An auditory memory test that measures holding, not hearing

0 The short answer

An auditory memory test measures how well you hold spoken material in mind and, in its better versions, how well you work on it, and the version worth taking reports a score against adult norms rather than a count from a game. A psychologist can give the auditory working memory tasks of the Wechsler scale in a supervised session, and ACIS offers two audio subtests, Digit Span and Alphanumeric Sequencing, which are in the Quick form (15 dollars, about 45 minutes, prices read October 29, 2026) and in the other forms. The decisive limit is that this is a memory test and not a hearing test: it makes no claim about hearing or about auditory processing disorders, and an online, unsupervised, non clinical score cannot diagnose anything.

1 What does an auditory memory test measure?

An auditory memory test measures the temporary holding of sound based material, most often spoken digits, letters or words, and a good one also measures how much control the person keeps over that material. The phrase "auditory memory" is a search term, not a standard score category. In the Cattell-Horn-Carroll (CHC) model the nearest constructs are short term memory (Gsm) and working memory capacity (Gwm), and the ACIS pages label both of their audio tasks with the pair Gwm/Gsm. The page on the CHC model lists the broad abilities and explains how they sit under a general factor, and the technical manual lists working memory as one of the six domains ACIS reports.

The cognitive machinery behind spoken span is usually described with Baddeley's phonological loop. Baddeley, Gathercole and Papagno, in their review in Psychological Review in 1998, describe the loop as a component of working memory, and they argue from studies of adults, children and neuropsychological patients that it plays a crucial role in learning the novel phonological forms of new words. They propose that its primary purpose is to store unfamiliar sound patterns while more permanent memory records are being built. A digit span task borrows that machinery for an artificial purpose: it asks for familiar items in a fixed order, with no help from meaning.

The capacity is small. Cowan reviewed the evidence in Behavioral and Brain Sciences in 2001 and concluded that a single central capacity limit averaging about four chunks is implicated, once rehearsal and grouping are blocked. His abstract notes that Miller, in 1956, had summarized evidence that people can remember about seven chunks, and that the number was meant more as a rough estimate than a real capacity limit.

Two distinctions keep this construct from collapsing into vague talk about memory. The first is storage versus manipulation. Repeating digits in the order heard mainly taps storage, while reversing or sorting them adds work on top of storage. The page on the short term memory test owns that distinction in full, the guide to the working memory test covers span tasks and norms in general, and this page uses both only where the formats require it. The second is memory versus the rest of a person's cognition. The page on IQ and memory explains that memory is a family of systems, that a person can reason very well and still forget names, and that a memory complaint does not establish lower intelligence.

This page is also not about hearing. An auditory memory test presents sound and asks what the person can do with it afterwards. It does not test whether the person detects tones or distinguishes speech sounds, and it does not test auditory processing, which CHC theory treats as a separate broad ability (Ga) concerned with perceiving, analyzing and synthesizing patterns among sounds, as the page on the CHC model describes. The American Speech-Language-Hearing Association, on its page on central auditory processing disorder (read October 29, 2026), states that the disorder is diagnosed by an audiologist and that a peripheral hearing evaluation is essential to rule out hearing loss first. A reader with concerns about hearing or listening should see an audiologist. No score on this page's tests answers that question.

2 Which formats do auditory memory tests use?

Auditory memory tests use five recurring formats, which are digit span forward, digit span backward, digit sequencing, letter number sequencing and spoken problem solving, and they differ mainly in how much they ask the person to do with what was heard. The shortest description of each follows, and none reproduces an item.

Digit span forward plays a string of digits and asks for the same string back, in order. It is the simplest format and mostly samples storage. Digit span backward asks for the string in reverse, which adds a transformation: the person has to hold the string and also reorder it before responding. Digit sequencing asks for the digits re-ordered into ascending sequence, which is a second transformation with a rule the person must apply. Gignac and Weiss, in Psychological Assessment in 2015, analyze the forward, backward and sequencing scores of the Wechsler Adult Intelligence Scale, Fourth Edition, as three separate Digit Span indicators. The ACIS manual describes its own Digit Span the same way, with three conditions: the examinee reproduces digits "in the order presented, in reverse order, and re-ordered into ascending sequence."

Letter number sequencing presents a mixed stream of numbers and letters and asks for them back sorted, with the numbers in ascending order and then the letters in alphabetical order. The ACIS manual calls this a dual rule task: the series has to be held while two ordering rules are applied to it. That is why the manual says the subtest separates adequate retention from efficiency once sorting and dual rule control are added.

Running or updating formats keep a changing set in mind while new items arrive. Pearson's product page for the Wechsler Adult Intelligence Scale, Fifth Edition (WAIS-5, read October 29, 2026) lists Running Digits as an auditory working memory task and describes it as measuring registration and maintenance of information, including updating items held in short term storage while other stimuli are processed. The page also lists, among the features of its working memory measurement, a "task that mimics real-world rate of speech."

Spoken problem solving is the fifth format. Here the material is a short word problem delivered as audio, with no written copy, and the person has to retain the terms while computing. It is classified as quantitative rather than as a memory task in ACIS, for reasons a later section sets out, but it belongs to the family of tests where hearing something once and holding it matters.

One more family is deliberately left out. Word list learning tests, in which a person hears a list over several trials and later recalls it, measure the encoding and retrieval of material over longer delays. That is a different memory system from the seconds long holding that span tasks sample, and this page does not review it.

3 What are the options for taking an auditory memory test?

You can take an auditory memory test from a psychologist, through research software, from a free game, or from a normed online battery, and the options differ more in what they report than in how they sound. The table sets them side by side. Prices and times were read on October 29, 2026 and can change.

The clinician route is the reference for diagnostic questions, and it is the only row where a professional watches the person respond. Pearson lists the WAIS-5 as published in 2024 with a qualification level of C, a complete print kit at 1,784 dollars, and administration times of 45 minutes for the 7 subtest Full Scale IQ and 60 minutes for the 10 primary index subtests. The kit price is what a professional pays, not what a client pays. A person with a clinical question about attention, memory or hearing should go to that route and not to this table. The page on WAIS tests online explains what the official online options are and what they are not.

Research software answers a different question. The Millisecond Software library page for its auditory Digit Span Test, read October 29, 2026, lists forward and backward versions with spoken digits, states that the combined test takes 15 minutes and requires Inquisit Lab, and says no normative data is provided. That is appropriate for an experiment, where the researcher compares groups, and it is a poor fit for a person who wants to know where they stand, because without a norm group a raw span has no reference.

Free games deserve a plain statement. A game that plays a sequence of sounds and asks you to repeat it can be enjoyable. If its page does not state a norm group, its result is a count and not an index, and the count says nothing about how adults of your age usually perform. The page on free IQ tests explains what a free result can and cannot support.

ACIS is a seller in this market, so treat this section as a disclosure. The comparison lists where the options differ: only the clinician route is supervised and clinically interpreted, and only the Wechsler and ACIS routes report a normed score for a person. Other online batteries exist, and the page on the best online IQ tests and the page on where to take an IQ test compare them without a focus on memory.

4 What do the ACIS audio subtests consist of?

ACIS presents two working memory subtests as audio, Digit Span and Alphanumeric Sequencing, and a third audio task, Arithmetic, that the manual classifies as quantitative reasoning. The technical manual, read on October 29, 2026, assigns the two memory subtests to the working memory capacity domain with an auditory short term storage component, and it describes the demand as attention, phonological encoding, serial maintenance and, when required, manipulation of ordered information.

The Digit Span page and the manual describe three conditions. In the first the examinee reproduces digits in the order presented, in the second in reverse order, and in the third re-ordered into ascending sequence. The manual's reading is that holding the material is common to all three, and that reversal and re-ordering add manipulation on top of storage, so the conditions differ in how much working memory capacity they demand beyond short term storage, and not in the material itself. The page for the live task says the digits play at a rate of one per second, recommends headphones or a quiet space, and allows no external aids, which it lists as calculators, notes, scratch paper and drawing tools.

The Alphanumeric Sequencing page describes a different task on the same footing. A mixed series of numbers and letters plays at the same rate, and the examinee gives it back with the numbers in ascending order followed by the letters in alphabetical order. The task includes an unscored practice section, and the instructions note that the sequence length increases later in the task. The page says a weaker result can appear even when simple span is adequate, because sorting adds executive demand. The manual puts the profile rule in its own words: if Digit Span is adequate but Alphanumeric Sequencing is weaker, the person may maintain information but have difficulty reorganizing it under rules, and if both are strong, the auditory working memory profile is more robust.

Those two subtests also define a composite. The manual lists the Auditory Working Memory Index (AWMI), described as auditory verbal serial storage and manipulation, and says it focuses on auditory working memory through Digit Span and Alphanumeric Sequencing. It also says composites are available when their subtests are available. The AWMI is a narrower question than the Working Memory Index (WMI), the primary index for registration, maintenance and manipulation of information in awareness. Visual Sequence, the visual counterpart, belongs to the WMI and not to the AWMI, and the section on spoken versus visual span returns to why that contrast matters.

Which form contains which subtest decides what the report can show. Quick (15 dollars, six subtests, about 45 minutes) contains both audio working memory subtests, together with Similarities, Vocabulary, Matrix Reasoning and Figure Weights. Optimized (30 dollars, 13 subtests, about 110 minutes) contains both among its 13, including Coding and Symbol Search from a different domain. Full Scale (50 dollars, 20 subtests, about 175 minutes) contains both and adds Arithmetic and Visual Sequence, among others, which are only in that form. Breaks are allowed, progress is saved between subtests, and the buyer has 30 days to complete. The Custom Test, at 5 dollars per subtest, lets a buyer pick any subtests, and its description states that "any domain you complete returns its full index score, and every subtest returns its scaled score." Two audio subtests at 5 dollars each is 10 dollars, and the Full Scale form costs the same as ten Custom subtests, so a buyer who wants more than ten should take the Full Scale form.

The free trial needs no card, is limited to one per person, and, when started from the Full Scale form, opens five subtests that include both audio working memory subtests. Finished trial subtests are kept if the person later unlocks a form that includes them. That makes the trial the lowest cost way to hear how the tasks sound before paying, and it is also a test of your own equipment, which matters more here than for a silent task. The page for adults describes the whole experience for a first time taker. A working memory result also reads best beside speed: the manual defines a composite, CPI, that "combines working memory and processing speed as cognitive proficiency," and the page on the processing speed test covers the Coding and Symbol Search side of that pair.

5 How does auditory working memory relate to IQ?

Working memory, measured with span tasks that also demand manipulation, shows a strong connection to fluid intelligence, while storage alone shows a much weaker one, and digit span relates to general intelligence in a largely linear way across the range of span, not only at the low end. The evidence is correlational and comes from latent variable studies, which model what several tasks share rather than treating one task as the construct.

Engle, Tuholski, Laughlin and Conway ran the study most often cited for the contrast. In Journal of Experimental Psychology: General in 1999, they report that 133 participants performed 11 memory tasks, some thought to reflect working memory and some short term memory, along with 2 tests of general fluid intelligence and the verbal and quantitative Scholastic Aptitude Tests. Their structural equation models suggested that short term and working memory are separate but highly related constructs, that many of the tasks used as working memory tasks reflect a common construct, and that working memory shows a strong connection to fluid intelligence while short term memory does not.

Kane and colleagues tested the same idea across modalities. Their study in Journal of Experimental Psychology: General in 2004 used 236 participants, with three span tests each of verbal working memory, visuospatial working memory, verbal short term memory and visuospatial short term memory, plus tests of verbal and spatial reasoning and general fluid intelligence. The working memory tasks largely reflected a domain general factor, while the short term memory tasks, based on the same stimuli, were much more domain specific. Working memory was a strong predictor of fluid intelligence and a weaker predictor of domain specific reasoning, and the reverse was true for short term memory.

For a reader choosing a test, the consequence is that the manipulating conditions matter. A forward span task is mostly storage, and storage is the more domain specific part. A backward or sorting task asks for the control that, in these studies, carries the link to reasoning. That is the reason the ACIS manual reads Digit Span as a working memory measure and not as a measure of reasoning, and the reason it notes that a low Digit Span with strong reasoning may show a profile in which reasoning is better than raw serial storage.

A second line of evidence concerns the shape of the relation. Wechsler contended in 1958 that, beyond an average span, there is little benefit to a greater span, and that view became clinical lore. Gignac and Weiss tested it in Psychological Assessment in 2015 on the normative sample of the Wechsler Adult Intelligence Scale, Fourth Edition (1,800 people aged 16 to 69). Using contrast analyses of means and bifactor models, they found that all three Digit Span indicators showed primarily linear associations with Full Scale IQ and general intelligence, so the position that span beyond an average level says nothing about intellectual functioning was not supported. Their title states the practical message: every extra bit of span counts.

Three limits keep these findings in proportion. The studies describe groups and associations, and they do not show that holding more digits causes higher reasoning or the reverse. The samples were specific, and the 2015 sample was a standardization sample of one battery. And a memory index is one part of a profile. The page on what IQ measures explains how working memory fits among the abilities a battery samples, and the page on the General Ability Index explains why some composites deliberately reduce the weight of working memory and speed. The manual describes the General Ability Index as higher order reasoning and knowledge with reduced working memory and processing speed emphasis, which is the reason a person can have a Full Scale IQ and a General Ability Index that differ.

6 Is spoken span the same as visual span?

Spoken span and visual span share the working memory construct but not the sensory material, and a person can differ between them, which is why a battery that samples both is more informative than either alone. The Kane study above is the main evidence that storage tasks are domain specific: the short term memory tasks, though based on the same stimuli as the working memory tasks, were much more domain specific, while the working memory tasks largely shared a general factor.

The ACIS manual builds that contrast into its design. Visual Sequence is described as a task in which the examinee watches a sequence of tiles flash and reproduces it by clicking the tiles, first in the order shown and then in reverse. The manual calls it the visual counterpart to Digit Span, because the material is spatial and not verbal, and says that lets visual spatial short term storage be read beside its auditory equivalent. It adds two interpretive rules. A high Visual Sequence score with a lower Digit Span may suggest stronger visual than auditory working memory. A low Visual Sequence score with adequate visual spatial reasoning may suggest that the weakness is not visual processing itself but visual sequence maintenance. The Visual Sequence page describes the task, and the page on the visual memory test owns the visual side of this contrast, including how a span task differs from remembering pictures over a longer delay.

A neighboring ability is worth separating so that it is not mistaken for a memory score. The page on the block design test covers a visual spatial assembly task, in which the examinee reproduces a design with blocks. It samples a different broad ability from span, and a person can be strong in one and ordinary in the other. The ACIS manual keeps Visual Sequence in the working memory domain, not in the visual spatial index, for a related reason: its defining operation is temporary storage and controlled sequence reproduction, even though the material is visual.

Musicians are a group for whom sound matters especially. Talamini, Altoè, Carretti and Grassi reported a meta-analysis in PLOS ONE in 2017 of 29 studies with 53 memory tasks. Musicians outperformed nonmusicians in short term memory (g of .57) and working memory (g of .56), with a smaller advantage in long term memory (g of .29). When the authors split by the type of material, the advantage was large with tonal stimuli, moderate with verbal stimuli, and small or null with visuospatial stimuli. That is a finding about groups and about the material of the task, and it does not show that musical training causes better memory, since the studies compared musicians with nonmusicians. The page on an IQ test for musicians takes up what a musician can and cannot infer from a score.

7 Why does a spoken arithmetic task count as quantitative and not as memory?

Spoken arithmetic loads working memory because the terms of the problem are heard once and must be held, but ACIS classifies Arithmetic as quantitative reasoning (Gq) because the answer depends on learned number concepts and operations. The Arithmetic page says the task measures mental quantitative problem solving, drawing on number knowledge, basic operations, attention and the ability to hold intermediate values while applying a sequence of steps. It also says that Gq appears as an index in its own right depending on the instrument, and that the Wechsler scales place arithmetic under working memory instead.

The format is what links it to this page. The manual says the examinee hears a word problem delivered in the listening area and computes the answer mentally, and, because the problem is presented aurally and no written copy remains on screen, the terms must be held while the solution is computed, so the format adds a retention demand to the arithmetic itself. The page for the live task says items can be replayed once while a 30 second timer keeps running, and the manual describes a 30 second response limit per item. The manual's interpretive rule is that a low Arithmetic score with adequate Mathematical Achievement may indicate that the person understands math but performs less efficiently under working memory demands, and that Arithmetic is sensitive to attention, anxiety, working memory and calculation fluency.

The research literature supports the idea that mental arithmetic uses working memory, with caveats. DeStefano and LeFevre reviewed the question in the European Journal of Cognitive Psychology in 2004. They describe their literature review as neither comprehensive nor systematic, and they conclude that all three components of Baddeley's working memory system (the central executive, the phonological loop and the visual spatial sketchpad) play a role in mental arithmetic, though under different conditions, and that mental arithmetic requires central executive resources even for single digit problems. They add that problems with multiple digits are likely to involve an interaction of all the components.

Two readers should care about this boundary. A person who scores lower on Arithmetic than on a written math task has two candidate explanations, a listening and holding problem or a calculation problem, and the manual's comparison with Mathematical Achievement is the way to separate them. And a person comparing two websites should notice that a "mental math" test can present its problems on screen or by voice, which changes what it measures. The sibling page on the mental math test covers calculation itself and keeps the arithmetic skills in the foreground, while this page covers the memory load that the spoken format adds.

8 Can auditory memory be trained, and what does a retake show?

Practice improves performance on the trained task, but the best available meta-analysis found no convincing evidence that working memory training improves intelligence or other far transfer measures, so a better second score is not proof of a better memory. Melby-Lervåg, Redick and Hulme pooled 87 publications with 145 experimental comparisons in Perspectives on Psychological Science in 2016. Studies had a pretest, a posttest and a control group, and they examined transfer to nonverbal ability, verbal ability, word decoding, reading comprehension and arithmetic. Immediately after training there were reliable improvements on measures of intermediate transfer, which were verbal and visuospatial working memory tasks. For far transfer there was no convincing evidence of reliable improvement when training was compared with a treated control condition, and the degree of improvement on working memory measures was not related to the size of far transfer effects. The authors conclude that the programs appear to produce short term, specific training effects that do not generalize to real world cognitive skills.

Two consequences follow for a person who takes a spoken span task. First, a low first score is not a verdict on ability, because performance on a given day depends on attention, fatigue, the room and the equipment, which the next section covers. Second, a higher score on a second attempt after practicing the same format cannot be read as pure ability change. Practice with a format plausibly makes that format easier, and the meta-analysis found that improvement on trained working memory tasks did not carry reliably to measures of reasoning. This page gives no retake gain figure, because no source opened for it supplies one for these tasks. A sensible plan is to take a span test once under good conditions, treat the interval around the score as the answer and not the single number, and compare any later attempt with that interval. The page on whether you can increase your IQ places working memory training among the other training claims.

9 What conditions does an unsupervised session need?

An unsupervised auditory test needs working sound, headphones or a quiet room, a stable connection and no outside aids, because every one of those conditions changes what the score measures. For a silent test, a noisy room is an annoyance. For a spoken test, it is a change to the stimulus itself.

The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014), in Standard 6.4, say that the testing environment should furnish reasonable comfort with minimal distractions to avoid construct irrelevant variance. The comment lists noise, disruption in the testing area and malfunctioning computers among the conditions to avoid, and it adds that technology based administrations should avoid distractions such as equipment or internet connectivity failures. Standard 6.1 asks that the procedures specified by the test developer be followed carefully, and its comment says the instructions to test takers, the form of item presentation or response, and the test materials or equipment should be strictly observed. For a spoken task, "form of item presentation" means the audio.

The ACIS pages give the following conditions for their own audio tasks. Digit Span and Alphanumeric Sequencing play items at one per second and need working sound, and the Digit Span page recommends headphones or a quiet space. Neither allows calculators, notes, scratch paper or drawing tools, and each item should be heard to its end before a response is typed. The page for Alphanumeric Sequencing includes a practice section that is not scored. The live pages show a loading step for the audio before each task. Practically, that means the following checklist before starting:

• Test the sound first, with headphones on, and set the volume so that a quiet sequence is clear without being loud.
• Close other tabs and applications that play sound, and silence notifications.
• Sit in a room where no one will speak to you.
• Keep a pen and paper out of reach, because the rule is no outside aids and writing would change the task.
• Take a spoken subtest when rested, using the breaks the platform allows.

Does remote administration change span scores? One small study suggests not much. Nisha and colleagues, in Frontiers in Digital Health in 2025, ran a pilot with 15 young adults aged 18 to 30 with normal hearing. Each completed forward digit span, backward digit span and an n-back task twice, once by tele-assessment and once face to face, three to five days apart, with digits played through headphones. They found no significant differences in performance between the two modes, and the backward digit span was the most difficult of the three tasks in both. The study is a small pilot of young adults, with the remote participants seated in a soundproof room, so it cannot show how an unsupervised session in a living room compares, and it is not evidence about older adults or people who are not native speakers. What it does show is that a spoken span task can be delivered remotely at all, and that the backward condition was the harder one in that study.

The APA's Ethical Principles of Psychologists and Code of Conduct make the same point from the interpreter's side. Standard 9.06, in the 2017 text read in a public copy on October 29, 2026, asks psychologists to take into account situational, personal, linguistic and cultural differences that might reduce the accuracy of their interpretations, and to indicate significant limitations. A noisy room is a situational difference. A reader who interprets their own score is not bound by the code, but the benchmark is useful: if the room, the equipment or the day was not normal, the score deserves a note.

10 What can a low score mean besides weak memory?

A low score on a spoken span task can come from poor listening conditions, divided attention, anxiety, language, hearing or a real limit in holding and manipulating the material, and a single score cannot say which. That is why the ACIS pages list several reasons for a weaker result and tell the reader to compare tasks.

The Digit Span page says a weaker result may reflect distraction, anxiety, sequencing difficulty or limits in active manipulation, and it recommends comparing the result with the alphanumeric and visual sequence tasks to see whether the pattern is verbal, visual or executive. The manual adds that a low Digit Span with strong reasoning may mean that reasoning is better than raw serial storage, and it says Digit Span should be interpreted as a working memory measure and not as a measure of reasoning. These are statements about patterns across tasks, which is the right unit of analysis.

Hearing deserves a plain statement. The tasks play spoken digits and letters, so a person who cannot hear them clearly will not be able to hold them, and the score will then reflect listening and not memory. This is a reason to check the sound before starting and to stop if the audio is unclear. It is not a reason to read a low span as a sign of a hearing problem, and the ACIS test makes no claim about hearing. A hearing concern is a medical and audiological question. The ASHA page cited above states that central auditory processing disorder is diagnosed by an audiologist and that a peripheral hearing evaluation is essential first, so a person with that concern should book that evaluation and not interpret a memory score as an answer.

Language is a second boundary. The ACIS tasks are in English, and the technical manual says verbal and knowledge tasks are affected by English proficiency and educational exposure. The Digit Span page notes that digit sequences are familiar, relatively neutral and easy to standardize, but letters, the alphabet order and instructions in English still require the language. A person who learned English as a second language, or who mostly works in another language, may perform differently from a native speaker for reasons that are not about memory.

Age is a third. The ACIS Digit Span page states that absolute performance on span and speed tasks falls across adulthood while standing within a birth cohort stays stable, so a raw span only becomes interpretable against an age reference group. That is the purpose of age norms, and it is why the same number of digits can mean different things at 25 and at 65. The page on how IQ scores are normed explains the logic of a norm group in general terms.

Finally, a group statistic is never an individual diagnosis, and a score on an online test cannot say why a person scored as they did. A person with concerns about attention, memory or hearing should talk to a clinician, who can use a supervised, individually interpreted assessment.

11 How do you read an auditory memory score?

You read an auditory memory score as a scaled score for each task, an index for the domain and a percentile with a confidence interval, and you treat a single task as a rough indicator and a pattern across tasks as the finding. ACIS reports subtest results as scaled scores from 1 to 19, with a mean of 10 and a standard deviation of 3, and indices and the Full Scale IQ on the standard scale with a mean of 100 and a standard deviation of 15, with a percentile and a 95 percent confidence interval. Adult norms cover ages 16 to 90.

A scaled score of 10 is the middle of the adult norm group, and each 3 points is one standard deviation. On the index scale, 100 is the middle and each 15 points is one standard deviation. The page on the standard deviation of 15 explains why that scale was chosen, and the page on IQ scores versus percentiles explains why the two should not be averaged.

The confidence interval is the honest part of the report. It reflects measurement error. The Standards for Educational and Psychological Testing say, in Standard 2.3, that for each total score, subscore or combination of scores to be interpreted, estimates of reliability and precision should be reported, and the comment says it is not sufficient to report them only for total scores when subscores are also interpreted, because total scores can be acceptably consistent while subscores have unacceptably low reliability. It adds that users need information about the reliability of composites formed from selected subtests. Standard 2.4 asks for reliability and precision data for differences between two observed scores, because the precision of a difference can be much lower than the precision of the separate scores. Both apply here. A gap between Digit Span and Alphanumeric Sequencing, or between spoken and visual span, is a difference score, and a small gap is within the noise. The page on reliability and validity explains these terms in plain language, and the technical manual is where the documentation for ACIS scores lives, which this page does not quote.

Three reading rules follow. Read the interval before the point estimate. Read the pattern across spoken and visual tasks before the single score, using the manual's rules quoted earlier. And read the Working Memory Index beside the other indices, because working memory is one domain among six. The page on index score differences describes how common a gap between indices is, and an uneven profile is common and is not a defect.

12 Where ACIS sits, and what a buyer should do

ACIS is a normed, online, unsupervised memory and cognitive battery, not a clinical instrument, and the right use of its auditory subtests is to learn how your spoken working memory compares with adult norms and with your own visual memory and reasoning. The limits are as important as the strengths. ACIS is online and unsupervised, it is not a clinical or diagnostic instrument, it is not for hiring, school accommodations or admission to high IQ societies, it is available in English only, and it makes no claim about hearing or auditory processing. The manual says verbal and knowledge tasks are affected by English proficiency and educational exposure. Prices were read on October 29, 2026 and can change.

The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) frame how any such score should be read. Standard 6.4 asks for minimal distractions to avoid construct irrelevant variance, Standard 6.1 asks that the developer's procedures be observed, and Standards 2.3 and 2.4 ask for precision data for the scores and the differences that are interpreted. The APA's Ethical Principles of Psychologists and Code of Conduct ask in Standard 9.02(a) that assessment techniques be used in a manner and for purposes appropriate in light of the evidence on their usefulness and proper application, and Standard 9.06 asks for attention to situational and personal factors and a statement of limitations. Applied to the reader: an unsupervised spoken score taken in a noisy room is a weak observation, and the same score taken with headphones in a quiet room, read with its interval and beside the visual task, is a defensible one.

A practical path follows from that. If the question is simply how your auditory working memory compares with adults of your age, take the Quick form (15 dollars, about 45 minutes), which contains both audio subtests. If you want the contrast with visual span, take the Full Scale form (50 dollars, about 175 minutes), which adds Visual Sequence and Arithmetic. If you want only the two audio subtests, the Custom Test costs 10 dollars for the pair. If you are undecided, the free trial from the Full Scale form needs no card and opens five subtests that include both audio ones, so you can test your sound and your reaction to the task before paying. 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."

Whatever the choice, plan for sound before you start, and read the report as a pattern. A person who wants a diagnosis, an accommodation or a hiring decision needs a clinician or an employer's validated process, not this page.

13 Sources Behind This Page

Every source below is cited in the body at the point where its figure or statement first appears.

• American Educational Research Association, American Psychological Association and National Council on Measurement in Education. Standards for Educational and Psychological Testing. American Educational Research Association, 2014. Standards 2.3, 2.4, 6.1 and 6.4.
• American Psychological Association. Ethical Principles of Psychologists and Code of Conduct, effective January 1, 2017. Standards 9.02 and 9.06, read in a public copy of the 2017 text on October 29, 2026.
• American Speech-Language-Hearing Association. Central Auditory Processing Disorder. Practice Portal, read October 29, 2026.
• Baddeley, A. D., Gathercole, S. E. and Papagno, C. (1998). The phonological loop as a language learning device. Psychological Review, 105(1), 158 to 173.
• Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences, 24(1), 87 to 114.
• DeStefano, D. and LeFevre, J.-A. (2004). The role of working memory in mental arithmetic. European Journal of Cognitive Psychology, 16(3), 353 to 386.
• Engle, R. W., Tuholski, S. W., Laughlin, J. E. and Conway, A. R. A. (1999). Working memory, short-term memory, and general fluid intelligence: A latent-variable approach. Journal of Experimental Psychology: General, 128(3), 309 to 331.
• Gignac, G. E. and Weiss, L. G. (2015). Digit Span is (mostly) related linearly to general intelligence: Every extra bit of span counts. Psychological Assessment, 27(4), 1312 to 1323.
• Kane, M. J., Hambrick, D. Z., Tuholski, S. W., Wilhelm, O., Payne, T. W. and Engle, R. W. (2004). The generality of working memory capacity: A latent-variable approach to verbal and visuospatial memory span and reasoning. Journal of Experimental Psychology: General, 133(2), 189 to 217.
• Melby-Lervåg, M., Redick, T. S. and Hulme, C. (2016). Working memory training does not improve performance on measures of intelligence or other measures of far transfer. Perspectives on Psychological Science, 11(4), 512 to 534.
• Millisecond Software. Digit Span Test (auditory), Inquisit library. Read October 29, 2026.
• Nisha, K. V., Devi, N., Vandana, B., Rashmi, E., Shende, S. A. and Mudar, R. A. (2025). Evaluating working memory in young individuals with normal hearing through tele-assessment and traditional assessment: A comparative study. Frontiers in Digital Health, 7.
• Pearson Assessments. Wechsler Adult Intelligence Scale, Fifth Edition (WAIS-5) product page. Read October 29, 2026.
• Talamini, F., Altoè, G., Carretti, B. and Grassi, M. (2017). Musicians have better memory than nonmusicians: A meta-analysis. PLOS ONE, 12(10), e0186773.

14 Frequently Asked Questions