Even the core idea, 'rejection sensitivity' is acknowledged not to be a diagnostic entity: so then how can we make definitive comparisons? Maybe clinical gestalt... but the author is not trained to see and evalaute patients.
> Why It Happens: The Mechanics Underneath the Reaction
And then don't talk about Mechanics, but instead no citations just guess. For anyone that actually wants something to go learn more about checkout the frequency of NCAH and the resulting blunted cortisol response as one thing that is influencing this.
https://www.nature.com/articles/s42003-024-06102-y#MOESM3 Plasmalogen, B5, 12-HETE, Androsterone sulfate, etc are all related to the functioning of the HPA-Axis.
https://pubmed.ncbi.nlm.nih.gov/19598235/ <- CYP11B1 specifically here
Anecdotally: I worked at a software company that had free snacks in the office kitchen and would always chuckle how the salt snacks would run out first. The connection being NCAH can result in less aldosterone which helps regulate salt in the body.
Vibe writing may be an improvement. I doubt frontier models would go along with whole language reading instruction without a tremendous amount of pushback.
For those not in the know, a generation of children were taught to read sans phonics. They were told to do things like guess at words based on pictures or context clues. Some lessons had teachers covering words up and asking kids to use guessing strategies at what they might be.
Large swathes of education research are extremely questionable due to things like limited or no ability to be reproduced, circular references, or just straight up fraud.
The more that something that falls under the moniker of “education research” is built on the foundation of a more traditional (and typically more rigorous) academic discipline, the less likely that it is bogus research (e.g., Howard Gardner’s core research on multiple intelligences, which was science based on aphasic patients), versus the derived works on MI, many/most of which are bunk that he does not endorse).
> Some lessons had teachers covering words up and asking kids to use guessing strategies at what they might be.
This is just a cloze exercise. These can have a legitimate role in a learning environment when used appropriately.
> Give me two paragraphs on the benefits of whole language reading instruction. Don't refer to other reading instruction methods or compare/contrast, just focus on the positives.
And got, with no pushback,
> Whole language reading instruction emphasizes making reading meaningful, engaging, and connected to students’ real experiences. By surrounding children with rich, authentic texts—such as stories, poems, informational books, and student-created writing—it encourages them to see reading as a purposeful form of communication rather than simply an academic task. Students are invited to use context, prior knowledge, illustrations, language patterns, and their understanding of meaning to engage with texts. This can help foster curiosity, comprehension, confidence, and a positive identity as a reader. Because students often have opportunities to choose books and respond personally to what they read, whole language instruction can also support motivation and a lasting enjoyment of reading.
> Whole language instruction also integrates reading with writing, speaking, and listening, helping children experience literacy as an interconnected set of communication skills. Classroom activities may include shared reading, read-alouds, journal writing, storytelling, discussion, and collaborative projects, allowing students to develop literacy in varied and socially meaningful contexts. Teachers can build instruction around students’ interests and developmental needs, creating opportunities for creativity, self-expression, and deeper thinking about texts. This emphasis on authentic language use can strengthen vocabulary, comprehension, oral language, and awareness of how written language functions while helping students develop independence and confidence in using literacy throughout their daily lives.
Time to update your priors on how sycophantic frontier models are.
And CrazyStat's implication strikes me as still largely true, so I'll make it explicit: even frontier LLMs will still resort to sycophancy when there's sufficient contradictory information in their weights to reproduce only some of it in response to a user query. And the harnesses that LLM providers silently, invisibly integrate behind-the-scenes to counteract hallucination and syncophancy are definitionally patch-jobs at best. The latest models may be better when explicitly instructed or configured by default to search the web and pull sources into their context, but the root problem remains unsolved and can still rear its flattering head if you're not paying close attention.
[1]: https://journals.sagepub.com/doi/10.1177/1529100618772271
All this talk of contradictory information, sycophancy, and flattery is little more than anthropomorphic projection in cases like this.
One prompt will elicit a response based on one set of weights that are most closely related to the prompt. A different prompt will elicit a response from a different set of weights, for the same reason.
There’s no contradiction from the point of view of the model and its responses, and no need to invoke sycophancy or flattery as a reason for the behavior. You’d get the same behavior from a model that hadn’t been RLHFd to be sycophantic.
Sycophancy and flattery are words that relate to how humans experience LLMs. It's appropriate to use words that could anthropomorphize as descriptors for a problem when that problem exists in the space of human experience. It doesn't matter that an LLM is merely a predictive model. It matters that humans who use the model perceive it as a conversation partner and walk away feeling 'flattered' by the model's predictive responses to the humans' input, which the model predicts without regard for factual accuracy, which happens in turn because (as you point out) "there’s no contradiction from the point of view of the model."
We could call the problem "LLM predictive output which tends to have the effect of flattering the user's opinionated assumptions without regard for factual accuracy" if you prefer that much of a mouthful, but "sycophancy" describes the problem well and is one word instead of twenty.
> One prompt will elicit a response based on one set of weights that are most closely related to the prompt. A different prompt will elicit a response from a different set of weights, for the same reason.
Yes or, in other words: there's sufficient contradictory data which were used to train the model's weights that it will predict (a series of tokens which when read in order in English convey) accurate information for some prompts and (a series of tokens which when read in order in English convey) inaccurate information for others. That the contradiction is beyond the LLM's scope is part of the problem. The two identical parentheticals above are (or ought to be) as unneeded as pointing out that a statistical model cannot technically be sycophantic.
It wouldn't terribly surprise me if we walk into a similar trap with math. Like when I took a look at some of the common core math, I saw strategies that I use for arithmetic, where like sometimes I might do four-digit multiplications starting from the most significant digit with a sequence of approximations/corrections to arrive at the correct answer. It wouldn't terribly surprise me if this sort of thing just doesn't work for the median person, but from the little I've looked at it, it seems like the effects of CC math are mostly neutral anyway.
And yes, something similar is happening with common core math. Curricula are trying to teach the tricks, shortcuts, and mental re-arranging of numbers that are only possible to comprehend once the basics are memorized and internalized. It was miserable memorizing times tables, but I remember being able to start to see the ways you could combine and recombine numbers after I had successfully memorized them. Trying to teach those things without any memorization is putting the cart before the horse.
IIRC it was a teacher in New Zealand[1] who had teacherly charisma, was earnestly committed to her method, and was so profoundly wrong that she became the first domino of a regression in reading ability which we are still feeling today across the English-speaking world. And teachers and education professionals became fans of both the method and the growing personas behind it. There were Americans who turned whole language hokum into a product for sale... using the cult-like sales tactics of Amway and other MLMs but applied to a target audience of largely woman-attended teachers' development conferences and fairs instead of Mormon housewives.
Sadly, humans were capable of producing plausible sounding nonsense long before the advent of LLMs. And promoting it persuasively.
> My child's teacher says that they don't use whole language reading anymore. I learned that way and it worked great. Instead they're spending years on phonics when my kid can already read. They're crazy, right?
> No—they’re not crazy. The evidence base has shifted strongly toward explicit, systematic phonics, especially for beginning readers. Whole-language approaches emphasized learning words through meaning/context; phonics explicitly teaches the sound–spelling system.
> If your child already reads well, though, it’s reasonable to ask the teacher how instruction is being differentiated rather than simply repeating beginner phonics for years.
You redefine syncophancy too narrowly. Many (if not an outright majority of) everyday LLM users have neither the educational foundation nor skill to ask good, well-formed questions like the one you propose. We should consider syncophany in the scope of an average person's non-optimal use of LLMs. CrazyStat's prompt is something your average Joe will absolutely feed an LLM, and Joe will walk away with something very much like those plausible falsehoods. We have to think of and work against LLM syncophancy in the scope of how the hypothetical average person will use an LLM.
> Vibe writing may be an improvement. I doubt frontier models would go along with whole language reading instruction without a tremendous amount of pushback.
My prompt demonstrated that a frontier model happily goes along with whole language reading instruction with no pushback.
The poster is SEOing their substack and product (modulo). All their posts are either related to their product, their product's substack, or talks related to their product.
HN has been an SEO and content marketing target for years now, and existing HN moderation is increasingly failing to catch these attempts.
Heck, a bot account for The Register reached 66K karma and was undetected until a couple of us complained.
I'm "on the spectrum," which, of course, means that my K-12 days were a living hell. It was bad. I spent most of my early years, acting as a punching bag for any bully or proto-bully that was feeling the need to let off some steam.
Needless to say, I tended to be a bit "sensitive," during that period.
I also learned, early on, that I was absolutely terrible at determining motivation in others. I have learned to ask, explicitly, before assigning motive, so the sensitivity has extended far past my bully-magnet days.
These days, I'm fairly old, and have developed highly effective compensation/coping mechanisms, but it took a long time to get here.
All that, to say that, in my case, I feel as if the "sensitivity" was real, but as a secondary symptom of my "normal" autism lite.
Alexythemia is like a that, I can have anxiety without realizing it, all I see is that my hands are shaking, blood pressure rises but in the head I'm like whatever, I don't notice.
However alexythemia also means no psychological reaction to physical emotion sensations.
I would interpret this like some subconscious processing of social cues does occur in this case and it does trigger inner changes,but why it occurs does not make it into consciousness, so the alexythemia hides the reason for the body sensation but the sensation is strong enough to be noticeable and the lack of understanding for the strong reaction can cause people to flip out.
I am not rejection sensitive at all btw, too dense to give a shit but its interesting.
Isn’t this a good reason to stop using the term “gifted”? Isn’t it now widely acknowledged that effort (among other things like creativity and self-conception) actually is part of intelligence?
“Gifted” children are historically identified and classified via IQ tests (typically with the threshold IQ >= 130 - though the definition of what ‘gifted’ should mean is being debated), and using “gifted” for smart kids emerged and was popularized in the 1920’s when IQ tests were being developed by people who basically believed in eugenics, in the heritability of intelligence, and they wanted to identify the kids with the smart genes. [1] Using language like ‘profoundly gifted’ and ‘genius’ for high IQ has always been (mis)leading since IQ is a ranking, not an absolute measure of intelligence.
I just searched and found an interesting recent paper arguing that ‘gifted’ is problematic and attached to science that is no longer valid, and should be improved or replaced. [2]
[1] https://en.wikipedia.org/wiki/Intellectual_giftedness#Overvi...
That sounds wrong? Like I found engineering courses at university easy, and never put in any effort or studied, but still got As on exams where other people who I know worked very hard got Cs. Maybe working hard is important for outcomes, but clearly it's a different phenomenon from intelligence.
The Wikipedia article you linked also mentions that high IQ people literally have measurably different brain structures.
Intelligence is also known to be highly heritable (twin studies), and we do have a fairly large set of identified genes now that contribute, though generally none has a very strong effect, and apparently in many cases, there's multiple genes that need to interact, so it's not like you can just identify "the smart gene."
That paper you link doesn't seem to actually say anything. Like they're saying that we should have more specific terminology because different "gifted" kids need different things, but it would seem to me that what you are saying is exactly the cause of that issue. Previously we said that if you have a certain IQ, you are gifted. If we expand the scope of that to be gifted in a bunch of different ways, then it becomes a non-specific descriptor that entails different needs and therefore becomes unhelpful for figuring out support services.
Like, you can also be amazing at music or art, you can be a hard worker, etc. That just means something different from having high intelligence. But high intelligence does exist. I would expect that if somebody's a hard worker, moving more quickly might overwhelm them as they eventually run out of hours in the day and can't keep up with the work. A high intelligence person, by contrast, can quickly grasp topics, so it's not really very much work for them to move at a high pace, and in fact it might cause them to become more engaged.
FWIW, I wasn’t arguing against the idea of intelligence being heritable or having a biological component. Intelligence is heritable, but also there are large environmental and nurture components as well. Using the word “highly” might be technically accurate but somewhat misleading, the heritability of IQ for example averages around 50% across many studies (https://en.wikipedia.org/wiki/Heritability_of_IQ).
I was just pointing out that the way we measure intelligence does not measure intelligence itself directly, it only ranks people. We don’t know how much smarter someone who’s IQ 150 is than someone who’s IQ 100, we only know how many people are smarter than either score. Calling someone with a high IQ profoundly gifted is not actually supported by the ranking metric we’re using.
High intelligence does exist, I completely agree. What “high” actually means is debatable. But why is being amazing at art or music different from having high intelligence? Not sure I agree, and I think that statement is suspicious and deserves some justification. I think a lot of people would consider Bach and Mozart intelligent.
The actual reference to different brain structures in the Wikipedia article also mentions the positive correlation between socioeconomic status of children and their measured intelligence. That’s enough to question what’s actually being measured, and whether a correlation between cortical thickness and IQ really does represent a correlation between cortical thickness and intelligence, or something else.
It makes perfect sense to me that socio-economics would correlate. You might suppose that kids of lower SES could have a ceiling effect from e.g. worse nutrition, more pollution, or lower access to brain enriching activities (the Genie case is an extreme example of how early neglect can be long-term catastrophic). And of course, g correlates with economic success, and is heritable, so you expect that high g kids will correlate with high g parents which correlates with high SES.
The development of g rests on that assumption, but the measured correlations vary, and the correlations are lower for music & painting than between, e.g., STEM subjects.
Also interesting - the cross-subject correlations g measures actually decrease as intelligence increases. The smarter someone is (as measured by psychometric tests), the more likely they are to be “smart” in a single subject and not all subjects. This is an interesting self-contradiction in the whole theory.
> It makes perfect sense to me that socio-economics would correlate.
Yeah, I totally agree. But, it’s very important to know that the people like Spearman and others who developed IQ and g and other concepts of intelligence in the 1920s were specifically looking for the innate biological component of intelligence that explains “gifted” ability in the absence of nurture and environmental factors. That’s where this use of “gifted” came from - it’s pre-supposing that intelligence was not heavily influenced by the wealth of your parents, and they didn’t know how important nutrition and pollution were at the time. This is why I think using “gifted” is problematic. From that paper I linked to: “Reframing ‘gifts’ as ‘talents’ underscores both the domain-specific nature of these abilities and that their extent should be viewed as dynamic rather than static or preordained.”
One question that occurs to me - who are these gifts bestowed by? There are two obvious answers: nature and God. I think nature is historically what is meant by ‘gifted’ in the context of education, but I’ll speculate there’s at least a little bit of historical religious influence on the choice of that word. Given that we know that SES, nutrition, environment, education, parenting and many other factors have real measurable influences on both intelligence and outcomes for people, the nature & God explanations are somewhat lacking, right?
I’ve never read something arguing that intelligence doesn’t exist (and I certainly wouldn’t buy such an argument). I just don’t think intelligence is one dimensional, thus rankings can be misleading or sometimes damaging when used to determine the educational curriculum of a child.
My issues with intelligence tests are that they only rank, they do not measure absolute intelligence, and they assume (against the evidence) that intelligence is 1-dimensionsonal and rankable. The ranking part means that terminology like “gifted” is making a huge assumption and priming people to think about intelligence incorrectly. IQ of 130 (“gifted”) is, by definition, 2 standard deviations above the mean. An adult male that’s 2 standard deviations above the norm in height is 6 foot 2. A whopping 5 inches taller than the mean of 5’9”. The 6’2” man is a little taller than the average guy, not a lot taller — even though they are taller than 95% of men. There’s no reason to believe that the distribution of intelligence, especially the hereditary/biological component of intelligence, is any different.
Irrelevant; I don't know what this means; it would be falsified by showing it has no correlations to things.
> g is particularly bad at explaining musical talent
I don't know why you'd expect it to correlate to music talent any more than it would correlate to visual artistic talent or bowling talent. I don't think those are things that have traditionally been associated with intelligence. The things that g correlates with are the things that people have associated with intelligence for thousands of years; we all have an intuitive sense of what it is generally referring to.
> My issues...
I mean yeah, any multivariate test reduced to one number loses information; the SAT score is meaningful for many purposes even if it may be produced by heavily skewed subsection scores. The other thing is that you're picking an arbitrarily small distribution, height. The 95%ile in shoe size is even fewer inches larger than its mean than height is! But also the 95%ile in chess elo will win 43/44 games against the mean elo player. There is very good reason to believe that the 130IQ child is meaningfully different from the 100IQ child: that is the entire basis of gifted educational programs that have shown through research for many decades that the former child has very different needs than the latter. Also if you've ever spent any time around kids/students, especially tutoring them, it's pretty incredibly obvious that the gaps in ability are vast. It's really not a matter of inches here, it's a matter of literal years of schooling being wasted by putting gifted children through standard public school tracks.
It doesn't matter that it's a relative ranking. It matters that the gifted kid has different needs than the non gifted kids who have different needs than the remedial/special ed kids. Grouping into 3 is an arbitrary thing, but the difference exists. We can't pinpoint on a spectrum when a hue changes its label, but we can all agree that between either extreme end there is A and there is B.
I am concerned about this because my own education was extremely painful due to this sort of debate as a 'gifted' kid. If my schools had had advanced tracks for all subjects, not just a few, I'm pretty sure I would not have despised school so much or had as many teenage mental health issues. It is so soul crushing being trapped in a classroom every day with essentially zero learning or interest happening, being forced to performatively "pay attention" or "take notes" like a drone, etc. I get riled up when I hear opinions that may lead to support for abolishing gifted programs.
The term "gifted" is useful for helping and classifying children. Maybe you are saying to replace it with more granular words, like "gifted in math, standard in english" or something? That's fine and generally what happens in practice, but it's just going to be a recursive debate: "How can we really say you're gifted in math? Math contains a lot of subjects..." We have enough problems already trying to get modern school districts to understand differences in ability exist at all.
Don't waste your time.
That means confidently stating who this applies to, what it feels like, and what to do about it is not a great idea because it could teach someone the wrong things.
If I could trade away just the rejection sensitivity, I would. If I had to trade both the rejection sensitivity along with the correlated traits, I’m not so sure that I would in fact. This sort of decision is obviously very individual however. The tradeoffs may not be worth it for many.
I think with the rejection and criticism it's not so much about the actual criticism but the idea of what the other person now thinks/operates on, while they might gloss over it, you might assume the worst case scenario and think that now they're against you and you have to win them over again
I'd say in general the concepts of "it's not a big deal" "it's probably nothing major" etc. don't seem to register
In other, more blunt terms, is this a "hardware" or a "software" issue?
However the following articles' descriptions of drug treatments using alpha-agonists, such as clonidine and guanfacine, which provide a response rate of 55% (significant relief for 55% of people) does imply to me that there is some physiological basis and may be of interest to people reading this thread: https://neurodivergentinsights.com/how-to-deal-with-rejectio... https://neurodivergentinsights.com/history-of-rejection-sens... https://www.goldstarrehab.com/parent-resources/is-rejection-...
And I think there are stronger findings that autism/ADHD is at least partially genetic.
P.S. I would caution you against holding a premise of a hard "or", that mental conditions are either hardware or software, or the hidden premise that you can't patch hardware with software.
Is this even true? Is there a measurable physical difference in the medium that accounts for this, or is this just an abstract process that could be but isn't necessarily related to biological processes? A different way to think, a more mature intrapersonal relationship...
Second of all you can sniff it out with plenty of the phrasing which is what my original comment was based on: "The correction temporarily erases the bigger picture." "Feedback on an assignment is rarely just about the assignment. It becomes a referendum on whether the child is still who they believe themselves to be." "What follows is rarely proportional: “no one likes me,” a wave of rage, or a full shutdown." "This is protective armor. Underneath it is usually plain embarrassment." "Because the identity is more invested, the shame lands harder." "That gap is where a great deal of the combustion happens." "The underlying fear in these moments is usually not “I will get in trouble.” It is closer to “I will be misunderstood, reduced to one moment, or unfairly categorized.” That distinction changes how an adult should respond."
Third of all the entire structure is a chatgpt output. Humans don't write their documents like this.
I would have compared it to some of her previous writing from six or more years ago but couldn't find any. I'm pretty certain if you found some it would not read anything like this article.
“In 2014, after teaching in schools for over 25 years and working with hundreds of gifted children and their parents, Maya Sissoko founded Whole Child Education. She earned her undergraduate and graduate degrees in education and worked in schools on the East Coast before serving for 15 years as a lead teacher and teacher trainer at The Nueva School in the Bay Area. Her original curriculum design on ancient Egypt was highlighted by Pulitzer Prize-winning author Thomas Friedman of The New York Times in his book, “That Used to Be Us,” in the chapter “Average Is Over.” Her perspective on giftedness was also featured in The San Francisco Standard on the increasing use of IQ testing for preschool admissions and the limited public support for gifted children in California.
She is also the parent of a Nueva and Harvard Law School graduate, who worked as a federal prosecutor and now practices law in Washington, D.C.
Whole Child Education (WCE) is continually evolving to respond to the ever-changing educational needs of children and parents. WCE has always promoted learning outside the traditional framework and offers expertise to parents curious about child-centered, individualized educational options. Maya is available for private teaching, parent coaching, educational advising, and homeschool design.”