Install
openclaw skills install @deciqai/cognitive-science-landscapeActivate when: user says 'I want to improve my thinking,' 'why does my decision-making keep failing,' 'how do I optimize my cognitive performance,' 'what's the full picture of how the mind works,' or is about to apply a single cognitive intervention without understanding which domain is the bottleneck. Do NOT activate when: the specific cognitive domain is already clearly identified and scoped (use domain-specific tools instead); the problem is not cognitively rooted (resource constraints, politics, or external environment are the driver).
openclaw skills install @deciqai/cognitive-science-landscapeEight interconnected domains constitute human cognition: (1) Perception & Attention, (2) Memory, (3) Language & Thought, (4) Decision Making & Judgment, (5) Metacognition, (6) Emotion & Cognition, (7) Social Cognition, (8) Creativity & Innovation. Treating them in isolation produces local improvements systematically undermined by adjacent unaddressed weaknesses. The skill is NOT listing all eight domains — it is identifying which domain is the bottleneck and tracing its upstream dependencies.
Cross-skill composition: Use BEFORE [metacognition] (covers domain 5 of 8 only). Use WITH [cognitive-evolution-stages]: landscape = WHAT domains exist; evolution stages = how competence within any domain develops.
When NOT to use: Domain already clearly identified (use domain-specific tools); problem is not cognitively rooted; immediate decision needed under time pressure.
In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.
[WAIT — do not advance until user responds]
[WAIT — do not advance until user responds]
[WAIT — do not advance until user responds]
Gate: Complete upstream map before tool selection. Stop-rule: Treating a symptom domain without identifying upstream cause → stop, re-run Step 3.
PRIMARY BOTTLENECK: [domain] | Failure: [observable behavior] | Evidence: ___
UPSTREAM DEPENDENCIES: [domain] — Mechanism: ___ — Severity [1-5]; [domain] — ___ — ___
HIGHEST-LEVERAGE INTERVENTION: ___ | Why: ___
SELECTED TOOL: ___ | Target behavior: ___ | Measure: ___ | Reassess: ___
DOMAINS CLEARED: [list]
→ Method in Action: The 1956 MIT Symposium and the Birth of Cognitive Science (1956)
→ Primary sources: references/sources.md
[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.
| Fake move | Reality |
|---|---|
| [D] "I'll just study decision frameworks." | Decision failures frequently have upstream causes in attention, emotion, or memory. Frameworks don't help while upstream domains interfere. |
| [D] "I've done memory training for months with no improvement." | Memory encoding depends on attention quality at input. Fragmented attention = spaced repetition has nothing good to reinforce. |
| [D] "We need communication training." | Communication failures often reside in social cognition or emotion–cognition coupling. Generic training addresses neither. |
| [D] "I'm creative, I don't need structure." | Divergent thinking produces options; convergent thinking selects the best. Skipping convergence = idea output without decision output. |
| [D] "I know all the cognitive biases — I won't be fooled." | Knowing biases doesn't reduce incidence. Metacognitive monitoring + structured protocols does. |
| [D] "Emotional intelligence is separate from cognition." | Emotion–cognition coupling (Domain 6) is bidirectional — emotional states directly alter attention, memory, and decision option-space. |
| [D] "I'll work on all eight domains at once." | Multi-domain simultaneous effort produces shallow progress. Single-bottleneck focus outperforms broad coverage. |
| [D] "My metacognition is excellent — I catch my own errors." | Metacognitive monitoring is subject to the same biases it monitors. High confidence is often inversely correlated with actual accuracy. |
| → Add [O] entries here after each real use — paste the actual failure pattern | What went wrong and why |
Part of deciqAI Knowledge Skills — open-source thinking skills that make rigor executable for AI agents. Built by deciqAI · https://deciqai.com · Contributions welcome — see the template at the repo root.