Study - Increased Auditory Capacity in Autistics

Post Reply New Topic

Highlights
•
Autistic individuals have an increased auditory perceptual capacity.
•
This higher capacity led to enhanced task performance on a selective attention task.
•
The same capacity led to increased distraction on an inattentional deafness task.
•
This offers an explanation for abilities and difficulties seen in autistic perception.
•
Reframing autistic attention in this manner suggests novel intervention approaches.


Autism Spectrum Disorder (ASD) has an intriguing auditory processing profile. Individuals show enhanced pitch discrimination, yet often find seemingly innocuous sounds distressing. This study used two behavioural experiments to examine whether an increased capacity for processing sounds in ASD could underlie both the difficulties and enhanced abilities found in the auditory domain. Autistic and non-autistic young adults performed a set of auditory detection and identification tasks designed to tax processing capacity and establish the extent of perceptual capacity in each population. Tasks were constructed to highlight both the benefits and disadvantages of increased capacity. Autistic people were better at detecting additional unexpected and expected sounds (increased distraction and superior performance respectively). This suggests that they have increased auditory perceptual capacity relative to non-autistic people. This increased capacity may offer an explanation for the auditory superiorities seen in autism (e.g. heightened pitch detection). Somewhat counter-intuitively, this same ‘skill’ could result in the sensory overload that is often reported – which subsequently can interfere with social communication. Reframing autistic perceptual processing in terms of increased capacity, rather than a filtering deficit or inability to maintain focus, increases our understanding of this complex condition, and has important practical implications that could be used to develop intervention programs to minimise the distress that is often seen in response to sensory stimuli.

Twenty autistic adults and 20 neurotypical adults (aged 17–34 years) were recruited through social networking websites and autism support groups around London. Sample size was determined by previous research using similar paradigms (Remington, Campbell, & Swettenham, 2012). Participants in the ASD group had received a clinical diagnosis of autism from a trained, independent clinician who used the criteria listed in the Diagnostic and Statistical Manual of Mental Disorders, Fourth or Fifth Edition (American Psychiatric Association, 1994 and American Psychiatric Association, 2013) and reached threshold for an ASD on Module 4 of the Autism Diagnostic Observational Schedule (ADOS; Lord, Rutter, DiLavore, & Risi, 2002). Three participants with a clinical diagnosis of ASD did not meet the ADOS criteria and were therefore excluded. The remaining group of autistic participants showed a mean ADOS score of 9.8 (SD = 2.0). None of the participants reported having any other mental or neurological disorder.

In order to improve group matching, one neurotypical individual was excluded due to an extremely high IQ (greater than 2 S.D. above the mean) (see Table 1 for resulting participant groups). The resulting 17 ASD (13 males) and 19 neurotypical adults (11 males) did not differ in IQ, as measured by the Wechsler Abbreviated Scale for Intelligence – Second Edition (Wechsler, 2011) (mean ASD IQ: 110, SD = 13.0; mean neurotypical IQ: 114.5, SD = 10.0; p = 0.26). The autism group was significantly older than the neurotypical group (mean ASD group age: 30 years, SD = 3.6; mean neurotypical group age: 23.6 years, SD = 5.0; p < 0.001). All participants had their audiometric thresholds measured prior to taking part in the study, following the procedure recommended by the British Society of Audiology (2004). Audiometric air-conduction thresholds were measured for the left and right ears for octave-spaced frequencies from 250 to 8000 Hz using a Kamplex Diagnostic Audiometer AD17 and Telephonics TDH39P headphones. All participants had normal hearing, defined as audiometric thresholds equal to or better than 15 dB HL for all frequencies between 250 and 8000 Hz in both ears. Participants took part in both Experiments in the same testing session, and the task order was counterbalanced.

6 Discussion
The results reported here provide evidence, from two separate attention tasks, that auditory perceptual capacity is increased in autistic adults. In the first, a dual-task paradigm with different levels of perceptual load, participants were asked to identify an auditory target and also perform a secondary auditory detection task. Both the autistic and neurotypical groups performed the primary target identification task to the same level (no significant group differences in RT or error rates). However, whereas neurotypical performance on the detection task dropped as load increased, autistic participants remained able to detect the critical stimulus under high levels of auditory load. This is reflected in significantly better performance by the autistic participants on the secondary task at the highest level of load. As such, these results suggest that the autistic adults have increased auditory capacity that, in this particular case, is an asset: capacity is available to perform both the primary and secondary aspects of the task.

In the second experiment, the same participants’ auditory capacity was tested on a more ecologically valid task that involved a binaural recording of an auditory scene. In this selective attention task, participants were asked to pay attention to a conversation between two women and answer a subsequent question about what they had said. In the middle of the scene an unexpected and unusual stimulus was presented: a man walking through the scene repeatedly saying “I’m a gorilla”. Immediately after listening to the auditory scene, participants were asked whether they had heard this unusual event. As in Experiment 1, both groups were able to successfully complete the primary task (answering the question correctly about the women’s conversation). With respect to the ‘gorilla’, 88% of neurotypical participants failed to notice the unexpected stimulus (replicating results of inattentional deafness in the mainstream population (Dalton & Fraenkel, 2012). Conversely, 47% of the autistic participants were aware of the ‘gorilla’, suggesting that they had the capacity to attend to the additional character as well as the central scene conversation. In this case, the increased capacity manifests as increased susceptibility to distraction, though not at the expense of task-performance (the ‘gorilla’ was task-irrelevant but did not conflict with the target response).

Our demonstration that auditory perceptual capacity appears to be enhanced in autism may allow a reinterpretation of some of the previous literature on auditory processing. Past research has reported a diverse set of atypicalities with respect to autistic audition, assessing aspects that range from basic physical properties (such as pitch) to more complex components (such as prosody) (see 8 for a review). Here, when considering auditory perceptual capacity, it is most relevant to focus on low-level auditory processing. It has been noted that a much higher proportion of autistic individuals show absolute (or ‘perfect’) pitch (5%, compared to 0.01–0.05% in the general population, (Rimland & Fein, 1988). Similarly, autistic children were more accurate at identifying and remembering musical notes (e.g. Heaton, 2003) and pitch discrimination (e.g. Bonnel et al., 2003, Heaton, 2005, Jarvinen-Pasley et al., 2008 and O'Riordan and Passetti, 2006). In adolescents and adults on the autism spectrum, these advantages were primarily noted for those individuals who also displayed language difficulties (Bonnel et al., 2010, Heaton et al., 2008 and Jones et al., 2009). In addition,

In many cases, however, this increased ability appears to be accompanied by a feeling of over-arousal (an overwhelming level of sensory input) or hyperacusis (where seemingly innocuous sounds are perceived as distressing) (Gomes et al., 2008). In the few studies examining loudness perception (despite many clinical observations), autistic children showed lower loudness discomfort thresholds (Khalfa et al., 2004) and were more likely to show discomfort to sounds below 80 dB(HL) compared to neurotypical individuals (Rosenhall, Nordin, Sandstrom, Ahlsen, & Gillberg, 1999). However, on tests of volume discrimination, autistic and non-autistic performance was equivalent (Bonnel et al., 2010). Closely related to this work, other research suggests autistic people may be more susceptible to distraction from auditory stimuli (Teder-Salejarvi, Pierce, Courchesne, & Hillyard, 2005) and appear to have a wider auditory filter than non-autistic individuals (Plaisted, Saksida, Alcantara, & Weisblatt, 2003). Consequently, there is mixed evidence regarding the ability of autistic individuals to extract speech from background noise, with some suggesting a reduced ability (Alcantara, Weisblatt, Moore, & Bolton, 2004) based on difficulties with stream segregation (Lepisto et al., 2009), while others highlight increased stream segregation abilities (Lin, Yamada, Komine, Kato, & Kashino, 2015). This issue is of particular interest when considering any links between auditory processing and social abilities.

In light of our current findings, we suggest that increased auditory capacity might, in part, underlie the superior performance observed on tasks of auditory processing. If autistic individuals were able to process more information at any given time, performance would be enhanced on tasks that require participants to memorize and discriminate pitch and melodies. Conversely, this same increase in capacity could be detrimental to task performance in other situations: giving rise to additional auditory processing that results in distractibility, over-arousal and hyperacusis. Indeed, our cross-experiment analyses suggested that, irrespective of diagnosis, performance on the two tasks were related: those who noticed the auditory ‘gorilla’ also showed greater ability to detect the car stimulus under high levels of load. As such, increased auditory capacity could offer an explanation for the mixed picture of superiorities and difficulties seen with respect to auditory processing in autism.

At this point, it is also important to consider the potential neural mechanisms that may underlie the increased auditory capacity. Electrophysiological studies (using various oddball paradigms) have identified shorter N1 latencies to simple tones (Ferri et al., 2003 and Oades et al., 1988), greater mismatch negativities (MMN) to pitch change and shortened MMN latencies to pitch changes (Gomot, Giard, Adrien, Barthelemy, & Bruneau, 2002) in children with autism, suggestive of more efficient early auditory processing components in autism. However, findings in this area have been mixed, with others showing the opposite pattern in response to more complex tones (see Haesen, Boets, & Wagemans, 2011, for discussion). On a neuroanatomical level, it has been shown that grey matter thickness was greater in the Heschl’s gyri of autistic individuals (compared to neurotypical controls), a region in the primary auditory cortex which is the first to process incoming auditory stimuli (Hyde, Samson, Evans, & Mottron, 2010). This perhaps reflects an increased availability of low-level auditory processing resources. Interestingly, this increased cortical thickness was also seen in other neural regions, including the visual and parietal cortices. This is therefore in line with our previous research showing increased visual perceptual capacity and superior performance on visual attention tasks in autism (Remington et al., 2012).

The importance of understanding of the mechanisms underlying autistic sensory processing should not be underestimated, given that it often causes a great deal of distress for those with the condition. In addition, difficulties with these basic perceptual processes risk disrupting numerous other areas of functioning as they may also reduce access to learning and employment opportunities.

Interestingly, research that has examined links between auditory discrimination ability and self-report sensory symptoms identified some associations between increased performance on auditory duration discrimination tasks and an increased level of reported sensory symptoms (Jones et al., 2009). However, in the same study intensity discrimination ability seemed to show the opposite relationship: with lower levels of sensory symptoms seen in those who were able to better discriminate between different auditory intensities.

While it is clear that further research is warranted, our findings reframe the altered behaviors in terms of increased capacity, rather than a filtering deficit or inability to maintain focus. This reinterpretation fits well with anecdotal reports from autistic people who describe their ears being “like microphones”, picking up all the surrounding sounds indiscriminately (Grandin, 1996). Our results may also offer suggestions to ameliorate auditory difficulties. To reduce the impact of unwanted distraction in autism that results from increased capacity, we need to both reduce background noise but also increase the level of perceptual load in a given task (to exhaust more of the processing capacity with task-relevant information). This somewhat counterintuitive prediction is at odds with the common view that tasks and stimuli should be simplified for autistic children in schools. It is also a prediction that needs to be carefully applied in order to avoid over-arousal. As such, we view the present findings as the starting point for an interesting line of subsequent experimental and applied research.