The Effect of Disruptions on Vigilance (2013)
Type of ContentTheses / Dissertations
Degree NameMaster of Science
PublisherUniversity of Canterbury. Psychology
AuthorsRoss, Hayden Aaronshow all
The goal habituation model of vigilance proposed by Ariga and Lleras (2011) posits that it is possible to attenuate the vigilance decrement (the decline in performance that occurs with time-on-task) through dis-habituation of a vigilance task’s goal. The goal in a vigilance task is to detect critical signals. Hence, a switch away from this goal should dishabituate the task goal. When a person resumes the vigilance task, the person’s performance should improve. Follow up studies to the Ariga and Lleras study have not found supporting evidence (Helton & Russell, 2011; 2012). The methods in these follow up studies differed from those of Ariga and Lleras in a number of ways. The present study attempts to replicate the original Ariga and Lleras (2011) work using methods that more closely follow the original study. The present research sought to uncover confounds in the original article through replicating the original task and manipulating variables in line with hypotheses made by attentional resource theory. Overall, the results of this research present a challenge to the goal habituation model. Rest breaks, not goal switching; lessened the magnitude of the decline in sensitivity with time on task, while task switching resulted in a temporary re-correction of increasingly conservative responding that occurred with greater time on task. We also found support for the context dependent relationship of task-unrelated thought, (TUT) and task demand. Given that the vigil was determined to impose relatively low workload, pre-task TUT was associated with average performance rather than post-task TUT. Finally, we discuss the dangers of over compliance with signal detection theory (SDT) measures. We explain that SDT sensitivity and bias measures are not independent given responding floor and ceiling effects during low demanding tasks such as the present. It is argued that this may have distorted the original conclusion arrived at by Ariga and Lleras (2011).