Imagine slowly forgetting what "dog" means. Not the word itself. The actual concept. What a dog is. First, you can't come up with the name when you see one. Then all dogs start becoming "big cats" or just "animals." Eventually, you genuinely can't tell a dog from a chair. They're both just... things in the world.
This is the terrifying reality for people with semantic primary progressive aphasia (PPA), and a study in Brain has mapped exactly how this dissolution of meaning corresponds to the brain literally shrinking away. The progression is predictable, it follows the architecture of how we store concepts, and that predictability might eventually help us do something about it.
The Disease With Too Many Letters
The study focuses on frontotemporal lobar degeneration with TDP-43 type C. That's a lot of syllables to describe something heartbreaking: a specific form of dementia that targets your brain's semantic system, the part that stores meaning.
This pathology invariably starts in the temporal pole, which sits at the front tip of your temporal lobes. Think of this region as your brain's "concept hub," the place where you store abstract knowledge about the world. What makes a dog a dog? What makes a chair a chair? The temporal pole helps bind these concepts together.
When the left temporal pole gets hit first, patients develop semantic PPA. Their grammar stays perfectly intact. Their speech production works fine. They can construct a grammatically beautiful sentence. It just might not mean anything coherent, because the concepts themselves are dissolving.
Your Mental Dictionary Burning Page by Page
The researchers identified three stages of semantic decline. Watching someone progress through these stages is like watching a well-organized filing cabinet slowly catch fire, with files becoming illegible in a predictable order.
Stage one is anomia. This is when you can't retrieve names for things, but you still understand the words when someone else says them. You point to a dog and say "that's a... um... you know, the thing that barks. With the tail." Someone says "dog" and you immediately know what they mean. The word is in there somewhere; you just can't grab it.
Stage two is intra-category blurring. Now things get more unsettling. You start confusing related concepts. A tiger becomes a "big cat" (which, fair enough, but the distinction matters). A hammer and a screwdriver become interchangeable. The meaning is technically still present, but the boundaries between similar things are getting fuzzy. Everything is losing resolution.
Stage three is inter-category blurring. This is when the system truly breaks down. Patients can't distinguish unrelated concepts. A dog might be confused with a lamp. A fork might be called a "thing for walking." The entire fabric of semantic memory has come unraveled. The filing cabinet isn't just damaged; whole sections are now blank.
Following the Shrinking Brain
The researchers didn't just describe these stages; they mapped which brain regions were most atrophied at each point. Using MRI data from 24 visits across their patient population, they tracked where the damage was worst as the disease progressed.
Early anomia corresponded to temporopolar atrophy. The front tip of the temporal lobe, the concept hub, showed the most shrinkage when patients were in that "I know what it is, I just can't find the word" phase.
As semantic impairment progressed to those middle and late stages, atrophy spread backward. The fusiform gyrus got hit, then lateral temporal regions, then temporooccipital areas. This progression isn't random. It follows the architecture of how we actually store concepts in the brain.
Think of it this way: abstract categorical knowledge (is this thing alive or not?) sits closer to the temporal pole. More specific detailed knowledge (what makes a golden retriever different from a labrador?) is stored in areas further back, closer to where visual processing happens.
The disease essentially erodes concepts from the top down, from abstract to specific, as it spreads through the semantic network. First you lose the word, then you lose the category boundaries, then you lose the ability to distinguish anything at all.
Why Predictability Offers Hope
There's something darkly reassuring about finding that a disease follows rules. If semantic PPA progressed randomly, with different brain regions failing in unpredictable sequences, it would be much harder to study and nearly impossible to target with treatments.
But the consistent pattern suggests the disease is exploiting specific vulnerabilities in how the semantic system is organized. Understanding those vulnerabilities in detail might eventually reveal intervention points.
Could we develop therapies that slow the spread from the temporal pole to more posterior regions? Could we shore up the remaining semantic network as the front begins to fail? These are the questions that predictable disease progression allows researchers to ask.
For now, this study provides a detailed map of a terrible journey. We know which signposts to expect, in what order. That's cold comfort for patients and families living through it. But in the long history of medicine, mapping the problem accurately is usually the first step toward solving it.
The words don't have to disappear forever. But first, we need to understand exactly how and why they fade.
Reference: Barbieri E, et al. (2025). Atrophy progression in frontotemporal lobar degeneration-TDP-C with primary progressive aphasia. Brain. doi: 10.1093/brain/awaf369 | PMID: 41032666
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.