ABSTRACT
The learning mechanisms (Figure 13.1) mainly include hierarchical tokenization and recursive patternization. We recall how we combine words into meaningful phrases, phrases into sentences, sentences into paragraphs, paragraphs into chapters, and chapters into a book. Hierarchical tokenization is like concept-embedment, i.e., using learned concepts (high-level tokens) to replace combinations of tokens, aiming at shortening event-strings for better understanding. Like language grammar, Patternization is the use of rules to describe the structural commonalities among multiple event-strings. Overview of the learning mechanisms A flowchart provides a simple overview of the learning mechanisms that contribute to dealing with novelty, similarity-learning with imitation and innovation, and frequency and distributive rewards. https://www.w3.org/1999/xlink"/> https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781003392422/09e8e534-13e8-48fd-8df1-6c217f5200c1/content/fig13_1.tif" xmlns:xlink="https://www.w3.org/1999/xlink"/>
