Dense Array EEG Neural Markers of Learning

Over the past decade, with research funded by the National Institutes of Health (NIH) and the Office of Naval Research, the EGI Science Team has employed dense array EEG to investigate the neural mechanisms that index the progression of learning, going from no knowledge of appropriate action in response to a given stimulus to full expertise. The hope is that with knowledge of the progressive changes, at a network level, associated with learning, this information can be used to guide development of training curricula. Looking forward, we see the possibility of using the learning related signals themselves to adaptively guide the learning process.

We have found (published in Luu et al., 2007; Luu et al., 2009, Luu et al., 2011) that learning progression is indexed by several brain process that appear to be responsible for the tracking of feedback information, stimulus-response mapping, and working-memory. Interestingly, there also exists a marker that seems to predict the ability of the learner to generalize their training to a similar but yet new learning task (Luu et al., 2011).

Neural Markers of Learning

Figure 1. Topographic maps of the medial frontal negativity (MFN, top) and time course of the MFN at a midline electrode (bottom, highlighted by yellow box). PreStim1 = Pre learning for in session 1 1; CorPostStim1 = Correct Performance Post Learning in session 1; PreStim4 = Pre Learning for session 4 (new but similar task to Session 1); CorPostStim14 = Correct Performance Post Learning in session 4.

More recently, we examined in more detail the working-memory component associated with such learning tasks. We found that working memory is controlled by several distinct networks, each involved in slightly different processes (e.g., attentional control) with unique time courses (see Figure 2, Luu et al., 2014).

Dense Array EEG Neural Markers 

Figure 2. Time courses visual cortex network (C1), lateral and medial frontal pole network (C7), and temporo-parieto-occipital junction network (C9) in working memory.


Luu, P., Caggiano, D. M., Geyer, A., Lewis, J., Cohn, J., & Tucker, D. M. (2014). Time-course of cortical networks involved in working memory. Front Hum Neurosci, 8, 4. doi: 10.3389/fnhum.2014.00004

Luu, P., Jiang, Z., Poulsen, C., Mattson, C., Smith, A., & Tucker, D. M. (2011). Learning and the development of contexts for action. Front Hum Neurosci, 5, 159.

Luu, P., Shane, M., Pratt, N. L., & Tucker, D. M. (2009). Corticolimbic mechanisms in the control of trial and error learning. Brain Research, 1247, 100-113. 

Luu, P., Tucker, D. M., & Stripling, R. (2007). Neural mechanisms for learning actions in context. Brain Research, 1179, 89-105.