Open your phone and you’ll find an archive of good intentions. You’ll notice hundreds of bookmarked articles, a fully choked folder of photos, and saved reels you swear you will watch eventually. PDFs have been downloaded “for future reference” along with newsletters waiting for that mythical free afternoon when you’ll finally catch up. The more ambitious version of the same impulse is what psychologists call the “Second Brain”. Apps like Notion, Obsidian, Roam, and Google’s NotebookLM guarantee the user that technology will store what our increasingly distracted minds cannot. The term was popularised by Tiago Forte’s 2022 book Building a Second Brain, though the underlying idea is of an external memory you can search, connect, and trust. It is obvious people are remarkably good at collecting knowledge, but are far less certain about what happens afterwards. Somewhere between saving an article and actually reading it, procrastination is usual. The Second Brain trope raises important questions, especially if you are a student or researcher seeking to build real expertise. Pressing “save” feels satisfying, even when you know you will probably never get back to examining or at least look at what you have saved.
Does storing information reduce the anxiety of forgetting it? Or is procrastination a disguise? As AI makes retrieval nearly effortless by surfacing, summarising, and connecting information at extraordinary speed, is the Second Brain getting to be more useful and a more seductive proposition? Suhas Motwani and Sanket Bajaj are two people who built an external memory system, only to discover its limitations the hard way. Morwani, all of 31 years old co-founded The Product Folks, an online community for professionals living in Bengaluru to connect. He didn’t like what he found. “I promised to do or get things done during meetings, intros, and follow-ups, but would end up dropping one-third of them,” he says. “Because things were falling significantly, I built an external memory.” The system worked because it caught all of what he was forgetting or dithering about. “Note-takers call this syndrome the collector’s fallacy,” he says. “My archive grew fast in months, but my thinking didn’t match up.” Motwani’s confession is relevant particularly for students. An archive that expands while your thinking stays the same is not learning; it is a storage system dressed up as learning. In March this year Sanket Bajaj, 25, who handed over his financial advisory firm, Viable to his sister after four years on the job. When he opened every folder to find a usable record of everything he had collected almost nothing was worthwhile. “No tool could encompass the knowledge and the context of what I had accumulated, connected and carried in my mind,” he says. “Unfortunately, you can’t transfer brains.”
An entire industry has been built on the premise that with the right tool, our collected knowledge can be useful in the end. Bajaj’s experience suggests a flawed premise from the beginning itself because the richest aspects of expertise exist in tacit, associative, and semi-conscious connections that are never written down because they didn’t feel like “facts” worth saving. The Second Brain has failed you.
The Psychology of the Save Button
In the digital age, devices have become emotional attics with voice notes, unused recipes, bookmarked essays and screenshots of restaurants we may never visit. All these document versions of ourselves we once imagined would be doing. This is not a fringe behaviour. A survey in December 2025 showed 69 per cent 2,000 Britons found themselves digital hoarders to some degree, while only 29 per cent regularly cleared out old files. A third of those surveyed of age 25 to 34 were emotionally attached to their digital clutter. Part of this hoarding tendency is a consequence of plenty. Physical clutter can be cleaned up: a cupboard fills up, a bookshelf buckles, a drawer refuses to close. A screenshot costs nothing on the phone or laptop, a bookmark will appear in a second, and cloud storage means accumulation of them can be postponed as long as you want. The device becomes an extension of memory where we can deposit information without checking out what they really mean, or whether we need them at all, or whether we will check out them later. Saving creates the feeling of having secured an opportunity because a recipe downloaded can be cooked later, the article can still be read and idea can still be converted into a viable project. Deleting it, by contrast, feels like closing a door.
Digital hoarding, in this light, is often not about the object itself but is about preserving the possibility represented by the object. The bookmarked essay promises to make the hoarded will become well-read, and the saved lecture notes will make the student to finally become organized while deleting the file can feel, irrationally, like abandoning that future version of themselves. Bengaluru-based consultant psychiatrist Dr Prerna Maheshwari who is associated with Amaha Health and Manipal Hospital suggests how to separate the healthy version of this behaviour from the unhealthy one. “Saving something because you genuinely intend to use it later can be helpful. But if we keep saving because it feels like we are being productive, while postponing actual decision-making is a sign of avoidance.” This distinction matters enormously for anyone engaged in academic or knowledge work. Cognitive offloading—deliberately delegating a task to an external system to free your working memory for higher thinking—is a well-documented and genuinely useful strategy. The tell is what happens next: will the saved item be retrieved, processed, and used later? Or does it simply accumulate in its folder? Researchers from Northumbria University, who surveyed 846 people, found that digital hoarding statistically explained 37 per cent of a person’s total anxiety score. The invisible clutter sits in the background as unfinished business: messages to answer, documents to read, links to investigate. This accumulation itself turns into a chronic stressor.
Retrievability is Not the Same as Comprehension
Here is the conceptual crux relevant to a student: knowing something and knowing where to find something are different. Dr Maheshwari names the swap directly: “Both are useful, but not equal.” The risk is that irretrievability starts to become a substitute for actual requires. You save the article, and know you can find it whenever you want, but at no point the content is accessed by you.
This is not a new problem. Anyone who has passed an exam using highlighted notes could not reproduce the info a week late. Highlighting a textbook feels like a student is studying because it creates the sensation of having engaged with the material. But highlighting something means recognition, not recall. Recognition is a less durable form of memory. Learning requires the effort of recalling an idea without looking it up, and then explaining it in your own words. Applying it to a new problem and getting it wrong doesn’t stop you from trying again. Retrieval repetition techniques are the exact opposite of saving, bookmarking, and screens shots. None of these actions require generating anything original which is exactly why the user feels effective though little is accomplished. A perfectly organised Notion vault full of highlighted PDFs and AI-generated summaries creates a powerful illusion of mastery. They know the information is retrievable: so chill. Exams, interviews, and real intellectual work demand understanding information.
Comprehension Debt and the Problem of Staleness
Chetan Mendhe, 33, Director of Product Development in Toronto, has a term for this syndrome: comprehension debt. “The knowledge is preserved, but my own map of it has thinned out,” he says. This is a suitable metaphor. Like financial debt, comprehension debt compounds without you even realising it: every unread article or unprocessed note is a loan against future understanding, only that there is interest to be paid when the information is required later. A related and more dangerous problem is that outdated information doesn’t mention it when it is outdated. Mendhe says bluntly: “Outdated information is the biggest peril, one even bigger than hallucination.” AI models have, by his account, largely caught up on hallucination as a failure mode, because a confidently phrased answer has no visible timestamp. Motwani frames the danger sharply: a fact retrieved confidently in 2026 but was true only in 2023 “is worse than no answer at all, because fluency reads as authority”. A fluent, well-formatted answer feels trustworthy regardless of whether it’s current or obsolete, which makes stale information dangerous. Bajaj has put the fix in his own system. Any personal pattern or fact without fresh supporting evidence in sixty days is automatically flagged “confirm or retire”. This discipline can be followed by any student.
So What is a Second Brain Actually Good For?
These are not arguments against note-taking or bookmarking. The tool is rarely the problem. Dr Maheshwari’s practical litmus test cuts through the confusion. “The best system is the one with the least friction between capturing something, finding it when needed and actually using it,” she says. By that standard, a notebook can beat an elaborate app, while a consistent habit can beat both. Complexity in the system substitutes for mental effort, precisely because building the system itself produces a satisfying feeling of progress. Dr Agrima Thakur, a global health researcher, draws a sharp line between what technology should and shouldn’t be trusted to do. “Translation apps will get me through any conversation today,” she says. “But only my memory can build a vocabulary. The human mind must and should remember things.” She compares memory to a muscle which must be exercised, applied, and continually updated to avoid the risk of atrophy regardless of the tools. On the growing habit of asking AI to summarise and synthesise, she is similarly direct: “A machine shouldn’t tell you how to think.”
The Question Behind the Question
The deeper question about the Second Brain is not really about apps or attention. It is about the kind of relationship you want with what you learn. If the goal is simply to have access to information whenever required, a well-organised archive and a good search function is all you need to do the job. But if the goal is to get the kind of understanding that lets you improvise, argue, teach, or apply an idea in an unfamiliar context, then no archive, however well-built, can substitute absorbing the material by your own mind. Building a better filing system is the easy part, but the harder, slower, less satisfying work is to decide, and article by article and idea by idea, on what information actually is relevant to your thinking process, followed by doing the hard work of making that happen. As Dr Maheshwari’s framework suggests, that is the whole game: not whether just to save something on your phone or hard drive, but whether you’re saving it for use, or to stop thinking about it. A Second Brain, no matter how sophisticated, cannot make that decision for you.
The famous Ebbinghaus forgetting curve has been replicated: retention falls as the interval between learning and recall increases. A 2015 replication tested intervals from 20 minutes to 31 days and found broadly similar patterns to the original work
Workers spend 28% of their workday on email. A Second Brain effectively tries to move useful information out of transient channels and into a persistent, searchable knowledge system
McKinsey’s research found that interaction workers spend roughly 19-20% of their time searching for and gathering information
McKinsey estimated that faster and more accurate access to information generated up to $117 billion in productivity gains for knowledge workers in the countries studied
McKinsey’s analysis estimated 69.9 million knowledge workers in India and valued the potential annual productivity gain from enterprise search at $5-8 billion in 2009 terms
Boundaries: What a Second Brain Should Never Hold
Most people do not notice that the Second Brain, somewhere around the second week of enthusiastic system-building, passwords, PINs, and anything genuinely confidential is not collected an saved. Second Brains, as tool, is not designed to keep extremely sensitive information securely. Institutions must exercise the same caution for proprietary organisational or financial data fed into external AI systems, since a company’s proprietary context is often its actual competitive edge. For a student, this applies to unpublished research, personal information, or anything that does not need indexing and is retrievable by outsiders.