Donald Trump: Cognitive decline or TDS?

« previous post | next post »

Sharon Begley, "Trump wasn’t always so linguistically challenged. What could explain the change?", STAT 5/23/2017:

STAT reviewed decades of Trump’s on-air interviews and compared them to Q&A sessions since his inauguration. The differences are striking and unmistakable.  

Research has shown that changes in speaking style can result from cognitive decline. STAT therefore asked experts in neurolinguistics and cognitive assessment, as well as psychologists and psychiatrists, to compare Trump’s speech from decades ago to that in 2017; they all agreed there had been a deterioration, and some said it could reflect changes in the health of Trump’s brain.

STAT may have reviewed decades of Donald Trump's on-air interviews, but what's presented in the article is a scant handful of anecdotes. There's one example of a verbal flub from an (unidentified) interview in May of 2017; 41 seconds of a Larry King interview from 1987; 13 seconds from another unidentified NBC News interview "earlier this month"; a hundred words of transcript from an unidentified "interview with the Associated Press last month"; and one or two other fragments. Begley asserts that

[L]inguistic decline is also obvious in two interviews with David Letterman, in 1988 and 2013, presumably with much the same kind of audience. In the first, Trump threw around words such as “aesthetically” and “precarious,” and used long, complex sentences. In the second, he used simpler speech patterns, few polysyllabic words, and noticeably more fillers such as “uh” and “I mean.”

Again, she doesn't give us clear citations for the interviews, much less links; and again, the description is entirely anecdotal.

I was particularly surprised by her assertion of  "noticeably more fillers such as 'uh' and 'I mean.'" That assertion might be true of those two interviews — though it would be nice to have some numbers — but as I've observed several times, one of the striking characteristics of Donald Trump's (recent) spontaneous political rhetoric is that he uses filled pauses much less frequently than most other politicians (and most other people in general).

For example in "The narrow end of the funnel" (8/18/2016), I noted that filled pauses were 8.2% of Steve Bannon's words (in a sample passage from a panel discussion on The Future of Conservatism), and 4.0% of Hilary Clinton's words in a Vox interview, while three of Trump's rally speeches had between 0% and 0.05% filled pauses, and in a CNBC interview, Trump used 74 filled pauses in 5329 words, for a rate of 1.4%.

[I'll see if I can track down those two Letterman interviews and check them out.]

So Begley and her mostly-unnamed "experts in neurolinguistics and cognitive assessment, as well as psychologists and psychiatrists" might be right to wave their hands at "a neurodegenerative disease or the normal cognitive decline that comes with aging". But the evidence that they offer is anecdotal at best, without even citations or links to let readers check out the context of the anecdotes. (And perhaps because of heavy reader load, my attempts to view the article's chosen video-clip anecdotes froze four browsers on two operating systems, making them even less impressive to me than they otherwise would have been.)

This strikes me as a reverse-image version of the right-wing fixation about neurological interpretations of Hilary Clinton stumbling. Pending some real evidence, I'm going to diagnose this as a case of TDS: Trump Derangement Syndrome.

Update — the STAT story has been reprinted at MedPage Today.



  1. Philip Taylor said,

    May 23, 2017 @ 4:09 pm

    Re "the chosen video-clip anecdotes […] froze four browsers on two operating systems", I am told : "A script on this page may be busy, or it may have stopped responding. You can stop the script now, open the script in the debugger, or let the script continue.


    Windows 7 Enterprise, Seamonkey 2.46 (64-bit).

    [(myl) Yes, I got the same message. I tried all three options — none of them helped.]

  2. Jonathan Lundell said,

    May 23, 2017 @ 4:13 pm

    Related commentary that includes links to eight of the Letterman interviews.

  3. Tom S. Fox said,

    May 24, 2017 @ 3:27 am

    I must say, this is by far the most balanced article I have ever read on what has essentially devolved into a Trump–bashing blog, but I must point out that Hillary Clinton didn’t merely stumble, she was chucked into a van like a side of beef:

    Also, there is much more evidence than that for her declining health:

  4. Bill Benzon said,

    May 24, 2017 @ 5:40 am

    FWIW, I had the same problem in trying to view the video clips.

  5. random lurker said,

    May 24, 2017 @ 5:49 am

    A long time ago there was a video showing a comparison between George W. Bush in 1994 and 2004. You can see it here:

    Granted, he was probably very well prepared in the earlier clip and it's been claimed that he later purposefully took on an act of an affable slow-ish fellow, and the person who put the clips together probably had his or her own agenda, but the effect of aging on neurolinguistics cannot be denied altogether.

    [(myl) The effects of age-correlated neurodegeneration on speech, language, and communication are more than amply documented. But there remain very large individual and contextual differences at every age. The National Institute on Aging cites a study estimating that "One in seven Americans age 71 and older has some type of dementia". But that means that 6 out of 7 don't — and the proportions for people on the younger end of the range are much smaller. According to "2014 Alzheimer’s Disease Facts and Figures" from the Alzheimer's Association, the prevalence of "Alzheimer's Disease and other dementias" in Donald Trump's cohort is more like 2.9%:

    Berisha et al. (2015) showed that there were some signs of Ronald Reagan's developing Alzheimer's Disease in his news conferences:

    Changes in some lexical features of language have been associated with the onset and progression of Alzheimer's disease. Here we describe a method to extract key features from discourse transcripts, which we evaluated on non-scripted news conferences from President Ronald Reagan, who was diagnosed with Alzheimer's disease in 1994, and President George Herbert Walker Bush, who has no known diagnosis of Alzheimer's disease. Key word counts previously associated with cognitive decline in Alzheimer's disease were extracted and regression analyses were conducted. President Reagan showed a significant reduction in the number of unique words over time and a significant increase in conversational fillers and non-specific nouns over time. There was no significant trend in these features for President Bush.

    The authors looked at detailed counts of various categories of words in complete transcripts 46 of Reagan's press conferences, compared with 101 press conferences from G.H.W. Bush. No one has done anything comparable for even one of Donald Trump's press conferences or interviews — the cited article gives us nothing but brief selected clips and quotes from a handful of events.

    So the general idea is a reasonable one, but in this case the implementation is unscientific, unserious, and irresponsible.]

  6. Bev Rowe said,

    May 24, 2017 @ 5:55 am

    This seems to me rather dangerous stuff. It will just make his supporters more suspicious of the liberal opposition and so more entrenched in their own views.
    I saw a clip of an an interview with an unnamed American woman who blamed the anti-Trump campaign of the "the left-wing medias". As a left-wing Brit I'm bemused by the concept of powerful left-wing medias in the US! (.Reminds me, as well, of an application I saw for a computing job where the writer boasted of his ability in handling datas,)
    Trump seems to be able to brush off any attack with his position among his supporters unimpaired. Enhanced, in fact. I'm afraid that the alleged Russian connexion is not going to inflict a serious enough wound.

  7. Haamu said,

    May 24, 2017 @ 6:29 am

    @Tom S. Fox, lots of things have devolved lately.

  8. J.W. Brewer said,

    May 24, 2017 @ 8:48 am

    I am fascinated to note that three out of the four things "indicative of dementia" according to Fraser et al as summarized in the Neurocritic blog post linked above sound exactly like things a fully cognitively-competent person might deliberately do in order to communicate more effectively to a comparatively unsophisticated audience of oh let's say swing voters in a swing state.

    "Semantic impairment – using overly simple words"
    "Acoustic impairment – e.g., speaking more slowly"
    "Syntactic impairment – using less complex grammar"

    I expect my own idiolect is different in all sorts of ways from Trump's earlier idiolect as exhibited in the vintage talking-to-Letterman clips, but I certainly suspect that I would not be a particularly effective giver of political stump speeches to an audience of median American registered voters, not least on account of how, left to my own devices, I almost certainly (as judged for optimal rapport with that sort of audience): a) use too many complicated/obscure words; b) talk too fast; and c) use unduly complex or convoluted syntax. If for some improbable reason I wanted to learn to communicate effectively to that sort of audience in that sort of context, I would need a lot of coaching to help me develop all three of those "impairments."

  9. James Kabala said,

    May 24, 2017 @ 9:18 am

    To me the most notable (apparent, not based on hard data) characteristic of Trump's speech is that he seems to jump quickly from topic to topic, as if (although it may be an act) he is saying whatever comes into his head. So it doesn't surprise me that he uses fewer space fillers.

    I see some of this style in the old clips as well – the sudden pivot to "the farmers, who are really going though hell right now" in the King clip, the emphatic "and they were real bad days" in the Rose clip. So these clips don't prove their point to me at all.

  10. Carol said,

    May 24, 2017 @ 11:45 am

    Trump called the Manchester bombers "evil losers." But he calls everybody he even mildly dislikes a loser of some variety. Hard to see this as an appeal to swing voters. I imagine that even his fans felt deflated.

    For an example of cognitive impairment, I suggest the video of Trump introducing Rudy Giuliani, who Trump is then unable to see sitting directly across the table from him. It was sort of a wide table, I guess.

  11. The Neurocritic said,

    May 24, 2017 @ 12:05 pm

    Begley gave the names of five different experts, which hardly qualifies as "mostly-unnamed":

    – Ben Michaelis, a psychologist in New York City, performed cognitive assessments at the behest of the New York Supreme Court and criminal courts and taught the technique at a hospital and university.

    – neuropsychologist Sterling Johnson, of the University of Wisconsin, who studies brain function in Alzheimer’s disease.

    – John Montgomery, a psychologist in New York City and adjunct professor at New York University (who is not a Trump supporter)

    – Dr. Robert Pyles, a psychiatrist in suburban Boston (who is a Trump supporter)

    – Northwestern University psychology professor Dan McAdams, a critic of Trump

    However, I do believe a quantitative analysis of a large corpus is essential before drawing any tentative conclusions about change over time. Beyond the Letterman interviews (of which there are 20), I linked to The Trump Archive – over 900 televised speeches, interviews, debates, and other news broadcasts related to President Donald Trump – in my post. It includes hundreds of appearances on Fox, for instance. The earliest entry is December 2009.

    [(myl) But Begley suggests consultations with a much larger number of experts: "STAT therefore asked experts in neurolinguistics and cognitive assessment, as well as psychologists and psychiatrists". Plural experts in each of four categories would be a minimum of 8 experts; but the cited phrase made expect feedback from a dozen or more. So I think that "mostly-unnamed" is fair.

    As you point out, lots of relevant recordings are easily available. And your excellent 2/18/2017 post points to a number of simple quantitative measures that have been shown to be relevant — someone who wanted to make Begley's case in a responsible way could have presented data (not anecdotes) from the Letterman interviews.]

  12. The Neurocritic said,

    May 24, 2017 @ 12:28 pm

    My last comment went into the ether. Briefly, Begley named five experts, which hardly qualifies as "unnamed". I do, however, believe that a quantitative analysis of large corpus is essential before drawing any tentative conclusions about possible change over time. Beyond the Letterman videos, I linked to the Trump Archive in my post. It now has 1,142 clips, beginning with a Fox appearance in Dec 2009.

    J.W. Brewer – anecdote or not, I'm unclear on the intended audience during this campaign stop in South Carolina:

    "Look, having nuclear—my uncle was a great professor and scientist and engineer, Dr. John Trump at MIT; good genes, very good genes, OK, very smart, the Wharton School of Finance, very good, very smart—you know, if you're a conservative Republican, if I were a liberal, if, like, OK, if I ran as a liberal Democrat, they would say I'm one of the smartest people anywhere in the world—it's true!—but when you're a conservative Republican they try—oh, do they do a number—that's why I always start off: Went to Wharton, was a good student, went there, went there, did this, built a fortune—you know I have to give my, like, credentials all the time, because we're a little disadvantaged—but you look at the nuclear deal, the thing that really bothers me—it would have been so easy, and it's not as important as these lives are (nuclear is powerful; my uncle explained that to me many, many years ago, the power and that was 35 years ago; he would explain the power of what's going to happen and he was right—who would have thought?), but when you look at what's going on with the four prisoners—now it used to be three, now it's four—but when it was three and even now, I would have said it's all in the messenger, fellas, and it is fellas because, you know, they don't, they haven't figured that the women are smarter right now than the men, so you know, it's gonna take them about another 150 years—but the Persians are great negotiators, the Iranians are great negotiators, so, and they, they just killed, they just killed us."

    [(myl) Geoff Pullum and I discussed that quotation two years ago — Geoff posted on "Trump's aphasia" and I responded with a post on "Trump's eloquence" (8/5/2015).

    From my response:

    Geoff Pullum uses terms like “aphasia”, and phrases like “I don’t think there’s any structure in there”, in describing a quoted passage from Donald Trump’s 7/21/2015 speech in Sun City SC. But in my opinion, he’s been misled by a notorious problem: the apparent incoherence of much transcribed extemporized speech, even when the same material is completely comprehensible and even eloquent in audio or audio-visual form.

    This apparent incoherence has two main causes: false starts and parentheticals. Both are effectively signaled in speaking — by prosody along with gesture, posture, and gaze — and therefore largely factored out by listeners. But in textual form, the cues are gone, and we lose the thread.

    There’s another issue in this case as well. The segment in question takes place about 35 minutes into Trump’s speech, and the earlier parts of the speech have featured repeated assertions and implications that recent American leaders are bad negotiators, and have therefore made bad deals with other countries, including Mexico, China, and Saudi Arabia. This is relevant to his candidacy because “The Art of the Deal” is part of his persona — and it’s relevant to the the quoted segment, which is discussing another example, the nuclear deal with Iran. Because of this background, he can afford to criticize the nuclear deal in an allusive way without confusing his audience. […]

    His parenthetical asides in this segment feature another of his frequent themes, namely the allegedly liberal media’s lack of respect for his intellectual credentials.

    I don’t have time this morning to label this passage’s false starts and parentheticals according to (for example) the SimpleMDE guidelines, but this would be a straightforward exercise. Overall, I think the passage is entirely comprehensible, and in the context of the speech as a whole, even eloquent. The false starts and parentheticals may actually make the speech better, at least for people who are open to liking Trump and endorsing his ideas, by giving an impression of enthusiasm and genuineness.

    It would be a mistake to underestimate his considerable effectiveness as a public speaker, even if he doesn’t speak in conventionally coherent textual paragraphs.

    As I've pointed out in other posts, you'll get exactly the same false impression of incoherence from many kinds of spontaneous speech — the monologues of stand-up comics are an excellent example. See e.g. "The rhetorical style of spontaneous speech", 8/16/2016, where I give an extended example from Louis CK that's at least as "incoherent" as anything Donald Trump ever produced. My conclusion:

    I guess it’s fair for reporters to complain that Donald Trump uses the style of a stand-up comic rather than the style of schoolteacher. But let’s not pretend that the result is incoherent or unintelligible — it’s just a skillful instance of human speech communication in its natural state.


  13. cameron said,

    May 24, 2017 @ 12:56 pm

    The current Trump persona, what he does on the campaign trail and in office, is a character he originally developed in a professional wrestling context. He started working with the WWE in the late aughts, and he's been maintaining kayfabe ever since.

    To compare his performances now with his performances in the past, you have to take into account that he's playing different characters in different contexts.

  14. peterv said,

    May 24, 2017 @ 4:46 pm


    "To compare his performances now with his performances in the past, you have to take into account that he’s playing different characters in different contexts."

    What is the name of the play Donald Trump is in now? How does it end?

  15. Jim said,

    May 24, 2017 @ 5:00 pm

    @cameron echoes my thoughts — if a speaker perceives his audience as "simpler", he adjusts his speech terms to match, so it doesn't surprise me to find that Trump catering to a wide and shallow audience would "dumb down" his terms.

  16. The Neurocritic said,

    May 25, 2017 @ 2:19 am

    @myl – Thanks for the pointer to the exchange between you and Geoff, I missed it at the time. Let's just say my personal opinion is closer to the aphasia view…

  17. Ivan said,

    May 25, 2017 @ 5:59 am

  18. David Marjanović said,

    May 25, 2017 @ 6:51 am

    A long time ago there was a video showing a comparison between George W. Bush in 1994 and 2004. […] the effect of aging on neurolinguistics cannot be denied altogether.

    Huh. That's the first time I see this blamed on aging. The only explanation I had encountered is that he spent a lot of time drunk.

  19. William Berry said,

    May 25, 2017 @ 1:08 pm

    It has been widely rumored (yeah, I know) that Trump is ADHD and has taken Adderall to treat the disorder. If he has done so for years, in possibly increasing doses as he developed an addiction, this would be equivalent to being a meth addict, and could certainly explain some degree of deterioration of cognitive faculties.

    It would be highly irresponsible not to speculate.

  20. Graeme said,

    May 28, 2017 @ 1:15 am

    Could it also be that at 45 he was still relatively close to a more intellectually curious youth – and doing fewer interviews, each one keen to impress an NY audience – than at 70 years? An age when boredom with years of routine money making and extra decades of self aggrandisement have made him what he is today, morally and intellectually? (Even leaving aside the deliberate shift in register of his calculated political persona, and the 'we become what we speak' possibility.)

RSS feed for comments on this post