An IQ test for autistic adults: what to take, what the research shows and how to read the profile

0 The short answer

An IQ test cannot detect, screen for or diagnose autism, so an autistic adult, or an adult who wonders, should take one only to see a profile of cognitive abilities, and should choose a test with clear instructions, low language load in the reasoning tasks and a setting the person controls. Group research finds autistic people across a wide range of measured ability, with profiles that differ from person to person, and a 2007 study found Raven's Progressive Matrices scores 30 percentile points above Wechsler scores, on average, in 38 autistic children, while a 2009 study that set out to replicate it found a gap of 9 percentile points. ACIS is one option, a normed online battery at 15, 30 or 50 dollars one time (prices read October 28, 2026) and about 45, 110 or 175 minutes, taken at home with breaks allowed, and its limits are that it is unsupervised, English only and not a clinical instrument, so it cannot replace an evaluation by a psychologist.

1 Can an IQ Test Tell You Whether You Are Autistic?

No: an intelligence test measures abilities such as reasoning, vocabulary, memory and speed, while autism is identified through a comprehensive assessment of social communication, repetitive behavior, restricted interests, developmental history and sensory sensitivities, and the two are separate questions.

A public statement of how adults are assessed is the NICE guideline CG142, Autism spectrum disorder in adults: diagnosis and management, which showed a publication date of June 27, 2012, a last updated date of June 14, 2021 and a last reviewed date of July 23, 2026 when we read it on October 28, 2026 (the page says that review will not update the recommendations on adapted trauma focused treatment). It comes from a United Kingdom institute, and other countries organize their pathways differently, but its content shows what an assessment contains. Recommendation 1.2.2 says to consider an assessment for possible autism when a person has persistent difficulties in social interaction or communication, or rigid and repetitive behaviors, resistance to change or restricted interests, together with a further difficulty such as problems obtaining or sustaining work or education. Recommendation 1.2.3 says that for adults with possible autism who do not have a moderate or severe learning disability, clinicians may consider the 10 item Autism-Spectrum Quotient (AQ-10), and that a score of 6 or above, or a suspicion of autism based on clinical judgment, should lead to an offer of a comprehensive assessment. The AQ-10 is a short questionnaire that helps decide who is offered that assessment. It is not an IQ test, and a score on it is not a diagnosis.

The comprehensive assessment itself, in recommendation 1.2.5, should be undertaken by trained and competent professionals, be team based and, where possible, involve an informant or documentary evidence of early development. Recommendation 1.2.7 lists what the clinician asks about: core autism features present in childhood and continuing into adulthood, early developmental history, functioning at home, in education or in employment, other conditions, and sensory sensitivities and attention to detail, with direct observation of core features. Recommendation 1.2.8 names formal tools for complex cases, among them the Adult Asperger Assessment, the Autism Diagnostic Interview Revised and the Autism Diagnostic Observation Schedule Generic. These are interviews, observation schedules and questionnaires, and none of them is an intelligence test. The American Psychiatric Association describes the condition on its public autism page, which cites the DSM-5-TR and which we read on October 28, 2026, as persistent challenges with social communication together with restricted interests and repetitive behavior. We use that source only as a pointer, and a diagnosis belongs to a clinician.

Cognitive testing does appear inside some autism evaluations. The Marquette University Autism Clinic page we opened says every evaluation there includes the Autism Diagnostic Interview Revised, the Autism Diagnostic Observation Schedule, cognitive testing with an age appropriate test and adaptive behavior testing, so the cognitive test sits beside the autism specific measures and describes abilities. NICE recommendation 1.1.3 points the same way: it says professionals who support autistic adults should have a broad understanding of the potential discrepancy between intellectual functioning as measured by IQ and adaptive functioning, reflected for example in difficulties in planning and performing activities of daily living, including education or employment. A study of 229 adults with autism spectrum disorder who completed the WAIS-IV in two Italian public health centers (Cicinelli and colleagues, 2022) reaches the same position in its abstract, which says that a test cannot replace diagnostic and clinical evaluation by experienced clinicians.

A high score does not suggest that a person is not autistic, a low score does not suggest that a person is, and any website that returns an "autism result" from an intelligence test is claiming something that none of the sources on this page supports. ACIS is not a clinical or diagnostic instrument and does not screen for autism. If you wonder whether you are autistic, the route is a conversation with a general practitioner or another clinician. NICE recommendation 1.2.10 notes that a comprehensive assessment also looks for other conditions, one reason a page on the IQ test for adults with ADHD sits beside this one, and the page on neuropsychological testing for adults explains the longer clinical route that includes cognitive testing.

2 Why Does This Page Say Autistic Adults, and Why No Functioning Labels?

This page uses identity first wording, "autistic adults", because a survey of the UK autism community found that term endorsed by a large percentage of autistic adults, and it avoids the labels high functioning and low functioning because a large study found that IQ, which such labels often rest on, is an imprecise predictor of functional abilities.

Kenny, Hattersley, Molins, Buckley, Povey and Pellicano (2016) surveyed 3,470 UK residents, among them autistic people, parents and their wider support network, about the terms they use for autism, and published the results in the journal Autism. The most highly endorsed terms were "autism" and "on the autism spectrum", with agreement across groups. The groups disagreed on others. According to the abstract, "autistic" was endorsed by a large percentage of autistic adults, family members and friends, and parents, but by considerably fewer professionals, while "person with autism" was endorsed by almost half of professionals but by fewer autistic adults and parents. The authors concluded that no single way of describing autism is universally accepted. This page therefore uses "autistic adults", which matches the wording of the NICE guideline, and says "on the autism spectrum" or "autism spectrum disorder" where a source does. Person first wording is a recognized preference too, and nothing below depends on it.

Alvares and colleagues (2020) studied 2,225 people aged 1 to 18 who were notified to a prospective register at diagnosis, grouped by the presence or absence of intellectual disability, and compared IQ with scores on the Vineland Adaptive Behavior Scales, a measure of everyday functional skills. After controlling for sex, IQ was a weak predictor of Vineland scores. For the group with intellectual disability the adaptive scores were close to the IQ scores, but for children without intellectual disability the adaptive scores fell significantly below IQ, and the gap stayed large with increasing age at diagnosis. The authors argued that "high functioning autism" is an inaccurate descriptor when it rests solely on IQ cut offs and that the term should be abandoned in research and clinical practice. That study was of children and adolescents at diagnosis, so it does not measure any adult's daily life, but it fits the point that NICE makes in recommendation 1.1.3: a measured IQ and daily functioning can diverge.

For a reader who is deciding whether to take an IQ test, the implication is practical. A score on a cognitive battery is an imprecise guide to how much support a person needs, in either direction, because the test does not sample adaptive skills, sensory load, executive demands or communication in daily life. The page on what IQ measures and what it leaves out sets out that boundary for every adult, and the research overview on autism and IQ covers the topic at length. The labels appear on this page only to explain why they are avoided and in the title of a source.

3 What Does Group Research Show About IQ Across the Autism Spectrum?

Group studies find autistic people across a wide range of measured ability, with a distribution that depends on who was studied and how, and none of those distributions says where one adult will score. The three studies in the table describe measured ability in autistic samples, and each answers a different question.

Charman and colleagues drew a sample from an epidemiological study and used a weighting procedure to estimate the characteristics of the whole population of children with autism spectrum disorder. In the abstract, 55 percent had an intellectual disability, defined as IQ below 70, 16 percent had an IQ below 50, 28 percent had average intelligence, defined as an IQ between 85 and 115, and 3 percent were above average. Our arithmetic on those figures leaves about 14 percent between 70 and 85. The authors concluded that autism spectrum disorder was less strongly associated with intellectual disability than traditionally held and that adaptive outcome was significantly impaired, in their words, even for children of average intelligence, and their results state that adaptive skills were significantly lower than IQ. Two further results matter for how a profile is read. A gap between performance and verbal IQ was not associated with a particular pattern of symptoms, and the evidence for a characteristic pattern of subtest peaks and troughs was mixed. The 55 percent figure describes weighted estimates for children aged 10 to 14, and it is not a rate for adults.

Velikonja and colleagues pooled 75 studies of people aged 16 or older that assessed at least one domain of neurocognitive functioning or social cognition with standard measures. Adults with autism spectrum disorder scored lower than neurotypical adults, on average, in several domains. The authors described the difference in processing speed (Hedges g minus 0.61, 95 percent confidence interval minus 0.83 to minus 0.38, from 21 studies) and in verbal learning and memory (g minus 0.55, from 12 studies) as medium sized, and the difference in working memory (g minus 0.23, interval minus 0.47 to 0.01, from 19 studies) as among the least altered.

Two cautions apply to all three. First, none of these samples consists of adults who chose to take an online test, so any group figure is a base rate and not a forecast. Second, a group description cannot be turned into advice about a person: a pooled difference in processing speed does not tell you that your speed is lower, and a population figure for intellectual disability tells you nothing about your reasoning. Cicinelli and colleagues wrote that few conclusions can be drawn from the Full Scale IQ alone, which is the argument this page makes for reading indices. The page on the CHC model of cognitive abilities explains the domain names used in these studies.

4 Do Autistic People Score Higher on Raven's Matrices Than on Wechsler Scales?

In a 2007 sample of 38 autistic children they did, by 30 percentile points on average, and a 2009 study that set out to replicate it found a gap of 9 percentile points, so the gap is large enough to take seriously and variable enough that nobody should assume it for one person, and the 30 point figure comes from children, not adults. Dawson, Soulières, Gernsbacher and Mottron (2007), writing in Psychological Science, assessed a broad sample of 38 autistic children on Raven's Progressive Matrices, which they called the preeminent test of fluid intelligence. Their scores were, on average, 30 percentile points higher than their scores on the Wechsler scales of intelligence, and in some cases more than 70 percentile points higher. Typically developing control children showed no such discrepancy, and the abstract adds that a similar contrast was observed when a sample of autistic adults was compared with a sample of nonautistic adults, without giving a figure for the adults, so the 30 percentile points describes the children. The authors concluded that intelligence has been underestimated in autistics.

Bölte, Dziobek and Poustka (2009) state in the Journal of Autism and Developmental Disorders that they sought to replicate and elaborate the 2007 results. They compared Raven's and Wechsler IQs in 48 individuals with autism, a mixed clinical group of 28 and a neurotypical control group of 25. Raven's IQ was higher than Wechsler IQ only in the autism group, but to a much lesser degree than previously reported, a difference of 9 percentile points against 30 in the results section, and only for individuals whose Wechsler IQ was below 85. The paper reports mean ages of 14.4 years for all participants with Wechsler IQs below 85 and 18.9 years for those above 85, and its analysis found no main effect of age group, defined as under or over 16 years. Neither figure describes adults alone. The authors also note that the 2007 sample had a higher average IQ than theirs. They recommended the Wechsler scales as the first choice IQ measure for people with average or above average IQ, because they give a multidimensional estimate, and said that additional testing with Raven's might be required for people with lower Wechsler IQs, the group in which the gap appeared.

Read together, the studies agree that when a gap appears it runs in one direction, and only in the autism group. They disagree on its size, 30 against 9 percentile points, and on whom it affects, and our reading is that the difference depends on the sample and on the level of Wechsler IQ, since the replication found it only below 85. Neither abstract explains why a Raven's format would raise scores, and this page offers no mechanism.

For a reader choosing a test, three practical points follow, each a conclusion of this page and not a finding of either study. First, a matrix reasoning task is a real view of fluid reasoning (Gf in the CHC model), and a person whose matrix result and vocabulary result differ is showing a profile, not an error. Second, a battery that reports a verbal index and a fluid reasoning index separately lets you see that difference, which a single Full Scale IQ would average away. Third, no single format suits everyone: Bölte and colleagues found the gap only below a Wechsler IQ of 85.

One boundary needs to be stated plainly. The ACIS Matrix Reasoning subtest belongs to the Raven style tradition of visual pattern completion, as its page describes, but it is not Raven's Progressive Matrices and it was not part of either study. A score on it cannot be compared with the 30 percentile point figure, and the professional version of the Raven family is a restricted instrument given by qualified examiners, which the page on Raven's 2 explains. If a Raven's versus Wechsler comparison matters to you, it is a question to bring to a psychologist who can administer both.

5 What Can an Autistic Adult Take, and What Does Each Option Cost?

Four kinds of option exist, a psychologist's evaluation, a clinic that evaluates for autism, a free quiz and a normed online battery, and they answer different questions at prices from nothing to several hundred dollars, so the choice should follow the question. The table sets out what we could verify on October 28, 2026, and prices can change.

The first row comes from the Pearson product page for the WAIS-5, which we opened on October 28, 2026. It states the age range, the completion times above, that the test is given on paper or on the publisher's Q-interactive platform, and that a Nonverbal Index and a Nonmotor Index are available. The intake interview, scoring and feedback come on top (our reading), as the page on the adult IQ testing process describes from first contact to report. The two clinic rows are examples, not a survey. The University of Alabama page says a flat fee applies to adult testing and that all fees range from 250 to 650 dollars, without saying which service costs which amount. The Marquette page describes a 400 dollar flat rate that can be paid in installments, and says plainly that the clinic was not taking new diagnostic clients when we read it, a reminder that a published price is not the same as an available appointment.

How to choose is mostly a matter of purpose. If you want to know whether you are autistic, only a clinical evaluation for autism can answer, as in the two clinic rows, and neither a WAIS-5 score nor an online score can on its own. If you want a documented evaluation for a program, workplace or benefit, that also runs through a psychologist. If you want a profile of abilities for your own understanding, a normed online battery with named indices is a reasonable choice, and a free quiz can serve as a rough first look if you check what it reports. The pages on where to take an IQ test, free IQ tests, the best online IQ tests and the IQ test for adults cover those routes for any adult.

Supervision cuts both ways, and the choice is personal. NICE recommendation 1.2.1 asks staff who assess autistic adults to adapt their procedures, including the setting, the duration and the pacing, which a supervised session makes possible. An unsupervised session lets you control the room, the time of day and the breaks, but nobody can then confirm that the conditions matched those of the people who set the norms.

6 Which Test Conditions Matter Most, and What Can You Control at Home?

The conditions that matter are clear instructions, predictable pacing, a sensory setting you can control and knowing in advance which tasks are timed, and the support for them here is guidance on test fairness and on assessing autistic adults, not a trial of online tests. This page does not cite a study that measured how home conditions change an autistic adult's online score, so what follows comes from standards and guidelines and is labeled as such.

The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014), the joint reference for test developers and users, say in Standard 3.1 that the testing process should be designed to promote valid score interpretations for the widest possible range of individuals. Its comment names universal design as one way to do that, including avoiding where possible characteristics such as test speededness that may compromise valid interpretations for some groups, and providing simple, clear and intuitive procedures and instructions. Standard 3.5 asks developers to document the provisions made to remove construct irrelevant barriers. A reader can use them as a checklist: does the test explain its procedures plainly, and does it say which tasks are timed?

Standard 3.4 and its comment add two cautions that bear on a home session. Test takers should receive comparable treatment, technology based testing raises extra concerns because examinees on older, slower equipment may be disadvantaged, and where unproctored administrations cannot ensure standardization, results should be interpreted with caution. The Standards' discussion of administration adds that the interpreter should be told if a test was unproctored, so if you later show an online result to a clinician, say so.

The environment guidance is more specific. NICE recommendation 1.1.8 says staff should take into account the physical environment in which autistic adults are assessed, and adjust where necessary the personal space, visual supports, colors of walls and furnishings, lighting (reducing fluorescent lighting or increasing natural light) and noise levels (reducing external sounds or advising earplugs or ear defenders). Where the environment cannot be adapted, the duration or nature of an assessment can be varied, including regular breaks. At home you control most of those: a quiet room, lighting you can tolerate, a device you know and the time of day.

Three ACIS conditions are fixed by design, and a buyer should know them before starting. First, three subtests present their items as audio and need working sound: Digit Span, Alphanumeric Sequencing and Arithmetic, and the Digit Span page says numbers play at one per second and advises headphones or a quiet space. Second, several subtests are timed, as the table shows. Third, the session is self paced between subtests, with breaks allowed, progress saved between subtests and 30 days to complete a form.

Changing a condition changes what the score means. The APA's Guidelines for Psychological Assessment and Evaluation (2020) say that when an evaluation requires accommodations or modifications to the administration that have not been researched, the psychologist seeks to adjust the interpretation and report that caveat. The same logic applies to you. Taking breaks between subtests is allowed. Pausing a clock inside a timed subtest or repeating an audio item is a change to the administration, so a score from an altered session should be read as an informal estimate, and ACIS is not for school accommodations. The neighboring page on a dyslexia friendly IQ test covers formats built around reading difficulties.

7 Which ACIS Subtests Carry Little Language, and What Do They Still Ask of You?

Three ACIS subtests whose pages describe a low language load are Matrix Reasoning, Visual Puzzles and Layer Rotation, but low language is not no language, and a nonverbal format does not make a test culture free, bias free or complete.

Layer Rotation is the most extreme case. The ACIS technical manual says it carries no words at all, and its page explains why: the instruction is carried by a mark on each layer and not by a sentence, so the task places very little load on language. The examinee establishes the orientation of each transparent layer from the mark, rotates each part mentally, combines the parts into one figure and selects the matching option. In CHC terms the Layer Rotation subtest falls under visual processing (Gv), in particular visualization (Vz) and speeded rotation (SR). Visual Puzzles belongs to the same broad ability. On its page, the Visual Puzzles subtest asks the person to imagine how separate pieces combine into a finished form, and the page describes the task as sampling visualization and spatial synthesis more than verbal explanation or factual knowledge. Matrix Reasoning, which an earlier section described, is a fluid reasoning (Gf) task whose page says it demands flexible rule discovery rather than learned verbal knowledge. Symbol Search is also kept simple so that speed, not language, is the main demand, according to its page, and the section on processing speed covers it.

The technical manual also lists a composite called the Nonverbal Index, which it describes as emphasizing nonverbal reasoning, visual spatial ability, working memory and speed with minimal reliance on verbal content. The manual names the eight subtests that feed it, Matrix Reasoning, Figure Weights, Visual Puzzles, Layer Rotation, Logic Grid, Visual Sequence, Symbol Search and Coding, and its form table reports the index on the Full Scale form only. None of the three audio subtests is among the eight, but at least five of the eight are timed, as the table in the section on test conditions shows, so a composite with minimal reliance on verbal content is not a composite without time pressure, and a reader who is sensitive to timed tasks should look at which subtests feed an index before relying on its name.

Three limits keep the claim honest. First, instructions and the rules of each task are still communicated in English, and ACIS is an English only test. The manual notes that verbal and knowledge tasks are affected by English proficiency and educational exposure, and a nonverbal format reduces that effect for the three subtests above without removing the need to read the instructions. Second, nonverbal describes how the problem is presented and not what the person brings to it: the page on nonverbal IQ tests makes the point that a nonverbal format is not culture free, education free or bias free. Third, a matrix, a puzzle and a rotation task cover only reasoning and spatial abilities, so a score from them is not a Full Scale IQ and says nothing about vocabulary, arithmetic or memory. The neighboring page on the pattern recognition test looks at that family of tasks as a stand alone format.

For an autistic adult who expects language heavy tasks to understate their reasoning, the practical use of this is to read the indices separately. If a verbal index sits lower than a fluid reasoning or visual spatial index, that pattern is worth examining with the three tests in the next section, but it is one possible reading. The Dawson and Bölte studies compared total Raven's and Wechsler scores, not these indices, so nothing here establishes that language demand explains any particular profile.

8 How Do You Read an Uneven Profile?

Treat a gap between two indices as a question with three tests, whether it is larger than measurement error, whether it is rare among people with the same overall level, and whether it predicts anything that matters outside the test, because gaps of some size are common even among typically developing adults.

The first test is about precision. The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) say in Standard 2.4 that when interpretation emphasizes differences between two observed scores of an individual, reliability and precision data, including standard errors, should be provided for those differences. The comment is direct: when interpretation centers on the peaks and valleys in the examinee's profile, the reliability of score differences is critical. Standard 1.14 adds that when profile interpretation is suggested, the rationale and evidence for it should be provided. In plain terms, a difference between two scores can be much less precise than either score alone, so a modest gap may sit inside the noise, which is why a confidence interval for each index is more useful than a number. The page on reliability and validity explains the concepts.

The second test is about base rates. A gap can be real and still ordinary. Takeda, Nakashima and Tsuji (2020) compared the WAIS-III profiles of 50 adults with ADHD with an age matched typically developing group, and noted that a flat profile is uncommon even in typically developing adults, since only a small portion of participants in either group lacked any significant gap between verbal and performance IQ or associated indices. Their study was not about autism, but it shows that a visible gap is not, by itself, evidence of a condition. Charman and colleagues make the point from the autism side: a gap between performance and verbal IQ was not associated with a particular pattern of symptoms. Six indices can be compared in 15 different pairs (our arithmetic), so a large looking pair is easy to find. The page on index score differences and base rates shows how to choose which comparison to examine before looking.

The third test is about meaning. A gap passes it only if it predicts something useful, such as a consistent difference across several measures. Tests also tend to correlate less with one another at higher ability levels, a pattern the page on Spearman's law of diminishing returns explains, so wide spread across indices in a high profile is not by itself a defect of the test or of the person. A composite can also mislead when the abilities it combines sit far apart, because an average of a peak and a valley describes neither. The manual lists composites built from predefined subtest sets, among them the General Ability Index, which the page on the general ability index compares with the Full Scale IQ.

A fourth habit is to keep scales straight. Indices and the Full Scale IQ use a mean of 100 and a standard deviation of 15, scaled subtest scores run from 1 to 19 with a mean of 10 and a standard deviation of 3, and a percentile is a different scale, so the two should not be averaged, as the page on IQ score versus percentile explains. For your own report, look at each index with its confidence interval, ask which tasks feed each index and whether a gap follows a task feature such as time limits, audio presentation or language load, and hold the result as a question to explore, remembering that overlapping intervals do not by themselves show that a gap is noise. The neighboring page on the IQ profile test covers profile reading in general.

9 What Does Processing Speed Mean in a Profile, and What Should You Not Conclude?

Processing speed is one of the domains where a meta-analysis of studies of autistic adults found a medium sized pooled difference, but a group mean is not a prediction, and a slow result on a timed task can reflect several things that a speed score cannot separate.

Velikonja, Fett and Velthorst (2019) pooled 21 studies of processing speed in adults with autism spectrum disorder and found the medium sized difference from neurotypical adults reported above (g minus 0.61). That figure says that, on average across studies, autistic adults scored lower on the speed measures used, and it says nothing about the cause, the spread within each group or any person.

A timed speed score is a composite of several things. The Coding page lists speed, attention, motor fluency, visual efficiency and comfort with timed performance as demands, and says a lower result can reflect motor speed, visual tracking, perfectionism, distraction or unfamiliarity with timed work. The Symbol Search page, whose subtest has an 80 puzzle set with a 2 minute limit and a penalty for wrong answers, makes the same kind of point about a cautious style and about reduced tolerance for time pressure. The penalty is a practical detail worth knowing: a person who works carefully and answers fewer items is making a trade between speed and accuracy that the score does not separate. That is why speed should be read next to accuracy and next to the reasoning indices, and why a low speed index in an otherwise strong profile is a pattern to describe, not a limitation to assume.

For choosing a form, the facts are simple. The Quick form does not include a processing speed subtest, so it returns no speed index. The Optimized and Full Scale forms include Coding and Symbol Search, and the page on the processing speed test describes the speed domain on its own. A Custom Test lets the buyer choose any subtests, but the Full Scale IQ needs the Full Scale battery, so a buyer who leaves out the timed subtests gives up both the speed index and the Full Scale IQ, a legitimate choice that should be made knowingly. Several ACIS subtests have time limits, as the table in the section on test conditions shows, and the subtest pages are the place to check each one before you buy.

10 Should You Take an IQ Test at All, and What Can a Score Do for You?

Take one if you want a measured profile of abilities for your own understanding, and do not take one expecting an answer about autism, support needs or your worth, because a score describes abilities under test conditions and nothing beyond them.

A normed profile can show where reasoning, vocabulary, memory, spatial ability and speed sit relative to a reference group of adults, which is information some people use to understand why some tasks feel easy and others do not. It can give you something concrete to bring to a clinician, along with the note that it was taken unsupervised. It can satisfy curiosity, including curiosity about whether your abilities are unusually high, which the page on the gifted test for adults approaches for any adult. What it cannot do is describe your daily life: the NICE guideline asks professionals to understand the potential discrepancy between IQ and adaptive functioning, and Alvares and colleagues found IQ to be a weak predictor of adaptive scores in children and adolescents at diagnosis.

Two popular images distort expectations in opposite directions. The first is the idea that autism implies exceptional isolated skills, which the page on savant syndrome covers. The second is the idea that a high score and a disability cannot coexist, which the page on twice exceptional people covers. A result is also easy to over read in either direction. If the number surprises you, check the conditions first (noise, fatigue, working sound, a timed task that rattled you), then the profile, and only then the number.

11 Which ACIS Form Fits, and What Does the Free Trial Show?

ACIS has three forms and a Custom Test, and the form should follow the question: Quick for a first look at verbal, reasoning and memory abilities, Optimized for five domains including processing speed, and Full Scale for all six domains and the broadest profile. The facts below were read on the ACIS home page on October 28, 2026, and prices can change. Each form is a one time payment with no subscription.

Every form reports a Full Scale IQ with its 95 percent confidence interval. Visual Puzzles is only in the Full Scale form, and Quick has no spatial or processing speed subtest. The Custom Test costs 5 dollars per subtest, and the buyer picks any subtests. In the product's own words, any domain you complete returns its full index score and every subtest returns its scaled score, while the Full Scale IQ needs the Full Scale battery.

The free trial needs no card, and the site allows one free trial per person, refusing a second from the same email address or the same browser. Trial subtests that you finish are kept if you later unlock a form that includes them, so the trial is a way to see how the format and the timing feel before paying, for the subtests that the trial includes. The guarantee reads: "Full refund within 5 days if the report does not deliver the features described at checkout or a technical issue prevents you from accessing your results." That is a conditional guarantee, and it is not an unconditional money back offer. Example reports for each form let you see the report before you buy.

12 Where ACIS Sits, and What a Buyer Should Do

ACIS is a normed online battery of adult cognitive abilities for personal use, and its strengths and its limits are one fact: it is online, unsupervised, English only and not clinical, so it can show a profile and cannot diagnose, document or accommodate anything. It has 20 subtests in six domains, reports a Full Scale IQ and six primary indices, gives percentiles and a 95 percent confidence interval, and its adult norms cover ages 16 to 90. It is not for hiring, school accommodations or admission to high IQ societies. This page makes no claim that ACIS has been validated for autistic adults specifically. The technical manual documents what has been established for the instrument, and a reader who needs that evidence should read it before relying on any result.

The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) give the reading rules this page has applied. Norms should refer to clearly described populations (Standard 5.8), unproctored administrations call for caution and the interpreter should be told that a test was unproctored, and when score differences or profiles are interpreted their precision and the evidence for the interpretation should be shown (Standards 2.4 and 1.14). The APA's Guidelines for Psychological Assessment and Evaluation (2020) add that the characteristics of the test taker in relation to the normative sample matter for valid interpretation. A reader who is not matched by the norm group in some way, whether because of age, language or something else, should hold a score more loosely. The page on how IQ scores are normed explains what a norm group is.

What to do next depends on your question. If you want to know whether you are autistic, speak to a general practitioner or another clinician, and bring any result you have, saying that it was taken unsupervised. If you want a documented evaluation, book a psychologist. If you want a profile for your own understanding, start the free trial, read what each form covers, pick the one that matches your question, set up a quiet room and a working sound setup, take breaks between subtests and read the indices with their intervals, not only the Full Scale IQ.

13 Sources Behind This Page

Every third party fact and figure on this page comes from the documents below, opened or verified on October 28, 2026, and any figure we derived ourselves is labeled in the text as our arithmetic. Prices and policies on commercial and clinic pages change, so check the page before you pay or book.

1. American Educational Research Association, American Psychological Association, and National Council on Measurement in Education (2014). Standards for Educational and Psychological Testing. Washington, DC: AERA. Open access edition: https://www.testingstandards.net/open-access-files.html
2. American Psychological Association (2020). APA Guidelines for Psychological Assessment and Evaluation. Approved by the APA Council of Representatives, March 2020. https://www.apa.org/about/policy/guidelines-psychological-assessment-evaluation.pdf
3. National Institute for Health and Care Excellence (2012, last updated June 14, 2021, last reviewed July 23, 2026). Autism spectrum disorder in adults: diagnosis and management (clinical guideline CG142), read October 28, 2026. https://www.nice.org.uk/guidance/cg142
4. Dawson, M., Soulières, I., Gernsbacher, M. A., and Mottron, L. (2007). The level and nature of autistic intelligence. Psychological Science, 18(8), 657 to 662. https://doi.org/10.1111/j.1467-9280.2007.01954.x
5. Bölte, S., Dziobek, I., and Poustka, F. (2009). Brief report: The level and nature of autistic intelligence revisited. Journal of Autism and Developmental Disorders, 39(4), 678 to 682. https://doi.org/10.1007/s10803-008-0667-2
6. Charman, T., Pickles, A., Simonoff, E., Chandler, S., Loucas, T., and Baird, G. (2011). IQ in children with autism spectrum disorders: data from the Special Needs and Autism Project (SNAP). Psychological Medicine, 41(3), 619 to 627. https://doi.org/10.1017/S0033291710000991
7. Velikonja, T., Fett, A.-K., and Velthorst, E. (2019). Patterns of nonsocial and social cognitive functioning in adults with autism spectrum disorder: A systematic review and meta-analysis. JAMA Psychiatry, 76(2), 135 to 151 (page range as listed in PubMed). https://doi.org/10.1001/jamapsychiatry.2018.3645
8. Cicinelli, G., Nobile, E., Brighenti, S., Bari, S., Tonella, E., Aresi, A., Attanasio, M., Mazza, M., Valenti, M., and Keller, R. (2022). Wechsler Intelligence Scale for Adults, Fourth Edition profiles of adults with autism spectrum disorder. Epidemiology and Psychiatric Sciences, 31, e67. https://doi.org/10.1017/S2045796022000506
9. Alvares, G. A., Bebbington, K., Cleary, D., Evans, K., Glasson, E. J., Maybery, M. T., Pillar, S., Uljarević, M., Varcin, K., Wray, J., and Whitehouse, A. J. O. (2020). The misnomer of "high functioning autism": Intelligence is an imprecise predictor of functional abilities at diagnosis. Autism, 24(1), 221 to 232. https://doi.org/10.1177/1362361319852831
10. Kenny, L., Hattersley, C., Molins, B., Buckley, C., Povey, C., and Pellicano, E. (2016). Which terms should be used to describe autism? Perspectives from the UK autism community. Autism, 20(4), 442 to 462. https://doi.org/10.1177/1362361315588200
11. Takeda, T., Nakashima, Y., and Tsuji, Y. (2020). Discrepancies in Wechsler Adult Intelligent Scale III profile in adult with and without attention deficit hyperactivity disorder. Neuropsychopharmacology Reports, 40(2), 166 to 174. https://doi.org/10.1002/npr2.12106
12. Pearson Assessments. Wechsler Adult Intelligence Scale, Fifth Edition (WAIS-5), product page, read October 28, 2026. https://www.pearsonassessments.com/en-us/Store/Professional-Assessments/Cognition-%26-Neuro/Wechsler-Adult-Intelligence-Scale-%7C-Fifth-Edition/p/P100071002.html
13. University of Alabama, Psychology Clinic. Adult Assessment and Testing, read October 28, 2026. https://psychologyclinic.ua.edu/services/adult/
14. Marquette University, Department of Psychology. Autism Clinic, read October 28, 2026. https://www.marquette.edu/psychology/autism-clinic.php
15. American Psychiatric Association. What Is Autism Spectrum Disorder?, read October 28, 2026. https://www.psychiatry.org/patients-families/autism/what-is-autism-spectrum-disorder

14 Frequently Asked Questions