Language that exercises the brain; poetry and gradations of understanding
« previous post | next post »
If you want to give your brain fortifying nutrition, what kind of language should you feed it?
"Complex, unfamiliar sentences make the brain’s language network work harder"
A new study finds that language regions in the left hemisphere light up when reading uncommon sentences, while straightforward sentences elicit little response.
Anne Trafton | MIT News (January 3, 2024)
—
"Complex, Unfamiliar Sentences Make the Brain's Language Network Work Harder." Science Daily, January 3, 2024. .
Discussing "Driving and Suppressing the Human Language Network Using Large Language Models." Tuckute, Greta et al. Nature Human Behaviour (January 3, 2024).
Fair enough, maybe, but what if you lose your auditor / lector altogether? It seems to me that the type of brain-food sentences cited in this study border on incomprehensibility.
With help from an artificial language network, MIT neuroscientists have discovered what kind of sentences are most likely to fire up the brain’s key language processing centers.
The new study reveals that sentences that are more complex, either because of unusual grammar or unexpected meaning, generate stronger responses in these language processing centers. Sentences that are very straightforward barely engage these regions, and nonsensical sequences of words don’t do much for them either.
For example, the researchers found this brain network was most active when reading unusual sentences such as “Buy sell signals remains a particular,” taken from a publicly available language dataset called C4. However, it went quiet when reading something very straightforward, such as “We were sitting on the couch.”
Hmmmmm…… As I was reading through this thought-provoking article, all of a sudden the question of mutual intelligibility popped into my mind. This is a topic that has often come up on Language Log when we discuss the differences between different dialects, topolects, and languages, as they grade / fade into each other on a continuum.
Let me give a personal example. I will start in a college classroom in Beijing, where most people are speaking MSM, and I can understand almost everything that is being said without making any special effort. In the evening, I go home to spend the evening in a hutong ("alley") where heavily colloquial Pekingese is being spoken, and I have a hard slog of it through all the retroflexion, slurring, and unique lexical items — e.g., der / dianr ("scram; skedaddle"). The next day I fly to Chengdu in southwest China. There I can understand maybe 80% on the street in the city, but the different tones put a strain on my ability to comprehend fully, and local lexicalisms such as modeile ("there aren't any") are impenetrable. Over the weekend, I take a trip to the big Buddha at Leshan and cross the river to climb Emei shan. Hiking along the way, I encounter many locals (they are Han speaking some kind of Sinitic, not "minorities"), but I can't understand a word they're saying. After a week in Sichuan, I fly to Hong Kong, but only make sense of a few terms that I had learned from living there for lengthy periods of time, such as "bin1 dou6" 邊度" ("where"), which is non-cognate with "nǎ'er 哪儿" ("where").
In such circumstances, does the brain react in similar ways to those described at the beginning of this post when it is confronted with different grades of intelligibility among diverse dialects, topolects, and languages? I think so. For example, when I first arrive in Uppsala, Sweden, I understand next to nothing of the local speech. After half a year, I start to pick up cognate lexical items. Then I go to spend a few weeks in Helsinki, and I begin to notice Swedish borrowings in Finnish, e.g., katu, which means "street" and derives from Old Swedish gata < Old Norse gata < Proto-Germanic *gatwǭ. I suspect that English "gate" might be related, but don't have time now to pursue that rabbit down the alley leading to IE ghē-.
Now back to the MIT study on unaccustomed language input to the brain.
“The input has to be language-like enough to engage the system,” says Evelina Fedorenko, Associate Professor of Neuroscience at MIT and a member of MIT’s McGovern Institute for Brain Research. “And then within that space, if things are really easy to process, then you don’t have much of a response. But if things get difficult, or surprising, if there’s an unusual construction or an unusual set of words that you’re maybe not very familiar with, then the network has to work harder.”
Fedorenko is the senior author of the study, which appears today in Nature Human Behavior. MIT graduate student Greta Tuckute is the lead author of the paper.
…
“We found that the sentences that elicit the highest brain response have a weird grammatical thing and/or a weird meaning,” Fedorenko says. “There’s something slightly unusual about these sentences.”
Aside from the stimulation from partial intelligibility among dialects, topolects, and languages that I mentioned above, I think even more pertinent for brain stimulation is poetry. Do you remember Seven Types of Ambiguity? In my estimation, poetry is the epitome of brain food.
Selected readings
- "Ambiguous Mandarin sentences" (1/23/14)
- "Psycho linguists?" (6/26/04)
- "Language Log literally changes your brain" (8/25/16)
[Thanks to Gene Hill and Ted McClure]
Victor Mair said,
January 7, 2024 @ 8:10 am
I have to add that memes are also good brain food.
Philip Taylor said,
January 7, 2024 @ 8:15 am
"poetry is the epitome of brain food" — this idea intrigues me, but worries me at the same time. I have had a life-long love for poetry, which probably started when I first encountered William Collins’ "How sleep the brave" in one of my mother’s many poetry books at the age of five or six [1], but this love was almost destroyed during my teens when I was routinely asked in English examinations "What did the poet mean when he wrote … ?". For me, what the poet meant was about the least important thing in the world — all that mattered was "Did I enjoy the poem ?". If I enjoyed it, that was sufficient — if I missed some obscure allusion because I did not fully understand "what the poet meant" was neither here nor there. So let me ask what I hope is a simple question : if "poetry is the epitome of brain food", does that require the reader of the poetry to be forever asking himself "What did the poet mean when he wrote …" ?
[1] https://discoverpoetry.com/poems/william-collins/how-sleep-the-brave/
This poem in fact exemplifies my assertion that one does not need to know "what the poet meant" in order to enjoy poetry, as for much of my life I believed that the opening line was a question. It was only when I read it again, much later in life, that I realised that it was in fact an exclamation (I don't think I really understood punctuation marks when I first read the poem), but whether it was a question or an exclamation was irrelevant to the fact that the poem moved me to my very soul …
I was similarly moved (and still am) when I read the epitaph by Flight Lieutenant Anthony Richardson, RAFVR, above the memorial to Pilot Officer J. K. G. "Curly" Clifton near to my former home in Chainhurst. It reads :
Roscoe said,
January 7, 2024 @ 8:42 am
Philip Taylor – most literature scholars since 1946 would concur with your distaste for asking “what the poet meant.” (See Wimsatt and Beardsley’s “The Intentional Fallacy.”)
Coby said,
January 7, 2024 @ 12:12 pm
Note only is gata related to 'gate', in northern England (e.g. York and Lincoln) there are actually streets called 'gate' (Castlegate, Eastgate, Westgate).
Philip Taylor said,
January 7, 2024 @ 12:54 pm
There is at least one in London itself :
[Source: Wikipedia; no plagiarism involved].
Gene Hill said,
January 7, 2024 @ 1:09 pm
We have recently discovered that the brain can rebuild synapse destroyed by injury or disease. Especially the deeply memorized items. Just yesterday, I was talking to my Granddaughter about the importance of memorization when you are young. And the importance of truth in memory. She chose the violin to learn a couple of years ago. (So did Einstein) Music is a blend of emotional mathematics, which enables memorization most efficiently. It would seem that the continual memorization broadens the mind, enabling access to telepathy and other forms of enlightenment. Like an evolving amphibian when first finding vision above the water, a higher level of consciousness.
Philip Anderson said,
January 7, 2024 @ 1:22 pm
@Philip Taylor
But in York -gate means street, from the Old Norse, and the entrance gates are called ‘Bar’; however, in London, Bishopsgate referred to an entrance gate, and the ward and street also referred to that gate. Although the modern name has the same form as northern street names, ‘gate’ didn’t change its meaning when Bishopsgate Street was shortened to Bishopsgate.
Chester Draws said,
January 7, 2024 @ 3:21 pm
"Working harder" isn't the same as useful exercise.
If I eat something that is difficult to digest, I do not improve my digestion.
Annie Gottlieb said,
January 7, 2024 @ 3:52 pm
That borrowing of the word for "street" from Finnish that cycles back through old Germanic appears in modern German as "Gasse," lane. Many streets in Germany have names like "Ludwigsgasse."
David Marjanović said,
January 7, 2024 @ 4:28 pm
Nothing against memorization, but at this point I have to ask for evidence.
Same as underwater, just blurry until you can flatten your lenses enough.
Well, vaguely. While Gasse usually* refers to smaller streets of the sort that is called "lane" in England, it does not have the other meaning of "lane", as in "a three-lane highway".
* Not in Vienna, where it's the default for any street in the city and occurs in most of their names; but that's an exception.
Philip Anderson said,
January 7, 2024 @ 7:25 pm
I wouldn’t consider translating ’Gasse’ as ‘lane’ to be vague, although it may refer to a specifically British usage. The narrow road meaning is surely the primary one, whether it’s bounded by hedges or buildings, and we have been primed by the -gate examples to expect an urban context; the artificial division to separate traffic is secondary, and there’s no reason to expect other languages to use same word for both.
Do American towns and cities even have the kind of narrow streets that historic European ones do, and if so what are they called?
bks said,
January 8, 2024 @ 8:26 am
Chester Draws, fiber is impossible to digest but:
https://www.hsph.harvard.edu/nutritionsource/carbohydrates/fiber/
Amy de Buitléir said,
January 8, 2024 @ 9:45 am
This is purely anecdotal, of course, but anyone who wants to give the language portion of their brain a workout might want to learn a sign language. I'm about to finish up my first online course in Irish Sign Language, and it has been *exhausting*. I'm used to listening intently in a class, but there was nothing to listen to. Instead I had to *watch* intently, which felt like using new brain "muscles". Very disconcerting, but also very rewarding.
Bloix said,
January 8, 2024 @ 12:08 pm
Five years ago, in a discussion of an English-language test administered to high school students in Germany,
https://languagelog.ldc.upenn.edu/nll/?p=38125
I used the word "defamiliarization," a translation of a term introduced in 1917 by the Russian-Soviet writer and critic Viktor Shklovsky in his essay Art as Device, and also translated as "estrangement:"
"The purpose of art is to impart the sensation of things as they are perceived and not as they are known. The technique of art is to make objects ‘unfamiliar,' to make forms difficult, to increase the difficulty and length of perception… A work is created 'artistically' so that its perception is impeded and the greatest possible effect is produced through the slowness of the perception."
Most pre-20th century art and much later art doesn't work this way. Poets like Wordsworth and Longfellow and Millay didn't write to be read slowly. But much modernist and post-modernist poetry is written to be difficult in Shklovsky's sense, and even short poems can take time to work out. Hart Crane's The Wine Menagerie:
Invariably when wine redeems the sight,
Narrowing the mustard scansions of the eyes,
A leopard ranging always in the brow
Asserts a vision in the slumbering gaze.
Then glozening decanters that reflect the street
Wear me in crescents on their bellies. Slow
Applause flows into liquid cynosures:
— I am conscripted to their shadows’ glow.
– is a description of an alcoholic nodding off in a bar. He didn't intend for you to get it on the first read.
There's a great deal of older English poetry, though, that was not written to be comprehended slowly, but today requires time and hard work to understand it because the language has changed – not only in its vocabulary but also in its grammar, which has become simpler and more dependent on word order. Notably this has happened to Shakespeare, because he's the oldest author whose original words are still widely read or listened to by non-specialists. Sentences that he clearly wrote to be understood swiftly require a lay reader to invest time and study if they are to be understood at all. Presumably Shakespeare is now sufficiently difficult as to provoke the reactions in the brain discussed in the post.
Linda Seebach said,
January 8, 2024 @ 2:33 pm
I'm not sure why this is surprising, or why it is assumed to be desirable.
Isn't the goal of effective practice to progress from conscious thought to automaticity over as broad a range as possible?
Chester Draws said,
January 8, 2024 @ 3:56 pm
Isn't the goal of effective practice to progress from conscious thought to automaticity over as broad a range as possible?
Yes, which is why I tell my students that they shouldn't need to do too much thinking in any Maths exam. If they are thinking for the easy questions, they'll never get to the hard ones.
But the point is surely to have as much automatic as possible, in order to then extend further.
My beef is that you don't get better at either thinking or language by being fed uncommon sentences. You merely get tired. It isn't practice, it is wasted exercise.
Bloix said,
January 8, 2024 @ 9:49 pm
Linda Seebach-
Sometimes people use language for the pleasure of it. A good joke often requires a bit more than automaticity, doesn't it? And it doesn't even have to be funny. To take an example from Shakespeare's era, since I brought him up, here's an epigram by John Donne (1572-1631), whose whole reason for existing is that it has to be read twice to be understood:
A BURNT SHIP
Out of a fired ship, which by no way
But drowning could be rescued from the flame,
Some men leap'd forth, and ever as they came
Near the foes' ships, did by their shot decay ;
So all were lost, which in the ship were found,
They in the sea being burnt, they in the burnt ship
drowned.
Did you understand it with "automaticity"? Yet once you did understanding it, wasn't it striking?
Or try this one, in which Donne uses the same tactics of contradiction and paradox, but to a much more serious purpose – should he have written it so that it could be understood with "automaticity"?
BATTER MY HEART
Batter my heart, three-person'd God, for you
As yet but knock, breathe, shine, and seek to mend;
That I may rise and stand, o'erthrow me, and bend
Your force to break, blow, burn, and make me new.
I, like an usurp'd town to another due,
Labor to admit you, but oh, to no end;
Reason, your viceroy in me, me should defend,
But is captiv'd, and proves weak or untrue.
Yet dearly I love you, and would be lov'd fain,
But am betroth'd unto your enemy;
Divorce me, untie or break that knot again,
Take me to you, imprison me, for I,
Except you enthrall me, never shall be free,
Nor ever chaste, except you ravish me.
awelotta said,
January 9, 2024 @ 1:22 am
Prof Mair links to "Language Log literally changes your brain" lampooning the use of logically irrelevant facts, but makes no explicit mention of the unsurprisingness (as Linda mentioned) of the present study's result as described by the headline.
From reading the linked study's abstract, the researchers used a "GPT-based" language model and found that it was able to be used to predict the brain activity and also be used to generate confusing sentences. And then they apparently found some properties of the weird sentences that could explain why they were eliciting this response.
re: Chester Draws
Agreed. Though generally it's best to work at the limits of one's ability in order to improve. In the case of language, it seems to me that it should suffice to just read what one wants to read, to improve in the relevant areas.
Rodger C said,
January 9, 2024 @ 10:54 am
Bloix, I understood both of those the first time, and I'd never encountered the first one before. To be sure, I grew up hearing the 1611 Bible every Sunday from the age of six.
Rodger C said,
January 9, 2024 @ 3:16 pm
Further thought: "automaticity" and "having to be read twice" aren't the only alternatives. I read the first slowly, with delays for construing the meaning, but still only once. To me, and I think to the Modernists, having to do this is what makes Donne interesting.