The inference of connectivity in brain networks has typically been performed using passive measurements of ongoing activity across recording sites. Passive measures of connectivity are harder to interpret, however, in terms of causality - how evoked activity in one region might induce activity in another. To obviate this issue, recent work has proposed the use of active stimulation in conjunction with network estimation. By actively stimulating the network, more accurate information can be gleaned regarding evoked connectivity. The assumption in these previous works, however, was that the underlying networks were static and do not change in time. Such an assumption may be limiting in situations of clinical relevance, where the introduction of a drug or of brain pathology, might change the underlying networks structure. Here, an extension of the evoked connectivity paradigm is introduced that enables tracking networks that change in time.
展开▼
