Research and methodology

Ritvo Autism Asperger Diagnostic Scale – Revised (RAADS-R)

The RAADS-R is an 80-item self-report questionnaire designed specifically for adults aged 16 and over. It was developed by Riva Ritvo and colleagues and published in 2011, with the explicit goal of providing a tool that could assist clinicians in identifying autism spectrum characteristics in people who had not received a childhood diagnosis.

The questionnaire covers four domains: Social Relatedness (39 items), Language (7 items), Sensory/Motor (20 items), and Circumscribed Interests (14 items). Its time-anchored response format asks whether traits were present in childhood, now, or both, which gives it more diagnostic specificity for adults than tools that ask only about the present.

The RAADS-R includes 17 normative items that score in the opposite direction to the remaining 63. These normative items describe common non-autistic experiences and act as a validity check. The total score range is 0 to 240, with a threshold of 65 from the original validation study.

Known biases and limitations

The original validation was conducted with a clinically referred sample, meaning people who had already been assessed for or diagnosed with autism. In that population the RAADS-R showed 97% sensitivity and 100% specificity. In community samples (people who have not been clinically referred), the false-positive rate is meaningfully higher. People with anxiety, depression, OCD, or other mental health experiences can score above 65 without autistic traits being the primary explanation. A high score is a signal worth discussing with a professional, not a result that stands alone. Similarly, autistic traits can sit below 65, particularly for people who camouflage; the CAT-Q is a useful companion measure for this reason. The RAADS-R was developed primarily in Western clinical settings and has been most extensively validated in English-speaking populations.

Camouflaging Autistic Traits Questionnaire (CAT-Q)

The CAT-Q was developed by Laura Hull and colleagues at University College London and published in 2019. It was the first dedicated measure of autistic camouflaging: the set of strategies autistic people use to conceal or suppress their autistic traits in social contexts, often at significant personal cost.

The questionnaire is 25 items across three subscales: Compensation (learning social rules explicitly, memorising scripts, mirroring others), Masking (monitoring and controlling body language and facial expressions), and Assimilation (forcing social interaction, performing a social role, avoiding situations). Total scores range from 25 to 175; higher scores indicate more frequent camouflaging.

The CAT-Q matters for Traitmark because camouflaging directly suppresses scores on trait-focused instruments like the AQ and even the RAADS-R. An autistic person who camouflages extensively may score below the threshold on those measures, giving a misleading impression. The CAT-Q makes the concealment itself visible.

Known biases and limitations

The CAT-Q does not identify autism. Many non-autistic people camouflage in social contexts, for reasons including anxiety, social pressure, or cultural expectations. High camouflaging scores are more informative when read alongside a broader autism trait measure. The validation sample was predominantly female and from Western English-speaking countries; cross-cultural replication continues. The instrument is most appropriate for adults aged 16 and over.

Perth Alexithymia Questionnaire (PAQ)

The PAQ was developed by David Preece and colleagues at Edith Cowan University in Perth, Australia, and published in 2018. It measures alexithymia: difficulty identifying and describing one's own emotions. The term comes from Greek and translates roughly as "no words for feelings."

Alexithymia is common in autistic people, estimated to affect around 50% of autistic adults in some research, compared to around 10% of the general population. It is not unique to autism, however; it also appears in people with histories of trauma, chronic stress, depression, and other experiences. The PAQ does not measure autism directly.

The PAQ is 24 items across five subscales covering difficulty describing negative feelings, difficulty identifying negative feelings, difficulty describing positive feelings, difficulty identifying positive feelings, and externally oriented thinking (a preference for concrete and practical thinking over emotional reflection). Its range is 24 to 168; higher scores indicate greater alexithymia.

The PAQ was chosen for Traitmark over the older Toronto Alexithymia Scale (TAS-20) because it explicitly covers positive emotions as well as negative ones. The TAS-20 was developed primarily around negative and painful emotional states, which made it a poor fit for autistic adults whose alexithymia often extends equally to joy, excitement, and other positive affects.

Known biases and limitations

The PAQ was developed and initially validated in Australian adult samples; subsequent cross-cultural validation studies have been conducted in Spanish and other populations. High scores reflect a self-report of difficulty with emotional identification and description; they do not indicate that a person lacks emotional depth or experience. The threshold of 118 for high alexithymia was established in the original Preece et al. (2018) paper.

Adult ADHD Self-Report Scale Part A Screener (ASRS-V1.1)

The ASRS Part A Screener is a 6-item questionnaire developed jointly by the World Health Organisation and Harvard Medical School, published in 2005 by Kessler and colleagues. It was designed as a rapid first-stage screening tool for use in the general adult population.

The six items were selected through psychometric analysis as the most predictive subset of the full 18-item ASRS. Items 1 to 4 cover inattentive symptoms; items 5 and 6 cover hyperactive and impulsive symptoms. Scoring uses item-specific thresholds: items 1 to 3 use "sometimes" as the threshold for a positive score, and items 4 to 6 use "often." A positive screen result requires 4 or more items to meet or exceed their respective thresholds.

ADHD is significantly under-recognised in women, girls, and people assigned female at birth. Inattentive presentations are frequently missed. Hyperactivity in female-presenting people often looks different from the visible restlessness that early clinical descriptions focused on. Many adults received diagnoses of anxiety or depression before ADHD was identified, because the inattentive features were misread as mood symptoms.

Known biases and limitations

At a 4-item threshold, the ASRS Part A has sensitivity of 68.7% and specificity of 99.5% in the original validation study. Its high specificity means a positive screen is a strong signal; its moderate sensitivity means a negative screen does not rule ADHD out, particularly for inattentive presentations and people whose ADHD manifests primarily as internal restlessness rather than visible behaviour. The screener is a starting point for a clinical conversation, not a diagnostic result.

Autistic Burnout Scale Revised (AABS-R14 / ABM)

The Autistic Burnout Scale Revised is a 14-item screening measure developed by the Academic Autism Spectrum Partnership in Research and Education (AASPIRE). It was developed by and with autistic adults, using community-based participatory research methods. The original burnout construct was described in Raymaker et al. (2020), with the scale development work by Nicolaidis and colleagues completed in 2026.

Autistic burnout is a period of pervasive exhaustion, increased difficulties with daily functioning, and reduced capacity for coping that results from the cumulative experience of navigating a world not designed for autistic people. It is distinct from general burnout and from depression, though it can co-occur with both.

The ABM asks about changes over the past three months compared to what is usual for you. It covers cognitive difficulties, emotional regulation, sensory sensitivity, stress responses, communication, self-care, daily activities, task completion, social withdrawal, avoidance, memory, and exhaustion. Scores range from 0 to 56; tentative AASPIRE cutoffs place 33 and above in the high probability range for clinically significant burnout.

Known biases and limitations

The ABM is recently developed, and the psychometric literature is still emerging. The cutoffs are described as tentative by AASPIRE. This is a self-report screening tool, not a clinical assessment of burnout severity. Because the ABM asks about changes relative to your personal baseline, it is sensitive to how clearly a person can identify what is typical for them. People in chronic burnout may not recognise how far they have shifted from their own baseline.

Autism Spectrum Quotient (AQ)

The AQ is a 50-item self-report questionnaire developed by Simon Baron-Cohen and colleagues at the Autism Research Centre, University of Cambridge, and published in 2001. It covers five domains: social skill, attention switching, attention to detail, communication, and imagination. It is one of the most widely used autism screening tools in research and clinical practice.

Each item uses a four-point agreement scale, but the AQ scores each item as either 0 or 1. Only agreement or disagreement counts, not the degree. The maximum score is 50. A score of 32 or above is the most commonly cited screening threshold; some research uses 26 when higher sensitivity is needed.

Known biases and limitations

This test was developed primarily with male university students. Research consistently finds that autistic women, non-binary people, and people who have learned to suppress or hide autistic traits score lower than expected on the AQ, even when many autistic traits are present in their lived experience. A study by Lai et al. (2015) found significant sex differences in how autistic traits present, and standard instruments including the AQ do not fully account for this. A score below 26 does not rule out autistic traits, particularly where camouflaging is present. Reading the AQ result alongside the CAT-Q gives a more complete picture for people who camouflage.

Empathy Quotient (EQ)

The EQ is a 60-item questionnaire developed by Simon Baron-Cohen and Sally Wheelwright at the Autism Research Centre and published in 2004. It contains 40 scored items and 20 filler items included for validity purposes.

Unlike the AQ, the EQ uses a graded scoring system: strong endorsement of the scored direction earns 2 points, slight endorsement earns 1 point. The maximum score is 80. In the original study, autistic adults averaged around 20 and non-autistic adults scored around 42 (male) and 47 (female). A score of 30 or below was associated with 81% of autistic adults in the original validation.

The EQ measures how people respond to items about empathy, social attunement, and caring for others. It captures one facet of how empathy-related responses show up in self-report. It does not measure empathic capacity overall.

Known biases and limitations

The EQ was developed within a theoretical framework that treated autistic people as having low empathy. That framing is not supported by current research and is rejected by most of the autistic community. Research on the "double empathy problem" (Milton, 2012) reframes social differences as a mismatch between different kinds of minds, rather than a deficit in autistic empathy. Many autistic people experience strong emotional responses and deep caring; the EQ's items may not capture the specific forms these take.

The EQ was normed predominantly on male samples. Lower scores in autistic women and non-binary people may reflect masking, different presentations, or measurement bias as much as genuine differences in empathic responding. No formal measurement invariance by gender has been established for the EQ-60.

Systemising Quotient Revised (SQ-R)

The SQ-R is a 75-item questionnaire developed by Sally Wheelwright, Simon Baron-Cohen, and colleagues, published in 2006. It measures the drive to analyse, explore, and construct systems governed by predictable rules. High scores are associated with autistic traits and with systematic cognitive styles more broadly.

The questionnaire covers mechanical, abstract, social, domestic, and natural systems. Items use a graded scoring system: strong agreement with agree-keyed items earns 2 points, slight agreement earns 1. The total range is 0 to 150. The SQ-R is unidimensional: factor analysis and Rasch modelling support a single-factor structure, so all 75 items contribute to one total score.

In the original normative sample, autistic adults averaged 77.8 (males) and 76.4 (females). Non-autistic adults showed a larger sex difference: males averaged 61.2 and females 51.7. This gap is one of the more notable features of the SQ-R data.

Known biases and limitations

Sex differences in SQ-R scores for the general population are substantial: males score higher than females on average by around 10 points. The reasons for this are debated; it may reflect genuine differences in cognitive style, socialisation differences, or measurement bias. A female-presenting person with a mid-range score may still have significant systematising tendencies. Gender measurement invariance has not been established for the SQ-R. The normative sample was drawn from UK university students and clinic-referred adults, which may not represent all populations. Several items reference technology from 2006 (video recorders, CD collections) that has dated; these items remain in the validated instrument and have not been reworded.

Friendship Questionnaire (FQ)

The FQ was developed by Simon Baron-Cohen and Sally Wheelwright at the Autism Research Centre and published in 2003. It measures individual differences in friendship style, focusing on depth of intimacy, social motivation, and empathic relating. A high score indicates the respondent reports enjoying close, empathic, supportive friendships; being interested in people; and enjoying interaction for its own sake.

The FQ has 35 items, but only 28 are scored. The format is unusual: items 1 to 15 use forced-choice multiple choice, items 16 to 27 use rated scales, and items 28 to 33 use categorical frequency or need scales. Items 30 and 34 use ranking. The total range is 0 to 140 (corrected from 135 in the original paper; the ARC's own scoring note records this correction). The FQ is not a measure of relationship quality or loneliness; it measures friendship style.

Known biases and limitations

The original normative sample was small (76 controls, 68 autistic adults) and drawn from a Western UK population. The original female control mean was 90.0 and the male mean was 70.3, a 19.7-point gap that reflects a real sex difference in this friendship style dimension. Current score bands do not stratify by gender, which means results for men may read as lower than they would with gender-adjusted norms. The FQ was developed within the "extreme male brain" theoretical framework of autism, which is contested. The questionnaire itself measures friendship style rather than pathology; the ARC is explicit that "a particular score on the FQ is not indicative of any need for intervention." A 2019 replication study by Holt et al. with 949 participants provided stronger normative data and found autistic adults averaged 56.15 and non-autistic adults averaged 79.09.

Glasgow Sensory Questionnaire (GSQ)

The GSQ was developed by Ashley Robertson and David Simmons at the University of Glasgow and published in 2013. It is a 42-item questionnaire that covers hyper- and hyposensory experiences across seven sensory modalities: visual, auditory, gustatory (taste), olfactory (smell), tactile, vestibular (balance and movement), and proprioceptive (body position sense).

Unlike most sensory questionnaires, the GSQ captures both directions of sensory difference. Hypersensitivity means a lower threshold: more easily overwhelmed by or reactive to sensory input. Hyposensitivity means a higher threshold: needing more input to register the same sensation. Both patterns are common in autistic people, often coexisting within the same person across different modalities.

The GSQ produces a total score (0 to 168), a hypersensitivity score (0 to 84), a hyposensitivity score (0 to 84), and 14 subscale scores covering each modality in each direction. The mean score in the original UK general population sample was 56.65 (SD 23.60).

Known biases and limitations

No Australian normative data exists for the GSQ. All published norms come from UK, Dutch, and German populations, which limits the applicability of published cut-offs to Australian users. Validation studies in multiple countries have found that the 14-subscale model is difficult to replicate; the two-factor (hyper/hypo) model fits better across populations. Several individual items have shown psychometric problems in German and Dutch validation studies, including items covering olfactory hyposensitivity. The GSQ's licence status is in the process of being confirmed with the original authors; the instrument is included on Traitmark with appropriate attribution.

Wender Utah Rating Scale 25-item (WURS-25)

The WURS-25 was developed by Ward, Wender, and Reimherr and published in 1993. It is a retrospective self-report questionnaire: adults look back and describe their behaviour and emotions before the age of 12. It was designed to establish childhood onset, which is a diagnostic criterion for ADHD under DSM criteria.

The WURS-25 covers inattention, impulsivity, emotional dysregulation, social difficulties, and academic problems as recalled from childhood. It scores from 0 to 100 across five 0-to-4 scale items. Three cutoff thresholds are supported by research: 36 (maximises sensitivity, catches 96% of ADHD cases), 39 (optimal balance of sensitivity and specificity), and 46 (original Wender threshold, high specificity at 99%).

Traitmark uses the WURS-25 alongside the ASRS Part A because together they approximate the two-part picture a clinician would want: current symptoms and childhood onset. Neither replaces a clinical assessment.

Known biases and limitations

Retrospective recall is imperfect by nature. Many people with ADHD were taught early to suppress or play down what they were struggling with, or had their patterns labelled in ways that did not name ADHD (lazy, defiant, distracted, not reaching their potential). Real childhood patterns can sit under a low score if the person has internalised these reframings. The original validation cohort was 74% female, which gives the WURS-25 better gender coverage than most ADHD instruments. Several items use 1990s clinical language that may read as harsh to a contemporary audience; the wording must be reproduced accurately for validity purposes. The licence status is being confirmed with the American Journal of Psychiatry rights desk prior to full public launch.

Sensory Perception Quotient (SPQ)

The SPQ was developed by Teresa Tavassoli, Rosa Hoekstra, and Simon Baron-Cohen at the Autism Research Centre and published in 2014. The version on Traitmark is the 35-item short form drawn from a longer 92-item full questionnaire. It measures sensory hypersensitivity across five modalities: touch, hearing, vision, smell, and taste.

The SPQ's scoring works in reverse to most instruments: lower scores indicate greater sensory sensitivity. Agreement with an item like "I can hear electricity humming in the walls" earns 0 points (very sensitive), while disagreement earns 3 points (less sensitive). The 5 hyposensitive items in the short form are reverse-scored so the scale direction is consistent.

In the original validation, autistic adults averaged 38.55 (SD 18.68) and non-autistic adults averaged 43.01 (SD 14.67). The difference was statistically significant, but the group overlap is substantial. No validated clinical cutoff exists for the short form.

Known biases and limitations

The short form primarily assesses hypersensitivity (30 of 35 items). People with predominantly hyposensitive profiles are less well captured. A revised scoring system (SPQ-RS, Taylor et al. 2020) separates hyper- and hyposensitivity into distinct dimensions and is recommended for gender-fair measurement; the original single-score approach has been found to mask sensory hypersensitivity in autistic women. The short form has not been revalidated against the SPQ-RS framework. No Australian normative data exists. The instrument was developed and validated in adults only.

Two-Factor Imagination Scale (TFIS)

The TFIS was developed by Jason Thompson and published in 2008 in Psychotherapy in Australia. It is a 22-item instrument measuring imaginative style across two factors: Spontaneous Imagining (imagination that arises without conscious effort, characterised by effortlessness, surprise, and instantaneity) and Controlled Imagining (deliberate, directed imagination, characterised by initiation, guidance, and termination).

The TFIS was developed in the context of research on alexithymia: Thompson was interested in how imaginal activity relates to emotional processing and whether differences in imaginative style could illuminate the experience of people with alexithymia. It is based on philosopher Edward Casey's theoretical framework of imaginal traits.

The TFIS uses a two-option response format (More Often True / Less Often True) rather than a Likert scale. The maximum score is 66; higher scores indicate more spontaneous imagining. Lower scores indicate more controlled imagining. Neither direction is more desirable; these are descriptions of how imagination tends to work, not assessments of creativity or capacity.

Known biases and limitations

The TFIS is the instrument on Traitmark with the smallest validation sample; the preliminary validation involved eight participants. Treat its psychometric properties as indicative rather than established. The instrument has not been validated in large-scale clinical or community samples, and no formal normative data exists. The bands in Traitmark's results are interpretive starting points rather than clinical thresholds. The TFIS measures imaginative style, not imagination in a general sense. It does not assess creativity, intelligence, or wellbeing. Its licence is being confirmed with the author prior to full public launch.