Anna Salamon
Anna Salamon is co-founder and (former) Executive Director of CFAR (Center for Applied Rationality), an organization focused on teaching rationality skills to people working on important problems — primarily in AI safety and EA. Before founding CFAR she worked at MIRI alongside Eliezer Yudkowsky. She has been a prominent voice in the rationalist community since the early LessWrong era.
Her writing tends to focus on the gap between knowing something and actually doing it — the difference between verbal understanding and motivational-system understanding.
Humans are not automatically strategic
Her most widely cited essay argues that the default human mode is not strategic. Most actions people take are either optimized for social signaling, habit, or local comfort — not for achieving stated goals. Strategic behavior (defining goals clearly, tracking progress, testing hypotheses) must be explicitly practiced; it does not emerge automatically.
Dylan Bowman's reading list singles this out as its single top recommendation — the most important piece on the entire list.
Eight non-automatic heuristics she identifies: ask what your actual goal is; define what success looks like and track it; be curious about what's actually true; gather information before acting; test your conjectures; focus on what's actually working; ask whether your stated goal is your real goal; engineer environmental supports. See humans-are-not-automatically-strategic.
The verbal-motivational gap: you can know something verbally (strategically, you should do X) while your motivational system runs a completely different program (socially, emotionally, habitually). Closing this gap is the actual work.