Skip to main content
NCBI home page
UKPMC Funders Author Manuscripts logoLink to UKPMC Funders Author Manuscripts
. Author manuscript; available in PMC: 2023 Apr 26.
Published in final edited form as: Autism. 2021 Jun 16;25(8):2209–2222. doi: 10.1177/13623613211014993

The effect of ambient sounds on decision-making and heart rate variability in autism

Rachel Bellamy 1, Howard Ring 1,2, Peter Watson 1, Andrew Kemp 3, Giles Munn 4, Isabel CH Clare 1,2,5
PMCID: PMC7614480  EMSID: EMS173941  PMID: 34132124

Abstract

Autistic people report difficulties with the demands of a neurotypical world, but little research has assessed the impact of the environment on such difficulties. We investigated the effect of ambient sounds on decision-making and heart rate variability. Adults without intellectual disability were allocated to autistic (n=38) or neurotypical (n=37) groups matched on age, intellectual functioning and self-reported gender. Participants completed a shopping decision-making task in three randomly ordered sound conditions: no sound, non-social shopping sound (e.g. fridges humming) and social shopping sound (e.g. people talking). Decision-making latency, decision-making consistency, and heart rate variability were measured. Participants also provided qualitative reports of their experiences. Qualitative analyses indicated that autistic participants experienced (1) the non-social and social sound conditions more negatively than the no sound condition and (2) the social sound condition more negatively than neurotypical participants. However, there were no statistically significant differences in decision-making latency, decision-making consistency, or heart rate variability across sound conditions and participant groups. Further research should consider alternative quantitative measures to explore subjective experience to help understand further which aspects of the environment autistic people are sensitive to, in turn, enabling more evidence-based autism-friendly changes to be made.

Lay abstract

Many autistic people report difficulties making decisions during everyday tasks, such as shopping. To examine the effect of sounds on decision-making, we developed a supermarket task where people watched a film shown from the shopper’s perspective and were asked to make decisions between different products. The task was divided into three sections and participants completed each section in a different auditory environment: (1) no sounds, (2) non-social sounds (e.g. fridges humming) and (3) social sounds (e.g. people talking). Thirty-eight autistic and 37 neurotypical adults took part. We measured decision-making by examining how long it took to make a decision and how consistent people were with their decisions. We also measured heart rate variability because this biological response provides a measure of anxiety. After the supermarket shopping task, participants told us in their own words about their experiences. Autistic participants said that they found the non-social and social sound conditions more difficult than the no sound condition, and autistic participants found the social sound condition more negative than neurotypical participants. However, decision-making and heart rate variability were similar for autistic and neurotypical participants across the sound conditions, suggesting that these measures may not have been sensitive enough to reflect the experiences the autistic participants reported.

Further research should consider alternative measures to explore the experiences reported by autistic people to help us understand which specific aspects of the environment autistic people are sensitive to. This, in turn, may enable more specific and evidence-based autism-friendly changes to be made.

Introduction

Difficulties in Shopping Environments

The ‘neurotypical (NT) world’ is challenging for autistic1 people (Crane, Adams, Harper, Welch & Pellicano, 2019; Krieger et al., 2018) and shops are particularly difficult: about two-thirds of autistic individuals (n=not provided) across the UK report they avoid shopping (National Autistic Society [NAS], 2018). There are likely to be many reasons, including those reflecting the social difficulties of autism (American Psychiatric Association, 2013).

Shopping-related challenges may be exacerbated by the sensory sensitivities, including sound sensitivity (Lucker, 2013; Stiegler & Davis, 2010), and anxiety often experienced by autistic people. Self-reports suggest unpredictable and uncontrollable sounds, competing sounds from multiple inputs and sounds that are loud, distracting or are very high or low frequency (Smith & Sharp, 2013) are particularly aversive. Autism is also associated with high rates of comorbid anxiety (Croen et al., 2015; Hollocks et al., 2019; Lai et al., 2019; Lugo-Marín et al., 2019), and a greater proportion of autistic than NT adults are prescribed anxiolytics (Vella et al., 2018). Self-reports and longitudinal analyses indicate sensory stimuli, specifically sound stimuli, can lead to anxiety in autism (Green, Ben-Sasson, Soto, & Carter, 2012; Landon, Shepherd & Lodhia, 2016).

Outcomes in Everyday Life: Decision-Making

Autistic individuals report that everyday decisions are problematic (Lawson, 2001; Sainsbury, 2000). Luke, Clare, Ring, Redley & Watson (2012) investigated the self-reported decision-making difficulties of 38 autistic individuals without intellectual disability compared to 40 NT individuals. Autistic individuals reported more frequent decision-making difficulties and greater avoidance of decision-making than NT individuals. Three kinds of decisions appeared particularly challenging: those that had to be made quickly, involved a change of routine and/or required talking to other people. Gaeth, Levin, Jain, & Burke (2016) found autistic adults rated everyday decisions as more difficult than NT adults.

Autism-Friendly Initiatives

The findings of focus groups involving 1,213 autistic people, their families, and clinicians have emphasised the importance of research-based evidence to promote environments that improve experiences for autistic people (Cusack & Sterry, 2016). The NAS organises a week each year during which businesses are supported in making their services more autism-friendly, such as by turning off music and dimming lights (NAS, 2019). However, there seems to be no experimental evidence to support these initiatives.

The Present Study

We used an experimental paradigm that modelled supermarket shopping to investigate how the auditory environment may impact two areas of difficulty in autism: decision-making and anxiety. Decision-making was measured using the attraction effect (AE) paradigm (Huber, Payne & Puto, 1982). The paradigm elicits a context effect, where the addition of an alternative option in a choice influences decision-making (Mourali et al., 2007) and contrasts with ‘rational’ decision-making (Luce, 1959, 1977; Tversky, 1972) (Supplementary Material 1). Two aspects of decision-making are measured: latency and consistency. Farmer, Baron-Cohen & Skylark (2017) found autistic adults had longer response latencies than NT adults when making AE choices. However, compared with NT adults, autistic people made significantly more consistent choices, suggesting they demonstrated more rational decision-making.

Interactions between anxiety and decision-making are well-known in the literature (eg. Remmers & Zander, 2018; Santos-Ruiz et al., 2012), and so is the high prevalence of anxiety in autism (Croen et al., 2015; Hollocks et al., 2019; Lai et al., 2019; Lugo-Marín et al., 2019). To better understand the role of anxiety in our shopping paradigm, heart rate variability (HRV) was measured to index general adaptive capacity to anxiety (Klusek, Roberts & Losh, 2015; Mertens, Zane, Neumeyer & Grossman, 2017). While HRV reduction in response to a stressor typically reflects an adaptive response, resting-state reductions reflect chronic arousal and autonomic inflexibility (Quintana, Alvares & Heathers, 2016). Autism has been associated with autonomic atypicalities (Benevides & Lane, 2015; Klusek et al., 2015), but the literature is contradictory. Some studies that compare HRV in autistic adults and NT adults report no differences in resting HRV (Dijkhuis, Ziermans, van Rijn, Staal & Swaab, 2019; Smeekens, Didden & Verhoeven, 2015) or HRV in response to a stimulus (Kuiper, Verhoeven & Geurts, 2017; Smeekens et al., 2015). However, others report reduced resting HRV in autistic adults (Cai, Richdale, Dissanayake & Uljarevic, 2019; Kuiper et al., 2017; Mathewson et al., 2011; Thapa et al., 2019) and reduced HRV reactivity in response to a stimulus (Dijkhuis et al., 2019), compared to NT adults. Some studies have assessed the effects of auditory stimuli on HRV in autistic individuals. For example, respiratory sinus arrhythmia (RSA), an index of HRV, was measured in autistic and NT children aged 6-9 years while they listened to different auditory tones (Schaaf, Benevides, Leiby & Sendecki, 2015); no differences in RSA were found. In contrast, Porges et al. (2013), found RSA to increase in autistic individuals (aged 6 to 21) during an auditory processing task. The effects of ambient environmental conditions on HRV in autistic adults are yet to be investigated.

We expected there would be differences between autistic and NT groups in the effect of different kinds of sounds on (i) decision-making latency; (ii) decision-making consistency; and (iii) HRV. Based on the greater sensory sensitivity of autistic individuals compared with NT adults (Crane et al., 2019), and the everyday decision-making difficulties (Gaeth et al., 2016; Luke et al., 2012) and anxiety in response to sensory stimulation (Landon et al., 2016) of autistic people, we also expected an interaction between these variables and the group (autistic/NT) to which participants belonged.

Methods

Ethics

The study was reviewed by XXXX (identifiable information removed).

Participants

Participants were at least 18 years old, were native English speakers, had not taken part in an AE pilot (Supplementary Material 2) and possessed corrected-to-normal vision and hearing. Participants had tested IQ scores ≥90 (Wechsler Abbreviated Scale of Intelligence, 2nd Edition [WASI-II]; Wechsler, 2011), with 20% of the assessments scored by two independent raters. Full Scale IQ was used to indicate IQ except for six autistic and six NT participants where Verbal Comprehension Index (VCI) was used because VCI and Perceptual Reasoning Index differed significantly.

Autistic participants (i) self-reported an autism diagnosis and (ii) scored above the cut-off for autism spectrum disorder on the Autism Diagnostic Observation Schedule-Second Edition (ADOS-2; Hus & Lord, 2014), Module 4. NT participants (i) self-reported they did not have an autism diagnosis and (ii) scored below 32 on the Autism-Spectrum Quotient (AQ; Baron-Cohen et al., 2001).

Power was estimated using Farmer et al. (2017). To detect a clinically meaningful difference of 10 seconds in decision-making latency between autistic and NT groups whilst listening to social sounds, 30 participants were needed in each group (α=0.05) allowing detection of a Cohen’s d of 0.88 (two-tailed) with 90% power (Altman, 1991).

Advertisements were distributed through the Cambridge Autism Research Centre, National Autistic Society, local student and community contacts, and autism support services.

All participants provided informed written consent for participation.

Materials

Shopping Decision-Making Task

The shopping task was based on that of Braeutigam, Stins, Rose, Swithenby & Ambler (2001; also see Ambler, Braeutigam, Stins, Rose & Swithenby, 2004). It comprised a film from a shopper’s perspective (Supplementary Material 3) and AE choices. The task was completed in no sound, non-social sound and social sound conditions.

The video and sound clips were recorded in a supermarket and further sounds were retrieved from sound banks. Piloting was completed to identify the choices and sounds to use (Supplementary Material 2). The task was programmed with Python.

The shopping task contained 54 AE choices across nine products (six choices per product). The task was split into three sections, each occurring in a different sound condition and comprising 18 choices, across three products. The order the products appeared was randomised. Each of the sound conditions began with the participant watching a video of a shopper entering a supermarket and approaching a product. The participant then made three pairs of choices about that product, before watching the shopper approach a second product and making three further pairs of choices. The participant then watched a video and made three pairs of choices about a third product, before seeing the shopper leave the supermarket. This process occurred three times, once in each sound condition.

Each choice was presented on a separate page and referred to one product (Supplementary Material 4). A picture of the product was given and remained the same for all six choices for that product. For each choice, there were three options: a target, competitor and decoy. Each option was described in terms of two attributes (Supplementary Material 4). The three options were presented in a triangle and the location of the options was randomised (Paulhus & Vazire, 2007). Participants could not change their choice once made.

Choices were grouped into pairs. Each pair of choices involved the same attributes, attribute descriptors and non-decoy options, but the decoy targeted a different option in each choice. The order of the choices was randomised, but all the first choices in a pair were shown before any of the second choices in a pair.

Ten-minute sound collages were produced using Audacity 2.3.0 (Audacity Team, 2018). All sound clips occurred in the first six minutes of the collage to ensure participants who completed the shopping task quickly were still exposed to most sounds. In the last four minutes, sounds were repeated.

Demographic Interview

Participants provided information on their demographics, diagnoses (including reading, listening and sensory impairments), and current medication.

Familiarity, Knowledge and Preference Questionnaire

Familiarity (Mishra, Umesh & Stem, 1993; Ratneshwar, Shocker & Stewart, 1987), knowledge (Malaviya & Sivakumar, 1998; Mishra et al., 1993) and preference of shopping products (Malkoc, Hedgcock & Hoeffler, 2013) may be important to AE decision-making, so were measured.

Participants indicated their familiarity and knowledge with each product on a four-point scale (scored 0-3) and their preference for each product on a five-point scale (scored 0-4). Participants received three scores by calculating the median values for familiarity, knowledge and preference across products. Higher scores indicated greater familiarity, knowledge or preference.

Subjective reports

Participants remained in the same room after the three sound conditions, so the first author could ask them ‘how did you find the experience?’. Participants were asked about each sound condition in the order it had been presented. The participants’ responses were transcribed verbatim.

Heart Rate Variability

Equipment and Protocol

A Zensor v0 Intelesens Ltd monitor recorded electrocardiogram (ECG) and respiration data using 3-lead monitoring, sampling at 360Hz (Laborde et al., 2017). Participants remained seated and upright, given the relationship between posture and HRV (Montano et al., 1994; Mukai & Hayano, 1995). Time of day of the recordings (morning, early afternoon, late afternoon or evening) did not differ between participant groups [Fisher’s exact test value=5.32, p=0.12].

HRV Questionnaire

To inform the HRV analysis, self-reported information about recent eating, sleeping and caffeine consumption was recorded, together with use of oral contraceptives, and urgency of need to use a bathroom.

Procedure

Participants attended one session lasting approximately 2.5 h. Sessions took place in one of two similar rooms. Participants completed the AQ and Product Familiarity, Knowledge and Preference Questionnaire before the session.

The order of the procedure ensured more and less demanding tasks were alternated, reducing fatigue (Lezak, Howieson, Loring, Hannay & Fischer, 2004). To begin, participants answered demographic questions and completed a hearing assessment: the sound of a phone ringtone was played and participants had to describe the sound to pass.

Participants then completed the WASI-II, followed by time-dependent questions in the HRV questionnaire. Subsequently, the HRV equipment was set-up. Participants were given an instruction sheet and the researcher left the room while the participant set-up the equipment, to ensure privacy.

Baseline heart rate measurement then took place, lasting 10 minutes. For approximately the first four minutes, the participant was given verbal and written instructions for the shopping task. Consistent language was used across participants when explaining how to complete the AE choices; ‘click on the one [option] you would prefer if you had to choose between them’ (Farmer et al., 2017). For the remaining time participants watched a nature film.

The shopping task was completed on the same computer by all participants with screen brightness, contrast and resolution all constant. Windows were covered to remove natural light variation and the lights were turned off. The researcher remained in the room, out of sight. The order of the sound conditions was randomised and the sound collages were played through speakers at approximately 78 decibels. Sound collages were looped if necessary. Between each condition, participants watched five minutes of the nature film. The researcher requested that the participant stayed seated. After the task was complete, participants were prompted to discuss each sound condition. Participants then finished the HRV questionnaire. The ADOS-2 was completed at the end of the session to ensure task order between NT and autistic groups remained as similar as possible.

All participants were paid, sent a letter of thanks, and those who had agreed to their contact details being retained were sent a layperson’s summary of the findings.

Analysis

Heart Rate Variability Pre-Processing

ECG processing was completed with Kubios 3.2.0, which uses an in-built QRS detection algorithm, based on the Pan-Tompkins algorithm (Pan & Tompkins, 1985). After peak detection, cubic spline interpolation of the RR time series occurred at four hertz and a smoothness prior detrending method (smoothing parameter, lambda=500) reduced the influence of very low-frequency components. The ECG trace was inspected manually to ensure correct QRS recognition. Participant information known to influence ectopic beats was considered during inspection. Ectopic beats and subsequent heart rate turbulence were corrected by altering R peak marks. Subsequent threshold-based artefact correction was applied on a participant-level basis if the RR interval trace showed outliers. The level of artefact correction was determined by applying the lowest threshold level which removed all spurious results. There was no difference in the percentage of heart beats corrected between autistic and NT groups [mean difference=-0.13, 95% CIs -1.11-0.83, U=313, p=0.72, d=-0.10].

For each participant, four 2.5 min samples were assessed (Esco & Flatt, 2014; Laborde et al., 2017; Munoz et al., 2015). The baseline sample was taken from the final 2.5 min of the baseline period to allow acclimatisation to the equipment (Quintana et al., 2016). For the sound conditions, samples were taken from the first 2.5 min of each condition. The high-frequency (HF) band was defined as 0.15-0.4Hz, as recommended for adults (Quintana et al., 2016). When, for the baseline recording, there was five percent artefact correction or more (Quintana & Heathers, 2014) or the electrocardiogram derived respiration (EDR) was outside the HF band, the sample was moved as close to the end of the baseline period as possible while allowing for artefact correction below five percent and HF band EDR placement. For the sound conditions, the sample was moved as close to the beginning of each condition as possible, while allowing for less than five percent artefact correction and HF band EDR placement. If a sample with less than five percent artefact correction and HF band EDR placement was not found in the last five minutes of the baseline period or within the sound conditions, the participant was excluded.

The root mean square of successive differences (RMSSD), high-frequency heart rate variability (HF-HRV) and heart rate were exported from Kubios to SPSS and RStudio. HF-HRV was calculated through autoregressive modelling (AR), since AR produces a spectrum with greater resolution than other methods when short samples are used (Laborde et al., 2017). AR was measured through absolute power, as advised by Laborde et al. (2017). Model order was 16 (Boardman, Schlindwein, Rocha & Leite, 2002; Dantas et al., 2012).

Baseline respiration was measured by manually counting the number of respiratory cycles in the respiratory waveform displayed by the Zensor+ software during baseline.

Statistical Analyses

Psychotropic medication and comorbid psychiatric conditions influence decision-making (Brand, Labudda & Markowitsch, 2006; Hauser, Moutoussis, Purg, Dayan & Dolan, 2018) so were considered as confounds in the decision-making latency and consistency analyses. For psychotropic medication, participants were given a score of 1 if they were currently taking any psychotropic medication and a score of 0 if they were not. For comorbid psychiatric conditions, participants were given a score of 1 if they currently had any psychiatric conditions; and again, a score of 0 if they did not. These categorical variables were included in a stepwise regression to assess their impact on decision-making latency and consistency. Product preference, knowledge and familiarity scores were also included in the stepwise regression for decision-making consistency, given the research to suggest these factors influence AE choosing behaviour (Malkoc et al., 2013; Mishra et al., 1993).

Given the impact of comorbid conditions, psychotropic medication and medication known to impact heart rate on HRV in NT (Laborde et al., 2017) and adult autistic groups (e.g. Thapa et al., 2019), these variables are often controlled for in HRV studies. We therefore assessed the impact of these variables on HRV through a stepwise regression. For psychotropic medication, participants were given a score of 1 if they were currently taking any psychotropic medication and a score of 0 if they were not. For medication known to impact heart rate, participants were given a score of 1 if they were currently taking any medication known to impact heart rate and a score of 0 if they were not. For comorbidites impacting HRV, participants were given a score of 1 if they currently had any comorbidites known to impact HRV and a score of 0 if they did not.

Variables which contributed to stepwise regression models at p≤0.05 were added as confounds to the ANOVAs testing the hypotheses.

Assumptions underlying the use of parametric tests comparing the autistic and NT groups were checked and, where required, transformations were made to the data or non-parametric tests were used. For variables where the data were transformed, the reported values reflect this, apart from means, standard deviations, mean differences and 95% CIs where raw values are reported. Tests were two-tailed and run at α=0.05.

Community Involvement

This research was developed to target one of the areas which the autism community (autistic people, their families and clinicians) highlight as needing more attention: research to better understand how environments can improve experiences for autistic people (Cusack & Sterry, 2016). In addition, one autistic individual provided feedback that was used to inform the wording of recruitment advertisements and payment for participants.

Results

Raw data and research materials are available on request to the corresponding author.

Participants

Forty-five autistic and 37 NT individuals participated. Of these, seven autistic participants were excluded: three with a tested IQ score below 90, three below the ADOS-2 cut-off, and one due to technical issues. Thirty-eight autistic and 37 NT participants were included in the analyses. More than three-quarters (79%) described themselves as White British (Supplementary Material 5). Twenty-two autistic participants self-reported a diagnosis of anxiety and 20 a diagnosis of depression, compared to four NT participants who self-reported anxiety and four who self-reported depression (Supplementary Material 6). Thirty autistic and 17 NT individuals self-reported taking medication, with 25 autistic participants reporting using psychotropic medication, compared to five NT participants (Supplementary Material 7).

Sixteen autistic participants and 13 NT participants were employed. Twenty-two autistic participants and 24 NT participants were unemployed. Twenty-four autistic and 33 NT participants had at least one degree or were currently completing a degree.

Table 1 shows the characteristics of the participants included in the analyses. Fisher’s exact test was used in the analysis of product familiarity, preference and knowledge because more than 20% of expected values fell below five. Product familiarity [Fisher’s exact test value=6.27, p=0.06] and knowledge [Fisher’s exact test value=1.39, p=0.61] did not differ between autistic and NT groups, but autistic participants gave significantly higher product preference scores than NT participants [Fisher’s exact test value=10.71, p=0.01] (Supplementary Material 8).

Table 1. Age, Intellectual Functioning (IQ), Autism Quotient (AQ) Scores and Self-Reported Gender across Participant Groups.

Variable Autistic
Mean		 Autistic
SD		 NT
Mean		 NT
SD		 95%
CIs		 df t-
value		 p-value Cohen’s
d
Age (years) 38.76 14.19 37.92 18.06 -6.62-8.31 73 0.23 0.82 0.05
IQ 113.7 12.00 112.8 10.54 -4.2-6.2 73 0.38 0.71 0.09
AQT 37.7 7.42 12.7 6.10 21.8-28.1 73 15.07 <0.001* 3.48
Variable Autistic Frequencies NT Frequencies df χ2value p-value Cramer’s
V
Self-Reported Gender Man=18,
Woman=20		 Man=16,
Woman=21		 1 0.02 0.90 0.041
Open in a new tab
T

Data were cube-root transformed due to raw data not being normally distributed.

*

p≤0.05.

Participants’ Self-Report

The responses of autistic (n=38) and NT (n=15) participants to the sound conditions (Figure 1) were analysed. A response was categorised as positive when the participant’s comments were entirely positive, for example, describing the experience as relaxing or easy. Comments that made no reference to positive or negative experiences, references to both positive and negative experiences or described the use of coping mechanisms were categorised as neutral. A negative categorisation was given when the participant’s comments were entirely negative, for example, describing difficulty or fear. Negative experiences were coded as one, neutral as two and positive as three. Responses were independently coded by two of the authors. Prior to a consensus meeting, Cohen’s kappa equalled 0.81 [95% CIs 0.71-0.90, p<0.001]. After the meeting, full agreement was met [k=1.00, p<0.001]. Sample comments are shown in Supplementary Material 9.

Figure 1. ““Proportion of Negative, Neutral or Positive Experiences Reported by Autistic and Neurotypical Participants””.

Figure 1

Open in a new tab

*=p<0.05. **=p≤0.001. All other differences are not statistically significant. Chi-square tests were used to assess significance, except when expected values<5, in which case Fisher's exact test was used.

A Friedman test indicated a significant difference across the sound conditions within the autistic group [χ2(2)=54.12, p<0.001]. Bonferroni-corrected post-hoc tests indicated responses to the no sound condition were significantly more positive than responses to the non-social sound [p<0.001, d=1.37] and social sound conditions [p<0.001, d=2.06], but no significant differences were seen between the non-social and social sound conditions [p=0.56, d=0.31]. To test, whether a group (autistic/NT) by sound condition interaction existed, a log-linear analysis was completed. A significant three-way interaction was found [χ2(4)=8.82, p=0.05], suggesting there was a difference in the qualitative ratings between the autistic and NT groups across the three sound conditions. Bonferroni-corrected post-hoc tests revealed there was no difference between the autistic and NT groups’ qualitative ratings in the no sound [Fisher’s exact test value=2.03, p=0.44] and non-social sound [Fisher’s exact test value=2.85, p=0.22] conditions, but the autistic group produced significantly more negative ratings than the NT group in the social sound condition [Fisher’s exact test value=16.51, p<0.001].

Decision-Making Latency

Figure 2 highlights decision-making latency in autistic and NT groups across sound conditions (Supplementary Material 10). The group x sound condition interaction [F(2, 146)=0.98, p=0.38, η2p=0.013], main effect for sound condition [F(2, 146)=2.44, p=0.09, η2p=0.032] and main effect for participant group [F(2, 73)=2.53, p=0.12, η2p=0.033]2 were non-significant.

Figure 2. ““Decision-Making Latency during No Sound, Non-Social Sound and Social Sound Conditions””.

Figure 2

Open in a new tab

Error bars represent one SE above and below the mean.

267x174mm (72 x 72 DPI)

A stepwise regression was completed with backward elimination of comorbid psychiatric conditions and psychotropic medication as categorical variables. Neither psychiatric comorbidities nor psychotropic medication significantly contributed to the model [p≥0.43] and so were not added to the ANOVA.

Decision-Making Consistency

Figure 3 shows decision-making consistency for autistic and NT groups across sound conditions3 (Supplementary Material 10). The group x sound condition interaction [F(2, 146)=2.82, p=0.06, η2p=0.037], main effect for sound condition [F(2, 146)=0.66, p=0.52, η2p=0.009] and main effect for participant group [F(1, 73)=0.21, p=0.65, η2p=0.003] were non-significant.

Figure 3. ““Decision-Making Consistency during No Sound, Non-Social Sound and Social Sound Conditions”“.

Figure 3

Open in a new tab

Error bars represent one SE above and below the mean.

257x167mm (72 x 72 DPI)

To assess those variables needing to be controlled in the ANOVA, a backward elimination of the following variables was completed: psychotropic medication, comorbid psychiatric conditions, product knowledge, product familiarity and product preference. Psychotropic medication significantly contributed to the final model [p=0.05], so the ANOVA was repeated with psychotropic medication as a factor. All effects were non-significant [F≤1.95, p≥0.15, η2p≤0.027].

Heart Rate Variability

HRV data were analysed for 30 autistic and 22 NT participants. Groups were matched on age, self-reported gender and IQ (Supplementary Material 11). Eighteen participants were excluded due to a lack of ECG trace; two due to data only being available for a 1.5 min sampling period in one or more sound conditions; one due to artefact correction being over five percent; and one due to ECG-derived respiration being outside the high frequency band. Failures were likely produced as the result of participants setting-up the ECG equipment independently. This procedure could not be altered due to participant privacy.

To establish whether exclusions from the HRV analysis were linked to participant group, a chi-square test was completed to compare HRV recording success (autistic: n=30, NT:n=22) and failure (autistic: n=8, NT: n=15) between groups; the difference was not significant [χ2(1)=2.50, p=0.11, K=0.21).

Forty-two individuals were included in the baseline respiration analysis (autistic: n=24, NT: n=18): baseline respiration could not be determined for five individuals because of a noisy respiratory trace and for another five due to a weak signal. HRV confounds were analysed for 24 participants (autistic: n=18, NT: n=6).

There were no differences in baseline heart rate, HRV or respiration between participant groups (Table 2).

Table 2. Baseline Heart Rate, HRV and Respiration across Participant Groups.

Variable Autistic
Mean		 Autistic
SD		 NT Mean NT SD 95% CIs df t- value p- value Cohen’s d
Baseline Heart 71.19 8.45 69.73 8.40 -3.29–6.22 50 0.62 0.54 0.17
Baseline
RMSSD		 37.75 23.90 42.52 23.67 -8.64–18.19 50 -0.72 0.48 0.20
(HRV, ms)T
Baseline HF-HRV (HRV, ms2)T 		893.94 1504.28 816.82 900.39 -647.73–801.96 50 0.39 0.70 0.11
Variable Autistic
Mean		 Autistic
SD		 NT
Mean		 NT SD 95% CIs u-value p-value d
Baseline Respiration (number of breaths during baseline)M 38.33 7.82 40.06 5.89 15.56–
-12.12		 176 0.31 -0.32
Open in a new tab
T

Data were transformed due to raw data not being normally distributed. Baseline RMSSD=square-root transformation. Baseline HF-HRV=log-transformation.

M

A Mann-Whitney U test was used since data transformation did not lead to normality

*=p≤0.05.

Figure 4 shows RMSSD across participant groups in the sound conditions (Supplementary Material 10). HRV was measured through log-transformed RMSSD values, since raw data violated the assumption of normality of residuals. The group x sound condition interaction [F(2, 100)=0.75, p=0.48, η2p=0.015], main effect for sound condition [F(2, 100)=1.42, p=0.25, η2p=0.028] and main effect for participant group [F(1, 50)=0.08, p=0.78, η2p=0.002] were non-significant. These results were replicated with HF-HRV (Supplementary Material 12).

Figure 4. ““Log-Transformed Root Mean Square of Successive Differences in Autistic and Neurotypical Groups during No Sound, Non-Social Sound and Social Sound Conditions””.

Figure 4

Open in a new tab

Error bars represent one SE above and below the mean.

299x176mm (72 x 72 DPI)

To understand the influence of medication and comorbidities on RMSSD (log-transformed), a stepwise regression was completed with backward elimination of three categorical variables: comorbidities believed to impact HRV, ‘medications that impact heart rate’ and psychotropic medications. Both the ‘medications that impact heart rate’ [p≤0.001] and psychotropic medication [p=0.006] variables had a significant influence on the final model, so were included as factors in the ANOVA; all interactions with RMSSD were non-significant [F≤1.63, p≥0.20, η2p≤0.036]. The results of these analyses (stepwise regression and ANOVA) were replicated with log-transformed HF-HRV [F≤2.76, p≥0.069, η2p≤0.059].

Discussion

Qualitative measures suggested the social sound condition was experienced more negatively by the autistic than NT participants. However, the hypotheses were supported: there were no statistically significant decision-making latency, decision-making consistency or HRV differences across sound conditions and participant groups.

The finding that autistic participants experienced the social sound condition more negatively than NT participants is expected given the social difficulties characterising autism (American Psychiatric Association, 2013). The difference in qualitative ratings between the autistic and NT groups in the social sound condition but not the non-social sound condition, suggests difficulties experienced by autistic individuals, as compared to NT individuals, in these ambient environments may be related to the social aspects of sounds, rather than a generalised sensitivity to all sounds. It is also possible the social sounds were characterised by features which autistic individuals self-report as particularly difficult, such as unpredictable, loud, or very high or low frequency sounds (Smith & Sharp, 2013). However, within the autistic group, both ambient sound conditions were rated more negatively than the no sound condition, suggesting benefit to reducing both non-social, such as checkout beeps, and social, such as loud conversations, sounds to make supermarkets more autism-friendly.

The lack of difference in decision-making and HRV measures between the non-social and social sound conditions in the autistic group is unexpected and noteworthy. This result is not in line with theories suggesting autism is due to inherent social difficulties (e.g. Baron-Cohen et al., 1985), which would predict the social sound condition would create more difficulty for autistic people than the non-social condition. Such results could be better explained by attentional theories of autism (e.g. Liss, Saulnier, Fein & Kinsbourne, 2006; Ploog, 2010; Tyndall et al., 2017) which suggest autistic individuals show broader atypicalities in attention, rather than specific social deficits. Prediction deficit models of autism (e.g. Sinha et al., 2014) may also better explain these results by suggesting autistic individuals rely less on prior expectations relative to sensory inputs, leading to broad perceptual atypicalities rather than atypicalities specific to the social domain. The presence of a participant group by sound condition interaction in the qualitative analysis should, however, be noted, since this interaction indicates that there are differences in the impact of social, but not non-social, sounds between autistic and NT individuals.

The absence of differences in decision-making between participant groups was unexpected based on self-reported decision-making difficulties in autism (Gaeth et al., 2016; Luke et al., 2012) and Farmer et al.'s (2017) findings of increased latency and consistency in autistic, compared to NT, adults. In the present study, the autistic participants’ reports highlighted the negative experience of the social and non-social sound conditions, suggesting the participants faced difficulties with the shopping task. Measures of latency and consistency may not have been sufficiently sensitive to reflect the subjective experience of autistic participants and participants’ difficulties may have gone undetected.

The outcome measures must be considered in terms of whether they reflect everyday decision-making. Taking a long time to make decisions is reported by autistic participants as an everyday difficulty (Luke et al., 2012), supporting the use of latency. However, decision-making consistency may not reflect decision-making in everyday life, since individuals rarely make several choices regarding the same product. There are methods to improve the everyday relevance of the AE paradigm, such as offering a no-buy option, but these are not validated with autistic participants (Hadar, Danziger, & Hertwig, 2018; Lichters, Sarstedt & Vogt, 2015; Stewart, 1989).

Our findings add to the inconsistencies in the HRV literature: some studies have found baseline/task-driven HRV differences between autistic and NT adults, but others have not (Cai et al., 2019; Dijkhuis et al., 2019; Kuiper et al., 2017; Mathewson et al., 2011; Smeekens et al., 2015; Thapa et al., 2019).

The absence of baseline differences in HRV between participant groups was surprising, given that 22 autistic participants self-reported ‘anxiety’ compared to only 4 NT participants (Supplementary Material 6). However, ‘anxiety’ is often used informally rather than according to standard diagnostic classifications (such as DSM-V, American Psychiatric Association, 2013; ICD-10, World Health Organization, 1992) to include a range of worries and fears, with different levels of impact. The relationship between these self-reports, which are very common in autistic people (Zaboski & Storch, 2018), and ‘anxiety’ as measured by HRV is uncertain. While we controlled for medication use and comorbidities known to impact HRV, it is possible that the signal may have been affected by variables that were uncontrolled, such as smoking (Kemp, Koenig & Thayer, 2017; Murgia et al., 2019), body mass index (Föhr et al., 2016; Koenig et al., 2014), physical activity (Kemp et al., 2017; May, McBerty, Zaky & Gianotti, 2017), and alcohol (Ralevski, Petrakis & Altemus, 2019) and water (Heathers et al., 2018) consumption.

During the task, the ambient sounds may not have induced anxiety, resulting in no HRV change. Since anxiety can impact decision-making (Remmers & Zander, 2018), this possibility may also account for the absence of decision-making differences across sound conditions. However, given that previous research suggests sound stimuli can lead to anxiety for autistic individuals (Landon et al., 2016), this would be surprising. It is possible there were differences in anxiety between participant groups at baseline and across the sound conditions that were not reflected through HRV measures. Such a possibility might reflect autonomic dysregulation in autistic individuals (Panju et al., 2015; Truzzi, Setoh, Shinohara & Esposito, 2016) where, despite self-reported negative affect, a physiological response is absent. For example, in Dijkhuis et al.’s (2019) study, there were no significant differences between baseline HRV and HRV during a speaking task in autistic adults, despite participant reports of negative affect during the task. Such dysregulation is problematic because it may result in reduced environmental adaptation, exacerbating feelings of negative affect (Dijkhuis et al., 2019).

Implications

Future research could expand on the present study’s paradigm to create a shopping experience with even greater ecological validity. Research could be conducted in a supermarket, though this may improve validity at the cost of experimental control. When Doyle et al. (1999) completed the AE paradigm in a supermarket, they noted several uncontrolled factors, such as the actions of other shoppers.

Several acoustic factors are known to influence sensory reactivity, including unpredictability and frequency (Robertson & Simmons, 2015; Smith & Sharp, 2013). However, in the present study, participants were not specifically asked whether the sounds were distracting to the task or whether certain aspects of the ambient environment were more difficult than others. Additional studies could ask participants to specify which aspects of the ambient environment impacted the task.

Additional physiological measures could be used in future research. For example, pupil diameter, an indicator of arousal, is affected by decision-making uncertainty and pupil-linked arousal also influences decision-making behaviour (Urai et al., 2017). Therefore, this measure may give insights into the relationship between arousal and decision-making behaviour during exposure to ambient sensory environments.

In the present study, participants had high levels of intellectual functioning. Future research could adapt the shopping paradigm to be inclusive of those with intellectual disability, since 50% of autistic people have co-occurring intellectual disability (n=24 studies) (Loomes, Hull & Mandy, 2017).

The context of this study was that of the increasing number of organisations that have begun creating autism-friendly environments. Our qualitative evidence indicated the non-social and social sound conditions were experienced more negatively than the no sound condition for the autistic group, suggesting such autism-friendly changes may help. Research studies should consider alternative quantitative measures and assess the impact of other environmental stimuli, such as lighting or proximity to others, to potentially allow for more detailed recommendations with applications to settings far beyond supermarkets such as classrooms (Zazzi & Faragher, 2018), the workplace (Johnson & Joshi, 2016) and healthcare (Morris, Greenblatt & Saini, 2019; Vohra, Madhavan & Sambamoorthi, 2016). By considering autism as a difference in neurodiversity (Kapp, Gillespie-Lynch, Sherman & Hutman, 2013) and understanding how environments can be altered to promote a good person-environment fit, autistic individuals may be better able to thrive (Lai & Baron-Cohen, 2015; Mandy & Lai, 2016).

Supplementary Material

Supplementary Material

Acknowledgements

Removed identifiable information.

Funding

Removed identifiable information.

Footnotes

1

Kenny et al. (2016) found a large percentage of autistic people and their families and friends endorsed the term ‘autistic’.

2

For one participant, the latency of one choice in the social sound condition was missing. Average latency for the social sound condition was calculated using the remaining 17 choices.

3

For one participant, the option chosen for one choice in the social sound condition was missing. Average decision-making consistency for the social sound condition was calculated using the eight complete choice pairs.

References

  1. Altman DG. Practical statistics for medical research. Chapman and Hall; 1991. [Google Scholar]
  2. Ambler T, Braeutigam S, Stins J, Rose S, Swithenby S. Salience and choice: Neural correlates of shopping decisions. Psychology and Marketing. 2004;21(4):247–261. doi: 10.1002/mar.20004. [DOI] [Google Scholar]
  3. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-5. American Psychiatric Publishing; 2013. [Google Scholar]
  4. Audacity Team. Audacity(R): Free audio editor and recorder. 2018. https://audacityteam.org/
  5. Baron-Cohen S, Leslie AM, Frith U. Does the autistic child have a “theory of mind”. Cognition. 1985;21(1):37–46. doi: 10.1016/0010-0277(85)90022-8. [DOI] [PubMed] [Google Scholar]
  6. Baron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E. The Autism-Spectrum Quotient (AQ): Evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. Journal of Autism and Developmental Disorders. 2001;31(1):5–17. doi: 10.1023/A:1005653411471. [DOI] [PubMed] [Google Scholar]
  7. Benevides TW, Lane SJ. A review of cardiac autonomic measures:Considerations for examination of physiological response in children with autism spectrum disorder. Journal of Autism and Developmental Disorders. 2015;45(2):560–575. doi: 10.1007/s10803-013-1971-z. [DOI] [PubMed] [Google Scholar]
  8. Boardman A, Schlindwein FS, Rocha AP, Leite A. A study on the optimum order of autoregressive models for heart rate variability. Physiological Measurement. 2002;23(2):325–336. doi: 10.1088/0967-3334/23/2/308. [DOI] [PubMed] [Google Scholar]
  9. Braeutigam S, Stins JF, Rose SP, Swithenby SJ, Ambler T. Magnetoencephalographic signals identify stages in real-life decision processes. Neural Plasticity. 2001;8(4):241–254. doi: 10.1155/NP.2001.241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Brand M, Labudda K, Markowitsch HJ. Neuropsychological correlates of decision-making in ambiguous and risky situations. Neural Networks. 2006;19(8):1266–1276. doi: 10.1016/_j.neunet.2006.03.001. [DOI] [PubMed] [Google Scholar]
  11. Cai RY, Richdale AL, Dissanayake C, Uljarević M. Resting heart rate variability, emotion regulation, psychological wellbeing and autism symptomatology in adults with and without autism. International Journal of Psychophysiology. 2019;137:54–62. doi: 10.1016/J.IJPSYCHO.2018.12.010. [DOI] [PubMed] [Google Scholar]
  12. Crane L, Adams F, Harper G, Welch J, Pellicano E. Something needs to change’: Mental health experiences of young autistic adults in England. Autism. 2019;23(2):477–493. doi: 10.1177/1362361318757048. [DOI] [PubMed] [Google Scholar]
  13. Croen LA, Zerbo O, Qian Y, Massolo ML, Rich S, Sidney S, Kripke C. The health status of adults on the autism spectrum. Autism. 2015;19(7):814–823. doi: 10.1177/1362361315577517. [DOI] [PubMed] [Google Scholar]
  14. Cusack J, Sterry R. Your questions: Shaping future autism research. 2016. https://www.autistica.org.uk/downloads/files/Autism-Top-10-Your-Priorities-for-Autism-Research.pdf .
  15. Dantas EM, Sant’Anna ML, Andreão RV, Gonçalves CP, Morra EA, Baldo MP, Rodrigues SL, Mill JG. Spectral analysis of heart rate variability with the autoregressive method: What model order to choose. Computers in Biology and Medicine. 2012;42(2):164–170. doi: 10.1016/j.compbiomed.2011.11.004. [DOI] [PubMed] [Google Scholar]
  16. Dijkhuis RR, Ziermans T, van Rijn S, Staal W, Swaab H. Emotional arousal during social stress in young adults with autism: Insights from heart rate, heart rate variability and self-report. Journal of Autism and Developmental Disorders. 2019;49(6):2524–2535. doi: 10.1007/s10803-019-04000-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Doyle JR, O’Connor DJ, Reynolds GM, Bottomley PA. The robustness of the asymmetrically dominated effect: Buying frames, phantom alternatives, and instore purchases. Psychology and Marketing. 1999;16(3):225–243. doi: 10.1002/(SICI)1520-6793(199905)16:3&#x0003c;225::AID-MAR3&#x0003e;3.0.CO;2-X. [DOI] [Google Scholar]
  18. Esco MR, Flatt AA. Ultra-short-term heart rate variability indexes at rest and post-exercise in athletes: Evaluating the agreement with accepted recommendations. Journal of Sports Science & Medicine. 2014;13(3):535–541. [PMC free article] [PubMed] [Google Scholar]
  19. Farmer GD, Baron-Cohen S, Skylark WJ. People with autism spectrum conditions make more consistent decisions. Psychological Science. 2017;28(8):1067–1076. doi: 10.1177/0956797617694867. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Föhr T, Pietilä J, Helander E, Myllymäki T, Lindholm H, Rusko H, Kujala UM. Physical activity, body mass index and heart rate variability-based stress and recovery in 16 275 Finnish employees: A cross-sectional study. BMC Public Health. 2016;16(1):701. doi: 10.1186/s12889-016-3391-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gaeth GJ, Levin IP, Jain G, Burke EV. Toward understanding everyday decision making by adults across the autism spectrum. Judgment and Decision Making. 2016;11(6):537–546. https://www.sas.upenn.edu/~baron/journal/16/16810/jdm16810.pdf . [Google Scholar]
  22. Green SA, Ben-Sasson A, Soto TW, Carter AS. Anxiety and sensory over-responsivity in toddlers with autism spectrum disorders: Bidirectional effects across time. Journal of Autism and Developmental Disorders. 2012;42(6):1112–1119. doi: 10.1007/s10803-011-1361-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hadar L, Danziger S, Hertwig R. The attraction effect in experience-based decisions. Journal of Behavioural Decision Making. 2018;31(3):461–468. doi: 10.1002/bdm.2058. [DOI] [Google Scholar]
  24. Hauser TU, Moutoussis M, Purg N, Dayan P, Dolan RJ. Beta-blocker propranolol modulates decision urgency during sequential information gathering. Journal of Neuroscience. 2018;38(32):7170–7178. doi: 10.1523/JNEUROSCI.0192-18.2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Heathers J, Quintana D, Angus D, Krygier J, Kemp A, de Rosnay M. Water consumption as a source of error in the measurement of heart rate variability. Open Science Framework. 2018 doi: 10.31219/osf.io/83exy. [DOI] [Google Scholar]
  26. Hollocks MJ, Lerh JW, Magiati I, Meiser-Stedman R, Brugha TS. Anxiety and depression in adults with autism spectrum disorder: A systematic review and meta-analysis. Psychological Medicine. 2019;49(4):559–572. doi: 10.1017/S0033291718002283. [DOI] [PubMed] [Google Scholar]
  27. Huber J, Payne JW, Puto C. Adding asymmetrically dominated alternatives: Violations of regularity and the similarity hypothesis. Journal of Consumer Research. 1982;9(1):90–98. doi: 10.1086/208899. [DOI] [Google Scholar]
  28. Hus V, Lord C. The Autism Diagnostic Observation Schedule, Module 4: Revised algorithm and standardized severity scores. Journal of Autism and Developmental Disorders. 2014;44(8):1996–2012. doi: 10.1007/s10803-014-2080-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Johnson TD, Joshi A. Dark clouds or silver linings? A stigma threat perspective on the implications of an autism diagnosis for workplace well-being. Journal of Applied Psychology. 2016;101(3):430–449. doi: 10.1037/apl0000058. [DOI] [PubMed] [Google Scholar]
  30. Kapp SK, Gillespie-Lynch K, Sherman LE, Hutman T. Deficit, difference, or both? Autism and neurodiversity. Developmental Psychology. 2013;49(1):59–71. doi: 10.1037/a0028353. [DOI] [PubMed] [Google Scholar]
  31. Kemp AH, Koenig J, Thayer JF. From psychological moments to mortality: A multidisciplinary synthesis on heart rate variability spanning the continuum of time. Neuroscience and Biobehavioral Reviews. 2017;83:547–567. doi: 10.1016/_j.neubiorev.2017.09.006. [DOI] [PubMed] [Google Scholar]
  32. Kenny L, Hattersley C, Molins B, Buckley C, Povey C, Pellicano E. Which terms should be used to describe autism? Perspectives from the UK autism community. Autism. 2016;20(4):442–462. doi: 10.1177/1362361315588200. [DOI] [PubMed] [Google Scholar]
  33. Klusek J, Roberts JE, Losh M. Cardiac autonomic regulation in autism and fragile X syndrome: A review. Psychological Bulletin. 2015;141(1):141–175. doi: 10.1037/a0038237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Koenig J, Jarczok MN, Warth M, Ellis RJ, Bach C, Hillecke TK, Thayer JF. Body mass index is related to autonomic nervous system activity as measured by heart rate variability — A replication using short term measurements. The Journal of Nutrition, Health & Aging. 2014;18(3):300–302. doi: 10.1007/s12603-014-0022-6. [DOI] [PubMed] [Google Scholar]
  35. Krieger B, Piškur B, Schulze C, Jakobs U, Beurskens A, Moser A. Supporting and hindering environments for participation of adolescents diagnosed with autism spectrum disorder: A scoping review. PLOS ONE. 2018;13(8):e0202071. doi: 10.1371/journal.pone.0202071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Kuiper MWM, Verhoeven EWM, Geurts HM. Heart rate variability predicts inhibitory control in adults with autism spectrum disorders. Biological Psychology. 2017;128:141–152. doi: 10.1016/J.BIOPSYCHO.2017.07.006. [DOI] [PubMed] [Google Scholar]
  37. Laborde S, Mosley E, Thayer JF. Heart rate variability and cardiac vagal tone in psychophysiological research - Recommendations for experiment planning, data analysis, and data reporting. Frontiers in Psychology. 2017;8:213. doi: 10.3389/fpsyg.2017.00213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lai M-C, Baron-Cohen S. Identifying the lost generation of adults with autism spectrum conditions. The Lancet Psychiatry. 2015;2(11):1013–1027. doi: 10.1016/S2215-0366(15)00277-1. [DOI] [PubMed] [Google Scholar]
  39. Lai M-C, Kassee C, Besney R, Bonato S, Hull L, Mandy W, Szatmari MD, Ameis SH. Prevalence of co-occurring mental health diagnoses in the autism population: A systematic review and meta-analysis. The Lancet Psychiatry. 2019 doi: 10.1016/S2215-0366(19)30289-5. [DOI] [PubMed] [Google Scholar]
  40. Landon J, Shepherd D, Lodhia V. A qualitative study of noise sensitivity in adults with autism spectrum disorder. Research in Autism Spectrum Disorders. 2016;32:43–52. doi: 10.1016/J.RASD.2016.08.005. [DOI] [Google Scholar]
  41. Lawson W. Understanding and working with the spectrum of autism: An insider’s view. Jessica Kingsley Publishers; 2001. [Google Scholar]
  42. Lezak MD, Howieson DB, Loring DW, Hannay HJ, Fischer JS. Neuropsychological assessment. 4th ed. Oxford University Press; 2004. [Google Scholar]
  43. Lichters M, Sarstedt M, Vogt B. On the practical relevance of the attraction effect: A cautionary note and guidelines for context effect experiments. AMS Review. 2015;5(1–2):1–19. doi: 10.1007/s13162-015-0066-8. [DOI] [Google Scholar]
  44. Liss M, Saulnier C, Fein D, Kinsbourne M. Sensory and attention abnormalities in autistic spectrum disorders. Autism. 2006;10(2):155–172. doi: 10.1177/1362361306062021. [DOI] [PubMed] [Google Scholar]
  45. Loomes R, Hull L, Mandy WPL. What is the male-to-female ratio in autism spectrum disorder? A systematic review and meta-analysis. Journal of the American Academy of Child & Adolescent Psychiatry. 2017;56(6):466–474. doi: 10.1016/J.JAAC.2017.03.013. [DOI] [PubMed] [Google Scholar]
  46. Luce RD. Individual choice behaviour: A theoretical analysis. Wiley; 1959. [Google Scholar]
  47. Luce RD. The choice axiom after twenty years. Journal of Mathematical Psychology. 1977;15(3):215–233. doi: 10.1016/0022-2496(77)90032-3. [DOI] [Google Scholar]
  48. Lucker JR. Auditory hypersensitivity in children with autism spectrum disorders. Focus on Autism and Other Developmental Disabilities. 2013;28(3):184–191. doi: 10.1177/1088357613475810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Lugo-Marín J, Magán-Maganto M, Rivero-Santana A, Cuellar-Pompa L, Alviani M, Jenaro-Rio C, Díez E, Canal-Bedia R. Prevalence of psychiatric disorders in adults with autism spectrum disorder: A systematic review and meta-analysis. Research in Autism Spectrum Disorders. 2019;59:22–33. doi: 10.1016/j.rasd.2018.12.004. [DOI] [Google Scholar]
  50. Luke L, Clare ICH, Ring H, Redley M, Watson P. Decision-making difficulties experienced by adults with autism spectrum conditions. Autism. 2012;16(6):612–621. doi: 10.1177/1362361311415876. [DOI] [PubMed] [Google Scholar]
  51. Malaviya P, Sivakumar K. The moderating effect of product category knowledge and attribute importance on the attraction effect. Marketing Letters. 1998;9(1):93–106. doi: 10.1023/A:1007976305757. [DOI] [Google Scholar]
  52. Malkoc SA, Hedgcock W, Hoeffler S. Between a rock and a hard place: The failure of the attraction effect among unattractive alternatives. Journal of Consumer Psychology. 2013;23(3):317–329. doi: 10.1016/j.jcps.2012.10.008. [DOI] [Google Scholar]
  53. Mandy W, Lai M-C. Annual research review: The role of the environment in the developmental psychopathology of autism spectrum condition. Journal of Child Psychology and Psychiatry. 2016;57(3):271–292. doi: 10.1111/jcpp.12501. [DOI] [PubMed] [Google Scholar]
  54. Mathewson KJ, Drmic IE, Jetha MK, Bryson SE, Goldberg JO, Hall GB, Santesso DL, Segalowitz SJ, Schmidt LA. Behavioral and cardiac responses to emotional stroop in adults with autism spectrum disorders: Influence of medication. Autism Research. 2011;4(2):98–108. doi: 10.1002/aur.176. [DOI] [PubMed] [Google Scholar]
  55. May R, McBerty V, Zaky A, Gianotti M. Vigorous physical activity predicts higher heart rate variability among younger adults. Journal of Physiological Anthropology. 2017;36(1):24. doi: 10.1186/s40101-017-0140-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Mertens J, Zane ER, Neumeyer K, Grossman RB. How anxious do you think I am? Relationship between state and trait anxiety in children with and without ASD during social tasks. Journal of Autism and Developmental Disorders. 2017;47(12):3692–3703. doi: 10.1007/s10803-016-2979-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Mishra S, Umesh UN, Stem DE. Antecedents of the attraction effect: An information-processing approach. Journal of Marketing Research. 1993;30(3):331–349. doi: 10.1177/002224379303000305. [DOI] [Google Scholar]
  58. Montano N, Ruscone TG, Porta A, Lombardi F, Pagani M, Malliani A. Power spectrum analysis of heart rate variability to assess the changes in sympathovagal balance during graded orthostatic tilt. Circulation. 1994;90(4):1826–1831. doi: 10.1161/01.cir.90.4.1826. [DOI] [PubMed] [Google Scholar]
  59. Morris R, Greenblatt A, Saini M. Healthcare providers’ experiences with autism: A scoping review. Journal of Autism and Developmental Disorders. 2019;49(6):2374–2388. doi: 10.1007/s10803-019-03912-6. [DOI] [PubMed] [Google Scholar]
  60. Mourali M, Böckenholt U, Laroche M. Compromise and attraction effects under prevention and promotion motivations. Journal of Consumer Research. 2007;34(2):234–247. doi: 10.1086/519151. [DOI] [Google Scholar]
  61. Mukai S, Hayano J. Heart rate and blood pressure variabilities during graded head-up tilt. Journal of Applied Physiology. 1995;78(1):212–216. doi: 10.1152/jappl.1995.78.1.212. [DOI] [PubMed] [Google Scholar]
  62. Munoz ML, Van Roon A, Riese H, Thio C, Oostenbroek E, Westrik I, De Geus EJC, Gansevoort R, Lefrandt J, Nolte IM, Snieder H. Validity of (Ultra-)Short recordings for heart rate variability measurements. PLoS ONE. 2015;10(9) doi: 10.1371/journal.pone.0138921. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Murgia F, Melotti R, Foco L, Gögele M, Meraviglia V, Motta B, Steger A, Toifl M, Sinnecker D, Müller A, Merati G, et al. Effects of smoking status, history and intensity on heart rate variability in the general population: The CHRIS study. PLOS ONE. 2019;14(4):e0215053. doi: 10.1371/journal.pone.0215053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. National Autistic Society. Autism hour highlights. 2018. https://www.autism.org.uk/get-involved/media-centre/news/2018-10-16-autism-hour-highlights.aspx .
  65. National Autistic Society. The autism friendly award. 2019. https://www.autism.org.uk/professionals/autism-friendly-award.aspx .
  66. Pan J, Tompkins WJ. A simple real-time QRS detection algorithm. IEEE Transactions On Biomedical Engineering. 1985;32(3) doi: 10.1109/IEMBS.1996.647473. [DOI] [PubMed] [Google Scholar]
  67. Panju S, Brian J, Dupuis A, Anagnostou E, Kushki A. Atypical sympathetic arousal in children with autism spectrum disorder and its association with anxiety symptomatology. Molecular Autism. 2015;6(1):64. doi: 10.1186/s13229-015-0057-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Paulhus D, Vazire S. In: Handbook of research methods in personality psychology. Robins R, Fraley C, editors. Guildford Press; 2007. The self-report method; pp. 224–239. [Google Scholar]
  69. Ploog BO. Stimulus overselectivity four decades later: a review of the literature and its implications for current research in autism spectrum disorder. Journal of Autism and Developmental Disorders. 2010;40(11):1332–1349. doi: 10.1007/s10803-010-0990-2. [DOI] [PubMed] [Google Scholar]
  70. Porges SW, Macellaio M, Stanfill SD, McCue K, Lewis GF, Harden ER, Handelman M, Denver J, Bazhenova OV, Heilman KJ. Respiratory sinus arrhythmia and auditory processing in autism: Modifiable deficits of an integrated social engagement system. International Journal of Psychophysiology. 2013;88(3):261–270. doi: 10.1016/j.ijpsycho.2012.11.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Quintana DS, Alvares GA, Heathers JAJ. Guidelines for Reporting Articles on Psychiatry and Heart rate variability (GRAPH): Recommendations to advance research communication. Translational Psychiatry. 2016;6(5):e803. doi: 10.1038/tp.2016.73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Quintana DS, Heathers JAJ. Considerations in the assessment of heart rate variability in biobehavioral research. Frontiers in Psychology. 2014;5:805. doi: 10.3389/fpsyg.2014.00805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Ralevski E, Petrakis I, Altemus M. Heart rate variability in alcohol use: A review. Pharmacology Biochemistry and Behavior. 2019;176:83–92. doi: 10.1016/J.PBB.2018.12.003. [DOI] [PubMed] [Google Scholar]
  74. Ratneshwar S, Shocker AD, Stewart DW. Toward understanding the attraction effect: The implications of product stimulus meaningfulness and familiarity. Journal of Consumer Research. 1987;13(4):520–533. doi: 10.1086/209085. [DOI] [Google Scholar]
  75. Remmers C, Zander T. Why you don’t see the forest for the trees when you are anxious: Anxiety impairs intuitive decision making. Clinical Psychological Science. 2018;6(1):48–62. doi: 10.1177/2167702617728705. [DOI] [Google Scholar]
  76. Robertson AE, Simmons DR. The sensory experiences of adults with autism spectrum disorder: A qualitative analysis. Perception. 2015;44(5):569–586. doi: 10.1068/p7833. [DOI] [PubMed] [Google Scholar]
  77. Sainsbury C. Martian in the playground: Understanding the schoolchild with Asperger's syndrome. Lucky Duck; 2000. [Google Scholar]
  78. Santos-Ruiz A, Garcia-Rios MC, Fernandez-Sanchez JC, Perez-Garcia M, Muñoz-García MA, Peralta-Ramirez MI. Can decision-making skills affect responses to psychological stress in healthy women. Psychoneuroendocrinology. 2012;37(12):1912–1921. doi: 10.1016/j.psyneuen.2012.04.002. [DOI] [PubMed] [Google Scholar]
  79. Schaaf RC, Benevides TW, Leiby BE, Sendecki JA. Autonomic dysregulation during sensory stimulation in children with autism spectrum disorder. Journal of Autism and Developmental Disorders. 2015;45(2):461–472. doi: 10.1007/s10803-013-1924-6. [DOI] [PubMed] [Google Scholar]
  80. Sinha P, Kjelgaard MM, Gandhi TK, Tsourides K, Cardinaux AL, Pantazis D, Diamond SP, Held RM. Autism as a disorder of prediction. Proceedings of the National Academy of Sciences of the United States of America. 2014;111(42):15220–15225. doi: 10.1073/pnas.1416797111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Smeekens I, Didden R, Verhoeven EWM. Exploring the relationship of autonomic and endocrine activity with social functioning in adults with autism spectrum disorders. Journal of Autism and Developmental Disorders. 2015;45(2):495–505. doi: 10.1007/s10803-013-1947-z. [DOI] [PubMed] [Google Scholar]
  82. Smith RS, Sharp J. Fascination and isolation: A grounded theory exploration of unusual sensory experiences in adults with Asperger syndrome. Journal of Autism and Developmental Disorders. 2013;43(4):891–910. doi: 10.1007/s10803-012-1633-6. [DOI] [PubMed] [Google Scholar]
  83. Stewart DW. On the meaningfulness of sensory attributes: Further evidence on the attraction effect. Advances in Consumer Research. 1989;16(1):197–202. [Google Scholar]
  84. Stiegler LN, Davis R. Understanding sound sensitivity in individuals with autism spectrum disorders. Focus on Autism and Other Developmental Disabilities. 2010;25(2):67–75. doi: 10.1177/1088357610364530. [DOI] [Google Scholar]
  85. Thapa R, Alvares GA, Zaidi TA, Thomas EE, Hickie IB, Park SH, Guastella AJ. Reduced heart rate variability in adults with autism spectrum disorder. Autism Research. 2019;12(6):922–930. doi: 10.1002/aur.2104. [DOI] [PubMed] [Google Scholar]
  86. Truzzi A, Setoh P, Shinohara K, Esposito G. Physiological responses to dyadic interactions are influenced by neurotypical adults’ levels of autistic and empathy traits. Physiology and Behavior. 2016;165:7–14. doi: 10.1016/_j.physbeh.2016.06.034. [DOI] [PubMed] [Google Scholar]
  87. Tversky A. Elimination by aspects: A theory of choice. Psychological Review. 1972;79(4):281–299. doi: 10.1037/h0032955. [DOI] [Google Scholar]
  88. Tyndall I, Ragless L, O’Hora D. Effects of perceptual load and socially meaningful stimuli on crossmodal selective attention in autism spectrum disorder and neurotypical samples. Consciousness and Cognition. 2017;60:25–36. doi: 10.1016/j.concog.2018.02.006. [DOI] [PubMed] [Google Scholar]
  89. Urai AE, Braun A, Donner TH. Pupil-linked arousal is driven by decision uncertainty and alters serial choice bias. Nature Communications. 2017;8(1):1–11. doi: 10.1038/ncomms14637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  90. Vella L, Ring HA, Aitken MR, Watson PC, Presland A, Clare IC. Understanding self-reported difficulties in decision-making by people with autism spectrum disorders. Autism. 2018;22(5):549–559. doi: 10.1177/1362361316687988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  91. Vohra R, Madhavan S, Sambamoorthi U. Emergency department use among adults with autism spectrum disorders (ASD) Journal of Autism and Developmental Disorders. 2016;46(4):1441–1454. doi: 10.1007/s10803-015-2692-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  92. Wechsler D. Wechsler Abbreviated Scale of Intelligence, Second Edition (WASI-II) NCS Pearson; 2011. [Google Scholar]
  93. World Health Organization. The ICD-10 classification of mental and behavioural disorders: Clinical descriptions and diagnostic guidelines. World Health Organization; 1992. [Google Scholar]
  94. Zaboski B, Storch E. Comorbid autism spectrum disorder and anxiety disorders: A brief review. Future Neurology. 2018;13(1):31–37. doi: 10.2217/fnl-2017-0030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  95. Zazzi H, Faragher R. ‘Visual clutter’ in the classroom: Voices of students with autism spectrum disorder. International Journal of Developmental Disabilities. 2018;64(3):212–224. doi: 10.1080/20473869.2018.1468619. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material
EMS173941-supplement-Supplementary_Material.pdf (542.6KB, pdf)

RESOURCES