🧡 Hey friends, if you want to keep the rabbit-hole rolling, these are some recommended 'watch next' videos, mentioned in this one! 👇

1. The Major Flaw in Autism Research (and How to Fix It)
https://www.youtube.com/watch?v=m4hhw9x7w8Y

2. Why Autistic Adults Deserve Better Sex Education (the last video - it's really, really, long and goes into a lot of detail about 'affective sex education');
https://www.youtube.com/watch?v=PzWWWfIxPkk

3. What Exactly is "Profound Autism"?
https://www.youtube.com/watch?v=-YR1G3lWaTA

▶ Binge the full playlist, “Debunking Bad Science 🧪,” here:
https://www.youtube.com/playlist?list=PLU8o7dwjyw6wLhxYuW8db3yTlknWKsC1m

🐈 And of course, if you'd like to become an official autisti-cat, please take a look at my Patreon. The tiers start at just $4 USD / month, and that includes exclusive videos, and posts, and access to The Scratching Post Discord Community - with nearly 300 lovely neuroaffirming members. You'll also help me keep the kettle on and the cats fed!
https://www.patreon.com/autisticaf

Take care!
- Mike 🧡

That woman infuriates and disgusts me.  Not only the fact that she considered such a thing, but in my opinion, so much worse:  That she said this aloud in media that will remain available where her kids may come across it later.  How will her autistic daughter feel when she sees herself in the video and hears that her mother considered killing her?  How will her NT child feel when they learn that they were used as the excuse for this not happening?  They'll carry that burden forever, and it could affect their relationships all their lives.  As someone with CPTSD, I can verify that a parent saying things like this will have horrific affects on those dear kids.

someone who can casually talk about thinking about ending their child’s life right in front of their kids is someone who scares me

I’m glad you take the time to find the baby in this bathwater. I often do not have the patience to find the enlightenment rationalism buried in the hyperbolic witchcraft.

I appreciate you pointing out how the SPARK recruitment pamphlet targets parents of autistic kids specifically. These organizations are treated like they're infallible, people put them up on a pedestal because they're supposedly 'helping autistic kids', but it's a major red flag and extremely telling that they never want to include autistic adults in the conversation. And it's not just that we're not being considered, seems more like they purposefully exclude us.

As someone who works in the AI field, the term "machine-led research" is pretty dubious, outside of pharmaceuticals, protein folding, genetics, radiology, image analysis and sound analysis.  This research was surely just a set of clinical observations, and putting AI into that kind of data analysis is a red flag for poor quality research.

I will admit I am only a few minutes into your video, but I am in Grad School and I also already have a masters in psychology and in political science. I am now getting one in Social Work and training to be a clinical therapist. I wrote a major project on the three levels of work (Micro, Mezzo, Macro) about how to best work with Autistic folks (as I am Autistic) in regards to Autism and Violence... There was a new treatment center opening up where I live and a couple friends of mine interviewed there and asked me why Autistic kids are so violent but all of the adults they know are fine because during the interview the entire focus was on the football pads they would wear, the wall padding, the poles and kick pads that they would have for working with the kids (under 14 I believe). Now this wasn't specifically violent kids, this was just ABA for all Autistic kids - So I spent 6 months deep diving into the research and REALLY LONG STORY short - Research MUST include Autistic researchers, decisions that say that something is a "best practice" needs to have Autistic people involved, and they need to ask autistic people their experiences - none of that has ever done. Second, there is a very high correlation between ABA and suicide rates especially for women with ASD (higher than veterans by over 100 times), Autistic individuals are LESS violent than allistic individuals but are accused of it for not making eye contact or verbally responding, people with ASD are much less likely to commit crimes or behave and clinically anti-social manners... Oh and they experience violence at a much higher rates than Allistic people and much higher than they commit. There is even a great deal of evidence where someone has been murdered and the killer's sentence was reduced because the victim had Autism, especially with infanticide.

I did write an extensive list of ways to counteract these issues on every level but have not developed it into a guide I could publish or send places because I don't know how but boy do I want to!!!

University of California Irvine has an autism program for children and research but the university disability services knows next to nothing about autism and has denied services through their incompetence. The university shows next to no interest in supporting autistic students in their graduate programs. While a professor and few students have started a neurodiversity support group in recent years, they do not have the ear of anyone in positions of power and have been able to little to support the success of the graduate students involved, other than to give the students a place to offer each other support. The students have experienced bullying, ostracization, manipulation, gaslighting, and threats from PIs and peers, with little to no support from the university. So, while “research” is being done about curing autism at the university, the autistic university students are completely abandoned.

Even with the earlier study, all FMRI studies need to be taken with a grain of salt. The few reliability/repeatability studies that have been done show that the same person doing the same activity on different days will often have varied regions of brain activity. Since FMRI are expensive, studies don't usually repeat test the people invlolved, so it's very hard to say how diffentiable individuals actually are from each other.

I haven't read these papers, but we have to be very careful trusting how ML is being used to 'discover' arbitrary correlations in any population. It's entirely likely the ML is effectively p-hacking and if you aren't very careful and have very good transparency, it's very easy to have an improper correction applied to your statistical significance, and in so doing have no clue if any of it is actually meaningful or just a random coincidence with this particular sample.

These are solvable difficulties, but I'm coming in with some skepticism about the science (above and beyond the ethical issues)

Since when does an MBA qualify anyone to talk about anything medical. Except maybe the selling of medical supplies 🙄

I'm all in for a more component based diagnosis. If you only have monotropism, you'll struggle switching tasks, you'll have trouble in society by being too attached to topics, or looking cold and indifferent during hyperfocusing on your interest. That alone is worthy of understanding both by people who have and who don't, so we don't misidentify that as narcissism or disinterest.

I've had PPD, so I understand where thoughts like the bridge thing that can come from... But I absolutely draw a line at her saying it on record with her kid playing in the background. Wtf.

“Fuzzy clusters”

I like that. 🤣

Ugh. When will these people FINALLY figure out that they can’t meaningfully refine categorization of a condition they don’t even understand?

Anyone who would put the onus on their other child as justification for not redruming her ND kid needs serious help, as does probably the NT kid. Hearing that was HORRIFYING. What if that NT kid gets unalived in an accident or by disease? Is that game on again for redmur/self-unaliving then? And do we think that the NT kid will NEVER feel that pressure to hold things together for crazy mommy? And to put that out on the internet or however publicly available for either child to see ad nauseum forever - is heinous in and of itself.

OMFG. I left off Autism Speaks a while back because of the divisiveness. After talking with my two therapists, tomorrow I'm talking to my primary care about being tested for autism and ADHD. I am 66 years old. In the last 20 years I cannot tell you how many people have asked me if I was Aspergers, or high functioning autism because of my tendency for stupid shit falling out my mouth a lot, and other behaviors. I am currently diagnosed with PTSD from child abuse. I'm not sure there is much of a difference. I did not show early developmental delays. I walked early, I learned to read early, I was described as very smart but not putting forth effort to succeed. I was always in trouble, and never knew why. I was always blamed for things I did not do. And definitely physically, emotionally and sexually abused. So I'm not sure how anyone would be able to tell the difference. Did my potential ASD and/or ADHD cause my abuse and resultant PTSD, or is my PTSD being used as a research product for ASD/ADHD? And I do not wish to go through another round of "let's try this drug" again. I refuse. I did ten years of that with no success and definite detriment. Mike, you are such a gem, I feel supported and comforted by your work here.

Thanks for an informative video, particularly for highlighting the biased samples. I find the amount of energy and resources directed into studies like this immensely frustrating. There are so many reasons why having a quicker, easier and less subjective diagnostic process would be beneficial (one thought that occurred to me recently: what happens to autistic elders? Do they have specific needs? Are those being met? We've barely started looking at middle aged adults and women, you can't tell me there aren't people in older age groups we've been missing/ignoring). Yet I wouldn't trust most research groups as far as I could throw them and this research is a great example of why.

I'm the autistic mum of an autistic 3 year old (I suspect we're both also ADHD - but that's a different story) and the journey of his diagnosis has been even more eye opening than mine. Even with doctors that are clearly trying to do better, the diagnosis is still so reliant on externally observable behaviors, research into diagnostic criteria that reflect internal autistic experience or even just a broader range of autistic experience would be so much more helpful. We might not be able to ask young kids about their internal experience, but that doesn't mean they don't have one.

ASD presents so differently in different situations (e.g. with or without ADHD, or by gender), I would readily believe there are different subtypes. Maybe they are partially socially determined (e.g. social conditioning, local/group social norms), and partially innate (e.g. hormones, co-occurring ADHD), I don't know. My son and I are so different it's almost like we have two different conditions (beyond the obvious age difference), but we don't. But the kind of research you've discussed here doesn't really get us any closer to understanding those differences.

Apologies for the slightly disjointed rant. I have a PhD in a public health/statistics related area and poor quality research just really irritates me, particularly when it's not even trying to serve the needs of the community. There are valid reasons to work with the data you have access too (exploratory analysis for example), but it's still garbage in = garbage out. (As an aside, the analysis method sounds like a new version of principal components analysis - that seems to underlie these dimensionality reduction methods).

Thanks for you insights and research.  I'll have to rethink my involvement with SPARK.  In two different questionnaires from themI was very disappointed that in their multiple-choice questions, over and over again they only had negative attributes of autism, how autism causes problems, etc.  I wrote to them and suggested that those questionnaires could only give a biased view of autism and I'd like at least a place for comments in the questions to express my misgivings about them.  I also gave them couple stories about the positive effects of autism in my family.  I can say that they responded quickly, thanked me for my point of view, acknowledged my feedback and would consider how to do better.  In that manner they were more accessible and receptive than most other organizations whose questionnaires I participate in.

If I remember correctly, they only asked the parents about their children's behaviour. While most parents know their kids well, I'm skeptical about that choice when it comes to bias. Especially with charities that focus on deficits and refuse to listen to the perspectives of autistic adults, instead centering parents of autistic children. Someone in the comments said that the questionnaire was very deficit oriented. It's interesting to see research like that, despite the paradigm shift in pedagogy and psychology towards a resource oriented approach. From my professional experience, working on new skills based on individual strengths is much more successful than constantly focusing on deficits.

2023 and 2025 study sounds like really good initial research. I proposed to one of the professors I was doing research on autism that there are most likely about 3-5 distinct types of autism based on my experience of autistic people, being around other autistic people, and my general knowledge about autism. However, like I tell everyone, it is really not useful to dissect autism into subtypes because it harms health insurance coverage and there is not really a good way to truly divide these groups perfectly.

These studies are a good example of the use of LLMs being used for their purposes (to answer a very rigid and restrictive research question) which is great. LLMs have been in research since I started my higher ed back in 2016. It is nice to see these models getting better. Now, like you said these studies are very limited.

First off, something you did not mention is that all of this research is correlational meaning you cannot draw a direct casual link between an autistic person's genetics and their behavior. Frankly, it could be that certain autistic behaviors influence the activation of certain genes that are linked to certain neuropathways in the brain. You just cannot simply know.

Second, this study is using a snapshot of various ages of autistic people. Because this study is focusing on trying to find subtypes, the various ages can actually be a detriment because the brain goes through multiple changes during adolescents to adulthood, all of which can be affected in many different ways. For example, pruning. A brain goes under pruning at least three significant times in your life. First happens around age 2, second happens during puberty, and finally a third time during 20-30 when specialization finally finishes. During pruning the brain gets rid of neuropathways that are no longer relevant to allow for new pathways to develop. However autistic people can express under and over pruning which has at least been associated with sensory problems in the first stage of pruning. This means that a brain that is between ages 4-10 could have a subtype that might experience more neuropathways associated with genetics that relate to sensory issues than brains 10-18 due to the second stage of pruning. This also means that said brains can have changes in their neuropathways during puberty which means genetic association factors might not be relevant. This is only possible brain change that occurs throughout your lifetime. Puberty does a bunch of neurological changes too. Needless to say, it might have been more beneficial to have longitudinal data to rule out changes of the brain in being the reason for differences in subtypes.

Third, these studies are very limited in their sample and do not really have good representation. You will need a lot more data from a variety of autistic people to get a study that better represents actual subtypes in autistic people. For example, the lack of women can pose its own problems because we know women tend to have different autistic traits express than men due societal gender standards. There is also the problem of incorrectly diagnosing autistic women with other mental disorders like BP, BPD, depression, etc. than an autism diagnosis. This data could substantially change subtypes found and this is only one other group. Racial and cultural groups would also have an impact.

Finally, as you mentioned the behavioral data is not really robust. It is parental reported behavioral data which can be beneficial in some studies but for a study trying to link neurological, genetic, and behavioral data together it is not. You need train clinicians using the same tools to get better data that is muddy with individual differences, good interrater reliability, and less self-report data. This does require more financial capital invested than parental report data which is most likely why this study decided to use parental reported data. A better study should look into actual clinical observations.

All in all, this sounds like a good initial study. I hope this does not devolve into trying to split autism in different diagnosis because the last time we tried this led to a bunch of autistic people not getting the care they need due to diagnostic biases and health insurance coverage problems. At the same point, it could bring a new understanding of why autism looks so different between people. It could be used to better make treatments and therapy targeting an autistic person's specialized weaknesses while also using their strengths to better understand them. However, it could also be easily used for eugenics. The line between helping people and eugenics is a fine line and based on the past and present, this research is more likely to be used for eugenics purposes than to actually help autistic people which sucks. Autism does not need a cure or be prevented. Neurological diversity like all other forms of diversity strengthens our population. Eugenics threatens to make humans weaker by limiting our genetic diversity. Sure, there are very very few disorders that genetic testing is great but that is due to the very very very few genetic edits that cause these mutations and few beneficial side effects said mutation brings. However, when you start to look at more complicated traits like autistic traits the way said genes interact with other functions becomes extremely complicated. Most genes associated with autism are found in neurotypical people at the same rate in autistic people too. Furthermore, these genes due multiple functions and the selective  selection of genes associated with autism would have devastating effects on our population that we could not know. This is why eugenics is bad without getting into the ethical concerns. It just makes humans worse than better because of less genetic diversity. Tangent aside, I hope there is going to be more research in subtypes but it needs to focus on improving autistic lives and not eugenics.

I am no scientist...but I do have a serious issue with what seem to be blind spots in diagnosing autism. And it is directly related to the human tendency to bully those who are "different." Children who are somehow different are generally targeted by other humans...even adults, sadly enough, "He's different; don't play with him" and so on. So these maligned children grow up being chronically abused from the get-go. And "normal" people who have grown up being chronically abused have similar symptoms to each other, enough so that there's even a term for this general grouping of symptoms: C-PTSD. No, it's not in the DMV--yet. All that said, how many autistic children ALSO have C-PTSD (which is NOT addressed/treated because of their Autism diagnosis), and could be more successful if they received therapy to address their PTSD? How many behaviors attributed to autism are actually trauma-response behaviors? This is muddled, but I hope  you understand what I mean? I have no answers, but I'd sure like to see some work done to help those who are or did grow up "different" and mistreated.