Psychology for Prompt Engineers: Cognitive and Behavioral Foundations of Human–LLM Interaction

Prompt engineering has emerged as a central practice in human interaction with large language models (LLMs). However, prevailing approaches frequently emphasize linguistic formulations, technical optimization, and model-specific instructions while paying insufficient attention to the cognitive and behavioral processes through which users formulate intentions, interpret generated outputs, regulate trust, and make decisions. This article proposes a psychologically informed perspective on prompt engineering, conceptualizing it as a process of cognitive interface design rather than merely instruction construction. Drawing on cognitive load theory, metacognitive regulation, cognitive offloading, trust calibration, anthropomorphism research, and human–computer interaction, the article examines how psychological mechanisms shape the effectiveness and consequences of human–LLM collaboration. Particular attention is given to goal representation, attentional allocation, anchoring, confirmation bias, perceived authority, iterative feedback, and the preservation of independent judgment. The article introduces the PACT framework (Purpose, Adaptive Scaffolding, Critical Calibration, and Transfer) as a proposed conceptual architecture for designing prompts that support both immediate task performance and the user's long-term cognitive competence. Practical prompt-design strategies, evaluation criteria, and testable research propositions are presented. The central argument is that effective prompt engineering should optimize not only the quality of model-generated responses but also the quality of human reasoning, learning, decision-making, and epistemic control throughout the interaction. Psychology is therefore positioned as an essential theoretical foundation for the development of responsible, adaptive, and human-centered prompt engineering practices

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-19
DOI
https://doi.org/10.5281/zenodo.24355756
Primary Topic
Human-Automation Interaction and Safety
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Psychology for Prompt Engineers: Cognitive and Behavioral Foundations of Human–LLM Interaction

Roman Shaban
11 links
Zenodo (CERN European Organization for Nuclear Research)
Human-Automation Interaction and Safety
preprint

Psychology for Prompt Engineers: Cognitive and Behavioral Foundations of Human–LLM Interaction

Roman Shaban
preprint en

Abstract

Prompt engineering has emerged as a central practice in human interaction with large language models (LLMs). However, prevailing approaches frequently emphasize linguistic formulations, technical optimization, and model-specific instructions while paying insufficient attention to the cognitive and behavioral processes through which users formulate intentions, interpret generated outputs, regulate trust, and make decisions. This article proposes a psychologically informed perspective on prompt engineering, conceptualizing it as a process of cognitive interface design rather than merely instruction construction. Drawing on cognitive load theory, metacognitive regulation, cognitive offloading, trust calibration, anthropomorphism research, and human–computer interaction, the article examines how psychological mechanisms shape the effectiveness and consequences of human–LLM collaboration. Particular attention is given to goal representation, attentional allocation, anchoring, confirmation bias, perceived authority, iterative feedback, and the preservation of independent judgment. The article introduces the PACT framework (Purpose, Adaptive Scaffolding, Critical Calibration, and Transfer) as a proposed conceptual architecture for designing prompts that support both immediate task performance and the user's long-term cognitive competence. Practical prompt-design strategies, evaluation criteria, and testable research propositions are presented. The central argument is that effective prompt engineering should optimize not only the quality of model-generated responses but also the quality of human reasoning, learning, decision-making, and epistemic control throughout the interaction. Psychology is therefore positioned as an essential theoretical foundation for the development of responsible, adaptive, and human-centered prompt engineering practices

Zenodo (CERN European Organization for Nuclear Research)
Human-Automation Interaction and Safety
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon