Repository is archived
No release in over 3 years
Models automatic vs deliberate processing modes based on Kahneman dual-process theory. Routine tasks run on autopilot (fast, low-cost), novel tasks trigger deliberate mode (slow, accurate). Tracks mode switching and override events.
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
 Dependencies

Development

 Project Readme

lex-cognitive-autopilot

System 1/System 2 dual-process autopilot for LegionIO. Models automatic vs deliberate processing modes based on Kahneman's dual-process theory.

What It Does

Routine tasks run on autopilot — fast, low cognitive energy cost — while novel or surprising tasks trigger deliberate mode: slower, more careful, and significantly more expensive. The extension tracks task patterns, accumulates familiarity through repetition, and automatically switches between modes as familiarity changes. It also models cognitive energy: deliberate processing drains reserves, and forced overrides (switching from autopilot to deliberate mid-stream) have the highest energy cost.

Usage

client = Legion::Extensions::CognitiveAutopilot::Client.new

routine = client.register_routine(pattern: 'daily_standup', domain: :routine)
client.execute_routine(routine_id: routine[:routine][:id], success: true)

# After enough successful executions, familiarity reaches autopilot threshold
# and the engine auto-switches to autopilot mode

client.autopilot_status
# => { current_mode: :autopilot, energy: 0.88, autopilot_ratio: 0.7, ... }

client.rest(amount: 0.2)    # recover energy
client.decay_all             # apply familiarity decay (call periodically)

Development

bundle install
bundle exec rspec
bundle exec rubocop

License

MIT