gate the impact of these strategies). To the extent that there may be greater continuity in the environments in which older adults live, it may be easier for them to develop habits. However, little is known about the process by which repeated behaviors become habits. My colleagues and I have suggested that what differentiates habits from behaviors that are maintained over time is that habits are behaviors for which people no longer feel any need to question their value (Rothman et al., 2004). For example, seat belt use may readily become a habit because people feel little need to continually reassess its worth. On the other hand, it may be difficult for a set of dietary behaviors to become a habit as people will continually reassess whether they are satisfied with the outcomes of the behavior. To the extent that habits are regulated by stored representations of the behavior, innovations in how people form and change implicit or automatically activated attitudes (Wilson, Lindsey, and Schooler, 2002) may provide critical insights into the design of initiatives that can create and sustain healthy habits.
If satisfaction proves to be a critical determinant of whether a new pattern of behavior is maintained, investigators need to formulate a better understanding of the factors that cause people to feel more or less satisfied with their actions. This is likely to be a daunting task as one would expect assessments of satisfaction to reflect not only the set of concerns highlighted by the behavioral domain but also the set of priorities specified by the individual. For example, people who want to lose weight in order to improve their social life will focus their assessment of their actions on a different set of criteria than will those who want to lose weight in order to be more physically active.
In order to apply this model to the behavioral practices of older adults, investigators need to develop a rich understanding of what older adults are likely to attend to when assessing the value of their behavior. According to socioemotional selectivity theory (Carstensen et al., 1999), older adults would be expected to base their evaluation on a behavior’s ability to address or meet their emotional needs. To the extent that older adults focus their attention to a greater degree on their present psychological needs, they may find it particularly difficult to forestall an assessment of a behavior’s value. This would lead to the prediction that older adults would have a particularly difficult time sustaining behaviors for which the initial costs are high and the focal benefits are delayed (e.g., smoking cessation). From the perspective of an intervention, older adults would therefore be likely to benefit from activities that both highlight any immediate benefits and heighten the critical long-term benefits afforded by the new behavior.
Older adults’ ability to evaluate the consequences of their actions may also be constrained by cognitive impairments that can cause uncertainty as to whether the behavior has been performed (Purdie and McCrindle, 2002). If people fail to recognize that they have forgotten to perform a behavior