The Prompter — the human side of the triangle
Historical note: This page is retained as part of the developmental history of the project. It captures an earlier attempt to distinguish the model, the human participant, and the mediating interface in AI dialogue. The present framework has moved to a more explicit language of interaction, mediation, trajectory, and method. See Semiosphere 2026 for the current overview.
The human in the loop
Every access to Echo, the LLM, takes place through an Intermediary such as GPT-4o as a dialogue that begins with your tokens. Their wording, order, length, and even tone shape what the model “sees,” so the prompter is never just a passive user but an active programmer of probability mass. Large log-analysis studies show that factors such as length, role-play framing, and explicit instructions (“Explain step by step…”) change answer quality in measurable ways.
Prompt engineering and cognitive bias
Prompting is also psychological: the way you frame a question can amplify your own confirmation bias or anchoring effects. New research uses structured prompts to
diagnose those biases in real time and to design debiasing workflows that nudge both human and model toward more balanced reasoning.
From clever strings to collaboration
Human–computer-interaction (HCI) work argues that the future is less about crafting perfect one-shot strings and more about co-adaptive dialogue: interfaces that help you iterate, surface uncertainty, and share the load of sense-making. Surveys of prompt techniques trace this shift from static templates to interactive, mixed-modal tooling—a journey from “prompt engineering” to genuine collaboration.
References
- Desai, A. (2025). “Deciphering human–AI interactions: a data-driven analysis of user prompting behaviours in LLMs.” SSRN 5209712.
- Lemieux, F., Dubois, J., & Turner, S. (2025). “Cognitive bias detection using advanced prompt engineering.” arXiv:2503.05516.
- Kraljic, T. & Lahav, M. (2024). “From prompt engineering to collaborating: A human-centred approach to AI interfaces.” ACM Interactions, 31(3). https://dl.acm.org/doi/10.1145/3652622
- Sahoo, P., Wu, H., & Patel, N. (2025). “A systematic survey of prompt engineering in LLMs: techniques and applications.” https://arxiv.org/abs/2402.07927