LLMs and the ZPD

TL;DR

This paper explores how large language models (LLMs) function similarly to Vygotsky's Zone of Proximal Development, emphasizing interaction and cognitive tools over hallucination or simple representations.

Contribution

It proposes viewing LLMs as engaging in primitive thinking through practices, highlighting interaction as central to human-like understanding.

Findings

LLMs are better understood as practicing primitive thinking rather than hallucinating.

Interaction is fundamental to human communication and cognition.

A new perspective on LLMs focusing on cognitive tools and practices.

Abstract

One hundred years ago Vygotsky and his circle were exploring the nature of consciousness and defining what would become psychology in the Soviet Union. They concluded that children develop "scientific thinking" through interacting with enculturated adults in Zones of Proximal Development or ZPDs. The proposal is that, contrary to the claims of some, the LLM mechanism is not doing thinking with "distributed representations," but rather the completion model is doing "primitive thinking" in terms of *practices*. Viewed from this perspective, it would seem our large language models don't hallucinate, but rather dream, and that what is needed is not "guard rails" but an investigation of the set of cognitive tools that enable us to do things that look like common-sense. The proposal here is that *interaction* is core to human communication rather than just an add-on to "real" understanding.