Type 1 versus type 2 processing
The defining contrast is autonomy versus working-memory-demanding decoupling.
What it means
In current dual-process theory, the cleanest way to distinguish the two kinds of thinking is by defining features rather than long correlated lists: Type 1 processing is autonomous, executing automatically once triggered and not requiring working memory, while Type 2 processing is non-autonomous, demanding working memory to sustain cognitive decoupling and hypothetical reasoning. Speed, consciousness, and emotionality are merely typical correlates, not defining criteria, and theorists distinguish parallel-competitive from default-interventionist architectures for how the two interact. On the dominant default-interventionist view, Type 1 supplies a rapid default response that Type 2 may or may not intervene to monitor and override, and most of the time it does not. This refined framing answers critiques that 'System 1 versus System 2' lumped together attributes that can dissociate. It sharpens what is really doing the explanatory work in two-process accounts of reasoning.
Examples
Reading a word is Type 1 — it happens whether you want it to or not — whereas deliberately checking an intuitive math answer recruits Type 2's working-memory-based override.
An experienced driver brakes at a red light without deciding to. Planning a diversion round a closed road is different work entirely, which is why the conversation in the car stops dead.
In the supermarket, 'two for £5' registers as a bargain before you have thought at all; holding the price per 100g in mind to compare tins is the override most shoppers skip.
First described in Evans & Stanovich (2013).