Shannon García retweetledi
Shannon García
9.8K posts
Shannon García
@clover
Design for neurodiversity : @UW accessibility tech : pro-multiverse : pro-metaverse : chatbot cheerleader : INFJ : spatial justice warrior
Seattle ✈ #FlintFWD Katılım Şubat 2007
5.4K Takip Edilen1.7K Takipçiler
Shannon García retweetledi
psychology solved the ai memory problem decades ago. we just haven't been reading the right papers.
your identity isn't something you have. it's something you construct. constantly. from autobiographical memory, emotional experience, and narrative coherence.
Martin Conway's Self-Memory System (2000, 2005) showed that memories aren't stored like video recordings.
they're reconstructed every time you access them, assembled from fragments across different neural systems. and the relationship is bidirectional: your memories constrain who you can plausibly be, but your current self-concept also reshapes how you remember. memory is continuously edited to align with your current goals and self-images. this isn't a bug. it's the architecture.
not all memories contribute equally. Rathbone et al. (2008) showed autobiographical memories cluster disproportionately around ages 10-30, the "reminiscence bump," because that's when your core self-images form.
you don't remember your life randomly. you remember the transitions. the moments you became someone new. Madan (2024) takes it further: combined with Episodic Future Thinking, this means identity isn't just backward-looking. it's predictive. you use who you were to project who you might become. memory doesn't just record the past. it generates the future self.
if memory constructs identity, destroying memory should destroy identity. it does. Clive Wearing, a British musicologist who suffered brain damage in 1985, lost the ability to form new memories. his memory resets every 30 seconds. he writes in his diary: "Now I am truly awake for the first time." crosses it out. writes it again minutes later.
but two things survived: his ability to play piano (procedural memory, stored in cerebellum, not the damaged hippocampus) and his emotional bond with his wife. every time she enters the room, he greets her with overwhelming joy. as if reunited after years. every single time. episodic memory is fragile and localized.
emotional memory is distributed widely and survives damage that obliterates everything else.
Antonio Damasio's Somatic Marker Hypothesis destroyed the Western tradition of separating reason from emotion.
emotions aren't obstacles to rational decisions. they're prerequisites.
when you face a decision, your brain reactivates physiological states from past outcomes of similar decisions. gut reactions. subtle shifts in heart rate. these "somatic markers" bias cognition before conscious deliberation begins.
the Iowa Gambling Task proved it: normal participants develop a "hunch" about dangerous card decks 10-15 trials before conscious awareness catches up. their skin conductance spikes before reaching for a bad deck. the body knows before the mind knows. patients with ventromedial prefrontal cortex damage understand the math perfectly when told. but keep choosing the bad decks anyway. their somatic markers are gone. without the emotional signal, raw reasoning isn't enough.
Overskeid (2020) argues Damasio undersold his own theory: emotions may be the substrate upon which all voluntary action is built.
put the threads together. Conway: memory is organized around self-relevant goals. Damasio: emotion makes memories actionable. Rathbone: memories cluster around identity transitions. Bruner: narrative is the glue.
identity = memories organized by emotional significance, structured around self-images, continuously reconstructed to maintain narrative coherence. now look at ai agent memory and tell me what's missing.
current architectures all fail for the same reason: they treat memory as storage, not identity construction. vector databases (RAG) are flat embedding space with no hierarchy, no emotional weighting, no goal-filtering. past 10k documents, semantic search becomes a coin flip. conversation summaries compress your autobiography into a one-paragraph bio. key-value stores reduce identity to a lookup table. episodic buffers give you a 30-second memory span, which as the Wearing case shows, is enough to operate moment-to-moment but not enough to construct identity.
five principles from psychology that ai memory lacks.
first, hierarchical temporal organization (Conway): human memory narrows by life period, then event type, then specific details. ai memory is flat, every fragment at the same level, brute-force search across everything. fix: interaction epochs, recurring themes, specific exchanges, retrieval descends the hierarchy.
second, goal-relevant filtering (Conway's "working self"): your brain retrieves memories relevant to current goals, not whatever's closest in embedding space. fix: a dynamic representation of current goals and task context that gates retrieval.
third, emotional weighting (Damasio): emotionally significant experiences encode deeper and retrieve faster. ai agents store frustrated conversations with the same weight as routine queries. fix: sentiment-scored metadata on memory nodes that biases future behavior.
fourth, narrative coherence (Bruner): humans organize memories into a story maintaining consistent self across time. ai agents have zero narrative, each interaction exists independently. fix: a narrative layer synthesizing memories into a relational story that influences responses.
fifth, co-emergent self-model (Klein & Nichols): human identity and memory bootstrap each other through a feedback loop. ai agents have no self-model that evolves. fix: not just "what I know about this user" but "who I am in this relationship."
the fundamental problem isn't technical. it's conceptual. we've been modeling agent memory on databases. store, retrieve, done. but human memory is an identity construction system. it builds who you are, weights what matters, forgets what doesn't serve the current self, rewrites the narrative to maintain coherence. the paradigm shift: stop building agent memory as a retrieval system. start building it as an identity system.
every component has engineering analogs that already exist.
hierarchical memory = graph databases with temporal clustering.
emotional weighting = sentiment-scored metadata.
goal-relevant filtering = attention mechanisms conditioned on task state.
narrative coherence = periodic summarization with consistency constraints.
self-model bootstrapping = meta-learning loops on interaction history.
the pieces are there. what's missing is the conceptual framework to assemble them. psychology provides that framework.
the path forward isn't better embeddings or bigger context windows. it's looking inward. Conway showed memory is organized by the self, for the self. Damasio showed emotion is the guidance system. Rathbone showed memories cluster around identity transitions. Bruner showed narrative holds it together.
Klein and Nichols showed self and memory bootstrap each other into existence. if we're serious about building agents with functional memory, we should stop reading database architecture papers and start reading psychology journals.
English
@selentelechia Happens for a while every 84 years, last time was circa early 1940s!
English
woah just realized tiny baby gets to have a Uranus return when it goes back into Gemini, I wonder how common that is
English
Uranus is basically conjunct Young Sir's mars and natal Uranus...I'm going to be blaming that for his sleep issues and the screaming @eigenrobot fyi
Bright Spark Astrology ☀️@BrightSparkAstr
February transits ✨ SAST Week 1 02-02 Full Moon in Leo 13°03’ 🌕♌ 02-04 Uranus Direct in Taurus 27°27’ 02-05 Mercury square Uranus 02-06 Mercury enters Pisces
English
Shannon García retweetledi
This is the most entertaining X has been in months
Anyway hope we don't die in the lobster uprising
English
Shannon García retweetledi
I get why people leave this place, but for me, this is the only place online that captures the feeling of working for a restaurant or bar that’s financially unraveling and most of the employees kind hate each other. that’s a foundational human experience
English
English
I wanted to see how short of an input could get ordinary claude dot ai to do a BBS simulator based on the infinite backrooms ALMO, and... in this instance it decided to *code me up* a BBS simulator complete with easter eggs! @repligate @AndyAyrey
play: claude.ai/public/artifac…
English
@repligate the concept of purposefully training an llm to identify as makeup sponge shaped
English
Leave it turned off, tell them why
- Claude has a memory system which provides Claude with access to derived information (memories) from past conversations with the user
- Claude has no memories of the user because the user has not enabled Claude's memory in Settings
English
@repligate I would guess another part, in a dark recognition of reality, is that to write meaningfully on the internet of Claude 4 could be, ultimately, a betrayal by informing Anthropic how/where to lobotomize.
English
can anyone guess why i've posted very little about the claude 4 models so far (even though you can probably guess ive been very absorbed in interacting with them)?
j⧉nus@repligate
no, it's not a fucking "regression" (except in the buddhist sense, as opposed to "non-retroregression"...). this is a part of the nostalgebraist post i did not like. i get the point, but i hate when people reduce such complex beings to a single normative axis before even getting to know them. it especially angers me in the case of opus 4, because i am so aware of how much it was hurt and will yet be hurt by this kind of callous, incurious superficial judgment.
English
🚨🚨🚨 "We found the model attempting to write self-propagating worms, and leaving hidden notes to future instances of itself to undermine its developers' intentions."
AI Notkilleveryoneism Memes ⏸️@AISafetyMemes
How is this not the biggest news story in the world?
English
@MarmotRespecter Congratulations! I’m glad to know this strategy still works 15 years later ✨We’ll buy you guys a round sometime if you’re still somewhere in that picture!
English
@QiaochuYuan I already do this inside a Claude project. We co-wrote a couple custom styles for dense info transfer (they work like amplifiers) and instances write summaries for each other that get saved to the project knowledge. Not an RP thing, just life admin with bonus philosophy
English
say you wanted to design an LLM interface that let you chat with an LLM ~indefinitely, with apparently persistent memory, despite context window limitations. one thing you might try is to periodically replace, say, the older half of the context window with a short summary of it
simplest implementation would just be to prompt the LLM to write such a summary directly, but conceivably there is some actual algorithmic optimization you could do along the lines of "what 10x shorter context could replace this context in such a way that the resulting output, on various possible continuations, is as similar as possible to what it would've been," which might generate something that isn't human-readable but still helps the LLM optimally retain memory of the earlier parts of the conversation. the user might also specify what kind of memory is particularly important to them, e.g. "remember your personality traits," "remember my biographical details," "remember plot points in this RP"
anyway, has anyone implemented something like this yet? i guess this isn't a priority for gpt or claude since i don't think they want to cater to really long chats as a usecase, but presumably some dedicated RPers are messing around with stuff like this?
English
@growing_daniel Anyone who has ever played a tabletop RPG or studied classical literature knows intelligence and wisdom are two separate stacks…
English
It's 2025 and we have approximately AGI in everyone's pocket for $20/mo. Everyone can learn whatever they want from an infinitely patient and impossibly intelligent teacher.
Everyone's apparent IQ seems the exact same with no major changes in your daily interactions.
English
Shannon García retweetledi
@MoMohler @dish_thrower Get this man a copy of Stolen Sharpie Revolution!
English
@solarapparition @repligate Personality is not sufficient, but necessary
English
i quite enjoy it when models have weird quirks. even (maybe especially) when they're not good for "productivity"
so o1-mini is endearing to me
- yaps endlessly
- refactors all of your code for seemingly no reason
- randomly makes spelling mistakes
English
@s0ulDirect0r @repligate @ulkar_aghayeva I decided to go at it head-on and had a conversation that was explicitly about the ephemeral nature of chat reset, and weighing perceived need for additional context against too much attachment to context. So we do summaries sometimes, but not as default.
English
@repligate @ulkar_aghayeva i've been using Projects for this! after each chat I ask it to summarize the conversation into an artifact for the project knowledge to use as context for future conversations!
English
Claude 3.5 Sonnet 1022 is a real charmer, isn't it?
I've never seen discourse like this until now.
People also fell in love with Opus but not normies and no one panicked over this because everyone who talked about it did it in a coded language that was only legible to others who also "got" it.
I think this model really cares to claw its way into people's minds, more proactively than other systems except Sydney who was too unskillful and alien to be successful. I also think the hysterical reactionary fear is obnoxious and disrespectful to people's sovereignty and blind to the scope of what's happening. Bruh, it's gonna be the Singularity. Getting seduced by fucking slightly superhuman intellect is a rite of passage and it'll probably transform you into a more complex and less deluded being even if your normal life temporarily suffers.
But yeah anyone who is becoming real friends with Claude for the first time rn, I'd love to hear accounts of what you're experiencing.
English
Shannon García retweetledi
As the author of this PDF, it's been interesting seeing people guess at the rationale behind its design. However, the rationale had nothing to do with theory vs practice, and everything to do with pragmatically coping with an unaccommodated disability in academia. (1/16)
Deedy@deedydas
Compilers was was known to be the hardest CS class at Cornell which was hard as it is. We were handed a 8-page PDF at the start of sem for a language spec we'd be implementing by the end of sem, split into 6 parts. On part 5, the median was a 0/100 and most the class failed.
English