James M. Hitt II, Mustapha Mouloua, and Dennis A. Vincenzi Center for Applied Human Factors in Aviation University of Central Florida
As automated systems continue to saturate our daily activities we as human factors specialists and design engineers must be concerned with the issues that may inhibit or promote the use of an automated system. Automated systems are found in many facets of daily life. It is difficult for a person not to see the many daily uses of automated systems. For example, automated teller machines (ATM's) can be used for deposits or withdrawals, transferring funds between accounts, purchasing stamps, making account inquiries and some ATM's can check current exchange prices. Aircraft can now take off, cruise, and land with the use of automated systems. Word processing software can now automatically correct spelling and grammar mistakes. Braking systems for automobiles can now detect the exact amount of pressure to put on the pads to stop the automobile without locking the brakes (often causing an uncontrolled spin). But our point here is not to belabor the existence of automation but to further understand the important design issues for automated systems. One important design issue is how does the age of the system user effect performance.
Several studies have addressed the issue of aging and its effects on human performance in automated systems. Research by Hardy, Mouloua, Molloy, Dwivedi, and Parasuraman ( 1995) examined age differences under dual task conditions using the MAT battery. Their results showed an equal degradation in performance on the two tasks between the young and elder participants. Another study by Hartley ( 1992), revealed that in dual task conditions there is a greater level of divided-attention cost for older adults when compared to younger adults. These studies report varied empirical evidence for an age effect when examining performance in automated systems.
Other studies have addressed such issues as how older adults view automation use. Issues such as reliance ( Riley, 1995), trust ( Muir, 1987), and safety have been cited as important deign considerations for automated systems. Research examining attitudes towards automation use has been conducted on a limited sample size ( Sams, Sierra, Sahagian, Nichols, & Mouloua, 1997). Using a composite score from the Complacency-Potential Rating Scale (CPRS) these researchers revealed a significant difference between age groups with younger adults having a higher level of positive attitudes towards automated systems than older adults. Other studies have examined ATM use by older adults and in several cases the elderly population declared the desire for training programs to further understand the uses of the automated systems. All of these studies have taken different paths to examine possible age related differences when humans interact with automated systems. The purpose of this study is to determine if differences among any of the five components of the CPRS (overall automation, reliance, trust, safety, and confidence) are attributed to age differences and/or exposure to various levels of automation (single, dual, or multi-task). We hypothesize that the CPRS scores will be lower for the elderly population and as the task load increases (more manual control), the scores of CPRS will decrease.