Discover how overly restrictive or complex prompts trigger the pigeonhole effect, causing LLM reasoning to collapse into brittle pattern-matching, and learn how objective-based prompting preserves model intelligence.