medRxiv preprint doi: https://doi.org/10.1101/2020.08.04.20168179.this version posted August 6, 2020.

The copyright holder for this

preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in
perpetuity.
All rights reserved. No reuse allowed without permission.

Descriptive analysis of the Autism Spectrum Quotient (AQ) in a sample of Brazilian adults

Ana Luíza Costa Alves1; Jonas Jardim de Paula1,2; Débora Marques de Miranda1,3; Marco Aurélio

Romano-Silva1,4

1 – Programa de Pós-Graduação em Medicina Molecular, Faculdade de Medicina, Universidade Federal de Minas

Gerais. Belo Horizonte-MG, Brazil

2 – Faculdade de Ciências Médicas de Minas Gerais, Belo Horizonte-MG, Brazil

3 – Departamento de Pediatria, Faculdade de Medicina, Universidade Federal de Minas Gerais. Belo Horizonte-MG,
Brazil

4 – Departamento de Saúde Mental, Faculdade de Medicina, Universidade Federal de Minas Gerais. Belo Horizonte-
MG, Brazil

ABSTRACT

Background: The Autism Spectrum Disorder is characterized by the presence of difficulties in social

interaction, inflexible, repetitive and/or stereotyped behaviors and interests. The Autism Spectrum Quotient

(AQ) is a self-report instrument with 50 items created to quantify autistic traits in individuals older than 18

years with average or above-average intelligence.

Objectives: The principal aim was to present a brief descriptive analysis of the AQ in a heterogeneous

sample of Brazilian adults, also, to measure the clinical validity of the scale.

Method: We recruited all the participants in Brazil (N=1024). Then we described the distribution in the

general population (N=385) and investigated the AQ accuracy in a sample of autistic adults (N=32).

Results: Our results suggested that autism traits were normally distributed in the population, but Brazilian

adults have shown a different profile from the original study. Further, we found that 24 adults from the

sample had a clinical score on the AQ, compatible with their previous autism diagnosis.

Discussion: Our population probably reports more symptoms compared to other because the original

clinical score represents a lower percentile in our sample. Also, future work will be required to adequate the

use of the AQ in the Brazilian population.

KEYWORDS: Autism Spectrum Disorder; Asperger syndrome; psychological assessment; psychometrics;

transcultural psychology
medRxiv preprint doi: https://doi.org/10.1101/2020.08.04.20168179.this version posted August 6, 2020. The copyright holder for this
preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in
perpetuity.
All rights reserved. No reuse allowed without permission.

INTRODUCTION

People with Autism Spectrum Disorder (ASD) show difficulties in two main areas: social interaction,

including verbal and non-verbal communication deficits, and inflexible, repetitive and stereotyped behaviors

and interests (APA, 2013). These traits could be perceived in the early development, the diagnosis is

currently made around the age of 3 years, although some behaviors can emerge as early as during the first

year. Those symptoms impact functionality in different domains, including relationships, professional life,

academic outputs, and mental health (Purpura et al., 2020). ASD is a heterogeneous condition seen as a

spectrum, where people may differ in how intense the symptoms are, which may range from mild to severe

presentation. This aspect will determine the impairments and the kind of support the person will need.

The prevalence of ASD around the world and across all ages is approximately 1% (APA, 2013) but

evidence also suggests that autism traits show a normal distribution in the typical population (Constantino

and Todd, 2003). Major difficulties with social behavior, which is a core symptom of autism, could be

common in people with no such diagnosis (Constantino and Todd, 2003; Rutter, 2011). The assessment of

ASD in childhood is well documented in the literature with many adapted and validated clinical instruments.

However, this scenario is different when we consider adults with ASD, especially when the clinical condition

is less severe, with average intelligence and no delay in language development.

With the proposal to quantify ASD traits in individuals older than 18 years old with IQ in the normal

range, Baron-Cohen et al. (2001) developed the Autism Spectrum Quotient (AQ), which is a self-report

questionnaire with 50 items that are divided into five different domains: social skills, imagination,

communication, attention switching, and attention to details. The cut-off score can identify the number of

symptoms the individual present, classifying for severity, and for the need and kind of support necessary.

Although the diagnosis of autism should be made by a team of multidisciplinary professionals, the use of

instruments and questionnaires have the objective to give support to the diagnostic process. Thus, the AQ

scale enables health professionals to identify some important traits of autism in the adult population. The

AQ was adapted for many cultures, including for the Brazilian Portuguese by Egito et al. (2018).

As the clinical importance of this instrument and the normal distribution of those traits in the general

population are known, we aimed to briefly present a descriptive analysis of AQ scores in a sample of

Brazilian adults without autism diagnosis or any other psychiatric condition, followed by validation of its

accuracy in a sample of adults with ASD.

medRxiv preprint doi: https://doi.org/10.1101/2020.08.04.20168179.this version posted August 6, 2020. The copyright holder for this
preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in
perpetuity.
All rights reserved. No reuse allowed without permission.

METHODS

This study was approved by the Research Ethics Committee of the Federal University of Minas

Gerais-UFMG and the consent was obtained from all volunteers before the study. The research was

organized in different stages. In the first one, we adopted an online platform for data collection. Subjects

were invited by direct mailing and in social media, especially targeting groups of researchers which work

with autism and groups of autism patients and relatives, along with other people who would like to volunteer.

We received 1024 valid form submissions in our server. A series of exclusion criteria were applied in this

initial sample: age below 18 years, self-reported history of mental disorders or neurological diseases, use

of psychotropic medication and scores above the cut-off score for mental disorders (>7) in the Self

Reporting Questionnaire-20 (SRQ-20) (Mari & Williams, 1985). The final sample of this study involved 385

individual (294 women), with mean age of 34.3 years (SD=11.3, range=18-68).

Then, we invited part of the sample who got a clinical score (>31, the international cutoff) on the

AQ or reported a previous diagnosis of ASD to a diagnostic interview, following the American Psychiatric

Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM-V) criteria. Only participants

which lived in Belo Horizonte-MG and were interested in the procedure, as reported in our online form were

invited. This subsample was formed by 32 voluntaries (23 women), with a mean age was 33.6 (SD=8,

range=20-50) and 16.5 (SD=3.5) years of formal schooling. The assessment of the 32 adults was performed

by ALC and discussed with a clinical neuropsychologist (JJdP) and a psychiatrist (MAR-S).

Descriptive data were computed using the mean as a central tendency measure from the sample.

Standard deviation, minimum and maximum values were also calculated. Normative values were defined

using percentile scores and to assess the AQ reliability the Cronbach's alpha was calculated.

RESULTS

Based on AQ responses, we computed descriptive parameters using standard scores and

percentiles following the original (Baron-Cohen et al., 2014) and Brazilian-adapted (Egito et al., 2018)

scoring systems. AQ scores showed a normal distribution according to histogram analysis, and the mean

score was 20.9 (SD=8.8).

medRxiv preprint doi: https://doi.org/10.1101/2020.08.04.20168179.this version posted August 6, 2020. The copyright holder for this
preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in
perpetuity.
All rights reserved. No reuse allowed without permission.

We found good internal consistency in both genders (α=0,85 and 0.87), which means that the items

reliably measured the same construct. Of the total of 91 men in our first sample, voluntaries with 32 points

or more (the cut-off proposed by Baron-Cohen and colleagues) scored higher than approximately 93% of

the control sample. Of 294 women, those who obtained 32 points or more, scored higher than approximately

97% of our sample (Table 1). In the original study, 32 points represented the 98th percentile (computed

from the mean and standard deviations reported in the original paper). The study conducted by Osorio

(Egito et al., 2018) examined the factor structure of the Brazilian version of the scale, and they proposed a

three-factor model instead of five, a reduced version (25 items) and a different way to correct it. Besides

that, they did not indicate a different cut-off for their sample. The distribution of AQ scores according to this

method is shown in Table 1.

Table 1

Descriptive data of Autism Quotient scores stratified by scoring method and sex

Cohen et al. (2001) Egito et al. (2018)

Male (N=91) Female (N=294) Male (N=91) Female (N=294)
Mean 25 20 62 54
Std. Dev. 8 9 12 13
Min 5 3 40 28
Max 43 45 86 94

Pc.5 5 5 40 38
Pc.10 10 7 44 41
Pc.20 14 11 47 44
Pc.30 16 13 50 46
Pc.40 18 15 52 48
Pc.50 20 17 54 50
Pc.60 22 19 56 52
Pc.70 24 21 58 54
Pc.80 27 23 60 52
Pc.90 30 27 64 59
Pc.95 34 30 66 62
Reliability 0.87 0.85 0.76 0.82

Pc.: Percentile
medRxiv preprint doi: https://doi.org/10.1101/2020.08.04.20168179.this version posted August 6, 2020. The copyright holder for this
preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in
perpetuity.
All rights reserved. No reuse allowed without permission.

In the second part of the study, based on the 32 score cutoff, all 32 adults got a clinical score on

the AQ scale, but 24 also had a previous autism diagnosis. Two of them did not present enough

characteristics to receive an autism diagnosis, but showed symptoms of other mental disorders (social

phobia and generalized anxiety). Six volunteers of this group were classified as non-clinical although they

showed borderline scores in AQ.

DISCUSSION

The Autism Spectrum Disorder was considered a diagnostic category, until the last edition of DSM

have proposed a dimensional view for the clinical condition. ASD varies according to symptoms, severity,

and need for support. It is expected that any population present from subclinical traits of ASD to very severe

condition.

As seen in other populations AQ scores were normally distributed in our sample, and people with

autism showed very high scores in the questionnaire (above the 90th percentile). AQ also showed good

reliability, both using the original (Baron-Cohen et al., 2001) and adapted (Egito et al., 2018) scoring

systems. In our second sample, which showed scores above the cutoff score or reported a previous

diagnosis of autism, we found 2 false-positive and 6 false-negative cases.

The availability of AQ as a screening instrument for autism traits in adults have huge importance,

especially in the diagnostic process of adults with fewer impairments and preserved intelligence. These

cases could be a challenge for health professionals. Besides that, knowing the level of autism symptoms

enables the clinician to predict impairments, to offer adequate support, and to provide better guidance to

the family. It is important to keep in mind that the scale alone is not enough for the diagnosis, but still a very

useful tool in the investigation.

Finally, our results suggest that the studied population have a different profile compared to the

original study, with clinical scores occuring at a lower percentile in our sample, which means that probably

Brazilian subjects report more symptoms than other populations. Despite this, we observed that those traits

were distributed in the typical population. Future studies are required to adequate the use of Autism

Spectrum Quotient in the Brazilian population, such as to define a cut-off score that will better consider its

culture and peculiarities.

medRxiv preprint doi: https://doi.org/10.1101/2020.08.04.20168179.this version posted August 6, 2020. The copyright holder for this
preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in
perpetuity.
All rights reserved. No reuse allowed without permission.

ACKNOWLEDGMENTS

Alves ALC received an scholarship from Conselho Nacional de Desenvolvimento Científico e Tecnológico

(CNPq). JJdP was the recipiente of a CAPES fellowship. DMM and MAR-S are CNPq research fellows.

Finantial support from CNPq and FAPEMIG.

REFERENCES

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-5).

American Psychiatric Pub.

Apicella, F., Costanzo, V., & Purpura, G. Are early visual behavior impairments involved in the onset of

autism spectrum disorders? Insights for early diagnosis and intervention. European Journal of Pediatrics,

1-10.

Baron-Cohen, S., Cassidy, S., Auyeung, B., Allison, C., Achoukhi, M., Robertson, S., ... & Lai, M. C. (2014).

Attenuation of typical sex differences in 800 adults with autism vs. 3,900 controls. PloS one, 9(7), e102251.

Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley, E. (2001). The autism-spectrum

quotient (AQ): Evidence from asperger syndrome/high-functioning autism, malesand females, scientists

and mathematicians. Journal of autism and developmental disorders, 31(1), 5-17.

Constantino, J. N., & Todd, R. D. (2003). Autistic traits in the general population: a twin study. Archives of

general psychiatry, 60(5), 524-530.

Egito, J. H. T., Ferreira, G. M. R., Gonçalves, M. I., & Osório, A. A. C. (2018). Brief Report: Factor Analysis

of the Brazilian Version of the Adult Autism Spectrum Quotient. Journal of autism and developmental

disorders, 48(5), 1847-1853.

Mari, J. D. J., & Williams, P. (1985). A comparison of the validity of two psychiatric screening questionnaires

(GHQ-12 and SRQ-20) in Brazil, using Relative Operating Characteristic (ROC) analysis. Psychological

medicine, 15(3), 651-659.

Rutter, M. L. (2011). Progress in understanding autism: 2007–2010. Journal of autism and developmental

disorders, 41(4), 395-404.