A study coordinated by Ricard Solé, an ICREA researcher, and Daniel Amor and Sergi Valverde, of the Complex Systems Lab research group in the Department of Experimental and Health Sciences (CEXS) at UPF, explains that technological innovations in technology are mainly produced by combination. The more innovations accumulate, the greater the probability that new inventions are obtained from previous designs.

The study, published on arXiv.org, presents a generalized model of technological evolution that takes into account two main properties - the number of prior technologies that are needed to create a new one, and the speed at which the technology concerned ages. 

When long-term memory is used and previous inventions are therefore available to create innovation, singularities, understood as the emergence of new technologies, may appear under certain conditions. However, a prior absence of innovations makes the rate of creation of inventions decrease until it becomes a linear regime.

Patent citations are important windows for corroborating this process, as they give an initial approximation of the increase in inventions, as well as their interaction over time. The accelerated growth in the pattern of patents shows that there is a highly linear process of innovation and the evidence suggests that this is at least partially associated with combinatorial processes.

The nature of technological innovation, an essential component of human creativity

Nature and the pace of innovation is a recurring and complicated topic. It has been the focus of attention from evolutionary biologists, physicists and economists in equal measure. Inventors find inspiration from previously existing designs, while they push to overcome the boundaries of innovation. So far, the idea that technological knowledge tends to infinity has been suggested, but the work of Costa, Amor and Valverde presents evidence that casts doubt on this.

Reference work

Ricard Solé, Daniel Amor and Sergi Valverde (2014), " On singularities in combination-driven models of technological innovation", arXiv.org, arXiv:1407.6890[physics.soc-ph]