Double-Loop Learning

Chris Argyris's distinction between single-loop learning (adjusting actions to achieve existing goals—a thermostat adjusting temperature) and double-loop learning (questioning whether the goals themselves are correct—asking whether this is the right temperature to target). Single-loop learning improves performance within the current framework. Double-loop learning questions whether the current framework is the right one. Most organizations do single-loop learning well (optimize existing processes) and double-loop learning poorly (question whether the processes should exist at all). The failure to double-loop produces organizations that get very efficient at the wrong thing.

When to use it

When optimization efforts produce diminishing returns (you may be optimizing within the wrong framework); when performance is strong on stated metrics but doesn't feel like genuine progress; when organizational learning addresses how but not why; when personal growth has plateaued despite continued effort within the current framework.

How it can help

Build double-loop learning into organizational and personal practice. After any significant outcome: don't just ask 'how can we do better next time?' (single-loop). Also ask 'are we pursuing the right goal?' and 'are we solving the right problem?' (double-loop). The practice: schedule regular reviews that explicitly question assumptions, goals, and frameworks—not just performance against them. In personal development: don't just optimize your approach to your current goals; regularly question whether the goals themselves serve your deeper values.