Working Memory: What It Is, Why It Slips, and How It’s Measured

Working memory is the mental notepad you use to keep a phone number in your head while you dial it, to follow a conversation while planning your reply, or to do a sum without paper. When it is strained, you walk into a room and forget why, lose the thread of a sentence, or re-read the same paragraph three times. It is one of the most commonly measured thinking skills because it touches nearly everything else, and because it is sensitive to so many things, from ADHD to a short night’s sleep.

Working memory versus short-term memory

The two terms are often used as synonyms, but researchers separate them. In her widely cited review of executive functions, developmental neuroscientist Adele Diamond puts it simply: working memory means holding information in mind and manipulating it, while short-term memory means just holding it. Repeating a list of digits back in the same order leans on short-term memory; repeating them backward, or putting them in numerical order, demands working memory.

Working memory is also counted among the three core executive functions, alongside inhibition and mental flexibility, which is why it reappears on our executive function page. Psychologist Alan Baddeley’s multicomponent model, refined over several decades, is the framework most researchers still use to describe how these systems fit together.

What a working memory problem feels like

People rarely describe it as “working memory.” They say things like:

• “I lose track halfway through instructions unless they’re written down.”
• “I can’t hold the question in my head while I search for the answer.”
• “In meetings I’m so busy remembering my point that I miss what people are saying.”
• “I start a task, get interrupted, and the plan is simply gone.”

These experiences are common and not automatically a sign of anything. They become worth investigating when they are persistent, out of step with how you used to function, or causing real problems at work, school or home.

How working memory is measured

Clinicians and researchers use a small family of task types, all of which ask you to hold a growing load until you can no longer keep up:

• Span tasks. You hear or see a sequence of digits or letters and repeat it. Forward span leans on storage; backward span adds manipulation. Moran’s 2016 meta-analysis grouped these as “simple span” tasks.
• Spatial sequences. The classic Corsi block test has you watch blocks light up in an order and reproduce it. Diamond names it as a widely used measure of visuospatial working memory.
• Updating tasks. In an n-back task you watch a stream of items and signal when the current one matches the one shown a set number of steps earlier.
• Self-ordered search. You look for hidden items without revisiting places you have already checked, which forces you to track your own progress.

Some longer batteries also include complex span tasks, in which you remember items while doing a second, unrelated job in between.

The Creyos tasks that tap memory

Creyos, the computerized assessment described in our Creyos patient guide, groups four of its 12 tasks under short-term memory and places Digit Span under verbal ability. In Creyos’s own labels: Token Search measures working memory (a self-ordered search), Number Ladder, formerly called Monkey Ladder, measures visuospatial working memory, Spatial Span is a web version of the Corsi blocks for spatial short-term memory, and Digit Span measures verbal short-term memory. Paired Associates measures episodic memory rather than working memory.

Test-retest correlations Creyos reports for these tasks are moderate (about 0.57 to 0.66 for Number Ladder, Digit Span and Token Search), so one sitting is a snapshot, not a fixed trait. Our task-by-task explainer shows what each task looks like.

ADHD and working memory

Working memory is one of the best-studied weaknesses in ADHD at the group level. A meta-analysis of 26 studies in children and teens (Martinussen and colleagues, 2005) found deficits across several components, with larger effects for spatial tasks than verbal ones. A larger review of 83 studies of executive function in ADHD (Willcutt and colleagues, 2005) found working memory among the strongest and most consistent effects.

The same review added an important caution: the effects were moderate and not universal, so executive weaknesses are “neither necessary nor sufficient” to explain every case of ADHD. A 2024 meta-analysis of Wechsler intelligence test scores found that children and adults with ADHD scored mostly in the age-expected range, with only slightly reduced working memory, and concluded that these profiles are not sensitive or specific enough to diagnose anything.

In practice, that means a working memory score is useful context in an adult ADHD evaluation, not a verdict. Our page on Creyos in an ADHD assessment explains how task results are combined with questionnaires and history.

Anxiety, depression and the crowded notepad

Worry takes up room. In a meta-analysis of 177 samples covering 22,061 people, Moran (2016) found that higher self-reported anxiety was reliably linked to poorer working memory performance, including on simple span tasks such as digit span. The average effect was modest but consistent across task types.

Depression shows a slightly different pattern. In the 2014 meta-analysis by Rock and colleagues, memory deficits were present during depressive episodes, but the memory difference was smaller and no longer statistically significant once mood had remitted, while attention and executive problems were more likely to linger. If low mood is part of your picture, see cognitive symptoms of depression and our anxiety care page.

Sleep: total deprivation versus habitual short sleep

Sleep loss is one of the most reliable ways to degrade working memory in the lab. Lim and Dinges’ 2010 meta-analysis of short-term total sleep deprivation (under 48 hours awake) found effects on working memory, processing speed and especially simple attention.

Everyday short sleep looks a little different. In a large online study of more than 10,000 people (Wild and colleagues, 2018), those who habitually slept less performed worse overall, yet short-term memory scores specifically were not affected. The two findings measure different things, a night with no sleep versus a habit of short nights, so it is fair to say that sleep loss affects thinking without assuming it always shows up as a lower memory span. Our pages on sleep and cognitive performance and ADHD, sleep and nutrition go further.

Can you train working memory?

Practising a working memory game reliably makes you better at that game. Whether it makes you better at life is a different question. A major review of the brain-training literature (Simons and colleagues, 2016) found extensive evidence of improvement on trained tasks, less on closely related tasks, and little that training improves everyday cognitive performance. An earlier study of 11,430 people doing six weeks of online training (Owen and colleagues, 2010) found no transfer to untrained tasks.

What tends to help day to day is reducing the load rather than enlarging the notepad: writing steps down, using one list instead of five, finishing one thing before starting the next, and protecting sleep. Our article on whether brain training works covers the evidence in more detail.

When to have it looked at

Consider an evaluation if working memory problems are new, getting worse, interfering with work or school, or arriving alongside changes in mood, sleep or attention. A psychiatric clinician can look at the likely contributors together, and at Willow & Stone a Creyos assessment may be part of that. The fee, and whether you would take it at home or in the Arlington office, are confirmed when you book through our consultation page.

Common Questions

Sources & Further Reading

• Diamond A. Executive functions. Annu Rev Psychol. 2013
• Baddeley A. Working memory: theories, models, and controversies. Annu Rev Psychol. 2012
• Martinussen R et al. A meta-analysis of working memory impairments in children with ADHD. J Am Acad Child Adolesc Psychiatry. 2005
• Willcutt EG et al. Validity of the executive function theory of ADHD: a meta-analytic review. Biol Psychiatry. 2005
• Moran TP. Anxiety and working memory capacity: a meta-analysis and narrative review. Psychol Bull. 2016
• Lim J, Dinges DF. Short-term sleep deprivation and cognitive variables: a meta-analysis. Psychol Bull. 2010
• Simons DJ et al. Do “brain-training” programs work? Psychol Sci Public Interest. 2016
• Creyos — Report Interpretation Guide (PDF)

This article is general education, not a diagnosis or advice for your situation. Creyos results support, and never replace, a clinical evaluation; only a clinician who has evaluated you can diagnose a condition or recommend treatment. Do not start, stop, or change a medication without talking to the clinician who prescribes it.