One possible way to address the dissociation between the behavior pattern and ERP activation, as discussed above, would be a finer grain analysis of the data separated across the three 20-trial blocks of the BART, as was done with the behavioral data. Unfortunately, a final shortcoming of the present study is that the number of trials in our paradigm was not sufficient to calculate ERPs for each block separately. The ERP data across all balloons already contained substantial noise levels and notorious inter- and intravariability. A finer grain analysis yielded to few data segments, which resulted in even higher noise levels. Consequently, the ERP data needed to be collapsed across the entire task to yield useful data. Future studies may further clarify this issue and should include more trials to perform trial by trial analyses. This would shed light on the question whether FRN is indeed related to short-term adjustments, rather than long-term strategic changes, as hypothesized.
