Exploring the use of artificial language learning (ALL) paradigms to differentiate intrinsic versus extrinsic sources of language impairment

Abstract

Purpose

The current investigation aimed to explore the potential of Artificial Language Learning (ALL) paradigms to distinguish between poor language abilities due to intrinsic difficulties in language learning as opposed to difficulties stemming from extrinsic factors such as environmental factors.

Method

Using a 2-step population-based screening procedure, 35 children from mid-high SES with presumed developmental language disorders (pDLD) were compared to an unselected sample of 33 children from low SES on measures of accumulated lexical and morphological knowledge and on tasks assessing lexical and morphological learning ability using artificial language learning (ALL) paradigms.

Results

No significant differences between the pDLD and the low-SES groups were found on lexical and morphological knowledge. However, the low-SES group outperformed the pDLD group in both learning an artificial morphological rule and learning new pseudowords.

Conclusions

These findings indicate that children from low SES, as a group, exhibit diminished language performance similar to that of children with pDLD when tested on conventional measures of accumulated linguistic knowledge yet outperform children with (persumed) DLD when assessed on their ability to learn and make use of new linguistic information to extract morphological rules and establish novel lexical representations. The potential of ALL paradigms in this population is discussed.