The subject may be informed on each trial whether their response was correct or incorrect. [0418] In one example, subjects may be trained to assess the level of an activity metric, such as the level of activation of a particular brain region, without being able to see information about that metric.