 males) with high-functioning ASD (with intelligence quotient (IQ) scores of 70 or above) were recruited to the study. Their chronological ages ranged between 23 years 9 months and 59 years 8 months. While the oldest adult in the ASD group was 7 years older than the oldest adult in the TD group, he was not an outlier on any of the background assessments or experimental measures. As the results did not change when his data were removed, they were retained in the analyses. All of the adults in the ASD group were recruited from local support groups or had previously participated in research at Goldsmiths College, London, UK. All ASD participants' preexisting diagnoses were confirmed by the first author using Autism Diagnostic Observation Schedule (ADOS) module 4.

Nineteen adults with typical development were matched to the ASD group on age, gender, receptive vocabulary, working memory, as well as on verbal, performance, and full-scale IQ scores (see following section). Their chronological ages ranged between 25 years 1 month and 52 years 8 months. All control participants' were screened for ASD using the Adult Autism Spectrum Quotient [AQ; Baron-Cohen, Wheelwright, Skinner, Martin, & Clubley, 2001]. Scores on this test ranged from 3 to 21, which is well below the cutoff score of 32 proposed by Baron-Cohen et al. [2001]. See Table 1 for the background data summary for the two groups.

Table 1. Participant Background Data Summary
 ASD N = 19TD N = 19P-values
Mean (SD)RangeMean (SD)Range
  1. a

    Weschler Abbreviated Scales of Intelligence (WASI), standard score [Wechsler, 1999].

  2. a1WASI verbal IQ; a2WASI performance IQ.

  3. b

    Peabody Picture Vocabulary Test (PPVT), standard score [Dunn & Dunn, 1997].

  4. c

    Working Memory Digit Span (WM), Weschler Adult Intelligence Scales, [Wechsler, 2008].

  5. c1WM Forward Digit Span; c2WM Backward Digit Span.

  6. d

    Communication Checklist—Self Report (CC-SR), raw score [Bishop et al., 2009].

  7. d1CC-SR language structure; d2CC-SR pragmatics; d3CC-SR social engagement.

  8. e

    Adult/Adolescent Sensory Profile (SP), [Brown & Dunn, 2002].

  9. e1SP low registration; e2SP sensation seeking; e3SP sensory sensitivity; e4SP sensation avoiding.

  10. f

    Adult Autism Spectrum Quotient (AQ) [Baron-Cohen et al., 2001].

  11. f1AQ social skills; f2AQ attention switching; f3AQ attention to detail; f4AQ communication; f5AQ imagination.

  12. g

    Autism Diagnostic Observation Schedule (ADOS), diagnostic total [Lord et al., 2001].

  13. g1ADOS communication; g2ADOS reciprocal social interaction; g3ADOS imagination and creativity; g4ADOS repetitive behaviors.

  14. ASD, autism spectrum disorder; CA, chronological age; N/A, not applicable; SD, standard deviation; TD, typically developing.

CA (months)482.79 (136.00)285–716459.79 (108.64)301–6320.568
Cognitive correlates     
WASI full scalea113.37 (15.27)78–133118.95 (10.84)87–1340.203
WASI verbala1111.16 (15.57)71–132115.58 (11.52)83–1350.326
WASI performancea2112.95 (12.97)92–129118.05 (12.21)96–1360.221
PPVTb105.63 (12.07)76–123106.05 (10.24)84–1250.908
WM-totalc19.68 (4.57)13–3019.16 (4.69)13–280.728
WM-forwardc111.32 (2.43)7–1611.53 (2.32)8–150.786
WM-backwardc28.37 (2.54)4–147.63 (2.98)4–130.418
Behavioral correlates     
CC-SR-totald67.84 (33.28)32–15922.00 (13.81)1–50<0.001*
CC-lang. struct.d114.58 (12.19)1–495.00 (3.97)0–16<0.01*
CC-pragmaticsd217.84 (11.35)0–395.89 (6.71)0–25<0.001*
CC-social eng.d335.42 (12.58)19–7111.11 (5.65)1–24<0.001*
Sensory Profile-totale179.58 (26.09)130–218131.89 (28.36)32–160<0.001*
SP-low reg.e143.42 (10.41)27–6226.16 (6.23)10–35<0.001*
SP-sensation seek.e243.79 (8.29)31–6347.58 (9.88)12–580.209
SP-sensory sens.e347.16 (10.19)23–6229.05 (8.18)4–39<0.001*
SP-sensat. avoid.e445.21 (9.54)31–6129.11 (9.31)6–48<0.001*
Clinical correlates     
AQ-totalf35.16 (7.59)21–4512.26 (5.45)3–21<0.001*
AQ-social skillsf16.72 (2.58)3–101.32 (1.38)0–4<0.001*
AQ-atten. switchf28.67 (1.37)6–103.26 (1.79)0–6<0.001*
AQ-atten. to detailf37.22 (2.13)1–103.58 (2.10)0–7<0.001*
AQ-commun.f46.50 (2.55)2–101.95 (1.39)0–5<0.001*
AQ-imaginationf56.22 (2.29)2–102.16 (1.98)0–7<0.001*
ADOS-diagnosticg9.58 (3.55)5–17N/AN/AN/A
ADOS-commun.g12.84 (1.54)1–6N/AN/AN/A
ADOS-soc. int.g26.74 (2.70)3–12N/AN/AN/A
ADOS-imag.g31.05 (0.70)0–2N/AN/AN/A
ADOS-rep. behav.g41.58 (1.02)0–3N/AN/AN/A

Cognitive Correlates

Weschler Abbreviated Scales of Intelligence

The Weschler Abbreviated Scales of Intelligence [Wechsler, 1999] was used as a measure of intellectual and cognitive functioning.

Peabody Picture Vocabulary

The Peabody Picture Vocabulary Test (PPVT) [Dunn & Dunn, 1997] was used as a measure receptive vocabulary.

Working memory

The ability to encode information into working memory plays an important role in the speed and accuracy with which one can rapidly process information [Stine et al., 1986; Tun, 1998]. In order to assess participants' working memory capacity, the backwards digit span subtest from the Weschler Adult Intelligence Scale—Fourth Edition [Wechsler, 2008] was used.

Behavioral Correlates

Communication Checklist—Self-Report

The Communication Checklist—Self-Report (CC-SR) [Bishop, Whitehouse, & Sharp, 2009] was administered to provide information on any difficulties in speech, language, or interaction that may affect the participants' communication abilities. The CC-SR is a 70-item self-report questionnaire that examines three factors of communication: language structure (“I make false starts or search for the right word.”), pragmatic skills (“I am told that I keep talking about things that others are not interested in.”), and social engagement (“I find it hard to know when people are upset or annoyed.”). Higher scores on the CC-SR indicate an increased level of communication difficulties. The ASD group scored higher than the TD group on all of the CC-SR measures, demonstrating a significantly greater level of self-reported communication difficulties.

Adult/Adolescent Sensory Profile

While sensory abnormalities are not currently included in DSM-IV Text Revision [DSM-IV-TR; American Psychiatric Association, 2000], they are widely prevalent ASD [e.g. Leekam et al., 2007] and may be implicated in language processing difficulties in ASD. Therefore, measures of sensory abnormalities using the Adult/Adolescent Sensory Profile (SP) test [Brown & Dunn, 2002] were also obtained. The SP is a 60-item self-report questionnaire that examines sensory processing patterns across six sensory processing categories including: taste/smell, movement, visual, touch, activity, and auditory processing. Participants' raw scores across the six categories are used to derive their quadrant scores identified as: low registration (“I don't get jokes as quickly as others.”), sensation seeking (“I like to wear colorful clothing.”), sensory sensitivity (“I am distracted if there is a lot of noise around.”), and sensation avoiding (“I stay away from crowds.”). Higher scores within each quadrant represent increased sensory abnormalities. The ASD group had higher scores than the TD group on three of the four quadrants, indicating a greater level of sensory abnormalities.

Clinical Correlates

AQ

In order to assess the self-reported levels of autistic traits in participants, the AQ [Baron-Cohen et al., 2001] was administered. The AQ is a 50-item questionnaire that examines five factors: social skills (“I would rather go to a library than a party.”), attention switching (“I frequently get so absorbed in one thing that I lose sight of other things.”), attention to detail (“I often notice small sounds when others do not.”), communication (“Other people frequently tell me that what I've said is impolite, even though I think it is polite.”), and imagination (“When I'm reading a story, I find it difficult to work out the characters' intentions.”). Within the AQ, autistic-like behavior is characterized by poor social, communication, or imagination skills, exceptional attention to detail, and either poor attention switching or a strong focus of attention [Baron-Cohen et al., 2001]. The ASD group scored higher than the TD group on all of the AQ measures, which demonstrates a significantly greater level of self-reported autistic traits in the participants with ASD.

Autism Diagnostic Observation Schedule

All of the ASD individuals who participated in the present study had previously been diagnosed by clinicians in accordance with Diagnostic and Statistical Manual of Mental Disorders [4th ed., rev.; American Psychiatric Association, 2000]. ASD participants' preexisting diagnoses were confirmed by administering the ADOS module 4 [Lord, Rutter, DiLavore, & Risi, 2001]. The ADOS provides a score representing autistic symptom severity in the areas of: communication, reciprocal social interaction, imagination and creativity, and repetitive behaviors. Of the 19 ASD participants recruited, two did not meet overall diagnostic criteria on the ADOS. However, as all participants had previously been diagnosed by a clinician and the results from the background assessments and the experimental task did not change if those individuals were excluded, they were retained in the final sample.

Experimental Methods

Experimental stimuli

Sentence stimuli consisted of 30, 15-word sentences randomly selected from the 60 sentences used by Tun et al. [1992] (e.g. “The setting of Greece and its ancient monuments make it a fascinating place to visit.”). The sentences were recorded by an adult British English-speaking female and manipulated using PRAAT [Boersma, 2001] to generate three different speed conditions: normal speech (140 words per minute (wpm)), moderate speed (200 wpm), and fast speed (280 wpm). Normal speech acted as the baseline condition and was only manipulated by adjusting the original sentences to the mean intensity (perceived volume) and a median pitch of 200 Hz. The moderate and fast speed conditions were generated using electronic time compression to reduce the normal speech sentences to 70% and 50% of their original length.

Procedure

Participants were administered three practice sentences, one under each condition, and asked to perform a verbatim recall immediately following the end of the recorded sentence. Participants were instructed to repeat as much of the sentence as they could in the order that they had heard it. Thirty experimental sentences followed in the same format. During the experimental trials, participants' responses were timed and recorded for later analysis. Participants received one point for each correct word. No points were awarded for words that were either incorrect or in the wrong order. Raw scores for each condition were converted to percentages for the analysis.

Analysis

Discrepancy scores were generated for each participant in order to account for any individual differences in working memory, language comprehension, or speech rate that may have affected their performance. Participants' percentage correct scores on the perceptual manipulation conditions were subtracted from their scores on the normal speech (baseline) condition in order to calculate their individual levels of change in response to speed manipulations.

A factorial analysis of variance was used to analyze the data with the within-subjects factor of speed condition (two levels; moderate speed and fast speed) and between-subjects factor of group (two levels; ASD and TD). The dependent variable was the discrepancy scores for each participant across the ten sentences in each speed condition. Further correlation and regression analyses were conducted to explore the relationship between cognitive, behavioral, and clinical correlates, and memory and recall for rapidly presented sentences. Because of the exploratory nature of the regression analyses and the lack of previous literature on which to base a priori predictions, a backwards stepwise entry method was employed.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Accuracy and Recall Time Analyses

Means (Ms), standard deviations (SDs), and ranges for the accuracy percentage correct scores and discrepancy scores as well as recall time raw scores and discrepancy scores across speed manipulations are shown in Table 2.

Table 2. Mean Percentage Correct and Discrepancy Scores, Standard Deviations, and Ranges
  ASDTD
Mean (SD)RangeMean (SD)Range
  1. Note. Negative accuracy discrepancy scores signify better performance on perceptual manipulation in comparison to baseline; negative recall time discrepancy scores indicate higher reaction times on perceptual manipulation in comparison to baseline.

  2. ASD, autism spectrum disorder; RT, reaction time; SD, standard deviation; TD, typically developing.

Accuracy percentage correct scoresNormal81.72 (15.77)40.67–98.6780.38 (12.26)51.33–95.33
Moderate80.63 (13.86)50.00–97.3382.42 (12.62)40.67–93.33
Fast77.16 (17.32)34.67–96.0078.07 (12.06)50.67–90.67
Total79.84 (14.94)44.22–97.3380.29 (11.45)48.89–92.44
Accuracy discrepancy scoresModerate1.09 (9.12)−16.67–18.00−2.03 (7.23)−18.67–10.67
Fast2.56 (8.14)−12.67–23.332.31 (8.56)−12.00–26.67
Recall time scoresNormal77.35 (21.48)51.50–142.7075.52 (68.60)47.10–111.10
Moderate76.00 (25.30)48.20–152.0068.60 (10.55)53.90–91.90
Fast68.46 (21.41)45.00–130.6071.43 (10.87)56.80–91.90
Total221.81 (62.26)151.40–353.40215.55 (32.23)166.00–274.80
RT discrepancy scoresModerate1.34 (17.92)−54.50–35.506.92 (13.86)−20.70–36.70
Fast8.89 (14.59)−23.30–50.704.09 (10.64)−14.40–21.80

There was a highly significant main effect of speed manipulation on participants' accuracy during sentence recall, F(1, 38) = 9.29, P < 0.01. Participants' performance indicated a significantly higher level of difficulty when encoding and recalling speech spoken at a fast rate of speed in comparison with moderate speed (M = −0.47, SD = 8.27 for moderate; M = 3.44, SD = 8.32 for fast speech). However, there was no significant main effect of speed manipulation on participants' sentence recall speed, F(1, 38) = 1.08, P = 0.305 (M = 4.14, SD = 28.01 for moderate; M = 6.49, SD = 23.08 for fast speed).

In order to examine whether participants experienced significantly more difficulty in the two conditions with speed manipulations than normal speech alone, a one-sample t-test were conducted. A mean value of 0, which would indicate identical performance when recalling perceptually manipulated speech and normal speech, was used. Accuracy and reaction time results revealed a significant difference between discrepancy scores on the fast speed condition (t(37) = 2.55, P < 0.05 for accuracy; t(37) = 3.12, P < 0.01 for recall time) and 0, but not on the moderate speed condition (t(37) = −0.35, P = 0.726 for accuracy; t(37) = 1.59, P = 0.121 for recall time) and 0. Thus, the results showed that individuals experienced significantly more difficulty with both accuracy and recall time during the fast condition in comparison with normal speech, but were equally able to encode and recall moderately fast and normal speeds of speech.

Although ASD individuals experienced slightly more difficulty with accuracy when encoding and recalling sentences with speed manipulations in comparison with TD participants (M = 2.82, SD = 7.62 for ASD, and M = 0.14, SD = 6.93 for TD) this was not statistically significant, F(1, 38) = 1.29, P = 0.264, and there was no significant speed manipulation by group interaction, F(1, 38) = 0.12, P = 0.735. There was also no significant main effect of group on the participants' sentence recall speed, F(1, 38) = 0.01, P = 0.925. However, results indicated that both groups were recalling the sentences faster during the speed manipulation conditions in comparison with their baseline recall speeds. (M = 5.12, SD = 14.13 for ASD, and M = 5.50, SD = 11.07 for TD).

There was, however, a significant speed manipulation by group interaction in the recall time analysis, F(1, 38) = 5.25, P < 0.05 (Fig. 1). In order to further examine the interaction, two post hoc t-tests were conducted. Results revealed a nonsignificant trend toward ASD participants recalling sentences more slowly during the moderate speed condition in comparison with the fast speech condition, t(18) = −2.00, P = 0.060. TD participants, on the other hand, tended to recall sentences slower during the fast speech condition, although not significantly so, t(18) = 1.12, P = 0.274.The second group of post hoc t-tests examined performance within each condition across groups. No significant differences were found (t(18) = 1.159, P = 0.254 for fast speed; t(18) = −1.073, P = 0.290 for moderate speed).

figure

Figure 1. Group × speed manipulation interaction. Note. Negative recall times indicate increased perceptual disturbance from speed.

Download figure to PowerPoint

Regression Analyses

While difficulties encoding and recalling temporally manipulated speech were not observed at the group level in the ASD sample, scores for this group did appear to be more variable. As an important aim of the study was to explore the extent that variations in performance on the speed manipulation paradigm were associated with specific clinical, cognitive, and behavioral factors (described in the methods section), correlation analyses were first carried out on the data. Both groups experienced more difficulty encoding and recalling fast but not moderate speech; therefore, the correlation and regression analyses focused only on the fast speech discrepancy scores. Significant correlations are shown in Table 3. All of the variables that were significantly correlated with performance on the experimental task were used in regression analyses. It is notable that whilst there were numerous correlations between the background measures and task performance in the ASD group, no such relationship was found in typically developing individuals, thus the regression analyses focused solely on the ASD group.

Table 3. Summary of Significant Correlations between Fast Speed Discrepancy Scores and Background Measures
 ASDTD
  1. Note. Negative correlations indicate a relationship between higher scores on the background measure and reduced perceptual disturbance from speed.

  2. *P < 0.05; **P < 0.01; ***P < 0.001 (two tailed).

  3. AQ, Adult Autism Spectrum Quotient; ASD, autism spectrum disorder; PIQ, performance IQ; TD, typically developing; WASI, Weschler Abbreviated Scales of Intelligence.

Cognitive correlates  
WASI  
PIQ−0.513*−0.030
Behavioral correlates  
Communication checklist  
Language structure0.654**0.197
Pragmatic skills0.481*0.273
Social engagement0.717**0.132
Total score0.675**0.243
Sensory profile  
Low registration0.654**0.125
Sensory sensitivity0.538*0.385
Total score0.544*0.248
Clinical correlates  
AQ  
Social skills0.480*−0.150
Imagination0.580*−0.007
Chronological age0.739***−0.149

A multiple linear regression was performed to examine the extent to which the significantly correlated cognitive, behavioral, and clinical factors shown in Table 3 explained the variance in encoding and recall of fast speech in the ASD group. The results revealed a significant linear relationship between ASD participants' accuracy discrepancy scores during the fast speed condition and the predictor variables with a multiple correlation of 0.96, (F(1,19) = 34.12, P < 0.001; adjusted R2 = 0.886) (Table 4). Thus, 88% of the variability in ASD participants' accuracy discrepancy scores during the fast speed condition was predicted by their chronological age and scores on the sensory sensitivity subscale of the SP. Higher chronological age and levels of sensory sensitivity predicted an increase in an ASD individual's discrepancy scores, indicating increased perceptual disturbance when encoding and recalling speech spoken at a fast speed.

Table 4. Multiple Regression Statistics for Predictor Variables to Accuracy Discrepancy Score of ASD Participants During Fast Speed Condition
 BSE BβtP
  1. **P < 0.01; ***P < 0.001.

  2. AQ, Adult Autism Spectrum Quotient; β, standardized beta coefficient; B, unstandardized beta coefficient; CC, Communication Checklist; SE B, standard error; SP, Adult/Adolescent Sensory Profile; t, t-test statistic.

Age0.040.010.627.090.000***
CC-social engagement0.190.090.282.130.053
SP-sensory sensitivity0.400.120.473.230.007**
AQ-social skills−0.700.38−0.22−0.180.092

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

One of the primary aims of the present study was to increase our understanding of the effect of temporal manipulations on speech encoding and recall in individuals with ASD. Although no overall group differences emerged within either the accuracy or recall time analyses, this is not necessarily surprising given that both groups possessed higher than average levels of intelligence and good verbal skills. This result does not appear to reflect insensitivity in the paradigm as there were significant differences between accuracy but not recall time discrepancy scores on the moderate and fast speed manipulations when the analysis was carried out on the data from the whole sample. Thus, individuals within both groups recalled significantly fewer correct words when the speed of speech was twice as fast as normal, but not when sentences were produced at a moderately fast speed. Furthermore, accuracy results indicated that individuals were experiencing significant levels of perceptual disturbance from the fast temporal manipulation in comparison with normal speech. Thus, individuals across both groups experienced more difficulty encoding and recalling sentences when the speed of speech increased. Recall time analyses also revealed a significant interaction between temporal manipulations and group. Trends suggested that ASD participants recalled sentences slower during the moderate speech condition in comparison with the fast speech condition, while TD individuals tended to recall sentences slower during the fast speech condition. It is unclear whether this result is a consequence of different auditory processing strategies in the two groups or indicative of subtle temporal processing abnormalities in individuals with ASD. Future studies utilizing electrophysiological methodologies may be able to address this question.

Another aim of the present study was to explore the cognitive, behavioral, and clinical correlates associated with auditory temporal processing in TD and ASD individuals. Individuals with ASD and typical controls demonstrated very different profiles on the regressions and correlations carried out on the experimental and background measures. First, regressions revealed a significant positive relationship between age and increased perceptual disturbance from fast speech in the ASD group but not the control group, despite the fact that the two groups were matched on mean age and range. Although previous research [Stine et al., 1986; Tun, 1998; Tun et al., 1992; Wingfield et al., 1985] found that elderly typical adults demonstrated steeper declines in rates of performance with increasing speech rate in comparison with younger individuals, their elderly cohort extended far beyond the age range of the participants tested in the present study. It is therefore unsurprising that the TD participants who completed the present experiment did not show the declines reported in these studies. The results showing an age-related decline in the ASD group may suggest that individuals with this disorder are more susceptible to age-related processing effects, such as cognitive slowing, than TD individuals. Furthermore, it is possible that such age-related effects are occuring at a younger age in adults with ASD compared to typical development. In light of Wallace et al.'s [2010] and Doyle-Thomas and Duerden's [2013] findings of a relationship between abnormal cortical thickness and age in individuals with ASD, the present study also provides evidence to suggest there may be behavioral effects of atypical neurological aging in this disorder.

Correlations also showed that individuals with ASD who reported higher levels of difficulty across all subscales of the communication checklist also experienced significantly higher levels of perceptual disturbance from speech spoken at fast speed. While a wealth of empirical data confirm that information processing abnormalities are a frequent, if not universal feature in ASD, they are not included in DSM-IV and are rarely considered in the context of core social and communication deficits ASD. The results from the current study indicate that there is a relationship between language and temporal processing abnormalities in individuals with ASD that extends to those on the very high-functioning end of the spectrum. A similar relationship was also found between sensory processing abnormalities, autistic traits relating to social skills and imagination, and accurate recall of sentences spoken at a fast speed. This finding was further supported by the regression analyses, which indicated that higher levels of self-reported sensory sensitivity in individuals with ASD predicted a significant increase in perceptual disturbance when encoding and recalling fast speech. These results suggest that auditory processing deficits in ASD may not just be a function of language impairment, but rather are indicative of an association with the sensory abnormalities and social and communication impairments characterizing the disorder.

The relationship between performance on the experimental task and performance IQ suggests that the absence of clear group differences in the present study may very well be due to fact that only high-functioning individuals were included in the study. The results from the study suggested that while sensory/temporal processing difficulties did not impact on speech encoding and memory in ASD during early to mid-adulthood, the decline in this skill, also observed in typical development, occurs at significantly earlier periods in ASD. Future research should consolidate and extend this finding in studies of high-function children and older adults with ASD. Questions about the extent that speech encoding and memory is disrupted in lower-functioning individuals with ASD also merit investigation.

One of the primary aims of the present study was to increase understanding of heterogeneity in language skills in high-functioning adults with ASD by identifying the cognitive, clinical, and behavioral correlates of deficits in speech perception. In order to address this question, a number of background measures were utilized, and on many of these variables (most notably the CC-SR and some of the subscales from the AQ) there was a much larger variance in the scores from the ASD group than the TD group. This is unsurprising given that a more heterogeneous profile, in terms of communication and autistic traits, would be expected within the ASD population. However, it is possible that this factor contributed to the absence of any significant correlations between performance on the experimental task and the background measures in the TD group. Therefore, the small correlations observed should be interpreted with caution and replication, especially focusing on sensory processing abnormalities and age should be sought.

Another limitation of the present study is the linguistic matching criteria of the sentence stimuli. The specific stimuli was selected because it had previously been used by Tun et al. [1992] to examine the age effects of rapid speech processing in TD individuals and thus provided the best opportunity for replication. While the stimuli used by Tun et al. were matched for numbers of words and provided a high cognitive load, other linguistic factors that can affect speech encoding and recall such as number of syllables and word frequency were not controlled. However, the present study randomized the sentences in each speech condition across participants in order to account for any variability within the individual sentences. Future research should attempt to take a more rigorous linguistic approach by taking additional matching factors into account.

While it may be argued that temporal manipulations of sentence stimuli are artificial and the results for the current study may not generalize to real-life situations, research has indicated that natural fast speech is actually more difficult to process than artificially time-compressed speech. As Janse [2004] suggests, naturally fast speech is characterized by more general prosodic change and segmental overlap than slower speech. It is therefore plausible to suggest that high-functioning individuals with ASD may experience more difficulties processing fast speech in their everyday lives than the current results suggest. Our results showed that speech processing deficits are most marked in older individuals and those with increased levels of self-reported sensory disturbance. Given the likely negative impact of such difficulties on the vocational and psychosocial opportunities of affected individuals, further study into temporal speech processing throughout the lifespan in high-functioning individuals with ASD is clearly merited.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This research was carried out with the support of the Baily Thomas Trust. We are very grateful to all the ASD and TD adults who participated in our study. We would also like to thank Mr. Ian Hannett for his assistance creating the stimuli.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association.
  • Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley, E. (2001). The Autism-Spectrum Quotient (AQ): Evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. Journal of Autism and Developmental Disorders, 31, 517.
  • Bishop, D.V.M., Whitehouse, A.J.O., & Sharp, M. (2009). Communication checklist—self-report. San Antonio, TX: The Psychological Corporation.
  • Boersma, P. (2001). PRAAT, a system for doing phonetics by computer. Glot International.
  • Brown, C., & Dunn, W. (2002). Adult/adolescent sensory profile. San Antonio, TX: The Psychological Corporation.
  • Crane, L., Goddard, L., & Pring, L. (2009). Sensory processing in adults with autism spectrum disorders. Autism: the international journal of research and practice, 13, 215228. doi: 10.1177/1362361309103794.
  • Doyle-Thomas, K., & Duerden, E. (2013). Effects of age and symptomatology on cortical thickness in autism spectrum disorders. Research in Autism Spectrum Disorders, 7, 141150.
  • Dunn, L.M., & Dunn, M. (1997). Peabody picture vocabulary test. Circle Pines, MN: American Guidance Service.
  • Fujikawa-Brooks, S., Isenberg, A.L., Osann, K., Spence, M.A., & Gage, N.M. (2010). The effect of rate stress on the auditory brainstem response in autism: A preliminary report. International Journal of Audiology, 49, 129140. doi:10.3109/14992020903289790
  • Gepner, B., & Féron, F. (2009). Autism: A world changing too fast for a mis-wired brain? Neuroscience and Biobehavioral Reviews, 33, 12271242. doi:10.1016/j.neubiorev.2009.06.006
  • Gepner, B., & Massion, J. (dir by), 2002. L'autisme: Unepathologie du codagetemporel ? Travaux Interdisciplinaires du Laboratoire Parole et Langage d'Aix-en-Provence (TIPA), 21, 177218. IN: Gepner, B., & Féron, F. (2009). Autism: a world changing too fast for a mis-wired brain? Neuroscience and biobehavioral reviews, 33, 1227–1242. doi:10.1016/j.neubiorev.2009.06.006
  • Gepner, B., & Tardif, C. (2006). Autism, movement, time and thought. E-motion mis- sight and other temporo-spatial processing disorders in autism. In M. Vanchevsky (Ed.), Frontiers in cognitive psychology (pp. 130). New York: Nova Science Publishers.
  • Gervais, H., Belin, P., Boddaert, N., Leboyer, M., Coez, A., et al. (2004). Abnormal cortical voice processing in autism. Nature, 7, 801803.
  • Gillberg, C., & Coleman, M. (2000). The biology of the autistic syndromes. London: John Wiley & Sons.
  • Howlin, P., Alcock, J., & Burkin, C. (2005). An 8 year follow-up of a specialist supported employment service for high-ability adults with autism or Asperger syndrome. Autism, 9, 533549.doi:10.1177/1362361305057871
  • Janse, E. (2004). Word perception in fast speech: Artificially time-compressed vs. naturally produced fast speech. Speech Communication, 42, 155173. doi:10.1016/j.specom.2003.07.001
  • Källstrand, J., Olsson, O., Nehlstedt, S.F., Skold, M.L., & Nielzen, S. (2010). Abnormal audi- tory forward masking pattern in the brainstem response of individuals with Asperger syndrome. Neuropsychiatry Disorders and Treatment, 6, 289296.
  • Klinger, L., Dawson, G., & Renner, P. (2002). Autistic disorder. In E. Mash & R. Barkley (Eds.), Child psychopathology (pp. 409454). New York: Guilford Press.
  • Kujala, T., Myllyviita, K., Tervaniemi, M., Alho, K., Kallio, J., & Näätänen, R. (2000). Basic auditory dysfunction in dyslexia as demonstrated by brain activity measurements. Psychophysiology, 37, 262266.
    Direct Link:
  • Kwakye, L.D., Foss-Feig, J.H., Cascio, C.J., Stone, W.L., & Wallace, M.T. (2011). Altered auditory and multisensory temporal processing in autism spectrum disorders. Frontiers in Integrative Neuroscience, 4, (January), 111. doi:10.3389/fnint.2010.00129
  • Laine, F., Rauzy, S., Gepner, B., & Tardif, C. (2009). Prise en compte des difficulte'sdetraitement des informationsvisuellesetauditivesrapidesdans le cadre de l'e'valuationdiagnostique de l'autisme. Enfance, 1, 133142. IN: Gepner, B., & Féron, F. (2009). Autism: A world changing too fast for a mis-wired brain? Neuroscience and biobehavioral reviews, 33, 1227–1242.doi:10.1016/j.neubiorev.2009.06.006.
  • Laine, F., Tardif, C., & Gepner, B. (2008). Slowing down biological motion and speech sounds enhances imitation and verbal comprehension in children and adolescents with autism. Poster at the Autism Neuroscience Conference, Royal Society, London, October.
  • 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, 894910. doi:10.1007/s10803-006-0218-7
  • Lepistö, T., Kuitunen, A., Sussman, E., Saalasti, S., Jansson-Verkasalo, E., et al. (2009). Auditory stream segregation in children with Asperger syndrome. Biological Psychology, 82, 301307. doi:10.1016/j.biopsycho.2009.09.004
  • Lepistö, T., Kujala, T., Vanhala, R., Alku, P., Huotilainen, M., & Näätänen, R. (2005). The discrimination of and orienting to speech and non-speech sounds in children with autism. Brain Research, 1066, 147157. Elsevier. doi:10.1016/j.brainres.2005.10.052
  • Lepistö, T., Silokallio, S., Nieminen-von Wendt, T., Alku, P., Näätänen, R., & Kujala, T. (2006). Auditory perception and attention as reflected by the brain event-related potentials in children with Asperger syndrome. Clinical Neurophysiology, 117, 21612171. doi:10.1016/j.clinph.2006.06.709
  • Lord, C., Rutter, M., DiLavore, P.C., & Risi, S. (2001). Autism diagnostic observation schedule—generic. Los Angeles, CA: Western Psychological Services.
  • Oram Cardy, J.E., Flagg, E.J., Roberts, W., Brian, J., & Roberts, T.P.L. (2005). Magnetoencephalography identifies rapid temporal processing deficit in autism and language impairment. Neuroreport, 16, 329332.
  • Rosen, S. (1992). Temporal information in speech: Acoustic, auditory and linguistic aspects. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 336, 367373. doi:10.1098/rstb.1992.0070
  • Stine, E.L., Wingfield, A., & Poon, L.W. (1986). How much and how fast: Rapid processing of spoken language in later adulthood. Psychology and Aging, 1, 303311.
  • Tardif, C., Thomas, K., Gepner, B., & Rey, V. (2002). Contribution a' l'e'valuation du syste'mephonologiqueexplicite chez des enfantsautistes. Parole, 21, 3572. IN: Gepner, B., & Féron, F. (2009). Autism: A world changing too fast for a mis-wired brain? Neuroscience and Biobehavioral Reviews, 33, 1227–1242. doi:10.1016/j.neubiorev.2009.06.006
  • Tun, P., Wingfield, A., Stine, E.L., & Mecsas, C. (1992). Rapid speech processing and divided attention: Processing rate versus processing resources as an explanation of age effects. Psychology and Aging, 7, 546550.
  • Tun, P.A. (1998). Fast noisy speech: Age differences in processing rapid speech with background noise. Psychology and Aging, 13, 424434.
  • Wallace, G.L., Dankner, N., Kenworthy, L., Giedd, J.N., & Martin, A. (2010). Age-related temporal and parietal cortical thinning in autism spectrum disorders. Brain, 133(Pt 12), 37453754. doi:10.1093/brain/awq279.
  • Wechsler, D. (1999). Wechsler abbreviated scale of intelligence. San Antonio, TX: The Psychological Corporation.
  • Wechsler, D. (2008). Wechsler adult intelligence scale—fourth edition. San Antonio, TX: Pearson.
  • Whitehouse, A.J., & Bishop, D.V. (2008). Do children with autism “switch off” to speech sounds? An investigation using event-related potentials. Developmental Science, 11, 516524.
  • Wingfield, A. (1975). Acoustic redundancy and the perception of time-compressed speech. Journal of Speech & Hearing Research, 18, 96104.
  • Wingfield, A., Poon, L.W., Lombardi, L., & Lowe, D. (1985). Speed of processing in normal aging: Effects of speech rate, linguistic structure, and processing time. Journal of Gerontology, 40, 579585.