In this commentary I take up thoughts about the sociology of time and Luhmann’s theory to consider whether automated future option maximization can be extended with a social dimension.

In order to capitalize on the self-referentiality of interpretation, as marked by Gasparyan, I suggest considering semiosis - the process of signification - as a systemic consequence of interacting….

Context: By proposing to regard objects as “tokens for eigenbehavior,” von Foerster’s seminal paper opposes the intuitive subject-object dualism of traditional philosophy, which considers objects to be instances of an external world Problem: We argue that this proposal has two implications, one for epistemology and one for the demarcation between the natural sciences and the humanities. Method: Our arguments are based on insights gained in computational models and from reviewing the contributions to this special issue. Results: Epistemologically, von Foerster’s proposal (...) suggests that what is called “reality” could be seen as an ensemble of eigenforms generated by the eigenbehavior that arises in the interaction of multiple dynamics. Regarding science, the contributions to this special issue demonstrate that the concept of eigenbehavior can be applied to a variety of disciplines from the formal and natural sciences to the humanities. Its universal applicability provides a strong argument for transdisciplinarity, and its emphasis on the observer points in the direction of an observer-inclusive science. Implications: Thinking in eigenbehavior may not only have implications for tearing down the barriers between sciences and humanities, a better understanding of eigenbehaviors may also have profound effects on our understanding of ourselves. This also opens the way to innovative behavior design/modification technologies. (shrink)

: In the commentaries, a number of intriguing projections of Future State Maximization on a multitude of different systems are suggested, which raised the need to clarify certain ….

Context: Many AI and machine-learning techniques are primarily focused on past-to-future extrapolations of statistical regularities in large amounts of data. We introduce a method that builds on an….