Language Log

https://youtu.be/XIhuNFv_P8U

There are dozens of articles Out There on "Zoom fatigue", with a wide range of ideas about causes and cures.

Gianpiero Petriglieri offered the BBC a couple of hypotheses about why "Zoom calls drain your energy":

Being on a video call requires more focus than a face-to-face chat, says Petriglieri. Video chats mean we need to work harder to process non-verbal cues like facial expressions, the tone and pitch of the voice, and body language; paying more attention to these consumes a lot of energy. “Our minds are together when our bodies feel we're not. That dissonance, which causes people to have conflicting feelings, is exhausting. You cannot relax into the conversation naturally,” he says.

Silence is another challenge, he adds. “Silence creates a natural rhythm in a real-life conversation. However, when it happens in a video call, you became anxious about the technology.” It also makes people uncomfortable. One 2014 study by German academics showed that delays on phone or conferencing systems shaped our views of people negatively: even delays of 1.2 seconds made people perceive the responder as less friendly or focused.

At the end of the May 1 episode of the NPR show "Milk Street", host Christopher Kimball interviews Dr. Aaron Carroll about a recent California court decision that could force coffee to come with a label warning that it contains a chemical known to cause cancer.

The chemical in question is acrylamide, and it's apparently created (in small quantities) whenever carbohydrates are heated above about 250 degrees farenheit — so bread, crackers, cake, cookies, pizza, pretzels, fried potatoes, corn chips, and lots of other things besides coffee that most people eat regularly. Dr. Carroll argues that the quantities of acrylamide involved are far too small to pose any measurable danger, and that warnings like this one have the bad effect of persuading people to ignore all such messages.

But this is Language Log, not Cancer Warning Over-Reach Log, so what's the linguistic point? It's the way that Dr. Carroll pronounces the name of the chemical in question.

On Twitter, @muffkin7 asks readers to "Ruin a film by inserting the word 'moist' into its title".

Ruin a film by inserting the word ‘moist’ into its title.

— muffkin (@muffkin7) April 27, 2020

Answers include "Gone moist with the wind", "All moist about Eve", "The good, the bad, the moist, and the ugly", "Little shop of moist horrors", and "Close encounters of the third moist kind".

A few months ago, I posted about a talk I gave at an Alzheimer's Association workshop on "Digital Biomarkers".

Overall I told a hopeful story, about the prospects for a future in which a few minutes of interaction each month, with an app on a smartphone or tablet, will give effective longitudinal tracking of neurocognitive health. […]

Speech-based tasks have been part of standard neuropsychological test batteries for many decades, because speaking engages many psychological and neurological systems, offering many (sometimes subtle) clues about what might be going wrong.

But I emphasized the fact that we're not there yet, and that some serious research and development problems stand in the way. […]

Some colleagues and I are starting a large-scale project to get speech data of this general kind: picture descriptions, "fluency" tests (e.g. "how many words starting with F can you think of in 60 seconds?"), and so on. The idea is to support research on analysis of such recordings, automated and otherwise, and to allow psychometric norming of both traditional and innovative measures, for both one-time and longitudinal administration, across a diverse population of subjects. We've got IRB approval to publish the recordings, the transcripts, and basic speaker metadata (age, gender, language background, years of education).

We've been testing the (browser-based) app across a variety of devices and users. When it's ready for prime time, this is one of many channels that we'll use to recruit participants — we're hoping for a few tens of thousands of volunteers.

We're finally ready to open this app to wider use, and you can contribute a few socially-distant minutes of your time by going to https://speechbiomarkers.org. And please tell your friends!

Listen to this sound, and describe it in the comments below:

You can learn what the sound is, and why I care how you hear it, after the fold.

Megan McArdle ("Four things Democrats need to understand about beating Trump", WaPo 1/31/2020) has something important to say about the style of Donald Trump's extemporized speeches:

Trump is a good public speaker. “Nails on a chalkboard” doesn’t quite capture how educated urbanites feel about Trump’s speaking style. A closer analogy would be having your teeth drilled — without Novocain.

His fragmented sentences, simplistic formulae (see those insults above) and rambling style would drive them wild even if the content and partisan ID were more to their taste. They like “polished” candidates who speak in complete sentences that read well when written down.

Trump, by contrast, sounds like … well, actually, he sounds a lot closer to how most people talk than a “good” public speaker. He speaks in short sentences and uses a small vocabulary. He makes up names for stuff to aid listener memory. He repeats himself. He digresses at random.

Trump talks, in short, the way people talk when they aren’t expecting their words to be written down. This informal approach horrifies those of us who love reading enough to do it on weekends. But one way to think about this is that it is not so much the difference between good and bad; it is the difference between an oral culture and a written one.

If you listened to the U.S. Senate proceedings yesterday, you may have been puzzled — as I was — by Jay Sekulow's discussion of "lawyer lawsuits":

And by the way,
lawyer lawsuits?
lawyer lawsuits?
We're talking about the impeachment of a president of the United States,
duly elected.
And the members,
the managers,
are complaining about
lawyer lawsuits?
The consitution allows
lawyer lawsuits.
It's disrespecting the constitution of the United States
to even say that in this chamber —
lawyer lawsuits!

Michael Ramscar, "The empirical structure of word frequency distributions", arXiv 1/9/2020:

The frequencies at which individual words occur across languages follow power law distributions, a pattern of findings known as Zipf's law. A vast literature argues over whether this serves to optimize the efficiency of human communication, however this claim is necessarily post hoc, and it has been suggested that Zipf's law may in fact describe mixtures of other distributions. From this perspective, recent findings that Sinosphere first (family) names are geometrically distributed are notable, because this is actually consistent with information theoretic predictions regarding optimal coding. First names form natural communicative distributions in most languages, and I show that when analyzed in relation to the communities in which they are used, first name distributions across a diverse set of languages are both geometric and, historically, remarkably similar, with power law distributions only emerging when empirical distributions are aggregated. I then show this pattern of findings replicates in communicative distributions of English nouns and verbs. These results indicate that if lexical distributions support efficient communication, they do so because their functional structures directly satisfy the constraints described by information theory, and not because of Zipf's law. Understanding the function of these information structures is likely to be key to explaining humankind's remarkable communicative capacities.

Following up on recent posts suggesting that speech-to-text is not yet a solved problem ("Shelties On Alki Story Forest", "The right boot of the warner of the baron", "AI is brittle"), here's a YouTube link to a lecture given in July of 2018 by Michael Picheny, "Speech Recognition: What's Left?" The whole thing is worth following, but I particularly draw your attention to the section starting around 50:06, where he reviews the state of human and machine performance with respect to "noise, speaking style, accent, domain robustness, and language learning capabilities", with the goal to "make the case that we have a long way to go in [automatic] speech recognition".

In the past few days, I've encountered some nice examples of the communicative interpretation of what I've suggested we ought to call "interpolations" rather than "disfluencies".

The latest episode of the new podcast Subtitle is about "Words we love to hate". Full disclosure: Kavita Pillay interviewed me for the program, and so you can hear my voice from time to time.

More later — I'm off to Washington DC for a workshop on "Digital Cognitive and Functional Biomarkers" organized by the Alzheimer's Association.

Meanwhile, you can find links to some Language Log posts on word aversion in "Word aversion science", 6/24/2015, and posts about word rage in "Annals of word rage", 5/2/2009.

Garrett Wollman writes:

Not sure if this really belongs in LL's misnegation files, but I found this sentence hard enough to parse (despite knowing exactly what the author meant) that I stumbled over it on a re-read:

"The really troubling thing," Zora says to the rain, "is that I can't convince myself I'm not in a life where knowing someone who can do that isn't purely a good thing."

Graydon Saunders, A SUCCESSION OF BAD DAYS

The context here is that one of the other characters makes a rather creepy magical barrier around the people in the scene while waiting for medical attention after a disease outbreak.  So what the character is (I believe intended to be) saying is that they think it's entirely good to know someone who can do that, but they are troubled by the thought.