Complexity and the uncertain impact of novel science

Novel scientific ideas drive progress, yet their outcomes are highly uneven. This paper examines whether problem complexity — the degree of interdependence among elements a study must coordinate — shapes the impact of novel research. I introduce a measure of complexity based on the NK framework and find that novel work addressing more complex problems is substantially more likely to achieve high impact, less likely to receive minimal citations, recognized earlier, and cited more broadly across fields. These patterns can be interpreted through the problem-solving view of science, the topology of search landscapes, and characteristics of the combinations themselves. The findings suggest that variation in the outcomes of novel research depends critically on the structure of the problems to which novelty is applied. Data on the complexity of papers are released at https://doi.org/10.5281/zenodo.20813592.