# Digital Dementia and the Smartphone Brain: What the Research Found > Digital dementia, or the smartphone brain, is an informal label for memory and concentration said to decline through heavy device use. It is not a medical diagnosis. A 2025 meta-analysis of roughly 411,000 adults found that digital technology use was associated with lower, not higher, risk of cognitive impairment. - URL: https://brainrank.jp/en/blog/digital-dementia - Published: 2026-08-18 - Publisher: BrainRank Editorial Team (https://brainrank.jp) - Tags: Cognitive abilities, Memory Digital dementia, often called the smartphone brain, is an informal label for memory and concentration said to decline through heavy device use. Two things are worth establishing first: it is not a medical diagnosis, and the direction large-scale research has found is the opposite of what the label implies. ## What digital dementia is: a label, not a diagnosis The term spread through Digitale Demenz, a 2012 popular book by the German psychiatrist Manfred Spitzer, and took hold in Japan alongside book titles about the smartphone brain. There are no standardized diagnostic criteria and no instrument, so there is no way to determine whether someone has it. In clinical use, dementia refers to a persistent decline in memory and judgment caused by brain disease, severe enough to interfere with daily life. Feeling foggy after a day of heavy scrolling differs from that in both cause and duration. The same discipline applies here as in [what cognitive abilities are](https://brainrank.jp/en/blog/cognitive-abilities): do not confuse a state with an ability. ## The 2025 meta-analysis pointed the other way The hypothesis was tested directly by [Benge and Scullin (2025, Nature Human Behaviour)](https://doi.org/10.1038/s41562-025-02159-9). They pooled 57 of the 136 papers that met inclusion criteria, covering roughly 411,000 adults with a baseline mean age of 68.7. Digital technology use was associated with an odds ratio of 0.42 (95% CI 0.35-0.52) for cognitive impairment and a hazard ratio of 0.74 (95% CI 0.66-0.84) for the rate of cognitive decline over time. The associations survived adjustment for age, education, income, and health. The authors present this as support for a technological reserve hypothesis rather than a digital dementia one. Two cautions on reading it. The pooled studies are observational, so causation is not established; healthier people with better-preserved cognition may simply take up new devices more readily. And the sample is adults averaging 68.7 years old, so the analysis speaks neither to children and adolescents nor to short-form video specifically. ## Why forgetting still feels worse Large-scale data pointing the other way does not make the experience imaginary. Two mechanisms account for much of it. The first is memory offloading. [Sparrow et al. (2011)](https://doi.org/10.1126/science.1207745) reported that when people know information will remain available, recall for the content drops while memory for where to find it strengthens. The capacity did not disappear; the target moved from contents to location. The second is attention. Encoding requires that attention reach the information in the first place. With notifications and short videos splitting attention, the encoding stage is never entered. In many cases the accurate description is not that you cannot retrieve it but that it never went in. [Working memory](https://brainrank.jp/en/blog/working-memory-training) absorbs that load, and it connects directly to the switching cost covered in [brain rot](https://brainrank.jp/en/blog/brain-rot). ## What actually helps Creating moments where attention reaches what you want to keep works better than cutting screen time. - Restate something once, without looking, right after you read it - Try for ten seconds to recall before you search (retrieval practice) - Fix set times to check notifications and switch them off otherwise - Prioritize sleep; consolidation happens during it - Offload figures and proper nouns freely, but assemble procedures and understanding yourself That last point is the same criterion used in [AI and cognitive offloading](https://brainrank.jp/en/blog/ai-and-cognitive-offloading). Offloading is not harmful in itself; the choice of what to offload is what separates the outcomes. ## If something has changed This article is a general explanation, not a diagnosis or medical advice. If you repeatedly forget dates or commitments, or lose the steps of familiar tasks, and it interferes with daily life, consult a medical professional rather than self-assessing. For a rough picture of your current tendencies, [BrainRank's free IQ test](https://brainrank.jp/en/start) is a starting point. It reports domain-level reasoning scores across 20 questions in about 10 minutes, with an estimated IQ on the mean-100, SD-15 scale. It does not detect memory disorders and is an estimate for self-understanding only. ## Frequently asked questions ### Is digital dementia a real diagnosis? No. It is neither a medical diagnosis nor a condition with a standardized test. The label comes from a 2012 popular book and needs to be kept separate from dementia as a clinical term. ### Does using a smartphone cause dementia? The current evidence points the other way. In a 2025 meta-analysis of about 411,000 adults with a baseline mean age of 68.7, digital technology use was associated with an odds ratio of 0.42 for cognitive impairment and a hazard ratio of 0.74 for cognitive decline. These are pooled observational studies, so they do not establish causation. ### Then why does it feel like I forget more? Memory offloading and divided attention explain much of it. Information you know you can look up later is encoded less well, and information you never attended to was never encoded at all. If age-related change concerns you, consult a medical professional. ## Sources - [Benge and Scullin (2025, Nature Human Behaviour)](https://doi.org/10.1038/s41562-025-02159-9) - [Sparrow et al. (2011)](https://doi.org/10.1126/science.1207745) --- Statistics and percentages are calculated from a normal distribution model with a mean of 100 and a standard deviation of 15. The tests on this site provide estimates for entertainment and self-understanding; they are not medical or clinical assessments, nor official intelligence tests. When citing, credit "BrainRank (brainrank.jp)" and link the article URL.