This paper examines how the design of evaluation architectures influences organizational sensing quality in volatile, uncertain, complex, and ambiguous (VUCA) environments. It introduces sensing resolution, which is the statistical fidelity with which a decision architecture preserves the variance of opportunity-resource interactions, and argues that pre-valuation categorical labeling systematically truncates this resolution. The study proposes the Recognition-Assessment-Selection (RAS) framework as a normative decision architecture grounded in the Dynamic Capabilities View and drawing on the Attention-Based View and information-processing theory.
Drawing on the Dynamic Capabilities View, the Attention-Based View, and information-processing theory, the paper combines deductive theory development with formal probabilistic modeling and comparative case analysis. Three propositions are derived to characterize the theoretical consequences of pre-valuation truncation. The framework is illustrated through a comparative analysis of Fujifilm and Polaroid during the analog-to-digital transition, serving as illustrative archetypes rather than empirical tests.
The formal model theorizes that pre-valuation categorical architectures systematically induce optimistic bias in perceived expected value, compress perceived variance, and amplify both distortions as environmental volatility increases, with further intensification theorized under heavy-tailed Paretian distributions. The Fujifilm-Polaroid contrast illustrates how differences in information architecture are consistent with divergent strategic decision processes in the same high-VUCA regime.
The formal model employs a Gaussian baseline that represents a conservative lower bound on truncation bias, and formal derivations for fat-tailed distributions remain for future research. The cases serve as theoretical archetypes rather than empirical tests of causal effects. Future studies could test the propositions through experimental designs examining RAS's de-labeling effect across teams with varying cognitive diversity, longitudinal field studies tracking dynamic capability gap detection, investigation of artificial intelligence-human collaboration in augmenting sensing resolution, and extension of the framework to public sector and non-profit organizational contexts.
For managers in VUCA environments, RAS offers four distinct contributions: a portfolio orchestration tool enabling cross-path resource and attention allocation across multiple simultaneous strategic initiatives; process design mechanisms – including independent pre-discussion assessment, authority-free evaluation sessions, and mandatory worst-case configurations – that structurally support valuation-neutral sensing; an early warning system for dynamic capability gaps when applied continuously; and explicit boundary conditions specifying when RAS's advantages are most pronounced and when simpler categorical tools remain appropriate.
This paper makes three contributions. Conceptually, it reframes sensing failures as functions of evaluation architecture rather than solely cognitive bias, extending the micro-foundational agendas of the Dynamic Capabilities View, the Attention-Based View, and information-processing theory. Formally, it derives the statistical consequences of pre-valuation truncation through a probabilistic model. Normatively, it proposes RAS as an operational decision architecture that translates the principle of informational fidelity into a structured, three-stage evaluation process with actionable guidance for high-stakes strategic decision-making under uncertainty.
