Calibrated Surprise: An Information-Theoretic Account of Creative Quality

2026-04-28

AI Summary

This study formalizes the core creative writing principle of “just-right surprise” through an information-theoretic lens, proposing a calibrated surprise framework that integrates three constraints: authorial intent, reader expectations, and narrative plausibility. For the first time, it incorporates Shannon’s mutual information $I(X;Y) = H(X) - H(X|Y)$ into modeling creative quality, leveraging conditional entropy and the chain rule to unify multidimensional textual constraints—ethos, mythos, lexis, and dianoia—within a coherent probabilistic structure. The work demonstrates a fundamental incompatibility in information architecture between high-quality creative output and mediocre content and introduces a dynamic constraint propagation mechanism that operates without manual weighting. Validated through case studies and log-probability evaluations using lightweight large language models, the framework offers a theoretically grounded and operational foundation for Creative Quality Alignment (CQA) and expert evaluation benchmarks.

Abstract

The essence of good creative writing is calibrated surprise: when constraints from all relevant dimensions act together, the feasible solution space collapses into a narrow region, and the surviving choices look least predictable from an unconstrained view. "Calibrated" has a precise meaning: the author's intent, the reader's reasonable expectation, and the logic of reality converge. When these three independent judgements agree on every dimension, the set of admissible writing choices is forced into a very small region. A mathematical corollary follows: full-dimensional accuracy and mediocrity are mutually exclusive -- two sides of one constraint structure, not separate goals. We use Shannon's mutual information $I(X;Y) = H(X) - H(X|Y)$ as our analysis tool. "Calibrated" corresponds to conditional entropy going to zero; "surprise" to entropy going up; mutual information is the precise measure of the joint quantity. The argument rests on two pillars. Static: when constraints from ethos, mythos, lexis, and dianoia are imposed together, the admissible set collapses sharply, and surviving solutions show up as low-probability choices from an unconstrained view. Dynamic: the chain rule shows each writing choice is constrained by what came before and constrains what comes after; macro-level decisions naturally contribute a larger share of information, removing the need for hand-tuned weighting. Through case studies and lightweight LLM-logprob computations, we show the framework is both analytically useful and operational, laying the theoretical groundwork for Creative Quality Alignment (CQA) and a professional evaluation benchmark.

Problem

• calibrated surprise
• creative quality
• mutual information
• conditional entropy
• constraint structure

Innovation

• calibrated surprise
• mutual information
• creative quality
• constraint collapse
• Creative Quality Alignment