Abstract
Humans are often blamed for errors in complex systems. A number of taxonomies exist for classifying human errors, but recent doubts have been raised as to whether these are actually errors or simply processes involved in both correct and incorrect decisions. A field test was designed to address some of these issues. During the field test, three geologists made assessments about the recent and past environment based on robotically collected data. The decisions were classified based on previous error taxonomy. Each conclusion was then checked for accuracy. From this data, the error rate for specific behaviors could be calculated. The results indicate many of the behaviors describe earlier as erroneous actually produce correct decisions most of the time. One behavior is singled out as producing a majority of the errors and further research is needed to better understand its causes.
Get full access to this article
View all access options for this article.
