Ever since its earliest years, information theory has enjoyed both a promising and complicated relationship with the social sciences… time to take inventory…
Overview of a promising and complicated history
Shannon himself applied his “mathematical theory of communication” to human communication early on, allegedly having his wife Betty estimating word probabilities to calculate the first approximation of the entropy of English. Five years later, he then warned of a “scientific bandwagon”, saying that information theory “is not necessarily relevant to such fields as psychology, economics, and other social sciences”. He added that he personally still believed “that many of the concepts of information theory will prove useful in these other fields—and, indeed, some results are already quite promising—but the establishing of such applications is not a trivial matter of translating words to a new domain, but rather the slow, tedious process of hypothesis and experimental verification”.
It turned out that in the social sciences, this translation process was indeed slow and tedious. After Miller’s famous “magical number seven” paper and Attneave’s groundbreaking “Applications of Information Theory to Psychology”, the emerging field of Communication adopted information theoretic concepts to study group decision-making, relational control in relationships, mass communication, and talk and silence sequences in conversations. A main historical account for why this work was discontinued was that “gathering everyday conversations … is nearly impossible … unless one carries a tape recorder around all day (a cumbersome and hardly practical endeavor)”. Additional culprits are the “adoption of approaches from other fields such as psychology that do not emphasize process as much as communication”, “the perceived scope of effort required from the researcher”, and those dynamics were “simply impractical to compute” before today’s computing power.
While these rather practical and computational limitations have been overcome in recent years due to the “big data” flood and omnipresent cloud computing, unfortunately, there have also been conceptual reservations to the applicability of information theory to the social sciences, especially during the 1990s and early 2000s. It was argued that information theory was supposedly “inappropriate to represent social processes” as it was allegedly a “misleading misrepresentation of the nature of human communication”. It is striking that all of these critics refer to Shannon’s channel logic of communication as the “Shannon–Weaver model”. In 1949, in his role as science advocate, Warren Weaver asked Shannon to reprint his two-part paper from 1948 in book format. He added a 28-page introduction for the 144-page book and changed the title from “A Mathematical Theory…” to “The Mathematical Theory…”. Weaver sees his introduction as “an interpretation of mathematical papers by Dr. Claude E. Shannon”. and not as an original contribution. Given this consistent misattribution of credits, it is questionable how familiar these critical social scientists indeed were with Shannon’s comprehensive framework and what became of it during the subsequent decades.
Reaching the year 2020, the increase in human interactions taking place in digital environments has led to a refound fascination with applying information theory in the social sciences. The new abundance of behavioral “big data” and our computational resources allow researchers to even calculate measures that converge rather slowly, while, at the same time, the maturation of the social sciences has led to an increased interest in more sophisticated nonlinear methods and measures.
This Special Issue compiles 11 creative research articles on innovative uses of information theory and its extensions to better understand human behavior and social processes. The articles in this Special Issue are proof of the abundant opportunities offered by information theory to better understand the nature of humans and its societal systems and dynamics…
https://www.mdpi.com/1099-4300/23/1/9/htm
For example, check out:
Hilbert, M., & Darmon, D. (2020). How Complexity and Uncertainty Grew with Algorithmic Trading. Entropy, 22(5), 499. https://doi.org/10.3390/e22050499
