http://www.guardian.co.uk/environment/2009/dec/22/copenhagen-climate-change-mark-lynas

http://languagelog.ldc.upenn.edu/nll/?p=1987

To follow up on our recent discussion of the effects of Alzheimer's disease on the writing of Iris Murdoch and Agatha Christie ("Literary Alzheimer's"; "Authorial Alzheimer's again"), I promised to post about the broader linguistic background, starting with a discussion of the normal effects of aging. With respect to lexical issues, there's a useful, if complicated, summary graph in a recent review paper coming out of the Language in the Aging Brain project (Mira Goral et al., "Change in lexical retrieval skills in adulthood", The Mental Lexicon 2: 215–240, 2007):

(Note that these are schematic plots from their statistical model, not the average values being modeled, much less the trajectory of any individual. The data comes from a "longitudinal data from 238 adults, ranging in age from 30 to 94, who were tested … over a period of 20 years".)

There are five lexical tests represented here, two picture-naming tests (the "Boston Naming Test" and the "Action Naming Test"), two list-generation tests ("FAS" and "Animals"), and the vocabulary subtest of the WAIS-R.

The results for the WAIS-R vocabulary test are plotted separately for two subgroups, those with a college or post-graduate degree ("VOC 16 yr", i.e. 16 or more years of education) and those with only a high-school education ("VOC 12 yr"). For (model fit for) the high-school group, this measure of vocabulary increases linearly throughout life. For the college group, the WAIS-R vocabulary measure doesn't change over the age-range tested.  (I presume that this is a sort of ceiling effect, and that a different vocabulary test would show increases with age for the college sub-group as well — though this is sheer speculation.)

In the animal-naming task, participants were asked to list as many animals as they could within one minute. This measure (in themodel fit) declined linearly thoughout the range of ages covered.

The "FAS" naming task requires subjects to list as many words as they can starting with the letter 'F' (excluding names and variant forms of words already listed), and then to do the same for 'A' and 'S'. They have one minute for each letter. On this task, (the model says that) subjects improved up to about age 55, and then declined.

The Boston Naming Test (BNT, Kaplan et al., 1983) comprises 60 line drawings of objects ranging from very common items, such as bed and tree, to less-common items, such as unicorn and protractor. Participants were presented with one picture at a time (e.g., a volcano) and were asked to say the label for it. If they could not name the picture within 20 seconds, they were given a semantic cue (e.g., a kind of mountain) and if they still could not name it they were given a phonemic cue (the first phonemes of the target word, e.g., /vә/). The dependent measure was percent correct before phonemic cues.

As you can see from the graph, the model showed an accelerating decline with age for scores on the BNT, and a difference between men and women (with men having higher scores overall).

The Action Naming Test (ANT, Obler & Albert, 1979) comprises 55 line drawings of actions ranging from very common activities, such as eating and reading, to less-common items such as proposing and knighting. As in the BNT, if needed, participants were given first a semantic cue and then (if needed) a phonemic cue. The dependent measure was percent correct before phonemic cues.

The model also showed an accelerating decline with age for the ANT, and also an advantage for males.

Some interactions reported in the paper but not shown in the graph: individuals with more education had higher scores on the FAS and animal-naming tasks; on the BNT, women with more education had higher scores, while there was no significant effect of education for men.

Again, it's important to recognize that these are interpreted parameters of a (rather complex) model fit to their data, not the data itself:

Initially, we analyzed available data for each of the five measures separately, to identify the “best” model for each measure. As fixed effects, we considered age, education, gender and their interactions; random effects consisted only of age (as linear and quadratic terms). The best model was that with the lowest value of Akaikes information criterion (AIC), a measure of overall fit. Subsequently, we combined the best model for each measure into a multivariate model, which estimated change in each outcome simultaneously and allowed us to examine correlations among levels and slopes for the measures. Such “correlated change” models have been used to investigate relations among changes in multiple measures of cognition over time. With a multivariate, correlated-change model, we can determine the extent to which change in one measure is associated with change in another.

The point is not that the model is inappropriate or wrong, or that the conclusions are questionable — the cited effects are confirmed in general terms by other studies — but that it's a summary of trends and tendencies in many measures on a large group of subjects, and not necessarily a fact about any particular individual.

The authors observe that different tasks show a range of different relationships to age, sex, and level of education, and specifically that "the tasks that required retrieval of unique lexical items (Boston Naming Test and Action Naming Test) yielded significant age-related decline that became more rapid in older age, distinguishing them from tasks that allowed for the retrieval of various lexical items", and also from tasks that required response to lexical items provided by the experimenter.

They suggest that some of these differences can be explained by "a cascaded progression of lemma and lexeme retrieval during word production". By lemma they mean (roughly) the concept associated with a word, while lexeme means the word itself. It may help to understand this distinction by supposing that when a word is "on the tip of your tongue", you've got the lemma but can't quite get the lexeme. (Warning: these terms are used somewhat differently in other areas of linguistics, e.g. by lexicographers or morphologists.)

The difficulty that some older adults experience with picture-naming tasks can be explained within the framework of the Transmission Deficit Hypothesis (TDH) developed by Burke and her colleagues based on D. MacKay’s Node Structure Theory (NST) (e.g., Burke et al., 1991). In the NST, the representational units, termed nodes, are interconnected. Lexical nodes are connected to semantic nodes in the semantic system and to phonological nodes in the phonological system. Word production involves activation of the relevant semantic nodes, which in turn prime the phonological nodes that are connected to the word. The strength of the connections between nodes will determine the amount of priming transmitted between them. According to the TDH, in older age the strength of these connections is reduced and so transmission of priming among nodes is reduced.

On the face of things, this theory also seems to predict age-related decline in vocabulary tests, since the same connections are involved in the opposite order. But perhaps learning new words overcomes any degradation in the connections for old words, or receptive connections are reinforced in ways that production connections are not, or something.

Anyhow, my main point here is a descriptive one: overall, as we get older, we're likely to have increasing difficulty with tasks that require retrieval of particular words from meanings.  It's comforting to observe that networked computer technology is increasingly able to assist our aging brains with such tasks.

http://languagelog.ldc.upenn.edu/nll/?p=1986

Here in the Edinburgh office of Language Log we are snowed in this morning. Thick, thick snow. (Though our language has only one word for it, we find that is quite enough.) There have been repeated falls overnight. This is unusual weather for Edinburgh. Part of the major London-to-Edinburgh highway, the A1, is being closed. Travel advisories of the don't-even-think-about-it type are being broadcast on the radio. And yet below the windows of our New Town apartment, cars and trucks and taxis belonging to those unable to understand broadcast warnings are sliding around and getting stuck on the snow-coated cobblestones of our street. People are digging spasmodically and hopelessly with rusty shovels they found in their basements to try and free these cars from their wintry doom. I saw one neighbour come out with an ice axe to try and free a truck that was unable to get up the hill. It was in vain. Linguists are helping too. We have teams out across the city doing comprehension tests: asking the drivers of stuck cars, "Which part of ‘unless absolutely necessary’ did you find hard to understand?"

http://feedproxy.google.com/~r/GoogleEarthBlog/~3/KHMQebg11Ik/happy_holidays_from_google_earth_bl_2.html

http://www.gearthblog.com/blog/archives/2009/12/happy_holidays_from_google_earth_bl_2.html

With Christmas just a few days away, we thought we'd share our annual holiday greeting card. You can download the KML file to view it yourself, or check it out via the Google Earth Plug-in.

Santa's NORAD tracker will be going live tomorrow, so be sure to check back for that.

Happy holidays to all!

http://feedproxy.google.com/~r/GoogleMapsMania/~3/CW6EraM1H10/google-maps-using-foursquare-api.html

http://feedproxy.google.com/~r/typepad/sethsmainblog/~3/Q2Mz0_dfjDY/first-organize-1000.html

Kevin Kelly really changed our thinking with his post about 1,000 true fans.

But what if you're not an artist or a musician? Is there a business case for this?

I think the ability to find and organize 1,000 people is a breakthrough opportunity. One thousand people coordinating their actions is enough to change your world (and make a living.)

1,000 people each spending $1,000 on a special interest cruise equals a million dollars.

1,000 people willing to spend $250 to attend a day-long seminar gives you the leverage to invite just about anyone you can imagine to fly in and speak.

1,000 people voting as a bloc can change local politics forever.

1,000 people willing to try a new restaurant you find for them gives you the ability to make an entrepreneur successful and change the landscape of your town.

Even better, coordinating the learning and connections of this tribe of 1,000 is not just profitable, it's rewarding. If you can take them where they want to go, you become indispensable (and respected).

What's difficult? What's difficult is changing your attitude. Instead of speed dating your way to interruption, instead of yelling at strangers all day trying to make a living, coordinating a tribe of 1,000 requires patience, consistency and a focus on long-term relationships and life time value. You don't find customers for your products. You find products for your customers.

http://news.bbc.co.uk/1/hi/scotland/edinburgh_and_east/8426940.stm

• Tram works to close Shandwick Place for the next 18 months.
  (tags: edinburgh trams)
• San Jose police test head-mounted cameras for officers
  Do David Brin and Charlie Stross get royalties?
  (tags: TransparentSociety police technology TheFuture)
• What fame is like for David Tennant. Very nicely put together long article/interview.
  "It just feels scary⦠all the time"
  (tags: doctorwho interview davidtennant)

http://www.irregularwebcomic.net/2523.html

http://www.mercurynews.com/ci_14030412

http://feedproxy.google.com/~r/BeijingSounds/~3/A0dr8ymockU/

http://www.bjshengr.com/bjs/?p=1119

The YU cafeteria is one of the best audio links to Old Beijing. Over the clink of dishes and the munching of radish tucked inside your bǐng (饼 = pan bread) you can listen idly to the deals that the city used to offer.

[Transcript linked to audio available on this page -- click on the asterisks at the left of each line to start the audio at that point.]

[See post to listen to audio]

Keep in mind, though, that the lessons don’t stop at the cafeteria door. First, there’s the reminder of what it means (phonologically and lexically) to be a native Beijinger (from the BJS Longtime Beijinger Checklist):

In everyday speech, a Beijinger…

Old hat, right? But what if we threw in…

Beijing-R 201

Beijing-R 201 is more properly known as Érhuàyīn 201. And actually you’ve been enrolled for years without being entirely aware of it. The subtitle is “Words that are hard to look up in dictionaries cuz the rhoticization muddies everything.”

The good news for today’s Zhonglish speakers is that, thanks to romanization, dictionaries make it pretty easy to look up a word you heard. (Try that with your beloved Hanzi!) Just learn some Pinyin, listen to a word, and off you go.

Still, there are tricks of the trade, and érhuàyīn (儿化音, i.e. rhoticization / “adding an R flavor” / the Beijing-R) plays some of the dirtiest tricks in the phonemic book: a rhoticized word, especially one where you only half-caught the tone or leading consonant, can be damn hard to look up in a standard dictionary.

In this snippet you get bāi yī bǎir, in line 4. That’s what the transcription team came up with in line 4: 掰一掰儿. Sure, it seems okay, but what if you had that bǎir in isolation, with the tone questionable and the final half of the syllable a muddle — would you be able to look it up?

It made me think of the following list, which is by no means complete and may not even offer the best examples; it’s just some flotsam plucked out from Beijing-R 201. Have more? Send your words this way and we’ll add them to the pile. After all, in Beijing a good rhotic syllable is like a good radish peel used to be: cheap AND delicious.

http://metamagician3000.blogspot.com/2009/12/helen-razer-on-internet-censorship.html

http://www.guardian.co.uk/tv-and-radio/2009/dec/20/david-tennant-best-doctor-who

http://ars.userfriendly.org/cartoons/?id=20091223

http://links.userfriendly.org/lotd?http://www.wired.com/autopia/2009/12/suzuki-motorcycle-assembles-itself/

Originally published at tansyrr.com. You can comment here or there.

Wow, so this domestic thing. People do this every day? I’ve been spending the last 3 days predominantly on housework, and I’m definitely over it. Still, the house is clean and tidy for Christmas, we’re way better organised than we were last year, and apart from the tree randomly falling over and the second batch of shortbread being sniffed at suspiciously by my daughter, there have been no catastrophes.

(ah, the gingerbread house fail of… too many years to count, ahh the Christmas I was 8 months pregnant and decided to make my own puddings but in my exuberance failed to read the recipe and ended up going to bed while my honey stayed up until 1am watching a boiling pot… ahh)

The guest room is really pretty, though, and I’m leaving in under an hour to pick up girliejonesanything</em> (apart from maybe paper towels, grumble mumble) which means for once I can have a relaxing Christmas Eve. Oh, and my obsessive podcast-habit means that from now on, baking shortbread and gingerbread daleks will always remind me of the Radio Free Skaro boys.

I’ve done ridiculously well with gifts received already this year. The latest is the most beautiful leather-cover edition of Room with a View, a book I love greatly, and have been meaning to replace my battered orange paperback with a shiny new orange paper back for some time.

I can’t tell you how gorgeous this book is – not only lovely in its own right, but wrapped in Penguin-stamped tissue paper and placed in an elegant cardboard book. It’s a lovely object I can’t stop looking at and, oh yes, I’ll be re-reading this book very soon. How can I not?

It occurs to me that this was possibly the last beloved classic novel that I didn’t have a really nice edition of – a friend once gifted me with a gorgeous set of ancient hardback copies of The Forsyte Saga that he found in an aunt’s attic, I have a lovely old green edition of Wuthering Heights (and matching Jane Eyre, bah, but I keep it cos it matches) that dates from my mother’s brief ‘old books’ hobby phase, and I bought myself a proper hardback of Pride & Prejudice from a uni bookshop. Now Room with a View.

Yep, those are the important ones. Which books do you love so much that, even if you already have a copy, a lovely hardback edition would be an awesome present for you?

• Tuesday, 0017: http://earwormhole.dreamwidth.org/14908.html Still earwormed.
• Tuesday, 0105: Madame O.G. - Flourish - Phantom of the Opera [Archive of Our Own] http://archiveofourown.org/works/34985
• Tuesday, 0913: http://www.theglobeandmail.com/blogs/bureau-blog/morning-buzz/article1408463/ Canada rocked by scandal: NDP leader Jack Layton seen smooching w/ fellow MP Olivia Chow! (See article.)
• Tuesday, 0917: @maureenjohnson Perhaps it is an office chair!
• Tuesday, 0929: longsufferingly: AoGG fic: Still He Offers the Sea Shell (Gilbert/Anne) http://longsufferingly.livejournal.com/9368.html
( read the other 17 )