With the advancement of Internet and information technology, the data generated across various disciplinary domains have witnessed exponential growth. The swift emergence, efficient organization and scientific processing of vast amounts of data have not only become potent instruments but also a novel foundation for academic research. This development has thus given rise to the so-called “data-intensive research paradigm”. The core characteristic of data-intensive research paradigm is its data-driven nature, which means computers are employed to uncover correlations within extensive datasets without the necessity of pre-established hypotheses, enabling the generation of more scientifically rigorous predictive outcomes. This paradigm has opened up new avenues for scientific research, facilitating the re-evaluation of numerous classical propositions on a more expansive empirical scale.

This research focuses on the age aspect within the scientific research life-cycle in the field of scientific sociology, delving deeply into novel conclusions regarding the traditional topic of “the statistical distribution of scientists” peak ages in scientific research. Scholars such as Zhao Hongzhou proposed that the age law of scientific discovery adheres to a Weibull distribution, and Liu Junwan, conversely, based on the ages of significant scientific discoveries of 435 Nobel laureates in natural sciences, put forward a corresponding log-normal distribution. However, in contrast to numerous previous studies, the sampling criteria of the self-constructed dataset in this paper have undergone substantial modifications.

Firstly, the scope of disciplines has been broadened to comprehensively incorporate a wide range of achievements in mathematics, mechanics, geology and computer science. Secondly, considerations have been given to the diversification of scientific elite types. Instead of being confined to scientists solely recognized by top international scientific awards, the scope has been extended to encompass all outstanding scholars who have left an indelible mark on the annals of human science and technology over the past few centuries. Upon conducting software analysis of current dataset, the authors have determined that the optimal fitting function is the “normal inverse Gaussian distribution”.

Further analysis reveals that the optimal fitting distribution of the present dataset exhibits distinct non-static characteristics. Variations in data sampling criteria can lead to disparities in the connotations and extensions of datasets. Regarding the classical topic of “the distribution of scientists’ peak ages in scientific research”, it appears that there is no fixed or static singular conclusion. Given the lack of a unified consensus among scholars regarding the definition of “outstanding scientists”, different datasets established accordingly may well correspond to diverse optimal fitting distributions. Consequently, despite the efforts to gather relevant information through web data mining and manual identification methods, the current dataset inevitably suffers from limitations in terms of sample size and type. In future research, if scholars can design datasets with larger sample sizes and greater sample diversity, it is highly likely that more profound and dynamic conclusions can be unearthed.

In addition to the above, this study also undertakes descriptive analysis and contemplation of local indicators such as the mean age, median age, minimum age, and maximum age of datasets from different epochs. Through observation of recent elections of academicians of Chinese Academy of Sciences, it has been noted that middle-aged and young academicians around the age of 50 are predominantly from the group of “National Science Fund for Distinguished Young Scholars” recipients. This finding further corroborates the distinct connection and isomorphic progression characteristics between “National Science Fund for Distinguished Young Scholars” recipients and “academicians”.

Based on the above, this study finally posits that the concept of “optimal age for scientific discovery” has exerted a profound and nuanced influence on China’s science and technology talent policies. This is evident in the explicit age limits stipulated for applications in various scientific research talent programs across different levels in China, which are set as stringent mandatory criteria. This has led to an increasingly prominent trend of youthfulness and precocity in the selection of scientific elites in China. However, considering the rich individuality and heterogeneous nature exhibited by each scientist during the innovation process, an overemphasis on age restrictions may potentially impede the full realization of individual scientists’ capabilities and runs counter to the intrinsic laws governing scientific and technological innovation. In light of that, the academic community should accord greater importance to the dialectical relationship between statistical regularities and microscopic case studies. While acknowledging the significance of general statistical patterns regarding peak ages in scientific research, it is equally crucial to delve deeply into the rich typical significance embedded within specific scientist cases.