Working memory is important for a wide range of high-level cognitive activities. Previous studies have shown that the dorsal lateral prefrontal cortex (DLPFC) plays a critical role in working memory and that behavioral training of working memory can alter the activity of DLPFC. However, it is unclear whether the activation in the DLPFC can be self-regulated and whether any self-regulation can affect working memory behavior. The recently emerged real-time functional magnetic resonance imaging (rtfMRI) technique enables the individuals to acquire self-control of localized brain activation, potentially inducing desirable behavioral changes. In the present study, we employed the rtfMRI technique to train subjects to up-regulate the activation in the left DLPFC, which is linked to verbal working memory. After two rtfMRI training sessions, activation in the left DLPFC was significantly increased, whereas the control group that received sham feedback did not show any increase in DLPFC activation. Pre- and post-training behavioral tests indicated that performance of the digit span and letter memory task was significantly improved in the experimental group. Between-group comparison of behavioral changes showed that the increase of digit span in the experimental group was significantly greater than that in the control group. These findings provide preliminary evidence that working memory performance can be improved through learned regulation of activation in associated brain regions using rtfMRI.

Recent studies demonstrated that working memory could be improved by training. We recruited healthy adult participants and used adaptive running working memory training tasks with a double-blind design, combined with the event-related potentials (ERPs) approach, to explore the influence of updating function training on brain activity. Participants in the training group underwent training for 20 days. Compared with the control group, the training group's accuracy (ACC) in the two-back working memory task had no significant differences after training, but reaction time (RT) was reduced significantly. Besides, the amplitudes of N160 and P300 increased significantly whereas that of P200 decreased significantly. The results suggest that training could have improved the participants' capacity on both inhibitory and updating.

Response inhibition is crucial for mental and physical health but studies assessing the trainability of this type of inhibition are rare. Thirty-nine children aged 10-12 years and 46 adults aged 18-24 years were assigned to an adaptive go/no-go inhibition training condition or an active control condition. Transfer of training effects to performance on tasks assessing response inhibition, interference control, working memory updating, task-switching, and non-verbal fluid intelligence were assessed during 3- and 6-month follow-up sessions and/or an immediate post-training session. Significant training improvements and positive transfer effects to a similar response inhibition task with other stimuli were observed for both children and adults. Reliable albeit short-lived transfer effects were only found for the children, specifically to working memory updating and task switching. These results suggest some potential for response-inhibition training programs to enhance aspects of cognitive functioning in children but not adults.

Recent studies suggest that the response inhibition ability of children can be modified through training. Based on the notion of embodied cognition, we investigated transfer effects of a 7-day training program using a game named “Wesley says” in 8- to 12-year-old children (n = 15). The game consists of providing commands for performing simple body actions, the actual execution of which is conditional upon the preceding verbal expression “Wesley says.” Training effects were assessed with a computer-based visual go/no-go task and the Stroop color–word interference task. Relative to a control group playing other games mainly involving physical exercise (n = 15), the trained group showed a performance improvement on the go/no-go task, but not on the Stroop task. These results suggest the potential of an easy-to-use and ecologically valid training game to improve the inhibition capacity of children on related response inhibition tasks but not on tasks measuring other aspects of inhibition, such as interference control.