Journal of Aging Studies 34 (2015) 1–9

Contents lists available at ScienceDirect

Journal of Aging Studies journal homepage: www.elsevier.com/locate/jaging

Differences in perception of gerotranscendence behaviors between college students and community-dwelling older adults Jeffrey A. Buchanan ⁎, Duc Lai, Donald Ebel Minnesota State University, Mankato, USA

a r t i c l e

i n f o

Article history: Received 17 December 2014 Received in revised form 11 March 2015 Accepted 12 March 2015 Available online xxxx Keywords: Gerotranscendence Gerontological theory Aging

a b s t r a c t Within the field of gerontology, several different theories have attempted to explain common psychological and social changes associated with the aging process. The Theory of Gerotranscendence is one such theory which purports that a shift in meta-perspective from a more materialistic and pragmatic view of the world to a more cosmic and transcendent one occurs as we age. Corresponding with this shift in meta-perspective, the individual exhibits certain behaviors that could be mistaken as signs of psychopathology if viewed based on the assumptions of more culturally-assimilated theories of aging. The purpose of this study was to examine the difference in perception of gerotranscendence behaviors between college students and older adults. Perceptions were quantified using an instrument that described many behaviors indicative of gerotranscendence within the context of a written narrative depicting an older adult living in an assisted living facility. Respondents were then asked to rate these behaviors in terms of how unusual they were and how concerning they were. As hypothesized, results indicated that several behaviors indicative of gerotranscendence were rated as more concerning and unusual by college students compared to older adults. Implications of these findings in terms of interactions between younger and older individuals occurring in the community and within healthcare settings are discussed. © 2015 Elsevier Inc. All rights reserved.

Introduction

Prominent developmental theories of aging

Within the field of gerontology, significant effort has been devoted to understanding the fundamental nature of how people change psychologically as they age. This includes attempts to explain how elders' self-perceptions change over time, how elders' relationships with others and society may change, and what is the “optimal” amount and type of activity in which one should continue to engage. However, this process has not proven to be easy, as evidenced by the data-rich, explanation-poor status of the field (Achenbaum & Bengtson, 1994). Over the past few decades, several theories have been developed to explain the normal process of aging.

One of the most prominent theories of aging is activity theory which posits that older adults have the same psychological and social needs as when they were younger (Havighurst, 1961). However, because society withdraws from the aging individual, people are forced to give up their roles (e.g., employee, parent) and decrease social interactions. When the loss of roles occurs, an individual can experience loss of identity, low self-esteem, and isolation. Therefore, in order to experience successful aging, the individual should remain productive in society and replace role losses with new roles and increase social interaction. Disengagement theory is similar to activity theory in that it focuses on the loss of roles and activity engagement as we age. However, in direct contrast to activity theory, disengagement theory proposes that an aging individual and society mutually withdraw from each other in order to prepare for the eventual

⁎ Corresponding author at: Department of Psychology, 23 Armstrong Hall, Mankato, MN, USA 56001. Tel.: +1 507 389 5824. E-mail address: [email protected] (J.A. Buchanan).

http://dx.doi.org/10.1016/j.jaging.2015.03.003 0890-4065/© 2015 Elsevier Inc. All rights reserved.

2

J.A. Buchanan et al. / Journal of Aging Studies 34 (2015) 1–9

exit of the individual from society (Cumming & Henry, 1960). As people age, they slowly give up the roles and functions they previously held in society and gradually transfer that role to younger generations. The process of withdrawal and role losses, while inevitable, is seen as positive and even welcomed by the individual. Thus, successful aging depends on an individual being able to effectively complete this process of disengagement. Erik Erikson's model of human development also addresses developmental concerns of older adults (1966). Most relevant to older adults is Erikson's eighth stage where the individual's primary task involves reflecting on their past experiences. If the individual is satisfied with his life, s/he successfully resolves this stage and attains wisdom and acceptance of death (i.e., ego integrity), whereas failure in this stage leads to regret, despair, and a fear of death. Finally, the Theory of Gerotranscendence was developed by Lars Tornstam (1989) as a means of addressing limitations in the other prominent developmental theories of aging. Tornstam believed that existing theories did not fully reflect the reality of many older adults' lives and experiences. In short, Tornstam (1997) describes gerotranscendence as the process of human aging that involves a progression toward maturation and wisdom as well as a shift in meta-perspective, from a materialistic and pragmatic view of the world to a more cosmic and transcendent one. According to Tornstam (1997), the individual experiencing gerotranscendence may experience ontological changes on three different levels, the first being the cosmic level which involves changes in the definitions of time and space. Behaviors indicative of changes in the cosmic level may include an increasing connection and attachment to earlier generations, the disappearing fear of death, acceptance of the uncertainty and mystery of life, and greater joy being derived from subtle experiences in life. The second level of change associated with gerotranscendence occurs in the self. These changes may involve the discovery of hidden positive and negative aspects of the self, decreases in self-centeredness, being less concerned about body image, a shift from egoism to altruism, rediscovery of the child within and ego-integrity (i.e., realizing that the pieces of life's jigsaw puzzle form a wholeness). Changes indicative of gerotranscendence may also occur in social and individual relationships. These changes may involve becoming more selective and less interested in superficial relationships, an increasing need for solitude, an understanding of the difference between self and the social roles one inhabits, the addition of innocence to maturity, the understanding of the gravity of wealth and the freedom of “asceticism”, realizing the difficulty in separating right from wrong, and withholding from judgments and giving advice. Gerotranscendence vs. other theories Based on this definition of gerotranscendence, it is clear that gerotranscendence overlaps somewhat with other theories of aging. First, certain signs that are indicative of gerotranscendence (i.e., increased need for solitude, greater selectivity in social interaction) could be interpreted by others as disengagement. However, while gerotranscendence is sometimes considered an extension of disengagement theory, Tornstam (1994) has articulated a fundamental difference

between the two. First, disengagement implies a turning inward while gerotranscendence implies a new definition of reality. Second, in contrast with disengagement theory, individuals with high gerotranscendence also experience an increase in preferred social activities as well as an increase in need for solitude. Finally, the degree of social activity becomes less essential for life satisfaction at higher levels of gerotranscendence (Tornstam, 1994). Aside from its roots in disengagement theory, gerotranscendence also appears to have similarities with Erikson's eighth stage of development in that gerotranscendence is a process that presumably ends with a higher state of maturity (Tornstam, 1994). In both cases, the mature state also includes contentment and a feeling of affinity with the past. However, the ego-integrity stage described by Erikson is a process of reflection and integration of the past, while gerotranscendence implies looking forward and outward accompanied by a fundamental change in how the self and world are perceived (Tornstam, 1999). Furthermore, there appears to be a spiritual component (i.e., increased feeling of cosmic communion) present in gerotranscendence that is not present in Erikson's theory. Because of this, gerotranscendence can be seen as an extension of Erikson's model (Erikson, 1997; Tornstam, 2005, p. 76). Finally, similar to activity theory, gerotranscendence proposes that older adults remain active and engaged both socially and otherwise. One significant difference between the two theories is that from the perspective of gerotranscendence theory, normal developmental changes in old age involve a noticeable change in the amount and type of activity. For example, there may be a greater preference for solitary activity that involves reminiscence and reflection on one's past. In addition, older adults may socialize less, but according to gerotranscendence theory, this is interpreted as a deliberate process of selection with the goal of maximizing meaningful and satisfying social interactions and minimizing unpleasant or superficial interactions. Behavior is always interpreted through the lens of assumptions, values and/or theories. In other words, in order to interpret the meaning of the behavior of others, in this case older adults, one must make these interpretations through the use of a preexisting belief system. Sometimes these beliefs systems are quite explicit in that the observers is aware of their assumptions while in other cases the observe may be unaware of their own assumptions about human behavior. In either case, these belief systems are learned (either explicitly or implicitly), at least in part, from the culture in which one lives and can impact how one overtly behaves toward other people. In relation to the behavior of older individuals, of the theories discussed above activity theory has likely been the most influential in shaping perceptions of what “successful” aging should look like in Western cultures (Tornstam, 1996; Tornstam & Törnqvist, 2000). Continued activity is often prescribed as the means for a healthy and satisfying old age. Furthermore, it is expected that older individual will maintain the same values, attitudes, perceptions (of self, others and the world) as they did in middle age. However, this theory may not capture the preferences or experiences of many older adults and may instead represent the values and desires of younger people (including gerontologists) concerning what an ideal old age should look like (Tornstam, 1992).

J.A. Buchanan et al. / Journal of Aging Studies 34 (2015) 1–9

In contrast, the Theory of Gerotranscendence reinterprets many behaviors of older adults as being developmentally normal when these same behaviors could be considered pathological from other perspectives. This reinterpretation is based on a body of empirical research suggesting that gerotranscendence is commonly experienced by older adults and that those who experience it exhibit greater life satisfaction (see Tornstam, 2005 for a review of this literature). The Theory of Gerotranscendence is generally less familiar than activity theory to those in Western cultures, however, so it is less likely to be used as an interpretive framework for understanding developmental changes in older adulthood. A specific example may clarify how the assumptions one holds may result in different interpretations of the behavior of an older adult. Let us suppose that an older adult over time has begun to engage in less social activity and spend more time in solitary activity. To an outside observer interpreting these behaviors through the lens of activity theory, it could appear that the individual is adjusting poorly and perhaps even is experiencing symptoms of depression. In other words, the reduction of activity is problematic and must be reversed. On the other hand, an observer interpreting these same behaviors based on the Theory of Gerotranscendence may interpret them as active choices on the part of the individual and, thus relatively normal and even desirable. The potential implications of these different interpretations of behavior are clear. The implication of the widespread adoption of the assumptions of activity theory among westerners is that younger individuals may impose these values and beliefs onto older individuals. This leaves open the possibility that care providers of older adults (both professional and family) will misinterpret behavior as abnormal or pathological that is considered to be developmentally normal through the lens of gerotranscendence. These behaviors may also be considered quite normal from the perspective of older adults themselves. The result may be unnecessary evaluations and interventions for problems that do not exist as well as implicitly communicating to an older person that what s/he is doing is abnormal or unhealthy, thereby stagnating normal growth and development.

Relevant literature Existing data suggests that the experience of gerotranscendence tends to be correlated with age such that older individuals reports having experiences consistent with gerotranscendence more so than do younger adults (Tornstam, 1997). This is particularly true with regard to the cosmic and self dimensions of gerotranscendence. Only a small body of research, however, has examined how people perceive and interpret behaviors indicative of gerotranscendence when they are displayed by someone else (Tornstam & Törnqvist, 2000; Wadensten & Carlsson, 2001). Tornstam and Törnqvist (2000) approached a group of nursing staff and asked if they had observed gerotranscendent behaviors and inquired about their interpretations of those behaviors. Most behaviors described by Tornstam and Törnqvist (2000) were regarded by staff as undesirable. Interpretations were either pathology-oriented or activity-oriented. Staff either felt that the behavior was due to some form of

3

pathology (i.e. dementia) or due to negative consequences of inactivity. In a similar study, Wadensten and Carlsson (2001) conducted qualitative interviews with nursing staff and asked them about behaviors indicative of gerotranscendence. The interviewer asked participants whether they had observed a particular behavior within healthy older adults. Then, participants were asked to provide a description of the behavior and how they interpreted these signs. The data was then coded in two stages. The first stage consisted of coding data that were clearly evident in the text, while the second stage consisted of coding data by providing interpretations what the staff had reported. Each behavior was classified as either pathological, normal, or unnoticed/invisible. The results of this study indicated that some behaviors of gerotranscendence may be harder to observe than others. Furthermore, among the more easily observable behaviors, the following specific behaviors appeared to staff as signs of pathology: changed perception of time and space, great capacity to take pleasure in the small things in life, new perspective on social contacts, withdrawal from social activities and preference for and satisfaction with sitting alone and thinking, and display of innocent behaviors by asking questions or doing things not common to adults (Wadensten & Carlsson, 2001). Overall, the studies by Tornstam and Törnqvist (2000) and Wadensten and Carlsson (2001) provide some evidence that behaviors indicative of gerotranscendence may be interpreted as signs of pathology and/or the adverse result of inactivity.

The current study Purpose The purpose of the current study is to extend the work of Tornstam and Törnqvist (2000) as well as Wadensten and Carlsson (2001) that examined how behaviors indicative of gerotranscendence are interpreted. However, this study utilized a different methodology than those used in previous research. In place of a qualitative approach to measuring perception, a quantitative model was proposed in order to quantify the perception of gerotranscendence and determine significant differences between groups. Furthermore, previous research done on gerotranscendence has examined behaviors taken out of context. One goal of this study was to describe these behaviors within the context of a narrative about an older adult so the behaviors are more clearly defined and are less esoteric. Lastly, instead of testing the perceptions of nursing staff, this study aimed to look at the differences in perceptions of gerotranscendence between a population of younger people (e.g. college students) and older adults. Older adults may be more familiar with gerotranscendence, either through observing it occur in peers or having experienced it themselves. Therefore, it is expected that significant differences between older adults and college students in perception of gerotranscendence will occur. This study specifically examined differences in perception concerning gerotranscendence behaviors that were perceived to be signs of pathology by Wadensten and Carlsson (2001). It is hypothesized that older adults will be less likely than college students to perceive these gerotranscendence behaviors as unusual or concerning.

4

J.A. Buchanan et al. / Journal of Aging Studies 34 (2015) 1–9

Participants There were a total of 284 participants in the current study. The sample was 16.2% male and 83.8% female. Regarding ethnicity, 90.9% of the sample identified as Caucasian, 4.0% African American, 2.3% Asian, and 3.0% identified their ethnicity as “other”. Regarding age, two different populations were sampled. The first group of participants consisted of younger adults recruited from undergraduate courses at a Midwestern university. A total of 170 younger adults participated in the study. The mean age of the young adult sample was 20.7, with a range of 18 to 31 years of age. The second group of participants consisted of older adults (defined as those over the age of 65). Older adults were recruited from various senior community centers and from a convent in a small metropolitan area. A total of 114 older adults participated in the study. The mean age of the older adult sample was 77.0, with a range of 65 to 98 years of age.

Materials and method Perceptions of behaviors indicative of gerotranscendence were measured using an instrument created by the second author's research team. For this survey, participants were first asked to read instructions about how the survey was to be completed. Following the instructions, a short introduction to a man named Mr. Smith, a resident living in an assisted living facility, was provided. This introduction was two paragraphs long and described Mr. Smith's social, occupational, and health history. After this introduction, a series of three short stories about Mr. Smith's current life in the assisted living facility were provided. Embedded within each story were descriptions of specific behaviors that Mr. Smith engaged in that were indicative of gerotranscendence. The behavioral examples of gerotranscendence were developed based on descriptions of gerotranscendence found in the existing research literature (e.g., Tornstam, 2005, pp. 55–69). As a means of determining the content validity of this instrument, a copy was sent to Lars Tornstam, developer of the Theory of Gerotranscendence, to review. In his correspondence, Dr. Tornstam indicated that the content of the instrument was indeed reflective of an individual experiencing gerotranscendence (L. Tornstam, personal communication, September 24, 2013). In total, there were 18 different behaviors in the three stories indicative of the three different dimensions of gerotranscendence (i.e., cosmic, self, and social/personal relations). Story one contained five behaviors, story two contained ten behaviors, and story three contained three behaviors indicative of gerotranscendence. After each story, participants were asked to provide four ratings about each behavior that was contained in the story they just read. On a scale of 1 to 4, where 1 indicated strongly disagree and 4 indicated strongly agree, participants were asked to rate each behavior using the following scales: “this behavior is common in other older adults”, “there is nothing unusual about this behavior”, “I have observed behavior similar to this”, and “someone should be concerned about this behavior”. After rating all 18 behaviors, participants were also asked to rate their perception of Mr. Smith's overall life satisfaction, mood, and how normal he is compared to others his age.

Data was collected in two ways. All younger adult participants completed an online version of the survey, while 10 older adults completed the electronic version of the survey. The remaining older adult participants completed a paper and pencil version of the survey.

Data analysis Exclusion criteria Because “older adult” was defined as age 65 and older for this study, data from individuals who reported an age between 50 and 65 were excluded from the study (three responses were excluded using this criterion). For all the responses collected electronically, responses that were completed in less than eight minutes were excluded from the study because it was determined that completing the survey in less eight minutes was unrealistic and that completing the survey this rapidly was likely due to random responding. Of the 219 electronic responses, 21 responses (all college students) were excluded using this criterion. Target behaviors examined In their qualitative study on the perception of gerotranscendence behaviors, Wadensten and Carlsson (2001) identified five groups of behaviors that nursing assistants indicated as being potential signs of pathology. In the instrument used for the current study, there were eight behaviors embedded within the survey that corresponded to these five groups. A description of these behaviors can be found in Table 1. For each of these eight behaviors, the researcher tested for between-group differences in responding on two questions: “there is nothing unusual about this behavior” and “someone should be concerned about this behavior”. This resulted in a total of 16 between-group analyses. Between-group differences for only these two questions were examined because they measure the constructs of “abnormality” (i.e., how unusual the behavior is) and “pathology” (i.e., how concerning the behavior is), constructs that were similar to the ones investigated by Wadensten and Carlsson (2001).

Statistical analyses Independent sample t-tests were used to test for group differences between older adults and college students. Because a total of 16 statistical tests were conducted, a Bonferroni correction was used to minimize the likelihood that a significant finding would occur by chance. This resulted in a critical alpha level of 0.003. Also, given the differences in the sample sizes between the two groups, the Levene's test of equality of variances was computed to determine the assumption of homogeneity of variances between the samples was met. Effect sizes (Cohen's d) were also calculated as an additional means for determining the magnitude of between-group differences. Additionally, for the purpose of data analysis, the item “there is nothing unusual about this behavior” was reverse coded such that a high score indicates perception of behavior as highly unusual.

J.A. Buchanan et al. / Journal of Aging Studies 34 (2015) 1–9

5

Table 1 Descriptions of targeted behaviors. Description of targeted behavior Changed perception of time and space Story 1, behavior 1 Mr. Smith described experiencing memories in great detail and that, “it is almost as if I am there again” and that, “it is like I can go back in time and relive parts of my childhood.” Great capacity to take pleasure the small things in life Story 1, behavior 5 His joy now comes from things such as listening to a thunderstorm, watching it snow, hearing the birds chirping in the morning outside his window, and reading the newspaper. He also very much looks forward to getting his mail and his occasional visits from his children and grandchildren. New perspective on social contacts Story 2, behavior 4 Mr. Smith now goes to lunch in sweatpants, a t-shirt, and baseball hat instead of his usual suit. He commented to his daughter that, “I don't need to impress anyone anymore.” Story 2, behavior 6 He regularly gives money to his grandchildren, buys gifts for them, and even calls them and writes to them. His daughter mentioned that, “he's showing more interest in my family than he ever has before.” Story 3, behavior 3 He wants to talk to people he already knows and likes and isn't interested in making new friends. He remarked, “I don't have the time or energy to make new friends, I'll stick with the people I already know I like.” Withdrawal from social activities and preference for solitude Story 3, behavior 1 Although he has many old acquaintances and colleagues in town, he does not visit them nor does he invite them to visit him. He also socializes little in the facility. Story 3, behavior 2 Staff always let him know when activities are, but he rarely agrees to participate. He has mentioned how he prefers his “alone time” so he can sit, think and reflect on his past. Display of innocent behaviors Story 2, behavior 9 He has been observed playing Barbies with one of his granddaughters, the Wii with his grandsons, and has taken up building model cars which was a hobby of his when he was a boy. He told his daughter that he doesn't care if others think his behavior is silly or undignified and that, “at my age I'm going to act how I want to act.”

Results According to Dalby (2006), there appears to be an increase in spirituality and some aspects of gerotranscendence with age. Because part of our older adult sample was recruited from a convent, the possibility that there would be differences in responding between the sample of nuns and communitydwelling older adults was investigated. A series of independent sample t-tests were conducted and resulted in no significant between-group differences for any of the target behaviors. Therefore, all data for our older adult sample was aggregated for the remainder of the statistical tests. For behavior one of story one (e.g., changed perception of time and space — “experiencing memories in great detail: ‘It's almost as if I am there again’”), the results showed a significant difference between older adults and college students on the item regarding how unusual the behavior was perceived to be, t(269) = 6.375, p b 0.00. The effect size for this analysis (d = 0.81) was found to exceed Cohen's (1988) convention for a large effect (d = .80). Results indicated that older adults (M = 1.46, SD = 0.727) were less likely than college students to rate the behavior as unusual (M = 2.05, SD = 0.761). There was also a significant difference between the two groups on the item indicating how “concerning” this behavior was, t(268) = 4.933, p b 0.00. A moderate effect size was found for this analysis (d = 0.63). Results indicated that older adults (M = 1.52, SD = 0.743) were less likely than college students (M = 2.01, SD = 0.809) to rate this behavior as concerning. For behavior five of story one (e.g., great capacity to take pleasure — “his joy now comes from listening to thunderstorm, birds chirping, reading the newspaper”), results did not show a significant difference between older adults and college

students on the item regarding how unusual the behavior was perceived to be, t(252.98) = 2.404, p N 0.003. A small effect size was found for this analysis (d = 0.29). There was also no significant difference between the two groups on the item indicating how “concerning” this behavior was perceived to be, t(267) = 1.254, p N 0.003. A small effect size was also found for this analysis (d = 0.15). For behavior four of story two (e.g., new perspective on social contacts — Mr. Smith going to lunch in sweatpants because he doesn't need to impress anyone), results showed a significant difference between older adults and college students on the item regarding how unusual the behavior was perceived to be, t(265) = 5.681, p b 0.00. The effect size for this analysis (d = 0.73) was moderate to large. Results indicated that older adults (M = 1.57, SD = 0.688) were less likely than college students to rate the behavior as unusual (M = 2.13, SD = 0.837). There was also a significant difference between the two groups on the item indicating how “concerning” this behavior was perceived to be, t(265) = 6.300, p b 0.00. A large effect size (d = 0.81) was found for this analysis. Results indicated that older adults (M = 1.54, SD = 0.771) were less likely than college students (M = 2.19, SD = 0.833) to rate this behavior as concerning. For behavior six of story two (e.g., new perspective on social contacts — regularly giving money to grandchildren and showing more interest in family members), results did not show a significant difference between older adults and college students on the item regarding how unusual the behavior was perceived to be, t(263) = −0.199, p N 0.003. The effect size for this analysis (d = 0.02) was small. There was also no significant difference between the two groups on the item indicating how “concerning” this behavior was perceived to be,

6

J.A. Buchanan et al. / Journal of Aging Studies 34 (2015) 1–9

t(265) = 2.300, p N 0.003. A small effect size was also found for this analysis (d = 0.30). For behavior nine of story two (e.g., display of innocent behavior — Mr. Smith plays Barbie dolls and Wii with his grandchildren), results did not show a significant difference between older adults and college students on the item indicating “unusuality” of the behavior, t(265) = 1.246, p N 0.003. A small effect size was found for this analysis (d = 0.16). There was also no significant difference between the two groups on the item indicating how “concerning” this behavior was perceived to be, t(265) = 2.134, p N 0.003. A small effect size was also found for this analysis (d = 0.28). For behavior one of story three (e.g., withdrawal from social activity and preference for solitude — Mr. Smith not visiting friends and socializes little within the facility), results showed a significant difference between older adults and college students on the item regarding how unusual the behavior was perceived to be, t(266) = 9.189, p b 0.00. A large effect size was found for this analysis (d = 1.14). Results indicated that older adults (M = 1.89, SD = 0.723) were less likely than college students to rate the behavior as unusual (M = 2.67, SD = 0.635). There was also a significant difference between the two groups on the item indicating how “concerning” this behavior was perceived to be, t(165.31) = 7.09, p b 0.00. A large effect size (d = 0.93) was found for this analysis. Results indicated that older adults (M = 2.05, SD = 0.941) were less likely than college students (M = 2.83, SD = 0.718) to rate this behavior as concerning. For behavior two of story three (e.g., withdrawal from social activity and preference for solitude — staff lets Mr. Smith know when activities are, but he prefers solitude), results showed a significant difference between older adults and college students on the item regarding how unusual the behavior was perceived to be, t(266) = 7.646, p b 0.00. A large effect size was found for this analysis (d = 0.96). Results indicated that older adults (M = 1.85, SD = 0.716) were less likely than college students to rate the behavior as unusual (M = 2.54, SD = 0.716). There was also a significant difference between the two groups on the item indicating how “concerning” this behavior was perceived to be, t(178.85) = 6.132, p b 0.00. A large effect size (d = 0.79) was found for this analysis. Results indicated that older adults (M = 2.08, SD = 0.929) were less likely than college students (M = 2.76, SD = 0.770) to rate this behavior as concerning. For behavior three of story three (e.g., new perspective on social contacts — Mr. Smith prefers interacting with familiar people instead of making new friends), results showed a significant difference between older adults and college students on the item indicating regarding how unusual the behavior was perceived to be, t(222.45) = 8.099, p b 0.00. A large effect size was found for this analysis (d = 1.01). Results indicated that older adults (M = 1.78, SD = 0.660) were less likely than college students to rate the behavior as unusual (M = 2.48, SD = 0.718). There was also a significant difference between the two groups on the item indicating how “concerning” this behavior was perceived to be, t(265) = 5.365, p b 0.00. A moderate effect size (d = 0.66) was found for this analysis. Results indicated that older adults (M = 2.01, SD = 0.904) were less likely than college students (M = 2.56, SD = 0.741) to rate this behavior as concerning. Table 2 includes a summary of results.

Table 2 Scores for group means, standard deviation, t-tests and Cohen's d. Mean College students

t

Cohen's d

Older adults

Story 1, behavior 1 2.05 (0.761) Unusuala Concerning 2.01 (0.809)

1.46 (0.727) 1.52 (0.743)

6.375⁎ 4.999⁎

0.81 0.63

Story 1, behavior 5 1.40 (0.728) Unusuala Concerning 1.40 (0.669)

1.21 (0.571) 1.30 (0.624)

2.404 1.254

0.29 0.15

Story 2, behavior 4 2.13 (0.837) Unusuala Concerning 2.19 (0.833)

1.57 (0.688) 1.54 (0.771)

5.681⁎ 6.300⁎

0.73 0.81

Story 2, behavior 6 1.58 (0.681) Unusuala Concerning 1.63 (0.731)

1.59 (0.727) 1.42 (0.671)

−0.199 2.300

−0.03 0.30

Story 2, behavior 9 1.68 (0.747) Unusuala Concerning 1.65 (0.719)

1.56 (0.729) 1.45 (0.718)

1.246 2.134

0.16 −0.28

Story 3, behavior 1 2.67 (0.635) Unusuala Concerning 2.83 (0.718)

1.89 (0.723) 2.05 (0.941)

9.189⁎ 7.090⁎

1.14 0.93

Story 3, behavior 2 2.54 (0.716) Unusuala Concerning 2.76 (0.770)

1.85 (0.716) 2.08 (0.929)

7.646⁎ 6.132⁎

0.96 0.79

Story 3, behavior 3 2.48 (0.718) Unusuala Concerning 2.56 (0.741)

1.78 (0.660) 2.01 (0.904)

8.099⁎ 5.365⁎

1.02 0.66

a The means represented here were based on reverse scores of the item. Standard deviations appear in the parenthesis next to the group means. ⁎ p b 0.003, two-tailed.

Conclusions Summary of findings The purpose of this study was to examine whether younger and older adults perceive behaviors indicative of gerotranscendence differently in terms of how pathological or unusual these behaviors are. The results of this study partially supported the hypothesis that older adults would be less likely than college students to perceive gerotranscendence behaviors as unusual or concerning. There were a total of five behaviors where college students were significantly more likely to rate the behaviors as unusual and concerning than older adults. These behaviors included Mr. Smith describing his past experiences and experiencing life in both the present and the past simultaneously, Mr. Smith underdressing because he doesn't feel like impressing people anymore, Mr. Smith withdrawing from social activities, preferring solitude to social activities, and being more selective in his social interactions. Overall, one of these significant findings involved behaviors belonging to the cosmic dimension, one belonging to the self dimension, and three involving the social and individual relations dimension of gerotranscendence. Aside from these significant findings, there were three behaviors where no significant differences were found between the two groups of college students and older adults. These behaviors included taking pleasure in the small things in

J.A. Buchanan et al. / Journal of Aging Studies 34 (2015) 1–9

life, taking greater interest in children and grandchildren, and displays of innocent behavior. Of these non-significant findings, one behavior belonged to the cosmic dimension, two involved the self dimension, with none belonging to the social an individual relations dimension. Findings in context of existing literature The current study extends previous research in this area in that perceptions of younger adults were compared with perception of older adults. Similar studies in the literature have examined only perceptions of younger adults and could not rule out the possibility that older individuals would find certain gerotranscendent behaviors equally pathological or unusual. Based on the Theory of Gerotranscendence one would predict that older individuals would perceive these behaviors as less problematic presumably because they are more familiar with these behaviors and interpret them differently (i.e., as developmentally normal). Furthermore, many younger adults living in Western cultures may implicitly adhere to the values and tenets of activity theory and thus find behaviors that are inconsistent with this theory to be concerning and possible signs of mental illness. Moreover, younger individuals may find behaviors consistent with gerotranscendence foreign simply because they are at a very different stage in life where they are completing different life tasks (e.g., establishing relationships, beginning careers, and starting families) and have more limited life experience. Results from this study lend more credence to the notion that a subset of gerotranscendent behaviors are likely to be perceived very differently by younger people and the older people who engage in these behaviors. The results of this study are relatively consistent with existing literature on this topic in that changes in social interactions and changes in the perception of time and space have been perceived as pathological in other studies. For example, Tornstam and Törnqvist (2000) interviewed a sample of 14 nursing assistants and found that behaviors such as transcendence of time and space and increased need for solitude were commonly observed and generally perceived to be negative or signs of pathology. Wadensten and Carlsson (2001) also found that nursing assistants (all age 59 or younger) tended to pathologize changes in perception of time and space, new perspectives on social contacts, and withdrawal from some social activities and a preference for solitude. One finding from this study that is not consistent with the existing literature is that in the current study rejoicing and taking pleasure in small things was generally not perceived as being abnormal or unusual and there were no between group differences in perceptions of this behavior. In fact mean ratings for the item related to rejoicing were 1.40 or less for both ratings in both samples, indicating general disagreement that this behavior was unusual and/or concerning. One possible explanation for the differences found in the current study and those from previous studies concerns the novel instrument used to measure perceptions/opinions of gerotranscendent behavior. It has been noted that some behaviors (e.g., changes in perception of time and space) indicative of gerotranscendence are somewhat abstract and may be difficult to understand without examples or context (Atchley, 1999, p. 143). This may be particularly true when these behaviors are simply described to a respondent either in

7

an interview or questionnaire format as has been done in previous studies. The current study attempted to contextualize these behaviors by embedding them within a vignette about an older individual. It was hoped that by contextualizing these behaviors, they would be more accessible and understandable to almost anyone and that opinions about these behaviors would be better informed, and thus more valid. It is possible that by contextualizing these behaviors, respondents had different (and perhaps better-informed) opinions about the normality of these behaviors. Another possible explanation has to do with the samples obtained in this study compared to other studies. For example, other studies examining perceptions of gerotranscendent behavior have sampled long-term care staff. The samples of younger individuals in the current study were all below the age of 31 and very few (12.3%) had any direct caregiving experience with the elderly. It is possible that individuals who are in a caregiving role that requires more direct contact with sick and frail older adults have different perceptions about these behaviors. When the findings of this study are put into the context of existing literature, it suggests that younger adults in care taking roles may be somewhat more likely to label a larger set of behaviors as pathological than are younger people who are not caregivers.

Limitations and future directions Although the current study produced some interesting results that extend the existing literature on perceptions of gerotranscendent behavior, limitations of the study must be acknowledged. As mentioned above, the survey completed by participants was created for this study and has some limitations that restrict the generalizability of the findings. For example, the vignettes were about a man with health problems residing in a long-term care facility. Responding could have been different had the character in these vignettes been female, relatively healthy, and/or living independently at home. In addition, ideally one wants to understand general perceptions of gerotranscendent behaviors as opposed to how they are perceived within a specific individual. Therefore, the fact that the survey used in this study embedded target behaviors in a context may have resulted in ratings that were specific to the individual in the vignette and his circumstances as opposed to general opinions about these behavior. Future research will need to address these limitations by comparing responses to different versions of the vignettes where variables such as gender, age, health status, and living situation of the protagonist in the vignettes are systematically manipulated. Similarly, gerotranscendence is a very broad construct that is composed of three levels or dimensions. Moreover, each level/dimension is composed of a number of sub-constructs (e.g., rejoicing, displays of innocence, changed perception of time and space, changed meaning and importance of relationships), each of which could be exemplified by a vast number of different specific behaviors. For example, there are a number of different examples of “rejoicing” that could have been described in the vignettes. Said another way, the specific behaviors that were embedded in the vignettes represent only one of an almost infinite pool of possible examples of each construct. Therefore, it is possible that if different examples had

8

J.A. Buchanan et al. / Journal of Aging Studies 34 (2015) 1–9

been chosen to represent different aspects of gerotranscendence, different results would have been found. Additionally, although the behaviors indicative of gerotranscendence were rationally-derived based on the work of Tornstam, the psychometric properties of this survey are unknown. It should be noted, however, that a form of content validation was obtained when Lars Tornstam reviewed this survey and indicated that the stories indeed describe an individual who is experiencing gerotranscendence. Lastly, the target behaviors used in this study and the behaviors described in previous literature may not represent the same groups of behaviors. Although the targeted behaviors used in the current survey are intended to exemplify the same behaviors previously described by Wadensten and Carlsson (2001), it is possible that the behaviors may be viewed differently. For example, behavior 4 of story 2 is intended to exemplify Mr. Smith's new perspective on social contacts by describing him as underdressing for social occasions because he doesn't need to impress other people. Participants rating this behavior may be basing ratings on his dressing behavior but not the change in perspective on social contacts. Another limitation of this study involves the sample that was obtained. For example, the sample lacked gender diversity. Previous research indicates that men may be less likely to experience gerotranscendence which may result in different perceptions of these behaviors (Tornstam, 1997). In addition, the older adult sample obtained for this study was primarily under the age of 85 (only 19 individuals were age 85 or older). Given that gerotranscendence appears to increase with age, individuals over the age of 85 may have even more accepting attitudes concerning these behaviors compared to older adults who are younger than 85 (Tornstam, 1997, 2003). The sample also lacked cultural and ethnic diversity, as most participants were Caucasian and recruited from the upper Midwestern United States. Finally, the older adult sample in this study was composed of individuals who lived in the community as opposed to long-term care facilities. Consequently, future studies need to examine more diverse samples in terms of age, gender, ethnicity, health status, and living situation. Future research should also explore the meaning attributed to various gerotranscendence behaviors to better understand why respondents rate behaviors as they do. Results of the current study only suggest that the two groups differ (or do not differ) with respect to how they perceive these behaviors, but do not address potentially important and interesting reasons as to why there are (or are not) differences. For example, as mentioned above, no between-groups differences were found with respect to the item related to rejoicing. Despite this lack of statistical difference, it may be that older and younger people provided the ratings they did for very different reasons. For instance, older individuals may perceive this behavior as relatively normal because they actually take greater pleasure from many “small” things and genuinely take less pleasure in life's “bigger” events. On the other hand, younger people may find this behavior relatively normal because they assume older people do not have many “big” events to look forward, so must take pleasure in small things instead. In fact, Tornstam and Törnqvist (2000) found that rejoicing may be perceived by younger people as a sign of loneliness, depression and isolation or reflects resignation in that the older person has nothing else to look forward to in life, so must find pleasure in small,

seemingly insignificant things. Expanding the survey used in this study or conducting qualitative interviews with subsamples of participants may assist in interpreting the reasons for between-group differences (or lack thereof). A final area for future inquiry is related to the abovementioned suggestion that younger adults in care taking roles may be more prone to over-pathologize a broader set of behaviors compared to those who are not caregivers. This conclusion was based on the findings from this study taken together with findings from previous studies. In previous studies that sampled individuals working in long-term care facilities (i.e., Tornstam & Törnqvist, 2000; Wadensten & Carlsson, 2001), a greater number of behaviors were identified as being cause for concern compared to the findings from the current study that was composed primarily of younger individuals who did not have elder care experience. First, this pattern of findings must be investigated further in future studies by directly comparing samples of younger people who work in long-term care to those who do not. Unfortunately the sample of younger people in this study that had caregiving experiencing was relatively low, making comparisons with non-caregivers unwarranted. Second, if this pattern of findings is replicated, research questions emerge concerning how those who are in caretaker roles are socialized to perceive pathology. Implications The differences in perception of specific gerotranscendent behaviors found in this study may have important implications for care of the elderly, whether it be care provided by family members or professional caregivers. Behaviors that are indicative of a normal developmental process such as gerotranscendence may be incorrectly pathologized by younger care providers who may primarily interpret behavior through the lens of activity theory (Hauge, 1998; Tornstam & Törnqvist, 2000). The process of pathologizing these behaviors may in turn affect the interactions between older and younger adults. For example, based on this perceived pathology, caregivers and family members may push the individual to engage in activities that they do not wish to participate in or recommend psychiatric medication that may not be necessary. An understanding of the Theory of Gerotranscendence may provide permission for caregivers to allow residents to choose solitude, reflection, or other more “passive” forms of activity (Hauge, 1998). The results of this study and others may provide some insight for family members and professional caregivers who work with older adults. Understanding that gerotranscendence is a natural developmental process experienced by many older adults may help caregivers provide more sensitive care. In fact, some researchers have developed educational programs and practice guidelines for long-term care staff members based on the theory of gerotranscendence with the goal of helping staff interpret certain behaviors in different ways (e.g., Tornstam, 1996; Wadensten & Carlsson, 2003). Tornstam (1996) found that 33% of care providers who attended a seminar on gerotranscendence reported changing their behavior toward residents and 43% understood residents in a new light. Qualitative responses revealed specific changes in staff perceptions and behaviors such as understanding the need for solitude, less insistence on participating in activities, listening to care recipients more, and not imposing one's own needs and values onto residents.

J.A. Buchanan et al. / Journal of Aging Studies 34 (2015) 1–9

Wadensten and Carlsson (2003) introduced the Theory of Gerotranscendence to three small groups of care providers and asked them to generate specific guidelines for staff to follow when they observe behaviors that could simultaneously be interpreted as signs of pathology or gerotranscendence. Some examples of recommendations included: accepting these behaviors as normal signs of aging; allowing and encouraging residents to reminisce about past events and how they have developed over time; listen when residents discuss death and ask questions to stimulate further thoughts; allow residents to be alone if they so choose; introduce residents to the Theory of Gerotranscendence as a positive form of aging. It also seems reasonable to recommend that care providers enter into a dialogue with an older individual about the meaning of their behaviors as opposed to assuming these behaviors are due to some pathological process such as depression or dementia. Results of Tornstam's (1996) study, however, indicates that recommendations for caregivers to engage older adults in conversations about gerotranscendence and the meaning of behavior may be difficult in that only 12% of their sample reported having these kinds of discussions after learning more about the theory. It may be that caregivers need more explicit guidance or practice (such as through role play exercises or viewing videotaped examples) about how to have these conversations. Summary This study attempted to determine if younger and older adults interpret specific behaviors indicative of gerotranscendence in different ways. It was hypothesized that older adults would be less likely to interpret these behaviors as unusual or abnormal. This hypothesis was partially supported in that analyses for five of eight behaviors showed significant between-group differences. Younger adults were more likely to pathologize behaviors related to changes in the perception of time and space and changes in social behavior. Tornstam (1996) succinctly explains the implications of these differences in perception when he states, “The Theory of Gerotranscendence suggests that we run the risk of misinterpreting effects of naturally increased transcendence as negative disengagement — disengagement being caused by a process of social breakdown.” A greater understanding

9

of the Theory of Gerotranscendence by those who provide care to older adults may prevent these misinterpretations from happening and could lead to more productive conversations that result in positive growth for all parties involved. References Achenbaum, W. A., & Bengtson, V. L. (1994). Re-engaging the disengagement theory of aging: On the history and assessment of theory development in gerontology. The Gerontologist, 34(6), 756–763. Atchley, R. C. (1999). Continuity and adaptation in aging. Baltimore, MD: Johns Hopkins University Press. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd Ed.). New York: Psychology Press. Cumming, E., & Henry, W. E. (1960). Growing old: The process of disengagement. New York: Basic Books. Dalby, P. (2006). Is there a process of spiritual change or development associated with ageing? A critical review of research. Aging & Mental Health, 10(1), 4–12. Erikson, E. H. (1966). Eight stages of man. International Journal of Psychiatry, 2(3), 281–300. Erikson, J. M. (1997). The life cycle completed: Extended version with new chapters on the ninth stage of development. New York/London: W.W. Norton & Company. Hauge, S. (1998). An analysis and critique of the theory of gerotranscendence. Retrieved from http://www.bib.hive.no/tekster/hveskrift/notat/1998-3/ Havighurst, R. J. (1961). Successful aging. The Gerontologist, 1(1), 8–13. Tornstam, L. (1989). Gero-transcendence: A reformulation of the disengagement theory. Aging Clinical and Experimental Research, 1(1), 55–63. Tornstam, L. (1992). The Quo Vadis of gerontology: On the scientific paradigm of gerontology. The Gerontologist, 32(3), 318–326. Tornstam, L. (1994). Gero-transcendence: A theoretical and empirical exploration. In L. E. Thomas, & S. A. Eisenhandler (Eds.), Aging and the Religious Dimension (pp. 203–226). Westport, CT: Auburn House. Tornstam, L. (1996). Caring for the elderly: Introducing the theory of gerotranscendence as a supplementary frame of reference for caring for the elderly. Scandinavian Journal of Caring Science, 10, 144–150. Tornstam, L. (1997). Gerotranscendence in a broad cross sectional perspective. Journal of Aging and Identity, 2(1), 17–36. Tornstam, L. (1999). Gerotranscendence and the functions of reminiscence. Journal of Aging and Identity, 4(3), 155–166. Tornstam, L. (2003). Gerotranscendence from young old age to old age. Online Publication from The Social Gerontology Group, Uppsala (URL: http:// www.soc.uu.se/publications/fulltext/gtransoldold.pdf). Tornstam, L. (2005). Gerotranscendence: A developmental theory of positive aging. New York: Springer. Tornstam, L., & Törnqvist, M. (2000). Nursing staff's interpretations of “gerotranscendental behavior” in the elderly. Journal of Aging and Identity, 5(1), 15–29. Wadensten, B., & Carlsson, M. (2001). A qualitative study of nursing staff members' interpretations of signs of gerotranscendence. Journal of Advanced Nursing, 36(5), 635–642. Wadensten, B., & Carlsson, M. (2003). Theory-driven guidelines for practical care of older people, based on the theory of gerotranscendence. Journal of Advanced Nursing, 41(5), 462–470.