Individual Submission Summary
Share...

Direct link:

Word Learning Differences in Children with Specific Language Impairment

Thu, March 21, 4:00 to 5:30pm, Hilton Baltimore, Floor: Level 2, Key 2

Integrative Statement

Introduction
Children with specific language impairment (SLI) have known deficits in word learning but what is missing is an understanding of why word learning is difficult for these children. Here we utilize ERPs and time frequency analysis (TFA) of the EEG to investigate the neural processes underlying word learning in children with SLI compared to typical language (TL) peers.

Hypotheses
For the ERP analysis, we predicted that, for both children with SLI and TL peers, the N400 would be a sensitive index of the increasingly detailed semantic representations of words over the course of learning. We also predict differences in the N400 effect between the two groups, suggesting that the TL group is better able to draw on semantic processing during word learning.
For the TFA analysis, we predicted changes in theta (4-8 Hz) and alpha (8-12 Hz) as a function of learning for both groups but that the SLI group would engage the two frequencies at different times during word learning compared to the TL group. In language processing research, theta is generally associated with semantic retrieval and alpha with attention/inhibition.

Participants
Here we report data from 14 8-10 year old children, 7 with SLI and 7 age-equivalent TL controls. Groups were equivalent on nonverbal IQ (WISC NVIQ) but the SLI group performed significantly lower on measures of receptive and expressive vocabulary knowledge (PPVT, EVT) and language comprehension and use (syntax, semantics; CELF).

Methods
All participants completed a word learning from context task in which they heard sentence triplets that either established a meaning for the target nonsense word (Meaning Plus condition) or provided a control for repetition of the nonsense word without providing a meaning (Meaning Minus condition). Analyses focused on N400, theta and alpha changes across the sentence triplet, time locked to the nonsense word onset.

Results
For the ERP analysis, visual inspection of the Meaning Plus condition indicates that the two groups showed similar N400 amplitude increases from sentence 1 to sentence 2 across centro-parietal sites but that the SLI group had less of an amplitude increase from sentence 2 to 3 than the TL group (Figure 1). This suggests decreased semantic processing for the SLI group during word learning. The TFA (Figure 2) revealed theta synchrony in sentence 1 of the triplet followed by desynchrony in sentences 2 and 3 for the TL group. The SLI group also showed theta synchrony in sentence 1 and 3, but desynchrony occurred in sentence 2. For alpha, the TL group showed synchrony in sentence 2 and desynchrony in sentence 3. The SLI group showed the same pattern of alpha but at different sentences - alpha synchrony occurred in sentence 1 followed by desynchrony in sentences 2 and 3. This pattern suggests group differences in when semantic retrieval and attention/inhibition are engaged during word learning.

Author