Elsevier

Cortex

Volume 134, January 2021, Pages 195-206
Cortex

Special Issue "Multisensory integration": Research Report
Reduced multisensory facilitation exists at different periods of development in autism

https://doi.org/10.1016/j.cortex.2020.09.031Get rights and content

Abstract

Atypical sensory processing is now recognised as a key component of an autism diagnosis. The integration of multiple sensory inputs (multisensory integration (MSI)) is thought to be idiosyncratic in autistic individuals and may have cascading effects on the development of higher-level skills such as social communication.

Multisensory facilitation was assessed using a target detection paradigm in 45 autistic and 111 neurotypical individuals, matched on age and IQ. Target stimuli were: auditory (A; 3500 Hz tone), visual (V; white disk ‘flash’) or audiovisual (AV; simultaneous tone and flash), and were presented on a dark background in a randomized order with varying stimulus onset delays. Reaction time (RT) was recorded via button press. In order to assess possible developmental effects, participants were divided into younger (age 14 or younger) and older (age 15 and older) groups.

Redundancy gain (RG) was significantly greater in neurotypical, compared to autistic individuals. No significant effect of age or interaction was found. Race model analysis was used to compute a bound value that represented the facilitation effect provided by MSI. Our results revealed that MSI facilitation occurred (violation of the race model) in neurotypical individuals, with more efficient MSI in older participants. In both the younger and older autistic groups, we found reduced MSI facilitation (no or limited violation of the race model).

Autistic participants showed reduced multisensory facilitation compared to neurotypical participants in a simple target detection task, void of social context. This remained consistent across age. Our results support evidence that autistic individuals may not integrate low-level, non-social information in a typical fashion, adding to the growing discussion around the influential effect that basic perceptual atypicalities may have on the development of higher-level, core aspects of autism.

Introduction

Our perceptual system is constantly receiving information from multiple sensory inputs. Thus, in order to obtain a unified and coherent representation of the external world, we must efficiently integrate multiple sources of information simultaneously, a process referred to as multisensory integration (MSI; Stein, 2012). In addition to other sensory differences (see Robertson & Baron-Cohen, 2017), atypical MSI has, increasingly, been found to be a defining characteristic of autism (see Zhou et al., 2018; Feldman et al., 2018). Autism1 (or Autism Spectrum Disorder) is a neurodevelopmental condition characterized by difficulties in social interaction/communication, repetitive behaviours and restricted interests. Atypical sensory processing is now a key feature of an autism diagnosis (APA, 2013), with reports estimating that sensory issues are prevalent in 69–95% of the autistic population (Hazen et al., 2014).

In the neurotypical population, MSI has been found to have an enhancing, or ‘facilitation’ effect on basic perceptual functions. For example, research has indicated superior performance on both lower- (target detection) and higher-level (speech perception) MSI tasks, where multiple sources of information are presented (i.e., audio and visual) compared to where just one source (audio or visual) is presented (e.g., Molholm et al., 2004, 2002; Ross et al., 2011). These results suggest that atypical multisensory processing will have consequences on the development of certain skills and behaviours that characterize autism (see Iarocci & McDonald, 2006). It has been theorized that altered multisensory processing is common in autistic individuals, and may contribute to the unique sensory profile in this population (Stevenson et al., 2018).

Recurrent demonstrations of altered MSI in autistic individuals have been published in recent years (for reviews see Feldman et al., 2018; Zhou et al., 2018; Stevenson et al., 2018) and, for the most part, the literature has been driven by findings from studies focussed on the integration of social information; assessing performance based on auditory (voice), visual (face) or audiovisual speech (voice and face) perception. This social-information approach is exemplified by demonstrations of atypical MSI performance in autism using the McGurk illusion (Woynaroski et al., 2013; Williams et al., 2004; Taylor et al., 2010; Stevenson et al., 2014a; Stevenson et al., 2014b; Saalasti et al., 2011; Saalasti et al., 2012; Foxe, et al., 2015; DePape, et al., 2012; Stevenson et al., 2018). Additionally, research has explored basic speech perception e.g., syllable recognition ((de Boer-Schellekens, et al., 2013); Ross, et al., 2011), word recognition (Smith and Bennetto, 2007) and sentence comprehension (Grossman et al., 2015). This focus on MSI for socially-relevant information is perhaps unsurprising, due to many of autism's behavioural characteristics being related to atypical social communication and interaction. However, in order to ascertain if MSI is a basic perceptual difference in autism, and not only related to social differences, MSI must be assessed in autism using stimuli and tasks void of social content.

Research assessing “non-social” MSI can generally be divided into two domains; sensitivity to temporal processing and susceptibility to visual illusions. For temporal processing, the time window within which auditory and visual stimuli appear simultaneous (Temporal Binding Window; TBW) has been demonstrated to be wider in autistic individuals compared to neurotypical individuals (Foss-Feig et al., 2010; Kwakye et al., 2011; Woynaroski et al., 2013 and see Wallace & Stevenson, 2014 for discussion). Similarly, sensitivity to the order in which audio and visual stimuli appear (Temporal Order Judgements; TOJ) has been found to be less pronounced in autistic individuals (de Boer-Schellekens, Eussen, & Vroomen, 2013; Kwakye et al., 2011).

Visual Illusions used to assess MSI in autism have included the Sound Induced Flash illusion (SIFI; Shams et al., 2002) and the Pip-Pop illusion (Van der Burg et al., 2008). These paradigms have been well-established in the neurotypical population, eliciting either strong illusory percepts requiring MSI (i.e., SIFI), or resulting in strong facilitatory effects on performance for multi- vs. uni-sensory conditions. For example, the SIFI illusion occurs when two visual flashes are presented simultaneously with one auditory beep, resulting in the illusory percept of one flash. Autistic children have been found to be less susceptible to the SIFI illusion as compared to neurotypical children (Stevenson et al., 2014c), suggestive of less efficient MSI. However, Foss-Feig et al. (2010) provide evidence for greater susceptibility to the illusion in autistic children compared to neurotypical children. In adolescents and adults, Van der Smagt et al. (2007) and Bao et al. (2017) found that autistic participants were just as susceptible to the SIFI illusion as neurotypical participants, suggesting that mixed findings using this task may depend on the age of the participant groups assessed, with results becoming more consistent age. During the Pip-Pop illusion task (Van der Burg et al., 2008), the presence of an auditory tone (or pip) facilitates the visual search of a target (or pop) when the auditory tone is presented at the same time as a change in colour of the target. This task facilitation is not observed in autistic adults, as response time to visual target detection does not improve with the addition of an auditory component (Collignon et al., 2013).

One of the simplest demonstrations of the facilitation effect of MSI is the redundant stimulus effect (RSE; Miller, 1982). This effect has been assessed experimentally via the Race Model Inequality (RMI) test for many years (Miller, 1982). Using this simple paradigm, MSI has been assessed in several clinical populations such as schizophrenia (Williams, et al., 2010; Wynn et al., 2014), dyslexia (Harrar et al., 2014) and Parkinson's (Fearon, et al., 2015). In the field of autism, two studies have used the RMI test. Ostrolenk et al. (2019) found reduced gain from the redundant stimulus effect in autistic adolescents and adults compared to neurotypical adolescents/adults. Brandwein et al. (2013) assessed RT MSI in autistic children 7 to 16 years old using the Race Model Analysis. They found no significant violation of the race model in their 2 autistic subgroups, while both neurotypical groups showed violations. Furthermore, they found that age had an effect on maximum race-model violation in the neurotypical group but not the autistic group. This result suggests that MSI in autistic individuals may not improve significantly in this age range. No research using the RMI in a sample of autistic children, teenagers, and adults currently exists, which is problematic given that autism is present from infancy and is a lifelong condition. Bridging these two studies by including participants from childhood to adulthood could paint a better picture of how MSI develops in autism.”

Although temporal and illusion paradigms provide informative evidence about the processing of MSI in autistic individuals, the complexity of such tasks may (i) confound conclusions that different MSI may be a core perceptual difference and (ii) render some of those on the autistic spectrum ineligible to participate, biasing participant samples (see Feldman et al., 2018 for discussion). Hence, non-complex, non-social MSI tasks provide an ideal platform upon which to assess basic MSI performance in autistic individuals. In addition to this, previous research has indicated that there may be a developmental difference between neurotypical and autistic individuals in MSI processing. For example, in a speech MSI task, Foxe et al. (2015) found that atypical MSI was found in school-aged autistic children (5–12-year-olds), but performance was equivalent to the neurotypical comparison group in older autistic children (13–15-year-olds). It is possible that performance on this type of social MSI task could change across age because of the development of social skills/interventions, hence, it is imperative that we understand if this developmental difference is found at a basic, non-social level of MSI. Equivalent non-social TOJs have been reported in autistic vs. neurotypical adults (de Boer-Schellekens, Keetels, et al., 2013; Poole et al., 2017) however, as far as we are aware, non-social MSI has yet to be tracked across age from young childhood to adulthood in autistic individuals.

The current study aimed to assess whether there is a reaction time advantage when simple visual and auditory stimuli were presented simultaneously vs. when they were presented alone, by using the sensitive Race Model Inequality test. Specifically, we aimed to explore (i) differences between autistic and neurotypical individuals and (ii) the influence of age group.

Section snippets

Participants

Our participant sample consisted of 156 individuals, 45 of which were diagnosed with ASD and 111 neurotypical individuals. The age of the participants ranged from 6-29 years old. Mean age in the autistic group was 15.4 (4.8) and mean age of the neurotypical group was 15.4 (5.8). We matched participants on age (t(154) = .02, p = .98) and performance IQ (t(144) = .23, p = .82). Prior to analysis we divided our groups by age (≤14 years; ≥15 years) -characteristic information can be found in Table 1

Per diagnostic group

We measured redundancy gain (RG) for each participant, a calculation of the gain in RT from the addition of a second modality. Overall, the autistic group had a mean RG of 12% (SD 9%) and the typical group had a mean RG of 17% (SD 1%). The range in the autistic group was an overall minimum gain of −11% (meaning unimodal was 11% faster than bimodal) and maximum gain of 30%. In the typical group the range was minimum 0% (where unimodal and bimodal had equal RTs) and maximum 44%.

Per age group

In order to assess

Discussion

This study aimed to assess the redundant stimulus effect (RSE) in autistic compared to neurotypical individuals, across younger and older ages. It is the largest study to assess RSE in autistic individuals using the race model and provides evidence for reduced multisensory facilitation in autism for simple, non-social information -a finding that was consistent across younger (≤14 years) and older (≥15 years) age groups.

Race model analyses indicated that neurotypical participants had

Funding

This study was funded by the Fonds de recherche du Quebec Santé (FRQS) Research Scholar Award -Junior 2 (# 33133) awarded to Dr. Bertone and the Quebec Autism Research Training (QART) grant (# 251605) awarded to Dr. Ainsworth.

Declaration of copyright

The work described in this manuscript has not been published previously and is not under consideration for publication elsewhere. This publication is approved by all authors and if accepted, it will not be published elsewhere in the same form (in English or in any other language), including electronically, without the written consent of the copyright holder.

Declaration of competing interest

The authors of this paper disclose no conflict of interest including any conflict of interest arising from any financial or personal means.

Acknowledgments

We would like to thank all participants and their families, who took part in our study. Their time and commitment made this research possible. We would also like to thank all the volunteers who helped on the project, with particular thanks to Samantha Scholes for her assistance with recruitment and data collection.

References (71)

  • S. Molholm et al.

    Multisensory auditory–visual interactions during early sensory processing in humans: A high-density electrical mapping study

    Cancer Biotherapy & Radiopharmaceuticals

    (2002)
  • J.P. Owen et al.

    Abnormal white matter microstructure in children with sensory processing disorders

    Neuroimage: clinical

    (2013)
  • E. Pellicano et al.

    When the world becomes ‘too real’: A bayesian explanation of autistic perception

    Trends in cognitive sciences

    (2012)
  • L. Shams et al.

    Visual illusion induced by sound

    Cancer Biotherapy & Radiopharmaceuticals

    (2002)
  • C. Thomas et al.

    The anatomy of the callosal and visual-association pathways in high-functioning autism: A DTI tractography study

    Cortex; a Journal Devoted To the Study of the Nervous System and Behavior

    (2011)
  • M.T. Wallace et al.

    The construct of the multisensory temporal binding window and its dysregulation in developmental disabilities

    Neuropsychologia

    (2014)
  • S. Wass

    Distortions and disconnections: Disrupted brain connectivity in autism

    Brain and cognition

    (2011)
  • L.E. Williams et al.

    Reduced multisensory integration in patients with schizophrenia on a target detection task

    Neuropsychologia

    (2010)
  • J.H. Williams et al.

    Visual–auditory integration during speech imitation in autism

    Research in Developmental Disabilities

    (2004)
  • H.Y. Zhou et al.

    Multisensory temporal binding window in autism spectrum disorders and schizophrenia spectrum disorders: A systematic review and meta-analysis

    Neuroscience and Biobehavioral Reviews

    (2018)
  • Diagnostic and statistical manual of mental disorders

    (2013)
  • V.A. Bao et al.

    Multisensory integration of low-level information in autism spectrum disorder: Measuring susceptibility to the flash-beep illusion

    Journal of Autism and Developmental Disorders

    (2017)
  • M.K. Belmonte et al.

    Autism and abnormal development of brain connectivity

    Journal of Neuroscience

    (2004)
  • A.B. Brandwein et al.

    The development of multisensory integration in high-functioning autism: High-density electrical mapping and psychophysical measures reveal impairments in the processing of audiovisual inputs

    Cerebral Cortex

    (2013)
  • A.B. Brandwein et al.

    The development of audiovisual multisensory integration across childhood and early adolescence: A high-density electrical mapping study

    Cerebral Cortex

    (2011)
  • S.M. Bury et al.

    “It defines who I Am” or “it's something I have”: What language do [autistic] Australian adults [on the autism spectrum] prefer?

    Journal of Autism and Developmental Disorders

    (2020)
  • V. Courchesne et al.

    Assessing intelligence at autism diagnosis: Mission impossible? Testability and cognitive profile of autistic preschoolers

    Journal of autism and developmental disorders

    (2019)
  • L. de Boer-Schellekens et al.

    Diminished sensitivity of audiovisual temporal order in autism spectrum disorder

    Frontiers in integrative neuroscience

    (2013)
  • A.M.R. DePape et al.

    Auditory processing in high-functioning adolescents with autism spectrum disorder

    Plos One

    (2012)
  • W. Dunn

    Sensory profile

    (1999)
  • C. Fearon et al.

    Audiovisual processing is abnormal in Parkinson's disease and correlates with freezing of gait and disease duration

    Journal of Parkinson's disease

    (2015)
  • J.H. Foss-Feig et al.

    An extended multisensory temporal binding window in autism spectrum disorders

    Experimental Brain Research

    (2010)
  • J. Foxe et al.

    Ten years at the multisensory forum: Musings on the evolution of a field

    Brain Topography

    (2009)
  • J. Foxe et al.

    Severe multisensory speech integration deficits in high-functioning school-aged children with autism spectrum disorder (ASD) and their resolution during early adolescence

    Cerebral Cortex

    (2015)
  • S. Girard et al.

    Impact of the spatial congruence of redundant targets on within-modal and cross-modal integration

    Experimental Brain Research

    (2013)
  • Cited by (10)

    • Wiring of higher-order cortical areas: Spatiotemporal development of cortical hierarchy

      2021, Seminars in Cell and Developmental Biology
      Citation Excerpt :

      Similarly, a transient increase in spine density specifically affecting association areas, notably the temporal cortex, is also observed in ASD patients, but normalize in late adulthood suggesting a delayed pruning [230–233]. Again, MSE in ASD patients presents a transient impairment, associated with a wider timescale of integration, which normalizes progressively to then become shorter in adulthood [220,234]. At the molecular level, increasing evidence support a global decreased of PV expression in ASD patients, notably in the PFC, while no such association have been found for SST interneurons [235].

    • Eating behavior in autism: senses as a window towards food acceptance

      2021, Current Opinion in Food Science
      Citation Excerpt :

      Both aspects must therefore be taken into consideration to adapt the intervention. Sensory particularities attested by neuroimaging studies in both humans and animals can be considered as fundamental and primary characteristics of the neurobiology of autism [63,64••]. These neurobiological specificities would even be at the origin of the social problems encountered by people with an ASD.

    View all citing articles on Scopus
    View full text