Motivation in Ethnography and Previous Research

Social engagement has profound health benefits, protecting against illnesses from the common cold to dementia [7, 8]. Our ethnographic inquiry highlighted barriers to social engagement in later life [10]. With retirement, many people lose opportunities for spontaneous contact as well as visibility of others' availability. Concerned about imposing on others or being rejected, many drift into isolation. These situational shifts, along with cognitive changes, such as difficulty recalling names or following rapid conversation, make many feel helpless about loneliness. As one formerly very social man put it "loneliness is a part of old age and there ain't a damn thing you can do about it."

We also noted significant variability in the way older adults described their experiences of loneliness. Current measures and interventions underestimate this variability and rely on characterological rather than situational explanations for differences in social engagement. But our interviews suggest that, like blood pressure or glucose levels, loneliness varies by the moment and therefore requires adaptive solutions.

These themes of helplessness and variability are illustrated in examples below:

• After the death of her husband, an 82-year-old woman moved across country to be near her daughter. She left behind a network of friends. She is socially gracious and very charming but uncomfortable initiating contact. She loves seeing family, especially her grandchildren, but feared imposing. A history of depression makes isolation a particularly serious risk.
• A 77-year-old divorced woman enjoys casual interaction with neighbors, choir practice, and creative writing during the week but dreads the weekends: "I just wish I could make them disappear." She retained an expectation that weekends should be spent with family and suffered chronic disappointment because her children were typically unavailable on weekends.
• An extremely bright and extroverted 91-year-old woman relocated to live near her son following an injury. She now struggles with quiet evenings in a retirement community. She fills her days with group activities and enjoys the weekends she spends with her son's family, but complained that, in the evenings "It's like a morgue around here."

Psychological research has demonstrated the benefits of mindfulness, or awareness of change. When people recognize fluctuations in negative circumstances, particularly their own ability to bring about positive change, they feel less helpless [9]. In this research, we tried to foster elders' social self-efficacy and empowerment by highlighting the dynamic qualities of social interaction.

The Research Prototype

We developed a platform of sensors and feedback displays to measure and encourage social engagement, for the prevention of cognitive decline. In the primary feedback display (Figure 4), friends and family rotate around the elder—planets which can be pulled in by a phone call or visit. This use of social networks as health feedback displays is described in [10].

Figure 4: Solar display of social activity. The elder, depicted as the sun in the center of the display, is surrounded by planetary representations of friends and family. Social interaction, measured by sensors, draws planets closer to the center.

click image for larger view
 

The Components of Embedded Assessment

Monitoring. A sensor network and online journal allowed continuous measurement of phone activity and other social interaction over several months. We first determined a baseline of social interaction (by monitoring only) and then measured changes in social interaction associated with the introduction of feedback displays.

Compensation. To encourage and reinforce social efforts, the displays provided real-time feedback of interactions with friends and family. It was important that the displays be intuitive and nonstigmatizing. They included the solar view shown in Figure 4, longitudinal graphs, visual cues of callers' names and faces, and a lamp that signaled the availability of a family member. Feedback was intended to help people compensate for isolated lifestyles and for cognitive changes in recall and processing of social information.

Prevention. This system was intended to protect against onset and progression of cognitive decline, by motivating social outreach and mitigating feelings of loneliness.

Responses to Solar Displays

A three-month-long in-home pilot study indicated that the feedback displays were valued by elders and their caregivers (see [10] for fuller description). We observed subtle and overt increases in social engagement. These behavioral changes not only improved the quality of life for elders and families, but they also set in motion a style of interacting that may protect against a range of illnesses in later life.

Revisiting case examples above:

• The 82-year-old woman who didn't want to burden her family started initiating plans, especially with her grandchildren. She also began volunteering as a teacher's aid for first-grade students. This activity was very gratifying for her, in part because of the positive impact she had on others. Her daughter exclaimed, "The kids love her ... she's in the yearbook and they've asked her to continue helping the same class next year."
• The 77-year-old woman who dreaded the weekends became more involved with friends. For example, she started seeing some members of her choir group outside formal practice sessions. These outings and those with other friends shifted her attention away from the unavailability of her children.
• The 91-year-old woman who spends weekends with her son remains very close with her family. But she also started reaching out to peers in the retirement community with a new level of interest. She even invited some neighbors into her apartment.

Observations from this study have implications for the model of embedded assessment. First, the feedback displays raised participants' enthusiasm for monitoring technologies that they initially experienced as intrusive and burdensome. Their enjoyment of the displays appeared to generalize to the monitoring: several participants even started to speak of the online journaling as a hobby. This observation provides support for the premise in embedded assessment that feedback will motivate monitoring. Second, elders and caregivers recognized opportunities in the feedback displays to catch early trends of isolation. This eagerness to catch early indicators, central to embedded assessment, most likely occurred because the visual displays invited people to objectively examine and discuss topics that were previously avoided or overlooked. Finally, we observed that the adult children who participated in the study as caregivers used the displays as an opportunity to reflect upon their own lives and set priorities pertaining to their health, professional pursuits, hobbies, and relationships. This finding supports the idea that embedded assessment technologies should be adopted in midlife to support early detection: tools used to help an elderly parent can simultaneously offer services to help a caregiver manage his or her own life. Baseline data can be gathered from implicit variables, such as the caregiver's typing speed or voice quality, as well as explicit entries in care-giving and self-health applications.