Developmental changes in visuo-spatial working memory in normally developing children: event-related potentials study

Working memory (WM) is very important for normal development. The fronto-parietal neuronal network supporting WM has already been well-studied. Less is known about the cortical activity changes during development of WM. We evaluated the maturation of visual WM network at the electrophysiological level in a group of normally developing children.

Multichannel (n=31) event-related potentials (ERP) were measured during a visuo-spatial backmatching task in 69 childrens (6-16 years old). One-backmatching (BM1) and two-backmatching (BM2) tasks were performed. Age-related changes in behavioral parameters (commission and omission errors and reaction times) and ERP parameters (peak amplitudes and latencies) were analyzed between different ages.

Clear improvement in performance from young childhood toward adolescence was seen at the behavioral level: decrease of errors and fastening of reaction times. At the electrophysiological level age-related changes were seen in peak latencies and especially in amplitudes. Different peaks have different dynamics in amplitudes and latencies: early peak amplitude decreased and latency shortened with age, which was not always seen in late peaks. This reflects developmental changes in intensity and speed of WM processing. Later peaks were more clearly seen over the right hemisphere in older children, illustrating hemispheric lateralization in visuo-spatial working memory.

Our results indicate that not only at the behavioral but also at the electrophysiological level clear age-related dynamics in WM processing can be seen. Furthermore, with ERP we showed that different WM components follow different developmental trajectories.

Our work demonstrates that age-related dynamics in intensity and speed of information processing during WM task is reflected in developmental changes in different ERP components. It also states that fronto-parietal visual WM network can be functional even before all its nodes are fully mature.