Curiosity Rooms

Phase1 - Orientation - Working Memory

8 min read

Phase 1 Orientation: Working memory, overload, and cognitive transformation

Source: ~/Downloads/working memory, cognitive overload and recursive redistribution - Google Search.md
Phase: 1
Previous: (none; first phase)
Next: Phase2 - Freeze
Status: hub-approved

Pipeline: CONSTITUTION - Publishable Asset Pipeline
Index: 00 - Downloads Batch Map Index
Slug folder: Work/working-memory-overload/

A. Human prompts only (concise restatement)

Source is a short Google-export thread (3 prompts). Full Phase 1 structure still applies.

#Concise version of the human prompt
P1Search / open frame: working memory, cognitive overload, and recursive redistribution. What are these, and how do they fit together?
P2Working memory is usually framed as seconds-to-minutes. What about unfinished work spanning days or weeks (pending tasks, open projects)? Do those occupy working-memory slots, sit outside WM but throttle it, or reduce real-time processing capacity another way? Stress and incomplete tasks as possible capacity thieves.
P3Maybe the literal question is wrong. Help reframe: how does information become thinking, expertise, intuition, problem-solving skill, professional identity, and long-term behavior change? Motivation: redesign personal knowledge management for cognitive transformation, not note accumulation. Do not solve the problem. Map the research landscape: disciplines, theories, mechanisms, models, frameworks, memory systems, skill/expertise/identity research, knowledge representation, systems ideas, and especially unknown unknowns (fields a pure cognitive-psychology search would miss). One-paragraph nodes, neighbors, map placement. Atlas / knowledge-graph shape, not a linear textbook answer. Expand search space; replace a too-narrow frame if needed.

B. Underlying intents / curiosities / conflicts (from prompts only)

Pulled from the human prompts (statement form):

  1. I need a clear, usable model of working memory as a limited real-time scratchpad and of cognitive overload when load exceeds that limit.
  2. I want “recursive redistribution” named and placed: how nested, multi-step reasoning recycles WM resources without collapsing.
  3. I am testing whether everyday language about “things on my mind” matches the science, or whether unfinished work lives outside WM and only raids it.
  4. I suspect open loops over days/weeks reduce available WM capacity even when the task is not on screen right now.
  5. I want the mechanism spelled out: occupation vs interruption vs stress/physiology vs long-term structures that stay “warm.”
  6. I am willing to abandon the first question if it is the wrong level; I care more about transformation than about a textbook WM definition.
  7. My real design problem is PKM: systems optimized for cognitive change (expertise, identity, behavior), not accumulation of notes.
  8. I need the boundary of the problem: which disciplines and theories even belong in the search space.
  9. I want unknown unknowns: concepts and fields an interdisciplinary researcher would check that a pure cog-psych path would miss.
  10. I prefer atlas / graph structure (clusters, neighbors, map location) over a single linear explanation.
  11. I want framing corrected before answers if my starting frame is too narrow (storage mind vs transforming agent).
  12. Conflict held in the thread: short-horizon capacity (WM limits, open loops) vs long-horizon becoming (intuition, identity, lasting change).

C. Coverage: which intents the AI side supports well vs weakly

Judged from what this file’s responses are structured to answer. Not inventing science beyond “was this the question the AI was answering.”

Intent #Coverage in this fileNote
1 WM + overload basicsStrongP1 response: capacity, load types, overload as total load over capacity
2 Recursive redistributionStrongDedicated section: intermediate results, serial bottleneck, chunking as relief
3 Open loops vs literal WM occupancyStrongExplicit “no, not sitting in WM / yes, they consume capacity” + Zeigarnik
4 Days/weeks unfinished work throttles capacityStrongTUTs, slot hijack, desk-and-warehouse metaphor
5 Mechanism split (interrupt / stress / LTWM)StrongZeigarnik, TUTs, LTWM energy cost, cortisol / PFC
6 Willingness to reframeStrongP3 treated as frame upgrade; storage vs complex adaptive system
7 PKM for transformation not accumulationStrong as framingAtlas repeatedly ties nodes back to PKM design implications
8 Disciplinary / theory boundary mapStrongMulti-cluster atlas (bio-spatial, computational, expertise, identity)
9 Unknown unknowns / adjacent fieldsStrongMET, ANT, FCA, PCT, autopoiesis, hippocampal spatial co-option, etc.
10 Graph / atlas formStrongClusters + acquisition/transformation matrix; not one essay line
11 Frame correctionStrongExplicit shift: note warehouse → evolving agent / structural coupling
12 Short-horizon capacity vs long-horizon becomingPartialBoth poles answered in sequence; little explicit synthesis of how open-loop WM mechanics sit inside the transformation atlas

Rule for skeleton: center on intents 1–11 (AI-backed in-thread). Treat 12 as the through-line / orientation frame so the asset does not feel like two unrelated articles stapled together. Do not invent a grand unified theory beyond what the thread already bridges.

D. Curiosities reshuffled for a normal audience (logical path)

Human order leaped: technical triad (WM / overload / recursive redistribution) → open loops over days/weeks → full research atlas for cognitive transformation and PKM.

Audience-friendly order (same curiosities, fewer leaps):

  1. What working memory is, and why its capacity is small
  2. What cognitive overload is (intrinsic, extraneous, germane load)
  3. Why multi-step, nested problems feel crushing (recursive redistribution and the serial bottleneck)
  4. How experts free slots (chunking, long-term structure) without magically enlarging the scratchpad
  5. The everyday puzzle: unfinished work from days/weeks ago still “uses up brain”
  6. Where that unfinished work actually lives (not inside the 3–4 / 5–9 slot desk)
  7. How open loops raid the desk anyway (Zeigarnik, intrusive reminders, mind-wandering)
  8. Stress as a second throttle (physiology, not only “another thought in a slot”)
  9. Why this matters for learning systems and PKM (capacity is for transformation, not only storage)
  10. Reframe: from note accumulation to cognitive transformation (expertise, intuition, identity, behavior)
  11. Atlas of the landscape: body/space, computation/structure, expertise/intuition, identity/behavior
  12. Unknown unknowns worth searching next (adjacent fields pure WM search would skip)
  13. Optional closer: how to use the map (pick a node, design for transformation, not hoarding)

E. Article skeleton only (outline)

Working title options (from thread orientation, not a new thesis):

  • Working memory, open loops, and the map from information to expertise
  • Why unfinished work steals capacity, and what that means for knowledge systems that actually change you

Mode lean: Authority-explanatory for sections 1–8; atlas / map mode for 9–12; light Experience frame only if open/close needs the human “wrong question / redesign PKM” steer.

0. Orientation (short, optional)

  • Thread started as a capacity question and became a design question.
  • Core conflict: limited real-time mind vs long-horizon becoming (expertise, identity, behavior).

1. Working memory as the bottleneck

  • Scratchpad metaphor; small capacity (classic 5–9 vs tighter modern estimates if the thread used both)
  • Bridge between sensory input and long-term memory

2. Cognitive load and overload

  • Intrinsic, extraneous, germane
  • Overload as total demand over capacity (fog, paralysis, sense-making failure)

3. Recursive redistribution under complex tasks

  • Holding intermediate results while running the next step
  • Serial processing bottleneck and self-correcting loops
  • Chunking / long-term structure as the practical relief valve

4. The days-and-weeks problem

  • Unfinished projects and pending tasks that are not “in the moment”
  • Everyday intuition: “they take up working memory”

5. Where open loops live, and how they still tax you

  • Not permanent residents of the conscious scratchpad
  • Long-term storage + periodic raids into WM
  • Zeigarnik / open loops; task-unrelated thoughts; reduced available slots
  • Stress / cortisol path as capacity shrink, not only slot theft
  • Long-term working memory: keeping a large project “warm” costs energy

6. Desk and warehouse picture (one clean metaphor section)

  • Small desk (WM) vs warehouse (LTM) vs warehouse manager throwing boxes onto the desk
  • Why simple tasks feel hard under chronic unfinished work

7. Bridge: capacity management is not the whole game

  • Offloading and closing loops free the desk
  • But the deeper design goal is not empty desk forever: it is transformation

8. Reframe: information → thinking, expertise, intuition, identity, behavior

  • Note accumulation vs cognitive transformation
  • Mind as evolving system, not only storage warehouse
  • PKM implication: optimize for change, not clip volume

9. Research atlas, Cluster A: body, space, materials

  • Material engagement, spatial/hippocampal mapping, human-plus-tool networks (as named in-thread)
  • Why passive text piles fail active cognition

10. Research atlas, Cluster B: computation and structure

  • Predictive processing / active inference
  • Formal concept structure; connectionist / distributed meaning
  • Why isolated notes stay “dead” without pattern and prediction pressure

11. Research atlas, Cluster C: expertise and intuition

  • Recognition-primed / naturalistic decision making
  • Cognitive flexibility across multiple lenses
  • Perception-control / cybernetic loops for behavior change

12. Research atlas, Cluster D: identity and long-term becoming

  • Communities of practice / legitimate peripheral participation
  • Ontological / transformative learning (who you are, not only what you know)
  • Autopoiesis and structural coupling: info as trigger, not forced overwrite

13. Using the map (skeleton closer, not a full how-to product)

  • Acquisition vs transformation matrix (as sketched in-thread)
  • Unknown unknowns as the real search expansion
  • Design heuristic: systems that force model rewrite, multi-lens reuse, and identity-relevant practice, not only storage

Explicit non-goals for this asset

  • No full article body in Phase 1
  • No inventing extra science outside the thread
  • No dumping the entire atlas as undifferentiated encyclopedia paste; Phase 2 freezes the best answer blocks under curiosity frames
  • No personal life-detail expansion; keep PKM motivation as design frame only
  • No forced merge into unrelated body-awareness or dual-N-back assets unless hub opens a separate cluster note

F. Phase 1 gate self-check (for hub)

  • Human prompts complete (3/3)
  • Intents feel human-driven (capacity mechanism + PKM transformation atlas)
  • Outline is audience-ordered (bottleneck → open loops → reframe → atlas)
  • No full article; no AI rewrite pass
  • Wikilinks: constitution + index; source path absolute-style under Downloads

Hub next: approve Phase 1 → brief Phase 2 Freeze worker on this slug only.

Graph View

Backlinks