Categories
Uncategorized

Small world of phonological words

It has been very fascinating to me how quickly infants learn a language and the importance of the processing of sounds in the adoption of a language. For that very reason, I am talking about a research study that examines the organization of words in the lexicon and how that may influence the acquisition and retrieval of phonologically similar words. Two words are phonologically similar if we can get one word by replacing, deleting, or inserting a phoneme into another word. For example, “hat” is a phonological neighbour of “cat”. In this study, phonological transcriptions of approximately 20 thousand words were added from the 1964 Merriam-Webster Pocket Dictionary. A node in the network corresponded to a phonological word from the dictionary and a link was present between two words if they are phonologically similar.

There were ten thousand “lexical hermits” in the network – basically nodes that do not have any neighbours; two thousand nodes with a few neighbours but isolated from the largest connected component in the phonological network. The largest connected component contained six thousand nodes. The average path length between any two nodes in the network was 6.05 comparable to a similarly sized random network – it was higher for the largest connected component at 9.105. The clustering coefficient for the nodes in the phonological network was 0.126 which shows neighbours of a word have a higher tendency to be connected with each other and it influences the spoken word recognition.

An external file that holds a picture, illustration, etc.
Object name is nihms66415f1.jpg
Degree distribution for the largest connected component of the phonological network

The average path length between phonological words and high clustering shows the small-world characteristics proposed by Watts & Strogatz in this phonological network. The small world characteristics of the phonological network mean less processing required to traverse through the words in the network and could potentially answer why lexical processing is so fast and accurate. This study attempts to explore the structure of the phonological network and provide insights about mental lexicon and lead to future studies that further dive into language understanding and processing.

Sources:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2535910/

https://www.verywellfamily.com/how-do-children-learn-language-1449116

Leave a Reply