ADHD and Working Memory: Why You Forget What You Were Doing

You walk into the kitchen to get something. You are standing in the kitchen. You do not know what you came for. You retrace your steps to the living room, hoping the original location will trigger the thought. Sometimes it does. Often it does not. You stand in the doorway between two rooms, belonging to neither task, holding nothing.

This is not aging. This is not absent-mindedness in the colloquial sense. If it happens to you several times a day, and has for as long as you can remember, it may be a working memory deficit, and it is one of the most consistent findings in ADHD research.

What working memory is

Baddeley’s model describes working memory as a multi-component system that holds a small amount of information in an active state while you use it [3]. It is not the same as long-term memory. You can have an excellent memory for facts, events, and trivia while having a working memory that drops information the moment something else claims your attention. The two systems are largely independent, which is why the experience is so disorienting: you can recall a conversation from three years ago in precise detail while losing the thread of one happening right now.

Think of it as a whiteboard with disappearing ink. You write something on it, and the ink starts fading immediately. In typical functioning, the ink fades slowly enough to finish the task. In ADHD, the ink fades faster, and anything that diverts your attention, a notification, a passing thought, someone speaking, accelerates the fade. The whiteboard is also smaller. The typical adult holds roughly four to seven items in working memory at once. The ADHD whiteboard holds fewer, and holds them less reliably.

The model distinguishes several components: the phonological loop (which maintains verbal information by rehearsing it), the visuospatial sketchpad (which maintains spatial and visual information), and the central executive (which coordinates the other components, directs attention, and manipulates the held information). ADHD does not affect all components equally, which matters for understanding why some tasks fail more reliably than others.

What the research shows

The evidence is consistent and substantial. Kasper and colleagues’ meta-analysis of 45 studies found moderate to large working memory deficits in ADHD across both verbal and visuospatial domains [1]. These were not borderline findings: the effect sizes were large enough that the deficits are among the most reliably replicated results in the ADHD literature. Martinussen and colleagues’ earlier meta-analysis of 26 studies found the same pattern, with spatial working memory more affected than verbal [5]. The spatial deficit is relevant because spatial working memory supports navigation, mental rotation, and tracking the layout of information, all of which affect daily functioning in ways that are hard to articulate but easy to feel.

Alderson and colleagues refined this further, distinguishing between deficits in the central executive (the component that manipulates information) and the phonological store (the component that holds verbal information) [2]. ADHD affects both, but the central executive deficit is larger. This means the problem is not only holding information; it is holding information while doing something with it. Following a recipe while cooking requires holding the instruction in mind while performing the action. Doing mental arithmetic requires holding the numbers while calculating. Having a conversation requires holding what the other person said while formulating your response. Each of these is a central executive task, and each is where the ADHD deficit bites hardest.

The deficit is present in both children and adults with ADHD. Adults may score better on lab tests than children because they have spent decades developing workarounds: subvocalizing, using environmental cues, writing things down. But the underlying capacity difference persists. The compensatory strategies make the deficit less visible without making it less present.

What this looks like in daily life

Losing track of multi-step tasks. You start a task with three steps. You complete the first step. By the time you finish, the second and third steps have dropped off the whiteboard. You stare at your work, knowing there was more, unsure what it was. You may complete step one and then start something else entirely, not because you decided to switch but because the remaining steps ceased to exist in your active awareness. Twenty minutes later you remember, and by then you have lost the context for where you were.

Forgetting what you were saying mid-sentence. The point you were building toward is gone. You can feel the shape of it, like a word on the tip of your tongue, but the content has dissolved. This is not a failure of intelligence or verbal ability. It is working memory failing to maintain the thread while you are constructing the sentence. The more complex the point, the more likely it is to vanish, because complex points require the central executive to hold multiple pieces in coordination, and that is exactly the component most affected.

Difficulty following conversations with multiple threads. Someone is telling you something while you are still processing the previous thing they said. The new information overwrites the old before you have finished with it. In group conversations, this effect is amplified: each speaker adds information that competes for the same limited space, and by the time the conversation circles back to you, the thing you wanted to say has been overwritten several times.

Re-reading the same paragraph. Your eyes move down the page, but by the time you reach the end, the beginning is gone. You go back. The same thing happens. This is not a reading comprehension problem. It is working memory failing to hold the accumulating meaning. Each new sentence pushes an earlier one off the whiteboard, so the paragraph never assembles into a coherent whole. You understand every sentence individually. You cannot hold them together.

The cascade failure. Working memory deficits rarely produce a single isolated mistake. They produce cascades. You forget step two of a task, so step three is done incorrectly, so the whole task needs redoing, so you are now behind schedule, so you rush the next task, so you make a different error. The daily experience is not one forgotten thing; it is a chain of consequences that follows from one dropped item.

What does not work

Working memory training programs, including commercial ones like Cogmed, improve performance on the trained tasks. But Cortese and colleagues’ meta-analysis found limited evidence that these improvements transfer to untrained tasks or to real-world ADHD symptoms [4]. You get better at the training exercises without getting better at remembering why you walked into the kitchen.

This is a consistent finding across cognitive training research: training improves the trained task, not the underlying capacity. The reason is that working memory capacity appears to be constrained by neurobiological factors, specifically the dopamine-mediated maintenance of information in prefrontal circuits, rather than by a lack of practice. The working memory deficit in ADHD is not a skill gap. It is a hardware difference. No amount of practice on a task changes the hardware.

The implication is important: strategies for working memory deficits in ADHD should focus on reducing reliance on working memory rather than trying to expand it. The whiteboard is the size it is. The question is what you put on it and what you put somewhere else.

What helps

External systems. Lists, timers, alarms, and written instructions compensate for what working memory cannot hold. The strategy is not to improve the whiteboard; it is to stop relying on it. If a task has three steps, write them down before starting. Not because you cannot remember them right now, but because you will not be able to remember them in four minutes. The list is not a crutch. It is a tool that does the work the prefrontal cortex is not doing reliably. The goal is to externalize everything that does not need to be held internally, so that the limited working memory space is reserved for the active manipulation that only you can do.

Medication. Stimulant medication improves working memory performance in ADHD, likely by increasing dopamine in the prefrontal circuits that support active maintenance. The improvement is real but partial: medication narrows the gap without closing it entirely. For many adults, the combination of medication and external systems is more effective than either alone, because the medication makes the executive coordination better while the external systems reduce how much coordination is needed.

Reducing cognitive load. Fewer open browser tabs. Fewer simultaneous projects. Fewer decisions queued. Each item held in working memory competes with every other item, and the ADHD whiteboard has less space. Reducing what needs to be held makes room for what matters. This is not about simplifying your life as a lifestyle choice. It is about recognizing that cognitive load is a finite resource and managing it accordingly. Close the tabs. Finish the current task before starting the next. Give your whiteboard a chance.

Environmental design. Keep frequently needed information visible: a physical checklist on the desk, the project brief pinned to the monitor, the relevant email printed and placed next to the keyboard. Each visible cue removes one item from working memory and puts it in the environment where it cannot be dropped. The more your surroundings remember for you, the less your working memory has to hold.

Where the evidence is thin

Most working memory research in ADHD has been conducted in children. Adult studies exist but are fewer, and the interaction between working memory deficits and the compensatory strategies that adults develop over decades is not well characterized. We do not know how much the measured deficit in a lab setting corresponds to real-world functional impairment after years of adaptation. An adult who has spent twenty years developing systems to compensate may show a moderate deficit on a lab test but experience significant real-world impairment when those systems are disrupted, or may show the same deficit but report less impairment because the systems are effective. The relationship between the lab score and the lived experience is not simple, and the research has not mapped it adequately.

We also do not know enough about how working memory deficits interact with other executive function deficits in ADHD. Working memory, inhibition, cognitive flexibility, and emotional regulation all share prefrontal resources. When multiple deficits co-occur, which they do, the combined effect is likely worse than the sum of the parts, but the interaction has not been well quantified in adults.

Getting a diagnosis or reviewing your options. If executive dysfunction is significantly affecting your work or daily life, speaking with a doctor about an ADHD assessment may be worthwhile. In the US, CHADD (chadd.org) has resources for adults seeking a diagnosis, including a provider directory. In the UK, the ADHD Foundation and AADD-UK offer guidance on the referral process.