Separating neural processes using mixed event-related and epoch-based fMRI paradigms

Abstract

Mixed event-related and epoch-based paradigms (mixed designs) have recently been introduced as viable experimental designs for fMRI studies. To date, mixed designs have been used only to evaluate transient and state changes that are associated with single event and blocked conditions, respectively. However, no study exists that demonstrates that the regional activity associated with two distinct processes with known activation patterns can be adequately separated. Rapid event-related experimental designs use the principles of linear summation and can identify multiple distinct processes that are not overlapping in time through linear regression. Events and epochs that are also separated in time can be easily distinguished using the same general linear model (GLM) and a multiple regression analysis. However, it may not be so obvious that events that are superimposed in time onto epochs can also theoretically be separated using this technique. We used the principles of linear summation and the GLM to demonstrate that regional activity from a mixed design with visual epochs and auditory events superimposed in time can in fact be separated. In addition, the flexibility of mixed designs is explored using a paradigm that includes events throughout the on and off blocks of the epoch-based paradigm. Mixed designs provide neuroscientists with the power and flexibility to address more complex cognitive hypotheses that were otherwise difficult to study with either epoch-based or event-related designs alone.