# Feature-rich multiplex lexical networks reveal mental strategies of early language learning

arXiv: https://arxiv.org/abs/2201.05061 · HTML: https://mlsanity.com/paper/2201.05061/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 94 references that are in the map.
- First posted on arXiv 13 Jan 2022.

## Builds on (most cited first, 12 of 12)

- [Distributional Semantics and Linguistic Theory](https://mlsanity.com/paper/1905.01896/) (2019)
- [Multiplex lexical networks reveal patterns in early word acquisition in children](https://mlsanity.com/paper/1609.03207/) (2016)
- [MultiVERSE: a multiplex and multiplex-heterogeneous network embedding approach](https://mlsanity.com/paper/2008.10085/) (2020)
- [Complex systems: features, similarity and connectivity](https://mlsanity.com/paper/1606.05400/) (2016)
- [Multiscale mixing patterns in networks](https://mlsanity.com/paper/1708.01236/) (2017)
- [The origins of Zipf's meaning-frequency law](https://mlsanity.com/paper/1801.00168/) (2017)
- [Multiplex model of mental lexicon reveals explosive learning in humans](https://mlsanity.com/paper/1705.09731/) (2017)
- [Conformity: A Path-Aware Homophily Measure for Node-Attributed Networks](https://mlsanity.com/paper/2012.05195/) (2020)
- [The polysemy of the words that children learn over time](https://mlsanity.com/paper/1611.08807/) (2016)
- [Patterns in the English Language: Phonological Networks, Percolation and Assembly Models](https://mlsanity.com/paper/1410.4445/) (2014)
- [Layer entanglement in multiplex, temporal multiplex, and coupled multilayer networks](https://mlsanity.com/paper/2004.01534/) (2020)
- [K-clique percolation in free association networks. The mechanism behind the 7 ± 2  law ?](https://mlsanity.com/paper/2110.09317/) (2021)

## Built on by (most cited first, 6 of 6)

- [Cognitive network science reveals bias in GPT-3, ChatGPT, and GPT-4 mirroring math anxiety in high-school students](https://mlsanity.com/paper/2305.18320/) (2023)
- [From Perils to Possibilities: Understanding how Human (and AI) Biases affect Online Fora](https://mlsanity.com/paper/2403.14298/) (2024)
- [The role of System 1 and System 2 semantic memory structure in human and LLM biases](https://mlsanity.com/paper/2604.12816/) (2026)
- [Cognitive modelling with multilayer networks: Insights, advancements and future challenges](https://mlsanity.com/paper/2210.00500/) (2022)
- [Towards hypergraph cognitive networks as feature-rich models of knowledge](https://mlsanity.com/paper/2304.06375/) (2023)
- [Early Language Learning via Spreading Activation and Category Exploration in Complex Networks](https://mlsanity.com/paper/2607.06258/) (2026)

## Abstract (arXiv)

Knowledge in the human mind exhibits a dualistic vector/network nature. Modelling words as vectors is key to natural language processing, whereas networks of word associations can map the nature of semantic memory. We reconcile these paradigms - fragmented across linguistics, psychology and computer science - by introducing FEature-Rich MUltiplex LEXical (FERMULEX) networks. This novel framework merges structural similarities in networks and vector features of words, which can be combined or explored independently. Similarities model heterogenous word associations across semantic/syntactic/phonological aspects of knowledge. Words are enriched with multi-dimensional feature embeddings including frequency, age of acquisition, length and polysemy. These aspects enable unprecedented explorations of cognitive knowledge. Through CHILDES data, we use FERMULEX networks to model normative language acquisition by 1000 toddlers between 18 and 30 months. Similarities and embeddings capture word homophily via conformity, which measures assortative mixing via distance and features. Conformity unearths a language kernel of frequent/polysemous/short nouns and verbs key for basic sentence production, supporting recent evidence of children's syntactic constructs emerging at 30 months. This kernel is invisible to network core-detection and feature-only clustering: It emerges from the dual vector/network nature of words. Our quantitative analysis reveals two key strategies in early word learning. Modelling word acquisition as random walks on FERMULEX topology, we highlight non-uniform filling of communicative developmental inventories (CDIs). Conformity-based walkers lead to accurate (75%), precise (55%) and partially well-recalled (34%) predictions of early word learning in CDIs, providing quantitative support to previous empirical findings and developmental theories.