Pragmatic Thinking And Learning - KeynesYouDigIt/Knowledge GitHub Wiki

Event theories are observable, construct theories are not (and are evaluated in terms of their usefulness)

Dreyfuss model

Experience is how much your thinking has been changed by doing tasks (first year ten times doesn't count). Don't force experts to follow rules. It may not feel scientific or repeatable, but you'll take away their key advantage. Give novices rules and experts context.

Novice

Just needs context-free rules, wants to accomplish a task. There are a limited range of tasks this applies to. You can't fully define everything, or you'll fall into infinite regression.

Advanced beginner

Can apply advice and formulate some basic principles. No troubleshooting skills, no big picture.

Competent

Forms conceptual models, can troubleshoot, and can solve novel problems. May not know which details to focus on, relies heavily on past experience.

Proficient

Need the big picture. Can reflect and revise performance, and can learn from others' mistakes. Can form and apply maxims contextually. More of a junior expert than an advanced competent.

Expert

Pure context and intuition, not reason. A primary source of knowledge.

Learning

10 years of practice to become an expert:

  • Well defined task
  • Appropriately difficult
  • informative feedback you can act on
  • Opportunities for repetition and correction
  • People learn by watching and imitating. Imitate, assimilate, innovate (Shu ha ri).
  • Take notes and do mind maps by hand to engage R mode thinking. Typing doesn't.
  • Playing with something needs to come before facts. Make something fun. Exploration needs to be safe.
  • Cultivate situational feedback (from inner game of tennis). No rules of the game, micro skills. Focus on how the correct movement feels. Be aware of what's happening, less on whether it's working.
  • Pressure kills your ability to think. R mode is shut out.
  • Something feeling dangerous is the same psychologically as it being dangerous, so cultivate safety

SQ3R

Transition from R mode to L mode

  • Survey - Scan the TOC and chapter summaries
  • Question - Write questions you have
  • Read - Read in its entirety
  • Recite - Summaries, take notes, and put into your own words
  • Review - Reread, expand notes, and discuss with colleagues

Systems

Beware tools, models, and formal methods (eg. process manuals).

  • Don't confuse the tool, model, or plan with reality
  • Don't devalue traits that can't be modeled effectively
  • Don't legislating behavior is at odds with individual autonomy
  • Tools alienate experts
  • Spelling out too much detail hides the important parts
  • They over oversimplify complex situations
  • They demand excessive conformity
  • Have no contextual nuance
  • They create confusion between when to follow rules and when to exercise judgment
  • The model starts eroding the thought process it originally embodied and the steps become meaningless

Brains

Have a slow/linear CPU and a fast/non-linear CPU. Only one can access memory at a time.

Slow/linear CPU (L mode, deliberate)

  • If it's not processing anything else, it generates an internal stream of verbal chatter (the voice in your head)
    • Listening to your idle chatter prevents your R-Mode from activating.
  • Focuses on details and steps

Fast/Non-linear CPU (R mode, undermind)

  • Super regular-expression search engine
  • Problem-solving, creativity, and intuition
  • Recognition, holistic patterns
  • No language
  • Not directly controllable, finds results at unpredictable times. Cannot be commanded, only invited
  • Required for expertise
  • Drawing a picture upside down disengages the L brain part that's trying to identify patterns by giving it a job it doesn't want, allowing you to just do a the thing concretely Walking a labrynth bores your L-mode, leaving your R-mode to run free
  • Right when you wake up, write 3 pages out long-hand, before you do anything else. This helps "dump" your R-mode contents.

Metaphor

  • Metaphor is where L and R meet. The right metaphor can engage both modes.
  • Metaphor means "to transfer"
  • The right metaphor helps generate new ideas
  • Humor helps find the boundaries of the metaphor

Habits

  • Once you start capturing ideas, your brain generates more of them
  • In a rich, rewarding environment, you grow new neurons and connections. In a drab one, you don't.
  • Your brain gets rewired to support the things that you do more (practicing music, immersing in a language, or writing code)
  • Engaging more senses improves retention and creativity
  • Climb a wall, then go to the class on climbing walls- give your brain an R-mode experience before L-mode instruction. "Write drunk, revise sober"
  • Pair programming does allows two people to be in all L or all R mode, at the same time
  • Multitasking kills 20-50% of your productivity because of context switching. Multitasking is doing multiple concurrent tasks at different levels of abstraction.

Cognitive Biases

  • Anchoring (the first result anchors subsequent ones)
  • Fundamental attribution error ("I was tired, but this person is dumb")
  • Self-serving bias ("If it worked it was me, if it doesn't it was the team")
  • Need for closure (planning upfront, masking inherent uncertainty)
  • Confirmation bias
  • Exposure effect (preferring things that are familiar)
  • Hawthorne effect (people behave differently when being studied)
  • False memory (every read is a write)
  • Symbolic reduction fallacy (the symbol doesn't capture all complexity)
  • Nominal fallacy (labeling something means you can explain or understand it)

The palest ink is better than the best memory

Generations

Howe/Strauss Generational archetypes:

  • Prophet (vision and values) Boomers
  • Nomad (liberty, survival, honor) Xers
  • Hero (community, affluence) Millenials
  • Artist (pluralism, expertise, due-process) Zoomers

Hedge your bets with diversity- different people bring different assumptions.

Training

Companies love sheep dip training- standardized, easy to purchase, easy to schedule, and easy to check boxes that it was accomplished.

Problems:

  • Learning is something you do, not something done to you
  • Knowledge without experience is useless
  • No goals and feedback produces random results