This post reprints from a reply to a comment to the last post. I thought it was interesting enough to feature. First, I want to express my gratitude for Dr. Dorothy Ross, who have has expressed consistent interest in this blog and have has commented thoughtfully and insightfully.
The question is, after a stroke, do neurons die, are damaged or merely dormant?
There was a moment within an hour of the stroke when recovery returned rapidly, much too rapidly for new neuron creation or rerouting, unless rerouting is virtually instant. Other aspects of language didn’t recover for weeks, even though I worked on those areas persistently, so, presumably, neurons don’t reroute instantly.
So I’d hypothesize that many neurons were simply waiting for blood flow to return immediately to normal function. So they were dormant. Still, the MRI apparently showed that these areas were affected. I haven’t seen all the MRI details yet.
Other neurons, not so lucky, were damaged or died.
Other linguistic functions were slower, over hours, some days and some weeks. Some have not fully recovered. So there may have been a gradient of damage, unless neurons reroute or rework within hours, days or weeks. This may seem obvious, but I assume that part of [the] area affected of the stroke was undamaged, part was damaged but recoverable in time, and part may have been completely lost (although I can look only at eighteen days so far).
It’s true that scripts and music scores and terrain can be memorized quickly and can cement permanently into the mind over night. But language is not exactly mere memorization of repetition, but a machine that can generate unbounded new sentences from a variety of relations. Similarly, memory retrieval also seems to have a machine structure (I think in Chomsky-Fodor rules and representations, partly because I don’t know parallel processes as well). Memory retrieval and grammar are both recalcitrant to recovery. It seems likely that more is required to structure language and memory retrieval than mere memory or rote.
So, I conclude that while neurons can make all sorts of new memories, language structure and retrieval require something more, some quantitatively or qualitatively different kind of work. That endorses the view that the neurons that improved at the first hour had not died or damaged, but were simply waiting for circulation. Likewise, the linguistic deficits that improved more slowly must have been damaged or died.
Here’s a thought: when we memorized a speech or a melody, we have before us the sensual evidence, the sound, the words. But grammar is not a sensual presence. Neither is retrieval. Both are methods, and more cryptic than, say, method of bike riding. Watching the route of a bicycle rider may be simpler than memorizing how to ride oneself, but at least learning how to ride is about concrete sensual evidence: feet and peddles and wheels and handlebars and hands. With grammar, most people don’t have a clue what the parts of grammar that would correspond to wheels and handlebars and legs and peddles and arms. In fact, linguists themselves do not entirely understand all the relations of grammar.
I mean that maybe there isn’t anything qualitatively different about grammar and retrieval at all. Maybe they’re just obscure to the senses.
Back to evidence. Within the first hour, I was aware of rapid improvement. Blurry vision lasted only a few minutes, less than a half hour. I was able to construct a sentence shortly after I’d taken a shower, within, I’d guess, forty minutes, at most, after the stroke. Improvement seem seemed at an asymptote, rapid improvement at first and increasingly slower improvement at an almost even decline, although I still notice marked renewal every day.
It may be that I have compensated cognitively, but, on the whole, I am so stubborn that I have not been focused on compensation. That is, I stubbornly want to regain old methods of sentence structure and vocabulary at the expense of skirting around or pushing through and not stumbling. So I’m inclined to believe that nearly all the renewed fluency is brain recovery, not cognitive compensation. So, either neurons have revived, or new neurons have taken their place, or networks have taken their place. The question is why some areas of grammar remain recalcitrant. The loss of “be” is one. Is it just obscure? Or is there a specific “be” network or link, and that died? Or is “be” a generally weak element of the language capacity? (The last should be easily established empirically — do many aphasics lose “be”?)
Regarding the grammatical sentences that were apparently meaningless, that I uttered at the beginning at the stroke: I know that I let loose with a fluent sentence, or part of a fluent sentence, but I couldn’t understand what I heard. It is possible that I said exactly what I intend intended to say, I just couldn’t hear what I had said. I was, however, quite shocked at the sentence because, whatever I had said, I didn’t understand it hearing.
So, here’s another Wernicke’s idea from an insider: it’s possible that a sentence begins appropriately to the speaker’s intention, but, not hearing the expectation, things derail quickly. I don’t imagine this is a thorough explanation for Wernicke’s grammatical-yet-meaningless fluency. But this may play into the problem. Lack of comprehension is not merely a problem of listening to others. Lack of comprehension includes one’s own heard speech. Just imagine that your own words sound completely meaningless. You’d wonder what you’d be saying, since you’d expect your own words to be exactly what you intend, and yet they don’t sound at all what you expect. It’s pretty shocking. But in fact your words may be exactly what you intend. Unfortunately, I have no record of the first utterances when the stroke was fresh and I had complete loss of comprehension.
aphasic diary 
January 22, 2010 at 4:20 am |
I think your analysis of the recovery process is very much on target. It definitely takes time for neurons to regenerate.
It is true that grammar allows us to generate an infinite number of sentences. To me what is even more interesting is that we tend to repeat the same sentences every day, with minor variations. One metaphor is a strong trunk of basic sentences that are used every day, with branches of less frequent phrases that are combined with basic sentences to make less common variations. Another metaphor is dance steps that are repeated over and over again in different combinations. But we tend to use certain combinations much more often. So even though grammar exists that can produce an infinite number of unique sentences, for all practical purposes, much communication in daily life occurs with stock phrases. Even if we can no longer play the grammatical game, we can still make some functional moves.
You asked a very interesting question about whether Wernicke’s says what they meant to say, but it does not sound right to them, so the process gets derailed. I don’t know what happened in your particular situation. But looking at an analogy – if I am touch typing and put my hands in the wrong position, I will be typing but nonsensical letters are coming out. So my response is to stop and re-evaluate the situation. If it kept happening, I would give up. That was your response also, as you mentioned in a previous post – you would not talk because you didn’t want to say the wrong thing.
However, the outstanding characteristic of Wernicke’s is the fluent, copious speech and apparent LACK of awareness that anything is wrong, that is so obvious to the listeners. To me that indicates that the internal monitor is not picking up that the output is incorrect. It also makes sense that one reason people with Wernicke’s talk so much is because the internal monitor is not giving them a signal that they have spoken enough.
The closest normal speaking situation that I can think of is the very minor occasional instance when the speaker substitutes a different word than the one he intended and does not notice it. Probably it would be even be apparent to the listener unless it drastically changes the meaning of the sentence.
It is a good question about whether “be” is simply weaker, or whether it has a specific location in the brain. You mentioned in another post that this grammatical form varies among some dialects of English. As far as I know, no one has examined the frequency of occurrence of a “be” deficit in aphasia, as compared to other grammatical deficits.