This is a special post for quick takes by flaritza. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.
"Just getting started learning" sounds unconvincing. Can you actually write a non-trivial computer proof? There are some math publications on the MIRI web page, can you imagine writing a computer proof for any of them? (If yes, contact the author first, to make sure someone else hasn't already done that.)
You can try your skills e.g. here. In my opinion (but I am not an expert), you should be able to complete at least half of it, before you have a chance to do something useful.
People on this forum gain reputation by writing texts that get upvoted, because readers consider the texts valuable.
Sorry if I'm too harsh, but it seems to me that you are trying to do things the wrong way: get reputation first, do awesome things later. It's the other way round: do something awesome, then maybe advertise it a bit, people will notice and maybe remember, and then they will give more weight to your later opinions.
I know of people who make high quality/novel posts only to have just one upvote or a comment, and no traction
You didn't provide specific examples, so it is hard to comment on this. Maybe it was bad luck. Maybe others did not judge the quality as high as you did. (If it was just a question of bad luck, you might have given the article second chance by mentioning it.)
before you have a chance to do something useful
That statement seems far too strong, at least if you aren’t just talking about a very narrow subset of AI safety research (part of MIRI’s agenda). at a glance, that website gauges a skillset associated with one flavor of proof-based mathematics. For proof-based AI safety work, i think that the more important and general skill is: can you make meaningful formal conjectures and then prove them?
I admit I am confused about what exactly "proof based math" means. I assumed that in general all math is proof based, so this specifically refers to computer proofs. If not, then of course my advice does not apply.
"Sorry if I'm too harsh, but it seems to me that you are trying to do things the wrong way: get reputation first, do awesome things later. It's the other way round: do something awesome, then maybe advertise it a bit, people will notice and maybe remember, and then they will give more weight to your later opinions."
I can see how it's confusing that I may have suggested I wanted to do this wrong order for myself--I was asking the second question for someone else, actually, who has put in a lot of time/effort to build up lots of ideas/share them on this forum but has gotten very little traction.
I feel pretty unhappy that many of the posts on LW and the likes that are geared towards seemingly sounding "universal" in their prescriptions for how to arrive at accurate beliefs, how to change your beliefs, how to recognize your biases, double-cruxing, so on, all assume this hard to define baseline emotional stability/technical intelligence that just isn't spelled out anywhere and if it isn't achieved, is used as a reason to discard certain individuals as those able to benefit from the posts.
There do actually exist individuals, not a novel mind-blowing statement I'm trying to make here, for whom every day to regulate their emotions to a level to which society does not deem unacceptable, as well as manage extremely difficult to regulate physical anxiety and fatigue symptoms, and very aberrant free-wheeling thought patterns, wish to do more technical thinking and feel safe to ask questions.
Now 1) where do these kind of individuals post "technically basic" questions, and 2) how do they learn to be of the personality type as to take criticism gently when their very emotional signals, all their life, have been considered "amplified/absurd"? Where do these individuals post without coming across "disorganized"?
I don't think the majority of authors here would claim universality, though many try to generalize observations to "most" or "aggregate" behaviors or values. https://www.lesswrong.com/tag/typical-mind-fallacy is well-known and often mentioned.
That said, I suspect you are nowhere near alone in your reactions, and I support and wish you well in your desire to become more rational and considered in your reactions to things. I don't know any good answers to your two questions - you're likely to get some sympathy, but also some hostility and not necessarily much useful advice here.
I'll predict that, to the extent you can ask ABOUT emotional reactions and reactions in the context of rationality, and to the extent that you can draft and edit things to minimize the disorganization, you can get some useful discussions here. If you're asking about other topics, especially fairly advanced or technical ones, before you're able to organize thoughts and handle a mix of responses which include some jerks and insensitive directness, you're probably not going to enjoy it.
Thanks for your reply. I am curious "you're likely to get some sympathy, but also some hostility and not necessarily much useful advice here." How are you defining hostile here and what may you think the hostility towards me would be in response to? Why would people be doing this if they aren't themselves emotionally shaken deeply to tears by something I write or say? Just for the fun of it?
I am wondering how thought organization is defined, and if there is a good template to assess if it has been achieved in a writing sample . Right now it just seems like an intuitive sense people have and use to put down others when they think others haven't achieved it.
Hostility may be too strong a word. It would appear mostly in the form of downvotes and helpful-sounding-but-not-really-actionable comments, for a post on a rationality-related topic that comes across as too fuzzy or incoherent. I don't think you'll find the kind of hostility you might on Reddit or Twitter, just a lack of embracing your ideas. It won't be personal attacks, people here are very good at suppressing their initial emotional reactions, even if it's a topic that triggers them.
Unfortunately, it's not something that can be templated - I'd advise reading a lot of posts, and starting with comments and shortform posts, to get a sense of the common styles and the kinds of reactions they get.
Response to first paragraph: Under the regime of "rationality is winning" (which is possibly being used as a motte) there is no room for being skeptical about rationality if rationality is simply what you would most prefer.
Response to second and third paragraph: such a person will come across as "disorganized" if they actually are so, but hopefully their interlocutor will be charitable after recognizing the goodness of the intent of that person.
EDIT: I prefer Dagon's comments above. Mine wasn't particularly helpful.
Under the regime of "rationality is winning" there is no room for being skeptical about rationality if rationality is simply what you would most prefer.
Rationality is patterns of systematically desirable/correct/winning/relevant cognition. It is (in part) about (or for) winning, the way a spoon is about eating, it's not winning itself. There is lots of room for being skeptical about spoons being useful for eating. If a pattern of reasoning doesn't offer systematic correctness/winning/etc., it's not part of rationality, that's how skepticism of rationality works.
Rationality is not what you would most prefer, the way a spoon is not the most eatworthy object, what you would prefer is captured by goals. Rationality is not just about achieving goals, patterns of reasoning conductive to achieving goals, it's also about correct/relevant understanding of facts, that are not tethered to any goals (epistemic rationality).
That's true, but I intentionally wrote "if rationality is simply what you would most prefer" i.e. if it really is the case that it is the most preferred means to the end. In your spoon example, it really may be the case that a spoon is the means you would most prefer to eat something with. A quibble, but yes.
I agree there is also a normative epistemic aspect to rationality, which could either complicate or be subsumed by the slogan "rationality is winning".
You were replying to someone new here, with statements that are very easy to read in very confused ways. So I attempted to reformulate them clearly, with a secondary objective of possibly catching something that you were also conflating in your own mind.
There is actually something directly relevant to flaritza's question in an expanded definition of rationality, though not useful. If rationality comprises patterns of thought that help with cognition and its many purposes, and there is a cognition-related problem like emotions acting up in nonstandard ways, perhaps interfering with usual communication norms or objecting to usual communication norms, this calls for patterns of emotional rationality that improve the situation. This is not useful because I don't have (references about) particular patterns of emotional rationality to offer. But there might be some discussion of this, and should be even if there isn't.
Do you mind expanding on "no room for being skeptical" and how that relates to my first paragraph? I am curious on the connection. Is it that the authors of the posts who are the subject of my criticism of not sounding "universal enough" in their posts, prefer rationality, and so they are not skeptical enough of it as to make it more accessible for other types of people? Or is it something totally different..
I meant that rationality is about systematically getting what you want, and is independent of moods/emotions. One could be a highly emotional rationalist if you are still making the right decisions.
I prefer Dagon's comments though. Mine wasn't particularly helpful.
As a preface I am trying to treat myself of extremely loose (by societal standards and "technical" crowds) thought associations that can cause strong emotions and disrupt my mental scratchpad. I have resignedly accepted it may be some sort of schizophrenic process, and not some latent form of creativity, because so far it has not produced any results in the outside world.
Specifically, I am looking for feedback on how to decide and develop a good method to decide the order to explore the questions in my below list, and if any questions jump out to anyone as either easy to answer or seem like the most obvious first questions to answer. Second, I'm also interested in hearing if anyone has good recommendations on some really good foundational concepts/texts/videos/blog posts/... to knock off many of these questions at once. And finally I'm interested in understanding what makes a good quality answer to any one of these questions.
Below is an example of the sort of questions I have:
Top questions I have
Some are based off of this video: Neural circuitry of cognitive dysfunction in schizophrenia - Dr. David Lewis - I find embeds distracting so this is url with a space: https://www.youtube.com/watch? v=k-pI5w-ElQ8
1. How does a system detect when something is off balance, and at which level to make compensations to restore "homeostasis"? Specifically, in the pyramidal-interneuron circuit in layer 3 DLPFC, why is the system trying to make "compensations" within layer 3, exclusively? As in, why isn't there a mechanism that extends beyond the layer 3 circuitry, namely, one further upstream; for example, an upstream mechanism that could increase the dendritic spine density on pyramidal cells, by detecting there is "insufficient" excitatory drive?
2. Through my extensive search, I have found there is not too much literature explaining/treating specifically the thought disorganization symptom of schizophrenia. Discussion on this topic, even in a social context, is limited and only the extreme caricatured instances are shed light on. I'm curious as to whether people have some references on the mechanism behind thought disorganization that can occur without any other psychotic-like symptoms, and the best way to treat this symptom.
3. Similarly to the thought disorganization symptom, I have not found much literature regarding (inner) musical hallucinations. I experience musical hallucinations in a mostly internal way, but the songs are often playing completely at random and bear no obvious relevance to anything in my immediate environment or even whether I have recently heard the song. Further, based on my interactions with others, the frequency to which the music changes in my head is much higher, and drawn from much more disparate sources, relative to my peers. This sort of phenomenon can be disruptive to maintaining an accurate mental scratchpad and engaging in deliberative thinking. I'm curious as to whether people know of the mechanism underlying this sort of fairly "random" musical hallucinatory thought.
4. I looked at a paper Dr. David Lewis authored with Vikaas Sohal's lab regarding new PAMs for targeting NMDA receptors in a cell specific way. From my understanding, his research identified that out of the 4 known NMDA receptors subtypes, GluN2D receptors are preferentially expressed on PV interneurons in the DLPFC and a GluN2C/GluN2D specific PAM called CIQ(+) depolarized the interneurons and increased firing rate in this cell type, in both normal and DLX5/6 het mice (that had a mutation in genes implicated in rule shifting deficits). There was no connection directly made to increased gamma band activity as a result of CIQ+ modulation, but I'm curious as to whether there are any clinical trials or planned ones centered around GluN2C/D modulators, and doing EEG follow ups to assess gamma band activity, in human subjects.
5. I see some literature testing the effects of morin in a mouse model exposed to an immune environment similar to that in schizophrenic humans. The study found that morin, a flavonoid, protected against dendritic spine density loss in pyramidal cells. I'm curious as to whether people know of further research regarding morin's capacity to regenerate dendritic spines, without concomitant administration with the dendritic spine density reducer, in chronically immunologically altered mice or even human models of schizophrenia. I'm asking because one of the important themes in Dr. Lewis' talk was that reduced excitatory drive to pyramidal cells in DLPFC was due to reduced dendritic spine density on these cells.
6. What are the best literature-based cognitive remediation protocols to reduce thought disorganization, purely as a symptom not necessarily existing with any other schizophrenia type symptoms?
Neurofeedback and other priority neuroscience questions
1. So how *do* you actually create a control for neurofeedback?
2. Why is it easier for brain states to enter hallucinatory ones rather than high IQ ones?
3. How do you detect if someone's emotion is feigned or reptilian--what is the difference in neural signature
4. When "normal people" listen to and can work in the midst of music in the background, how it is different in their brains compared to when schizos like me hear music in their head, even at a soft volume?
5. What were the seminal papers leading to the fitting of the PING model to the neural circuitry in the DLPFC?
6. What is the mechanism of gamma oscillations from a mathematical perspective, in certain regions of the brain?
7. Is the fact that individuals with ADHD exhibit gamma-band responses during stimulus encoding, that are uncorrelated with task performance, compared to controls, correlated to autistic traits?
8. What is the “auditory oddball task” and why is it important for comparing gamma oscillation differences in SZ and control patients?
9. What is the action of the medial prefrontal cortex vs dorsolateral prefrontal cortex and are the deficits more easily reversible in one region vs another? Which is more strongly implicated in SZ? Which in OCD?
10. What is the significance of 40 Hz activity in the human brain and why is activity in this band during working memory tasks correlated to increased performance?
11. Why are occipital beta responses reduced by valproate in bipolar individuals?
12. Why are SZ brains showing high gamma band oscillations for “simple tasks” compared to controls?
13. Why does 7-repeat isoform of DRD4 polymorphism enhance auditory-evoked gamma responses for SZ patients? Does this have anything to do with musical creativity?
14. Why does lower alpha rhythm differentiate bipolar euthymic and bipolar mania?
15. Why is there late theta response in SZ compared to controls in tasks involving significant WM and rule-switching tasks?
16. Why is visual evoked gamma oscillation deficit common in SZ but not auditory one?
17. How do you reduce PTSD like flashbacks of traumatic events and avoid painful memories/people from speaking to you in your head?
18. Has anyone actually logged their internal jumbled thoughts, not just the polished ones that make it out to the world as coherent sentences, with amazing depth and consistency?
Other questions
1. Is dopamine sensitization responsible for the cognitive disorganization I experience under the influence of cannabis?
2. Is any brain capable of reaching a temporarily "psychotic" state? If so what are the shared features in this universal psychotic state and which aren't?
3. Why not use similar mechanism to adenosine receptor upregulation from chronic caffeine use and more receptor availability for adenosine to bind to, to increase gaba receptor density for chandelier cell connection to pyramidal cell connection, to depolarize pyramidal cells more and make them ore likely to fire?
4. Why isn't the orbitofrontal cortex stimulated in FDA approved rTMS protocol if it is the most implicated in OCD? Why the DLPFC instead?
5. Why do some people decide to supplement with 5-HTP or tryptophan over SSRI?
6. Why is subverbal thought correlated with neural efficiency compared with vocalization/mental rehearsal thought? Similarly, swap in "mental rehearsal" with "non contextual (in a describable sense) memory recall (e.g. music :))"
7. Are intrusive thoughts in schizo-OCD an epiphenomenon or central to the cause of impaired working memory?
8. what is the balance of nmda agonists, antagonists (where partial, uncompetitive, allosterically modifying etc) to achieve "optimal" neurotransmission?
9. What are all the dopamine circuits in the brain and why do certain circuits result in psychosis while others in improved cognition? how does nmda/gaba transmission interface with these circuits?
10. What are all the pathways responsible for the superior efficacy of clozapine in schizophrenia other than the obvious ones and why aren't there studies replicating its effects with sarcosine?
11. Why do some people take individual b vitamins while others take b-complexes?
12. Why are reuptake and MAO inhibitor that are indirect adrenergic agonist like cocaine and amphetamine considered addictive whereas direct adrenergic agonist are not considered addictive?
13. Why is mitochondrial DNA inherited from the maternal lineage?
14. Why is silencing a gene difficult: no cure for Huntington's yet?
15. What is the right amount of learning on a subject to create a useful memory?
16. How are memories encoded in the brain?
17. How can alcohol both inhibit yet increase anger in different populations?
18. Explain the functional differences between D1, D2, D3 and D4 receptors, why cariprazine targets D3 receptors and why it has a superior clinical profile to other antipsychotics in terms of actually improving cognition
19. If glutamate is enhanced in DLPFC (https://pubmed.ncbi.nlm.nih.gov/12684737/)during mania, which is implicated in learning, why does this study (https://pubmed.ncbi.nlm.nih.gov/11465675/) say that “acute episode of mania can cause damage to learning and memory systems”?
20. Why is low GABA activity present in both bipolar/depressed individuals?
21. Why does ketamine induce long term desensitization of KOR receptors (implicated in suppressing dynorphin activity which suppresses working memory and such) but PCP does not, even though both are NMDA antagonists?
22. How effective is treating Alzheimer’s patients with psychedelics (as it can enhance acetylcholine release in hippocampus and PFC) vs the usual acetylcholinesterase inhibitors?
23. What are the strategies for schizophrenics to improve performance on the continuous performance test? Are working memory deficits more or less easily treatable than attentional deficits?
24. Apart from NMDA receptor agonism, as it seems that NMDA receptor structure itself is disrupted in SZ as are protiens involved in binding to NMDA receptors, what are the approaches for treating these additional differences/is it important?
25. Why does knocking out NR1 subunit of NMDA receptor (that is elevated in SZ patients) post-adolescence vs during early postnatal development lead to less chance of SZ phenotypes?
26. Knocking out NR1 subunit as mentioned in previous question also leads to decreased parvalbumin levels, yet this knockout early enough does not lead to SZ phenotypes, while PV levels are involved in generation of gamma oscillations, so what compensatory mechanisms are present in brains that have the early NR1 knockout that prevent SZ phenotype development?
27. How to increase the inhibitory output of PV interneurons as to not have disorganized output in SZ patients?
28. People talk about the benefits of antioxidant, anti-inflammatory and anti-VEGF (anti-angiogenic).. but what is the right amount of these effects and what happens if there is too much?
29. What are the different mechanisms by which different antioxidants work? Is it good to diversify one's sources? What happens when you overdose on them?
30. What is the difference between taking Vitamin D supplement and using turmeric to "activate Vitamin D receptors"?
31. Why are there contrasting results that "Some reports also suggested that β-carotene supplementation can produce significant increase in stroke incidence and overall cardiovascular deaths." from [The Role of Antioxidants in Human Health](https://pubs.acs.org/doi/pdf/10.1021/bk-2011-1083.ch001) vs decreased [Vitamin E and beta carotene supplementation in high risk for stroke: a subgroup analysis of the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study](https://pubmed.ncbi.nlm.nih.gov/11030804/) risk of stroke and cardiovascular deaths? Is there a difference in the dosages/patient populations in the studies here? What is going on? Is the first source reporting increased risk even legitimate since they didn't add a citation?
32. Are there subtypes of schizophrenia that involve high levels of SAM-e (evidenced by higher than normal levels of dopamine/serotonin along with pathways that shuttle these NTs to the non cortical pathways) in which case taking a B12 supplement that would increase SAM-e would *worsen* the positive symptoms of psychosis?
33. SAM-e is involved in converting cyanocobalamin to methylcobalamin, but also in the creation of creatine. How does the body know how to balance which pathway to utilize SAM-e for?
I am making this post because I don't see this topic discussed often or at all. I am interested in learning proof based math. Currently I am working through Linear Algebra Done Wrong and Tao Analysis texts and doing the exercises. However, I also struggle with faint musical hallucinations at times, as well as brain fog. To correct the brain fog I take caffeine/sarcosine, however that can agitate me physically causing extreme shaking and cold sweats while still increasing my working memory, so it becomes a difficult trade off. I also find it difficult to consistently work out the steps of a proof--when I page in information into my working memory, often times music in my head or a random thought will scrap my mental scratchpad and I have to keep rebuilding the structure in my head. I'm wondering if there are any people on this forum who have had this experience/have any advice short of "this is psychiatric, so deal with it accordingly"--obviously I am aware of that, but I am not interested in taking mentally blunting medications as to reduce the noise in my head, but want to preserve cognition and be able to do proofs. I also don't know if I'm underestimating the amount of noise/thought diversions in people's minds when they do proofs and needlessly self persecuting.
I've known many people who experience similar difficulties, and have had some of them myself. I have not found any reliable approaches that work for everyone, or even work for a majority. You're likely to need to try a bunch of stuff, and figure out what works for you. This takes years (as it's often weeks or months before you know whether a partial technique is actually useful for you), and changes over time, so it never ends. But for many, some idiosyncratic combination of some the following help a lot. This is not in order, just a brainstorm of things to try:
I have a proof that this is a solvable problem, but it won't fit in this margin.