Responsive image
Responsive image

Models

Lain Model (github)

The Lain Model implements a theoretical model that couches the acquisition of vowel normalization in terms of an infant's generation and alignment of cognitive models of the self and others developed in my dissertation. The theoretical model is situated within a fuller conceptual framework sketched in Plummer and Beckman (2015).

The Lain Model implementation focuses on formulating basic aspects of the theoretical model, including the vocal interaction between caretaker and infant agents, how each agent represents this interaction, and how these representations structure phonological acquisition. The implementation is divided into

  • a base model, consisting of a core set of functions that implement the computations central to the modeling framework, and
  • a set of modules that provide task-specific extensions to the base model (e.g., demos, models of sensory space alignment, models of dyadic vocal exchange, etc.).
The base model and a demos-module are currently available on github. The remaining modules that make up the acquisition model will be made available shortly.

On the name: I was told that I needed to give the model a name, so I chose an homage to the main character in the late 90's anime Serial Experiments Lain, which was about identity and computation, but a sciencey retcon might be something like Laplacian infant Model, OWHY.

Currently under development

FLAMES Model (concept)

Super-intelligent cyborg kitty model