Early signs

Is It Anxiety or Dementia? Why Stress Can Cause Memory Slips

By Viktor Stevanovic · Published 4 August 2026 · 10 min read

Short answer: Anxiety, stress and low mood are common causes of memory slips, and they are often treatable. But anxiety and dementia cannot be separated by self-assessment. They overlap, they can be present together, and worry about your memory is a reason to get checked — not a reason to relax.

Educational, not a diagnosis. Nothing here can tell you what is or isn't happening in your case, and nothing here can estimate your personal odds. If your memory worries you, the next step is an appointment.

Stress and anxiety really do affect memory

This isn't "all in your head" in the dismissive sense. Cortisol, the main stress hormone, crosses into the brain and binds to receptors in the hippocampus, prefrontal cortex and amygdala — regions involved in forming and retrieving memories (Lupien et al., Nature Reviews Neuroscience, 2009).

There is controlled evidence that this changes performance, though it is narrower than it sounds. In a randomised, double-blind study of 51 healthy adults aged 18–30, only the higher dose — approximating cortisol levels seen in major physical stress, and given to about a third of the sample — produced reversible decreases in verbal declarative memory; the lower, mild-stress dose had no measurable effect (Newcomer et al., Archives of General Psychiatry, 1999).

What it measured was recall of spoken paragraphs, not the attention-type slips people usually describe — losing the thread mid-sentence, blanking on a familiar word — which it specifically found were unaffected. It was also done in healthy young adults under experimental conditions, so it says nothing about the cause of any one person's memory problem — and cortisol reversibly lowering memory scores is not the same as treating anxiety restoring memory.

Why you can't sort this out from the inside

Worrying about your memory is not evidence against dementia. Researchers call this subjective cognitive decline: you notice a change, but formal tests come back normal. In a meta-analysis of roughly 29,700 people followed for an average of 4.8 years, those with subjective memory complaints were about twice as likely to develop dementia as those without (relative risk 2.07); the annual conversion rate was about 2.3%, and around 14% progressed to dementia over four or more years (Mitchell et al., Acta Psychiatrica Scandinavica, 2014). Both halves matter: most people in those studies did not develop dementia in that window, and the complaint was still a real risk marker. Neither figure tells you anything about you personally.

Worry itself doesn't point away from it either. In a meta-analysis of 12 longitudinal studies of people with subjective cognitive decline, anxiety or worry about the memory changes was associated with a 40% higher risk of progressing to measurable impairment (RR 1.40, 95% CI 1.20–1.63). Depressive symptoms were not (Desai et al., Ageing Research Reviews, 2021). That is a marker, not a mechanism you control: worry may simply be a signal that the change someone noticed is real. Nothing in it suggests you are harming yourself by worrying, or that you could help by worrying less. Pooled data from six cohorts (10,394 people) similarly put the relative risk of dementia in people with anxiety at 1.29 (95% CI 1.01–1.66), with the same caveat from the authors: it is unclear how much is anxiety as a cause versus anxiety as an early sign (Santabárbara et al., Maturitas, 2019).

And mood-related impairment genuinely does improve for many people. A systematic review of 18 studies covering 284 patients with so-called pseudodementia — cognitive impairment alongside a psychiatric illness, usually depression rather than anxiety — found 53% no longer met criteria for dementia at follow-up, while 33% were later diagnosed with dementia (Connors, Quinto & Brodaty, Psychological Medicine, 2019). These were patients already assessed as cognitively impaired and being treated for a psychiatric illness — a much smaller and more unwell group than people who are simply worried about their memory, so neither percentage transfers to a reader.

Together those say something calm and ordinary: this question is answered from the outside, not from introspection. A first assessment doesn't always settle it — normal results and a repeat check in six to twelve months is a common and reasonable outcome — but it moves you from guessing to a documented starting point.

What clinicians look at — and why it isn't a test you can take

These are tendencies seen in groups. Anxiety and dementia often co-occur, each can look like the other, and matching a column rules nothing in and nothing out.

Often seen with anxiety / stress Often seen with dementia
Awareness Acutely aware of the lapses, distressed by them Varies — family sometimes notice first, but many people early on are very aware
Onset Fluctuates with stressful periods Gradual change over months to years
Type of lapse Concentration, "tip of the tongue", forgetting under pressure Forgetting recent events, repeating questions, getting lost in familiar places
Daily function Usually coping, even if it feels harder Growing difficulty with familiar tasks, money, navigation
When rested Often eases with rest and lower load — though a single rested day settles nothing Less likely to lift with rest

If you found yourself matching a column, that's normal and it isn't a result — plenty of people match both, and one good or bad day tells you nothing. And one correction worth stating outright, because the opposite circulates widely: being aware of your own memory lapses does not rule dementia out — and doesn't rule it in either. Awareness isn't diagnostic in either direction, which is why this needs someone else's assessment. Take the description to the appointment, not the conclusion.

Treatable causes worth ruling out

Several conditions produce memory and thinking problems that resemble dementia, and some are simple to check. This is the genuinely useful part of getting seen.

Be straight about the numbers, though. Among people who already meet the criteria for dementia, fully reversible causes are uncommon: a meta-analysis of 39 studies covering 7,042 patients, 5,620 of whom had dementia, found a potentially reversible cause in 9% of dementia cases and actual reversal in 0.6% (0.31% fully) (Clarfield, Archives of Internal Medicine, 2003). That applies to people already diagnosed — not to the much larger group who are simply worried. Thyroid, B12, medication, mood and sleep are worth checking because they are common and easy to test, not because they usually explain dementia.

When to make the appointment, and what happens there

Arrange an evaluation — without alarm, but without putting it off — if the changes are steadily worsening rather than coming and going with stress; if other people have started noticing; if you're repeating questions, getting lost somewhere familiar, or struggling with routine tasks; or if it's affecting work, finances or daily life. Those features make an assessment more time-sensitive, not less. If none apply and you're still worried, that is also a reasonable reason to go.

The appointment is more ordinary than most people expect. A doctor will ask about your symptoms and everyday tasks, examine you, and may arrange blood and urine tests plus a short memory and thinking test. If questions remain, you may be referred to a specialist or memory clinic for more detailed testing and sometimes a scan (NHS). Book first — you can prepare while you wait. Three things make it go better: written-down examples collected between now and the visit, your full medication list, and someone who knows you well in the room. If you're reading this in the middle of the night, writing down the first example is a reasonable thing to do now; nothing else has to be decided before morning.

Where the worry can usefully go

While you arrange that, there's something constructive to do with the energy. The 2024 Lancet Commission estimates that around 45% of dementia cases worldwide could potentially be prevented or delayed by addressing 14 modifiable risk factors, among them depression, social isolation, physical inactivity, hearing loss, high blood pressure and smoking (Lancet Commission, 2024). Several overlap with what helps mood and stress. That isn't a substitute for being assessed.

That's what Solenna is for. The free, 3-minute profile turns vague dread into a clear picture of your own modifiable risk factors and the steps that move them. It does not diagnose anything and cannot tell you whether you have dementia.

Check your dementia risk profile — free
Common questions

Can anxiety really feel like dementia?

Yes — and that is why it can't be settled at home. Stress and anxiety affect concentration and memory, and controlled research has shown cortisol at major-stress levels producing reversible decreases in verbal memory in healthy adults (Newcomer et al., 1999). But how the slips feel does not distinguish the two: they overlap, each can look like the other, and they often occur together. Feeling anxious is not evidence that nothing else is happening, and noticing your own lapses is not evidence that something is. Telling them apart takes a clinical assessment, not a self-check.

Does memory loss caused by stress or low mood go away?

It often does, though not always, and not predictably. In a systematic review of pseudodementia — cognitive impairment alongside psychiatric illness, mostly depression rather than anxiety — 53% of 284 patients no longer met dementia criteria at follow-up, while 33% were later diagnosed with dementia (Connors et al., 2019). Those patients were already assessed as cognitively impaired and treated for a psychiatric illness, so neither figure transfers to someone who is simply worried about their memory, and no comparable long-term data exists for anxiety specifically. Which pattern applies in one person is what an evaluation is for.

Does worrying about my memory mean it's probably just anxiety?

No — worry doesn't make dementia less likely, which is why it's worth getting checked rather than waiting. In studies, people who noticed memory changes were about twice as likely to develop dementia as those who didn't (Mitchell et al., 2014) — though most people in those studies did not develop dementia in that period. Those are group averages and say nothing about any one person, including you. Worry isn't information in either direction; an assessment is.

How can I tell whether it's stress or something more serious?

You can't reliably tell on your own, and no online checklist or symptom table can either — anxiety and dementia overlap and often occur together, so matching a description rules nothing in and nothing out. That is not a reason to panic or to relax; it's why the question gets answered from the outside. Most people who go and ask this question do not turn out to have dementia, and stress, low mood, sleep, thyroid problems and medication effects are common and checkable explanations — but which one applies to any individual is exactly what an assessment is for, not something to settle at home. Write down concrete examples, note anything other people have mentioned, bring your medication list, and take it all to your doctor — sooner if the changes are steadily worsening, if others have noticed, or if daily tasks are affected.

Should I see a doctor if I'm just anxious about my memory?

Yes — that on its own is a reasonable reason to go, and you don't need to wait for things to get worse. An appointment can look for treatable contributors such as thyroid problems, B12 deficiency, medication effects, sleep disorders and mood, and it replaces guesswork with a documented baseline. It may not settle the question in one visit — normal results and a repeat check in six to twelve months is a common and reasonable outcome — but it is the route to an answer that self-assessment cannot give. Start with your primary care doctor or GP.

Related reading
Sources

How we research and review our content →

Educational content only. This article is not medical advice, diagnosis, or treatment: it does not diagnose anything and cannot tell you whether you have dementia. The studies cited — Mitchell et al. (2014), Desai et al. (2021), Santabárbara et al. (2019), Connors et al. (2019), Clarfield (2003), Newcomer et al. (1999) and Wang et al. (2020) — report averages and associations across groups of people, not conclusions about any individual, and none of them can tell you what is happening in your case. Solenna is not a medical device and does not diagnose, treat, prevent or cure any condition. Risk reduction means lowering probability, not eliminating it. If you are struggling with anxiety, low mood or memory concerns, please speak with a qualified clinician.