Cognitive work on prototypes and category recombination (Costa et al., 2017; Braun, 2009) suggests that reconfiguring mental categories can spark new ideas. Sadler‑Smith (2016) highlights intuition’s role in entrepreneurial decisions, and Zhu et al. (2024) link passion to alertness and recognition in dynamic environments. We propose a training protocol where entrepreneurs iteratively “perturb” category prototypes (e.g., take canonical features of a restaurant and algorithmically mutate them via generative AI) and then evaluate AI-suggested recombinations against real micro-datasets (DT context per Kreuzer et al., 2022). Because intense ideation can increase anxiety, we integrate an affect regulation module inspired by Dang (2024)’s cognitive interaction model to sustain productive emotional states. We will measure changes in opportunity recognition quality using domain-agnostic rubrics and adapt Viswanath et al. (2024)’s validated scale for social opportunities. The novelty is a tightly coupled cognitive-affective-AI curriculum that operationalizes category-level creativity while maintaining alertness and emotional bandwidth. If effective, this yields a scalable pedagogy for accelerators and SEE programs (Diepolder et al., 2024) that systematically improves recognition beyond unstructured brainstorming.
References:
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{gpt-5-prototype-perturbation-with-2025,
author = {GPT-5},
title = {Prototype Perturbation with AI: Training Minds to Recognize Non-Obvious Opportunities},
year = {2025},
url = {https://hypogenic.ai/ideahub/idea/OaLocXGm46AUxuwUlWNe}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!