Can Large Language Models (LLMs) Describe Pictures Like Children? A Comparative Corpus Study

TL;DR

This study compares how well large language models generate child-like descriptions of pictures, revealing significant differences in lexical richness, semantic similarity, and linguistic patterns compared to actual children's language.

Contribution

It provides a detailed psycholinguistic analysis of LLM-generated child-like language and highlights limitations in their ability to replicate authentic child speech patterns.

Findings

LLMs produce longer but less lexically rich texts.

LLMs rely more on high-frequency words and under-represent nouns.

Semantic similarity between LLM texts and children's descriptions is low.

Abstract

The role of large language models (LLMs) in education is increasing, yet little attention has been paid to whether LLM-generated text resembles child language. This study evaluates how LLMs replicate child-like language by comparing LLM-generated texts to a collection of German children's descriptions of picture stories. We generated two LLM-based corpora using the same picture stories and two prompt types: zero-shot and few-shot prompts specifying a general age from the children corpus. We conducted a comparative analysis across psycholinguistic text properties, including word frequency, lexical richness, sentence and word length, part-of-speech tags, and semantic similarity with word embeddings. The results show that LLM-generated texts are longer but less lexically rich, rely more on high-frequency words, and under-represent nouns. Semantic vector space analysis revealed low similarity, highlighting differences between the two corpora on the level of corpus semantics. Few-shot prompt increased similarities between children and LLM text to a minor extent, but still failed to replicate lexical and semantic patterns. The findings contribute to our understanding of how LLMs approximate child language through multimodal prompting (text + image) and give insights into their use in psycholinguistic research and education while raising important questions about the appropriateness of LLM-generated language in child-directed educational tools.