Modeling language contact with the Iterated Learning Model

Seth Bullock,Conor Houghton

2024-08-26

Abstract:Contact between languages has the potential to transmit vocabulary and other language features; however, this does not always happen. Here, an iterated learning model is used to examine, in a simple way, the resistance of languages to change during language contact. Iterated learning models are agent-based models of language change, they demonstrate that languages that are expressive and compositional arise spontaneously as a consequence of a language transmission bottleneck. A recently introduced type of iterated learning model, the Semi-Supervised ILM is used to simulate language contact. These simulations do not include many of the complex factors involved in language contact and do not model a population of speakers; nonetheless the model demonstrates that the dynamics which lead languages in the model to spontaneously become expressive and compositional, also cause a language to maintain its core traits even after mixing with another language.

Computation and Language

What problem does this paper attempt to address?

The paper aims to explore the issue of language stability during language contact. Specifically, the researchers used an Iterated Learning Model (ILM) to simulate what happens when two mature languages, which already possess expressiveness and compositionality, come into contact. By introducing a Semi-Supervised Iterated Learning Model (Semi-Supervised ILM), the researchers found that even in the absence of explicit population factors, languages can still maintain their core characteristics during the contact process. The main conclusions of the study include: 1. When both languages contribute equally to the input for novice learners, the resulting language often does not completely resemble either of the original languages. 2. If one of the languages dominates the input, that language or a language very similar to it usually re-emerges after a period of language instability. 3. The same mechanisms that lead to the development of expressiveness and compositionality in languages can also stabilize languages during contact. Through these simulation experiments, the researchers demonstrated that even simple models can reveal some fundamental dynamics of the language contact process and provide a theoretical foundation for further research.

Modeling language contact with the Iterated Learning Model

An iterated learning model of language change that mixes supervised and unsupervised learning

Bias Amplification in Language Model Evolution: An Iterated Learning Perspective

Spatial community structure impedes language amalgamation in a population-based iterated learning model

An Ising-like model for language evolution

Computational approach to the emergence and evolution of language - evolutionary naming game model

Modelling Language Evolution: Examples and Predictions.

Language Evolution with Deep Learning

A Simulative Study of the Roles of Cultural Transmission in Language Evolution

Searching for Structure: Investigating Emergent Communication with Large Language Models

Computational Modeling on Language Emergence: A Coevolution Model of Lexicon, Syntax and Social Structure *

Mathematical Modeling of Learning from an Inconsistent Source: A Nonlinear Approach

When LLMs Play the Telephone Game: Cumulative Changes and Attractors in Iterated Cultural Transmissions

Iterated Learning Improves Compositionality in Large Vision-Language Models

A bifurcation threshold for contact-induced language change

On an Interaction Model of General Language Change

Cultural evolution in populations of Large Language Models

Large Language Models: A Historical and Sociocultural Perspective

Imitation and Large Language Models

Computational exploration on language emergence and cultural dissemination

When Learners Surpass their Sources: Mathematical Modeling of Learning from an Inconsistent Source