What's your prompting style?

Liu & Yao (2026) surveyed 211 professionals and identified four distinct interaction styles: directive, conversational, and two co-creative hybrids. Each correlates with a different mental model of AI. Enthusiastic advocates overwhelmingly prefer co-creative approaches (86%). Those who see AI primarily as a tool favor directive prompting (46%).

No style is universally best. Sprague et al. (2024) surveyed over 100 papers and found that structured prompting helps on math and symbolic reasoning (+12-14%), but gains on all other task types are minimal (+0.7%). For complex, ambiguous problems, iterative refinement consistently outperforms single-shot commands by 20-40%. The optimal approach depends on the task.

Schneider (2025) observed that experienced users naturally shift from structured to conversational prompting over time. This tool helps you see your starting point and where your style serves you well or holds you back.