Human Behavioral Complexity Peaks At Age 25 Interview with:
Dr. Hector Zenil

Information Dynamics Lab
Unit of Computational Medicine, SciLifeLab
Center for Molecular Medicine
Karolinska Institute, Stockholm, Sweden What is the background for this study? What are the main findings?

Response: The generation of randomness is known to be related to cognitive abilities. It has also recently been shown that animals can recur to random behaviour to outsmart other animals or overcome certain situations. Our results that humans can best outsmart computers in generating randomness at a certain age (25). The results correspond to what it was suspected, that cognitive abilities peak at an early age before declining and that no other factor was important.

We quantified a type of mathematical randomness that is known to be the true type of randomness as opposed to e.g. ‘statistical randomness’. Something that is random is difficult to describe in a succinct way. Unlike ‘statistical randomness’, ‘algorithmic randomness’ does not only produce something that appears random but also that is very difficult to generate or produce. Conversely, something that may look random for the standard of statistical randomness may not turn out to be truly random. What should readers take away from your report?

Response: No formal approach had been able to quantify and reveal all this, especially over the length of the tests applied to a large number of people. More surprising, this puts everyone in the same situation, having not found any other factor determining this particular ability, neither gender, language spoken, beliefs or education level (or type) had any impact other than age.

The decline may have several factors, one of which may not only be a biological one—at least not at the beginning—but the consequence of a greater knowledge of the structured nature of the world.

Our results show that we are less bad at behaving randomly at a certain age and that it seems that more brain power is used and needed to generate true randomness, peaking at 25 years old. In other words, there is some form of ‘more sophisticated computation’ that the mind performs in order to generate ‘better randomness’, the more exceptional the more random. What recommendations do you have for future research as a result of this study?

Response: We think this knowledge and novel tools can be taken further to study how diseases can be profiled and can change the way in which we perceive and generate randomness connected to differences in cognitive capacities in cases of neurological disorders. Is there anything else you would like to add?

Response: The research is a kind of reverse Turing test where humans are required to behave like computer programs trying to simulate the best possible randomness. What we found is that humans can best outsmart computers when they are 25. Thank you for your contribution to the community.


Human behavioral complexity peaks at age 25

Nicolas Gauvrit,Hector Zenil ,Fernando Soler-Toscano,Jean-Paul Delahaye,Peter Brugger

PLOS Published: April 13, 2017

Note: Content is Not intended as medical advice. Please consult your health care provider regarding your specific medical condition and questions.

More Medical Research Interviews on



Leave a Reply

Your email address will not be published.