Short answer: yes.

I spent years trying to solve the same problem you're describing. Better internal compression, more efficient encoding, tighter mental models. And I kept hitting the same wall, because the premise has a ceiling. Lossless compression inside a system with unreliable buffer fidelity is an engineering problem with no clean solution. You can optimize the codec all day, but if the buffer drops frames, the data still degrades. (I work in software and this analogy is the one that works for me.)

The mind might be the wrong place to solve this, though. Clark and Chalmers' Extended Mind thesis argues that cognition doesn't stop at the skull. When external systems reliably store, process, and make information accessible in ways that functionally replace internal memory, they are part of the cognitive system. Not metaphorically. Functionally.

So, instead of building a better compressor, I started building better architecture around the mind. External systems that let MY brain do what it's actually optimized for (pattern recognition, synthesis, novel connection) instead of what it does poorly (sequential storage, reliable recall, lossless buffering).

A few pieces where I dig into this:

• The Browser Tab Brain (aigavemeautism.substack…) covers why 300+ open tabs aren't chaos but distributed cognition in action

• The Adaptive Neurocognitive Architecture of ADHD (jonmick.ai/writing/adhd…) is my 67-citation research paper on why some brains are architecturally optimized for exploration over exploitation, and what that means for memory systems.

• My Phone is a Mirror (aigavemeautism.substack…) maps 213 apps as extended mind infrastructure. It's an easy example of the concept using today's tools.

• 80% of My Cognitive Life Is Spent Translating for Brains That Aren't Mine (aigavemeautism.substack…) addresses the hidden bandwidth cost most people in our situation never account for. You're not simply compressing data. That’s painful for some minds. You're running a real-time translation layer between your native processing format and neurotypical communication protocols. That tax alone eats capacity you think you're losing to compression failure.

I'd be curious what systems you've been building. The language you're using (transform coding, holographic encoding, distributed representation) tells me you're already thinking architecturally. You might be closer to the answer than you think, just pointed inward when the solution is partially outward.