More problems after difficult problems? Behavioral and electrophysiological evidence for sequential difficulty effects in mental arithmetic

This study investigated whether sequential difficulty effects emerge during processing of a mixed set of small, easy and large, more difficult arithmetic problems. Furthermore, we assessed if these sequential difficulty effects are reflected in event-related (de-)synchronization (ERS/ERD) patterns. To this end, we analyzed data of 65 participants, who solved two separate blocks (additions and subtractions) of arithmetic problems while their EEG was recorded. In each block, half of the problems were difficult problems (two-digit/two-digit with carry/borrow), and the other half were easy problems (one-digit/one-digit). Half of the problems were preceded by a problem of the same difficulty (repeat trials), and half were preceded by problems of the other difficulty (switch trials). In subtractions a sequential difficulty effects pattern emerged. Participants solved easy repeat trials faster than easy switch trials, while difficult repeat trials were solved slower and less accurately than difficult switch trials. In the EEG, we found the strongest effects in left hemispheric beta band (13–30 Hz) ERD. Specifically, participants showed a stronger beta band ERD in easy switch trials than in easy repeat trials. Furthermore, beta band ERD was stronger in difficult problems than in easy problems within repeat trials, but stronger in easy problems than in difficult problems within switch trials. In summary, our results are in line with the presence of sequential difficulty effects, as processing of easy and difficult problems was impaired if they were preceded by a difficult problem. Furthermore, these sequential difficulty effects are reflected in ERD patterns.