Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Oct 1.
Published in final edited form as: Soc Sci Med. 2023 Aug 19;334:116174. doi: 10.1016/j.socscimed.2023.116174

The relationship between adolescent reading habits and older adult social engagement – a longitudinal cohort analysis

Viji Diane Kannan a, Kelly Peters b, Benjamin P Chapman a
PMCID: PMC10561696  NIHMSID: NIHMS1927493  PMID: 37633113

Abstract

Objectives –

To investigate an access point during youth with the potential to have a positive impact on social engagement in later-life. Our social selves begin long before older adulthood, a life-stage during which people face extensive changes to their social milieu. Adolescence is a sensitive period for sociocultural processing and adolescent activities have the potential to impact social engagement in older adulthood. We examine reading since, in adolescence, it promotes social-cognitive skills which facilitate social engagement.

Methods –

Our main goal was to study the relationship between adolescent reading and older adult social engagement. We used longitudinal cohort data from Project Talent and ordered logistic regression as well as two-wave, cross-lagged panel model to analyze this relationship. Ancillary analyses examined recognized relationships between reading and social-cognition and between social-cognition and social engagement in both adolescence and older adulthood.

Results –

Adolescent reading was related to more frequent older adult social engagement with both family and friends. This relationship was independent of reading during older adulthood and independent of social-cognition at both life-stages. Adolescent social engagement was not associated with older adult reading in the two-wave, cross-lagged model. Reading was positively related to social-cognition in adolescence, but was insignificant in older adulthood. And, social-cognition was positively related to social engagement in both adolescence and older adulthood.

Discussion –

Reading is declining for both young and older Americans. Young Americans are also socializing less than in previous decades. These trends could have a detrimental influence on social engagement and social isolation in future cohorts of older adults. We recommend continuing to look for activities and experiences during youth, especially the sensitive period of adolescence, that could stimulate social engagement over the life-course and into older adulthood.

Keywords: Social Brain Hypothesis, Social-Cognition, Life-Course, Project Talent

1. Introduction

High levels of social isolation are concerning because of the associated risk for cardiovascular disease, stroke, infectious disease, diabetes, and cognitive decline and dementia as well as all-cause mortality rates (Holt-Lunstad, Smith, Baker, Harris, & Stephenson, 2015; U.S. Surgeon General, 2023). Risk associated with social isolation is comparable to or exceeds many common health hazards (U.S. Surgeon General, 2023) in a dose response manner (Yang et al., 2016). Approximately half of U.S. adults report experiencing social isolation (U.S. Surgeon General, 2023). And, older Americans experience substantially more social isolation (Kannan & Veazie, 2023) and less social engagement (Vogelsang, 2021) than other age groups.

However, socioemotional selectivity theory (Carstensen, 2021) proposes that older adults intentionally prune their social networks to create space for more emotionally meaningful relationships. And, activity theory (Lemon, Bengtson, & Peterson, 1972) suggests that older adults should and do adapt to changing social circumstances by refocusing their social roles and engaging in new and different activities. Indeed, empirical findings show that older adults adapt to social network loss by cultivating new confidant relationships, socializing with neighbors more frequently, and increasing community involvement (Cornwell, Goldman, & Laumann, 2021). As informal social participation decreases with age, formal social participation increases (Ang, 2019). Further, older adults engage in relatively higher levels of companionship (i.e., shared leisure for enjoyment and intrinsic satisfaction) (Kannan & Veazie, 2023). These findings suggest that not all life-course trajectories inevitably lead to severe isolation and lack of social engagement as a consequence of growing older (Cudjoe et al., 2020). Moreover, these findings suggest that older adults who are more socially competent (i.e., who are better able to navigate the changing social circumstances that characterize older adulthood) would have more opportunities for social engagement.

Our social selves begin long before older adulthood. Both the convoy model and continuity theory propose that experiences and circumstances over the life-course contribute to older adults’ social lives (Antonucci, Fiori, Birditt, & Jackey, 2010; Atchley, 1989). Certain experiences over the life-course or at specific sensitive life-stages may confer to the individual greater social competence that lasts into older ages. In this study, we examine the relationship between adolescent reading habits and older adult social engagement.

1.1. Adolescence and the Social Brain

Human social dynamics are more complex than that of other social species and include strategic social processes such as coalition formation and tactical deception. The evolution of large brains in humans is attributed to the increasing complexity of human social networks over millennia and the associated increase in cognitive demands (Dunbar, 2009). The term ‘social brain’ refers to the network of brain regions that are involved in understanding other people (Frith, 2007). The social brain is primarily concerned with skills required to manage social relationships (Burnett, Sebastian, Kadosh, & Blakemore, 2011; Frith, 2007), skills that would be useful in navigating the changing social landscape older adults face.

Adolescence is a sensitive period for sociocultural processing (Blakemore & Mills, 2014). Sensitive periods refer to life-stages characterized by rapid changes during which the effect of an exposure on an outcome is stronger than at other life-stages and is expressed throughout the life-course (Kuh, Ben-Shlomo, Lynch, Hallqvist, & Power, 2003). During adolescence, we experience large and rapid development hormonally, physically, psychologically, behaviorally, and socially (Blakemore & Mills, 2014; Driver et al., 2023; Viner et al., 2015). And, the social brain undergoes important structural changes and functional reorganization during adolescence (Andrews, Ahmed, & Blakemore, 2021; Burnett et al., 2011; Mills, Lalonde, Clasen, Giedd, & Blakemore, 2014).

Accumulating evidence points to adolescence as a sensitive period for adapting to one’s social environment and for processing social signals (Blakemore & Mills, 2014). Adolescents are more sociable, form more complex and hierarchical peer relationships, are more sensitive to acceptance and rejection by their peers, and they put more effort into managing peer social relationships compared to children or young adults (Blakemore, 2008; Blakemore & Mills, 2014). Adolescents and young adults spend, on average, 3-fold more time with friends than do middle and older adults (Kannan & Veazie, 2023).

Patterns that begin during adolescence track strongly into adult life (Viner et al., 2015). Thus, variation, during adolescence, in the structural and functional changes to the social brain may have an impact on social engagement in subsequent life-stages. And, experiences during adolescence that contribute to changes in the social brain, shaping social competence, might have future effects on social engagement later in life.

Indeed, findings show that some adolescent experiences have long reaching effects on social engagement well into older adulthood. For example, different types of high school activities are associated with greater social activity decades later in mid- and later-life (Vogelsang, 2021). Involvement in sports, performance arts like drama or orchestra, pep squad, student government, publications like student newspaper or yearbook, and service projects are all positively related to older adult social engagement (Vogelsang, 2021). Perhaps, involvement in these activities during adolescence impacts the social brain in a way that enhances one’s ability to adapt to changing social contexts and to manage social relationships, which in turn, over the life-course, promotes social engagement through later life-stages. In this study, we examine the relationship between adolescent reading habits and older adult social engagement.

1.2. Reading

Leisure reading during adolescence contributes to social-cognitive development. Indeed, both processing the real social world and narrative-processing require the same cognitive mechanisms (Gerrig, 1993). Social-cognition is a set of social skills which develop early in life and evolve over successive life-stages (Lutfey & Mortimer, 2003). Social-cognitive skills allow us to more adeptly process social signals “necessary to navigate complex social interactions and to respond appropriately” (Batini, Lpuerini, Cei, Izzo, & Toti, 2021; Mumper & Gerrig, 2017). Social-cognitive skills also promote prosocial helping behaviors that encourage and maintain harmonious, satisfying interpersonal interactions (Simpson & Beckes, 2010). That is, social-cognitive skills engender greater social competence. And, empirically, higher social-cognition is related to lower levels of objective social isolation (Lutfey & Mortimer, 2003; Piejka et al., 2021).

Literary theories link reading to prosociality and social-cognition through conceptual frameworks of transportation and simulation (Johnson, 2012; Lenhart, Dangel, & Richter, 2020; Mar, 2018; Mar & Oatley, 2008; Oatley, 2016). According to these theories, reading emotionally transports readers into the story by fusing attention, imagery, and feeling. Once transported, readers immerse themselves in a variety of social simulations. Readers vicariously experience a wide range of emotions through many different characters, each with their own unique desires and motivations. These simulated social encounters allow the reader to privately develop and train social-cognitive skills. When the reader returns to the real world, she is subtly transformed by those social simulations.

Empirical evidence supports the connection between reading and social-cognition. Greater lifetime fiction reading was found to be associated with higher sensitivity to non-verbal social cues (Fong, Mullin, & Mar, 2013). Experimental studies showed that greater lifetime and concurrent fiction reading were both associated with higher tendencies toward prosocial helping behaviors (Johnson, 2012; Johnson, Cushman, Borden, & McCune, 2013). And, a 2016 meta-analysis found that larger amounts of fiction and/or non-fiction reading was related to higher scores on empathy (understanding and relating to the emotional states of others), theory of mind (the ability to infer from observation other people’s desires, beliefs, thoughts, emotions, and intentions as separate and different from one’s own), and perspective taking (viewing situations from the perspective of the other person) (Mumper & Gerrig, 2017). However, participants in these cross-sectional and “lifetime” reading studies were (almost exclusively) children, adolescents, and young-adults in their 20s (Batini et al., 2021; Lenhart et al., 2020; Mumper & Gerrig, 2017), representing only the early portion of the human life-span.

1.3. The Present Study

The impetus for this study was to consider an access point during youth that might stimulate social engagement in later-life, thereby, taking advantage of “the long-arm of childhood.” Employing access points in youth that influence social engagement decades later could potentially lessen the urgency and burden for geriatric interventions by having a longer lead-time prior to the life-stage when social isolation is most prominent and individuals are less malleable. Additionally, scholars and government officials have called for an increase in population-wide social isolation prevention strategies over the current predominance of individual level approaches (M. L. Smith et al., 2023; Smith, Holt-Lunstad, & Kawachi, 2023; U.S. Surgeon General, 2023). While reading is an individual pursuit, the creation of reading initiatives, especially for youth, can be accomplished at the population level through schools, libraries, community centers, and public service announcements (among other initiatives). Thus, the public health motivations for this study were to examine an access point during youth that could promote social engagement decades later and have population-wide application.

In this study, we examine the association between adolescent reading and older adult social engagement. Given that experiences that begin during adolescence track strongly into adult life, we propose that adolescent activities could have an enduring effect on older adult behaviors. Further, adolescence is a sensitive period for sociocultural processing; it is a time when the social brain develops through structural changes and functional reorganization. Thus, adolescent activities could specifically have an enduring effect on social engagement. Finally, reading has a strong influence on expressing social-cognitive skills and prosocial behaviors which are necessary for social engagement. Thus, we hypothesize that adolescent reading is positively related to older adult social engagement. We also do ancillary analyses that examine these recognized relationships in our sample. Since previous studies examining the association between reading and social-cognition were primarily conducted in younger age samples, we examine this relationship in our sample participants both when they were adolescents and as older adults. And, we examine the recognized relationship between social-cognition and social engagement, in our sample among both adolescents and older adults.

2. Methods

2.1. Sample

Project Talent (PT) is a longitudinal cohort study developed by the American Institutes for Research. The base year survey (PT60) was conducted in 1960 on a nationally representative sample of American high school students i.e., 9th – 12th graders (n= 377,016). The survey captured a variety of items on students’ characteristics, academic and cognitive capacity, family background, interests, and behaviors, including reading books for leisure. PT60 served as the sampling frame for the 2018 Project Talent Aging Study (PTAS18). The PTAS sampling frame included a total of 22,584 randomly selected participants from PT60. From this sampling frame, subsamples were randomly selected and given different surveys. The sample and items we draw on for this study were constrained by the stage 2 subsample Module A (n=2,038) which had a 79.7% response rate. PT60 and PTAS18 supply the adolescent and older adult variables for this study, respectively — resulting in 58-years between the adolescent exposure and the subsequent older adult outcome. Students identified as having a reading disability were dropped from the study, resulting in n=369,539 for PT60 and n=2,017 for PTAS18.

2.2. Outcome

Older Adult Social Engagement.

In PTAS, older adult social engagement was assessed for family and for friends with the following two items: (1) How often are you in contact with members of your Family (brothers, sisters, parents, or children) who do not live with you; including visits, phone calls, letters, or electronic mail messages? and (2) How often are you in contact with your Friends including visits, phone calls, letters, or electronic mail messages? Responses ranged across eight options from ‘several times a day’ to ‘never or hardly ever’.

2.3. Exposure

Adolescent Reading.

Adolescent leisure reading habits were captured in PT60 by the item:

How many books have you read (not including those required for school) in the past 12 months? Don’t count magazines or comic books. The number of books read ranged from 0 to 90. Quartiles of reading were created for which cutoff points fell at 15, 28, 45, and 90 books.

2.4. Covariates

Covariates included personal attributes of age, race, and sex as well as both adolescent and older adult characteristics. During adolescence, all participants were in high school, thus education level at the time was ‘some high school’ and grade level was closely related to age. We capture the highest level of attained education in the set of older adult characteristics. Additionally, adolescent family socioeconomic status was captured by a composite index consisting of total family income, possession of various high-value items, education-level of both mother and father, father’s occupational status, home value, and, importantly, the number of books in the home. Older adult covariates include income level, attained education, retirement status, marital status, number of people in household, and depression.

Adolescent Social-Cognitive Skills.

Adolescent reading could increase older adult social engagement by improving social-cognitive skills during adolescence which then indirectly increases older adult social engagement through increasing adolescent social engagement and/or older adult social-cognition. PT60 has a scale called ‘social sensitivity’ (defined as empathy, sensitivity to others’ feelings, and propensity for perspective taking (Pozzebon et al., 2013)) which includes the following nine items: I never hurt another person’s feelings if I can avoid it. I seem to know how other people will feel about things. I sympathize with my friends and encourage them when they have problems. People consider me a sympathetic listener. People consider me very helpful in dealing with other people. I am sympathetic. I am considerate. People consider me understanding. And, I like to tease people [reverse scored]. A score of ‘1’ was assigned to items that described the participant Extremely well or Quite well. A score of ‘0’ was assigned to items that described the participant Fairly well, Slightly well, or Not very well. The total social sensitivity score equaled the sum of all nine items. Cronbach α internal consistency reliability estimates for this nine-item social sensitivity scale was 0.85 and test–retest reliability was 0.83 (Damian, Spengler, Sutu, & Roberts, 2019; Pozzebon et al., 2013).

Adolescent Social Engagement.

Adolescent reading could increase older adult social engagement because it increases adolescent social engagement (directly and/or indirectly through adolescent social-cognition) and this effect on social engagement then carries over the life-course into older adulthood. PT60 contains a social life index constructed from the following five items: How many days per week are you allowed to go out for fun 8pm or later during the school year, How many evenings do you go out a week during school, Age of first date, Average number of dates per week, and How many steadies past three years?

Older Adult Reading.

Adolescent reading could increase older adult social engagement because it leads to more reading in older adulthood which influences older adult social engagement directly and/or indirectly through older adult social-cognition. Older adults’ reading habits were captured in PTAS with the item: On average, how much time do you spend reading books, magazines, and newspapers? (Include audio books and reading articles online). Responses ranged across six choices from ‘4 or more hours a day’ to ‘I never do this’.

Older Adult Social-Cognitive Skills.

Adolescent reading could increase older adult social engagement because it improves social-cognition in older adulthood through adolescent social-cognition and/or through older adult reading. PTAS measured older adult social-cognitive skills with an index using five of the nine PT60 social sensitivity items. The Cronbach α internal consistency reliability estimates for this five-item social sensitivity scale was 0.85 and the test–retest reliability was 0.80 (Damian et al., 2019).

2.5. Analysis

Sample characteristics and descriptions of reading, social-cognition, and social engagement variables are summarized. We started with the ancillary analyses examining the recognized relationships that motivate our main analysis: the relationship between reading and social-cognition in both adolescence and older adulthood followed by the relationship between social-cognition and social engagement in both adolescence and older adulthood. These analyses were conducted using multivariate ordered logistic regression with all adolescent covariates, as well as all older adult covariates in analyses with an older adult outcome. One exception was analysis of the adolescent social-cognition and adolescent social engagement association for which multivariate linear regression was used.

In our main analysis, we first established the total, unadjusted effect of adolescent reading on older adult family social engagement and on older adult friend social engagement using ordered logistic regression. Then, we examined this relationship independent of the listed covariates in multivariate ordered logistic regressions. Predicted probabilities of the first and fourth reading quartiles are presented for older adult social engagement once a week or less and for greater than once a week.

Next, we fit two-wave, cross-lagged panel models, one for family social engagement and one for friend social engagement in older adulthood. These models involved same wave correlations between adolescent reading and adolescent social engagement as well as between older adult reading and older adult social engagement. Multivariate linear regression was used to analyze the association of reading and of social engagement across both waves. Similarly, multivariate linear regression was used to analyze the relationship between reading in wave 1 and social engagement in wave 2 as well as the association between social engagement in wave 1 and reading in wave 2. These multivariate linear regressions include all covariates.

3. Results

Characteristics of the sample are summarized in Table 1. Ages ranged from 70 to 78 years, with an average age of 73.5 years. Men and women made up half the sample each and 79% of the sample was white. The family socioeconomic index in adolescence ranged from 67 to 126 with a mean of 97. Just under half the sample graduated from college and three-quarters made under $100,000. Almost all were retired, nearly three-quarters were married, and, on average, respondents lived in households containing about two people. Most respondents were rarely, if ever, depressed.

Table 1.

Descriptive Statistics of PT60 andPTAS18 Module A.

Characteristic Mean (SD) [Range] or Frequency
PT60 (n= 369,539) PTAS18 (n=2,017)
Age 15.8 (1.25) [13–20] 73.5 (1.26) [70–78]
Female 51% 51%
White 90% 79%
Black 7% 7%
Other Race 3% 14%
Socio-Economic Index 97.9 (10.2) [59–131] 97.2 (9.9) [67–126]
Income
 < $50,000 39%
 $50,000 up to $100,000 37%
 ≥ $100,000 24%
Education
 Some High School 0.6%
 High School Graduate 7.7%
 Some College 24.3%
 Associate / Technical Degree 21.8%
 Bachelor Degree 22.7%
 Master Degree 17.1%
 Doctorate / Professional Degree 5.8%
Marriage
 Married 71%
 Divorced, Separated, Widowed 25%
 Never Married 4%
Number in Household 1.9 (0.8) [1–8]
Retired 97%
Depression
 Rarely or none of the time (< 1 day) 76%
 Some or a little of the time (1–2 days) 16%
 Occasionally or a moderate amount of the 6%
 time (3–4 days)
 Most or all of the time (5–7 days) 2%
Open in a new tab

Descriptions of the exposure, outcome, and mediator variables are presented in Table 2. As adolescents, participants read on average 32 and a median of 28 books in a year. As older adults, 40% of participants read for two or more hours a day. As shown in previous studies (Oh, Chopik, Konrath, & Grimm, 2020), social-cognitive skills increased with age from an average score of 4.6 (0–9 range) in adolescence to an average score of 4.1 (0–5 range) in older adulthood. The adolescent social life index ranged from 0 to 79 with a mean of 42. In older adulthood, 30% of the sample were socially engaged with family and 25% were socially engaged with friends, at least once a day, while 21% and 22% were engaged less than once a week with family and friends, respectively.

Table 2.

Reading, Social-Cognition, and Social Engagement Variables.

Characteristic Mean (SD) [Range] or Frequency
PT60 (n= 369,539) PTAS18 (n=2,017)
Reading
Adolescent Reading (# of books) 32.2 (21.1) [0–90]
 Quartile 1 0 – 15
 Quartile 2 16 – 28
 Quartile 3 29 – 45
 Quartile 4 46 – 90
Older Adults Reading (time spent reading)
 I never do this 1%
 Less than 1 hour a week 8%
 A few hours a week 21%
 About 1 hour a day 30%
 2–3 hours a day 32%
 4 or more hours a day 8%
Social-Cognitive Skills (SCS)
 Adolescent SCS 4.6 (2.4) [0–9] 4.7 (2.3) [0–9]
 Older Adults SCS 4.1 (0.6), 0 – 5
Social Engagement
 Adolescent Social Life Index 45.3 (15.9) [0–83] 42.2 (16.2) [0 – 79]
 Older Adult Family Social Engagement
  Several times a day 13%
  About once a day 17%
  Several times a week 30%
  About once a week 19%
  2 or 3 times a month 10%
  About once a month 4%
  Less than once a month 4%
  Never or hardly ever 3%
 Older Adult Friend Social Engagement
  Several times a day 9%
  About once a day 16%
  Several times a week 33%
  About once a week 20%
  2 or 3 times a month 11%
  About once a month 4%
  Less than once a month 4%
  Never or hardly ever 3%
Open in a new tab

Recognized relationships between reading and social-cognition and between social-cognition and social engagement yielded expected results. Reading and social-cognition were positively related during adolescence (the life-stage primarily observed in previous studies) (OR=1.20, p<0.001, 95%CI[1.19,1.20]). However, during older adulthood the reading and social-cognition relationship was not statistically significant (OR=1.02, p=0.686, 95%CI[0.92,1.14]). In adolescence, social-cognition was positively related to social engagement (β=0.62, p<0.001, 95%CI[0.59,0.63]). In older adulthood, social-cognition was positively related to both family social engagement (OR=1.14, p=0.002, 95%CI[1.05,1.23]) and friend social engagement (OR=1.26, p<0.001, 95%CI[1.16,1.36]).

3.1. Ordered Logistic Regression

Results from the ordered logistic regression analysis are presented in Table 3. Unadjusted odds ratios for adolescent reading were positive (OR=1.16, p<0.001, 95%CI[1.08, 1.25]) for older adult family social engagement and positive (OR=1.14, p=0.001, 95%CI[1.06, 1.22]) for older adult friend social engagement. Adjusted for all covariates, odds ratios for adolescent reading were positive (OR=1.10, p=0.047, 95%CI[1.00, 1.22]) for older adult family social engagement and positive (OR=1.10, p=0.057, 95%CI[1.00, 1.22]) for older adult friend social engagement. These odds ratios point in the hypothesized direction suggesting that adolescent reading is positively related to older adult social engagement across nearly six decades. Adolescents in the first reading quartile were 7 percentage points more likely to be socially engaged with family and friends once a week or less than those in the fourth reading quartile. And, adolescents in the fourth reading quartile were 7 percentage points more likely to be socially engaged with family and friends greater than once a week than those in the first reading quartile.

Table 3.

Ordinal Logistic Regression Odds Ratios for Older Adult Social Engagement

with Family with Friends
OR p 95% CI OR p 95% CI
Unadjusted
Adolescent Reading 1.16 <0.001 [1.08, 1.25] 1.14 0.001 [1.06, 1.22]
Adjusted
Adolescent Reading 1.10 0.047 [1.00, 1.22] 1.10 0.057 [1.00, 1.22]
Age 0.99 0.804 [0.91, 1.08] 0.96 0.333 [0.88, 1.04]
Female 1.53 <0.001 [1.22, 1.93] 1.23 0.079 [0.98, 1.56]
White 1.27 0.090 [0.96, 1.66] 1.48 0.005 [1.12, 1.94]
Adolescent
 Socio-Economic Index 0.87 0.025 [0.78, 0.98] 1.04 0.544 [0.92, 1.17]
 Social-Cognitive Skills 1.03 0.175 [0.99, 1.09] 0.98 0.427 [0.93, 1.03]
 Social Life 1.00 0.590 [1.00, 1.01] 1.00 0.742 [0.99, 1.01]
Older Adult
 Reading 1.15 0.005 [1.04, 1.27] 1.10 0.056 [1.00, 1.22]
 Social-Cognitive Skills 1.12 0.006 [1.03, 1.21] 1.27 <0.001 [1.16, 1.38]
 Attained Education 0.93 0.069 [0.86, 1.01] 1.11 0.010 [1.03, 1.21]
 Family Income 1.01 0.114 [1.00, 1.02] 1.00 0.927 [0.99, 1.01]
 Married 1.12 0.414 [0.86, 1.45] 0.76 0.050 [0.58, 1.00]
 Number in Household 1.03 0.727 [0.87, 1.21] 0.93 0.363 [0.79, 1.09]
 Retired 0.78 0.465 [0.41, 1.51] 1.11 0.738 [0.59, 2.10]
 Depressed 1.01 0.927 [0.86, 1.18] 0.86 0.065 [0.74, 1.01]
Open in a new tab

Social Engagement for both family and friends:

once a week or less Quartile 1 > Quartile 4 by 7 percentage points

greater than once a week Quartile 1 < Quartile 4 by 7 percentage points

Table 3 also displays the results for covariates used in each analysis. The individual attributes female sex and white race have strong positive relationships with older adult social engagement. However, in contrast to other characteristics observed during adolescence such as socioeconomic status, social-cognitive skills, and, importantly, social engagement itself, adolescent reading has a statistically significant and positive relationship with older adult social engagement. Interestingly, socioeconomic status in adolescence is negatively related to family social engagement in older adults.

Independent of each other, reading in adolescence and reading in older adulthood were both positively related to older adult social engagement. Adolescent reading was related to older adult family social engagement at a similar level to older adult social-cognitive skills and was related to older adult friend social engagement at a similar level to highest attained education. In older adulthood, better social-cognitive skill has a strong positive relationship, while being depressed and being married have strong negative relationships with friend social engagement. In older adulthood, higher family income, having a higher number of people in the household, being retired, and being married (family social engagement) were statistically insignificant.

3.2. Two-Wave, Cross-Lagged Panel Models

The two-wave, cross-lagged models are shown in Figure 1(a) for family social engagement and Figure 1(b) for friend social engagement. Reading remained relatively stable over time with older adult reading being positively associated with adolescent reading. In contrast, social engagement in older adulthood was not associated with social engagement in adolescence indicating that social engagement evolves over the life-course as suggested in the convoy model and continuity theory. Adolescent reading positively predicted older adult family social engagement (β=0.09, p=0.039) and friend social engagement (β=0.12, p=0.007) while controlling for all covariates including adolescent levels of social engagement. However, adolescent social engagement did not predict older adult reading (β=0.0006, p=0.756).

FIGURE 1.

FIGURE 1.

Open in a new tab

Two-wave, cross-lagged relationships between reading and social engagement.

4. Discussion

This study examined the association between adolescent reading and older adult social engagement. We chose the adolescent life-stage because it is a sensitive period for developing social-cognitive skills and for socio-cultural processing. And, we chose the activity of reading due to its recognized effect on social-cognition during youth.

Our results confirmed previous findings that, among adolescents, reading has a positive association with social-cognitive skills. However, this association did not hold among older adults, perhaps accounting for the dearth of literature on reading and social-cognition with an older sample. Adolescence may be a sensitive period for the effects of reading on social-cognition. Our study also confirmed that social-cognition facilitates social engagement at both the adolescent and older adult life-stages.

Adolescent reading was associated with a greater frequency of social engagement with family and with friends in older adulthood. This association between adolescent reading and older adult social engagement was independent of other attributes such as social-cognitive skills, socioeconomic status, and social engagement in adolescence. This association was also independent of reading and social-cognitive skills (among other attributes) in older adulthood, suggesting that mechanisms other than sustained reading habits and social-cognition over the life-course may be contributing to the relationship. Another mechanism might be that reading increases verbal and communication skills, which in turn positively influences older adult social engagement. Alternatively, reading might increase social engagement by decreasing known predictors of social isolation (e.g., depression, anxiety, cognitive decline). Evidence indicates that reading improves cognition independent of the reverse pathway in which cognition drives reading (Bavishi, Slade, & Levy, 2016).

Other adolescent attributes were either statistically insignificant or, as in the case of adolescent socioeconomic status, negatively associated with older adult social engagement with family. Given that the span of time between adolescence and older adulthood was nearly six decades, our results that most of the adolescent characteristics were insignificant is unsurprising. However, given this time span, the study also highlights that studies regarding activities engaged in during adolescence, in this case reading, are worth pursuing to reduce social isolation over the life-course.

Given our study’s longitudinal cohort design, participants were adolescents in the 1960s. This was necessary in order to study the life-course trajectories of older adults who are currently in their 70s. Recent trends show a decline in adolescent reading habits. One-third of adolescents read zero books for pleasure in 2016, three-times greater than in 1979 (Twenge, Martin, & Spitzberg, 2019). In the late 1970s, 60% of adolescents read every day. However, by 2016, only 16% of adolescents read every day (Twenge, Martin, et al., 2019). Thirteen-year-olds who read for fun almost every day declined from 35% in 1984 to 17% in 2020 (Schaeffer, 2021). Over the same time frame, 13-year-olds who almost never read for fun increased from 8% to 29%. Seventeen-year-olds showed a similar pattern. And, the length of time spent reading per day for adolescents in 2019 and 2020 was no more than 9 minutes (U.S. Bureau of Labor Statistics, 2021).

By contrast, daily use of social media by 10th graders increased from 58% during the late 2000s to 83% in 2016 (Twenge, Martin, et al., 2019). And, adolescents today may be substituting online, digital social interaction for in-person, face-to-face social engagement (Kannan & Veazie, 2023; Twenge & Spitzberg, 2020; Twenge, Spitzberg, & Campbell, 2019). Indeed, since 2003, time spent in companionship and in social engagement with friends has plummeted for 15 to 24 year olds (Kannan & Veazie, 2023). Meanwhile, social isolation in this age group has increased faster than any other age group by almost two-fold (Kannan & Veazie, 2023). Thus, the current cohort of young people will age having had both a lesser reading habit and lower in-person social engagement than previous youth cohorts.

In our model, we find that reading during older adulthood has some positive association with older adult social engagement, independent of reading habits during adolescence. However, in America, the quantity of books read has declined most steeply for older adults (Jones, 2022). Additionally, in 2021, 28% of US adults over 50-years had read zero books whether print, electronic, or audio and whether in whole or in part for an entire year (Gelles-Watnick & Perrin, 2021). Nonetheless, currently, older Americans spend substantially more time in daily leisure reading than do young and middle aged Americans (U.S. Bureau of Labor Statistics, 2021). Although, this may be a cohort effect since youth in previous decades read more than the current generation of young people (Twenge, Martin, et al., 2019).

Older adult reading levels have far reaching health effects as shown in longevity outcomes (Bavishi et al., 2016; Jacobs, Hammerman-Rozenberg, Cohen, & Stessman, 2008). Nonetheless, other access points in youth should be explored in future research. High school participation is associated with greater social activity decades later in mid- and later-life (Vogelsang, 2021). Research investigating adolescent activities related to increased social-cognition, environmental mastery, communication skills, emotional health, and cognitive health could identify potential pathways for improving later-life social engagement.

4.1. Limitations

The limitations of this study are important to note. Although we use a longitudinal design and show that reading early in life is related to social engagement later in life, this association is only captured within a single cohort. This study does not provide evidence for such a relationship in more recent cohorts. But, access points during youth that might stimulate social engagement in older adulthood should continue to be studied in successive generations. Due to the long time frame between 1960 and 2018, some PT60 participants would have died and, thus, would not have been available for the PTAS18 random sample. Given that reading is also associated with mortality in older adults (Bavishi et al., 2016; Jacobs et al., 2008), this may introduce some bias to the study.

This study captured amount of reading, but did not delve into the quality or nature of the reading material (i.e., how simplistic or philosophical or abstract) due to data limitations. However, the 2016 meta-analysis (Mumper & Gerrig, 2017) found that regardless of whether a person engaged in fiction or non-fiction reading, it was the amount of reading that predicted greater social-cognition. Additionally, people’s reading abilities differ as do their interest in type of stories and reading content (e.g., horror, sci-fi, romance). Reading for leisure will necessarily skew toward each individual’s preferences and abilities.

Our data for both reading and social engagement is self-reported. However, the data delineates specific time ranges to express frequency of social engagement rather than, for example, more vague expressions such as “a lot” or “sometimes”. Similarly, the reading variables use numeric measures of reading in terms of number of books during adolescence and distinct ranges of time spent reading in older adulthood.

The PT data does not include reading, social-cognition, and social engagement variables in follow-up studies conducted during the intervening years between 1960 and 2018. Thus, we are unable to trace the trajectories of these variables over the entire 58-years. Although, the analysis does not identify a causal relationship, the temporal ordering of our data and theory suggest that adolescent reading has some influence on older adult social engagement.

  • Reading is positively related to social-cognition in adolescence.

  • Adolescent reading is positively related to older adult social engagement.

  • This relationship is independent of older adult reading.

Funding:

This work was supported by the National Institute on Aging (R01AG053155)

Footnotes

We have no conflicts of interest.

We have no financial interests to disclose.

Declaration of conflict of interest: None.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Andrews JL, Ahmed SP, & Blakemore S-J (2021). Navigating the social environment in adolescence: The role of social brain development. Biological Psychiatry, 89(2), 109–118. [DOI] [PubMed] [Google Scholar]
  2. Ang S. (2019). Life course social connectedness: Age-cohort trends in social participation. Advances in Life Course Research, 39, 13–22. [Google Scholar]
  3. Antonucci TC, Fiori KL, Birditt K, & Jackey LM (2010). Convoys of social relations: Integrating life-span and life-course perspectives. In Lerner RM, Lamb ME, & Freund AM (Eds.), The Handbook of Life-Span Development (Vol. 2, pp. 434–473). Hoboken, NJ: Wiley. [Google Scholar]
  4. Atchley RC (1989). A continuity theory of normal aging. The Gerontologist, 29(2), 183–190. [DOI] [PubMed] [Google Scholar]
  5. Batini F., Lpuerini V, Cei E, Izzo D, & Toti G. (2021). The association between reading and emotional development: a systematic review. J Educ Training Stud, 9(1), 12–50. [Google Scholar]
  6. Bavishi A, Slade MD, & Levy BR (2016). A chapter a day: Association of book reading with longevity. Social Science & Medicine, 164, 44–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Blakemore S-J (2008). The social brain in adolescence. Nature Reviews Neuroscience, 9(4), 267–277. [DOI] [PubMed] [Google Scholar]
  8. Blakemore S-J, & Mills KL (2014). Is adolescence a sensitive period for sociocultural processing? Annual review of psychology, 65, 187–207. [DOI] [PubMed] [Google Scholar]
  9. Burnett S, Sebastian C, Kadosh KC, & Blakemore S-J (2011). The social brain in adolescence: evidence from functional magnetic resonance imaging and behavioural studies. Neuroscience & Biobehavioral Reviews, 35(8), 1654–1664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Carstensen LL (2021). Socioemotional Selectivity Theory: The Role of Perceived Endings in Human Motivation. The Gerontologist, 61(8), 1188–1196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cornwell B, Goldman A, & Laumann EO (2021). Homeostasis revisited: Patterns of stability and rebalancing in older adults’ social lives. The Journals of Gerontology: Series B, 76(4), 778–789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cudjoe TK, Roth DL, Szanton SL, Wolff JL, Boyd CM, & Thorpe RJ Jr (2020). The epidemiology of social isolation: National health and aging trends study. The Journals of Gerontology: Series B, 75(1), 107–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Damian RI, Spengler M, Sutu A, & Roberts BW (2019). Sixteen going on sixty-six: A longitudinal study of personality stability and change across 50 years. Journal of Personality and Social Psychology, 117(3), 674. [DOI] [PubMed] [Google Scholar]
  14. Driver C, Moore L, Mohamed A, Boyes A, Sacks DD, Mills L, . . . Hermens DF (2023). Structural connectivity and its association with social connectedness in early adolescence. Behavioural Brain Research, 440, 114259. [DOI] [PubMed] [Google Scholar]
  15. Dunbar RI (2009). The social brain hypothesis and its implications for social evolution. Annals of human biology, 36(5), 562–572. [DOI] [PubMed] [Google Scholar]
  16. Fong K, Mullin JB, & Mar RA (2013). What you read matters: The role of fiction genre in predicting interpersonal sensitivity. Psychology of Aesthetics, Creativity, and the Arts, 7(4), 370. [Google Scholar]
  17. Frith CD (2007). The social brain? Philosophical Transactions of the Royal Society B: Biological Sciences, 362(1480), 671–678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gelles-Watnick R, & Perrin A. (2021). Who doesn’t read books in America? Retrieved from https://www.pewresearch.org/fact-tank/2021/09/21/who-doesnt-read-books-in-america/ [Google Scholar]
  19. Gerrig RJ (1993). Experiencing Narrative Worlds: On the Psychological Activities of Reading: Yale University Press. [Google Scholar]
  20. Holt-Lunstad J, Smith TB, Baker M, Harris T, & Stephenson D. (2015). Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspectives on Psychological Science, 10(2), 227–237. [DOI] [PubMed] [Google Scholar]
  21. Jacobs JM, Hammerman-Rozenberg R, Cohen A, & Stessman J. (2008). Reading daily predicts reduced mortality among men from a cohort of community-dwelling 70-year-olds. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 63(2), S73–S80. [DOI] [PubMed] [Google Scholar]
  22. Johnson DR (2012). Transportation into a story increases empathy, prosocial behavior, and perceptual bias toward fearful expressions. Personality and Individual Differences, 52(2), 150–155. [Google Scholar]
  23. Johnson DR, Cushman GK, Borden LA, & McCune MS (2013). Potentiating empathic growth: Generating imagery while reading fiction increases empathy and prosocial behavior. Psychology of Aesthetics, Creativity, and the Arts, 7(3), 306. [Google Scholar]
  24. Jones JM (2022). Americans Reading Fewer Books Than in Past. Retrieved from https://news.gallup.com/poll/388541/americans-reading-fewer-books-past.aspx [Google Scholar]
  25. Kannan VD, & Veazie PJ (2023). US trends in social isolation, social engagement, and companionship ⎯ nationally and by age, sex, race/ethnicity, family income, and work hours, 2003–2020. SSM-Population Health, 21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kuh D, Ben-Shlomo Y, Lynch J, Hallqvist J, & Power C. (2003). Glossary - Life Course Epidemiology. Journal of Epidemiology and Community Health, 57(10), 778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lemon BW, Bengtson VL, & Peterson JA (1972). An exploration of the activity theory of aging: Activity types and life satisfaction among in-movers to a retirement community. Journal of gerontology, 27(4), 511–523. [DOI] [PubMed] [Google Scholar]
  28. Lenhart J, Dangel J, & Richter T. (2020). The relationship between lifetime book reading and empathy in adolescents: Examining transportability as a moderator. Psychology of Aesthetics, Creativity, and the Arts. [Google Scholar]
  29. Lutfey K, & Mortimer JT (2003). Development and socialization through the adult life course. Handbook of social psychology, 183–202. [Google Scholar]
  30. Mar RA (2018). Stories and the promotion of social cognition. Current Directions in Psychological Science, 27(4), 257–262. [Google Scholar]
  31. Mar RA, & Oatley K. (2008). The function of fiction is the abstraction and simulation of social experience. Perspectives on Psychological Science, 3(3), 173–192. [DOI] [PubMed] [Google Scholar]
  32. Mills KL, Lalonde F, Clasen LS, Giedd JN, & Blakemore S-J (2014). Developmental changes in the structure of the social brain in late childhood and adolescence. Social Cognitive and Affective Neuroscience, 9(1), 123–131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mumper ML, & Gerrig RJ (2017). Leisure reading and social cognition: A meta-analysis. Psychology of Aesthetics, Creativity, and the Arts, 11(1), 109. [Google Scholar]
  34. Oatley K. (2016). Fiction: Simulation of social worlds. Trends in cognitive sciences, 20(8), 618–628. [DOI] [PubMed] [Google Scholar]
  35. Oh J, Chopik WJ, Konrath S, & Grimm KJ (2020). Longitudinal changes in empathy across the life span in six samples of human development. Social Psychological and Personality Science, 11(2), 244–253. [Google Scholar]
  36. Piejka A, Krawczyk M, Schudy A, Wiśniewska M, Żurek K, & Pinkham A. (2021). Owner of a lonely mind? Social cognitive capacity is associated with objective, but not perceived social isolation in healthy individuals. Journal of Research in Personality, 93, 104103. [Google Scholar]
  37. Pozzebon J, Damian RI, Hill P, Lin Y, Lapham S, & Roberts BW (2013). Establishing the validity and reliability of the Project Talent Personality Inventory. Frontiers in Psychology, 4, 968. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Schaeffer K. (2021). Among many U.S. children, reading for fun has become less common, federal data shows. Retrieved from https://www.pewresearch.org/fact-tank/2021/11/12/among-many-u-s-children-reading-for-fun-has-become-less-common-federal-data-shows/ [Google Scholar]
  39. Simpson JA, & Beckes L. (2010). Evolutionary perspectives on prosocial behavior. In Mikulincer M. & Shaver PR (Eds.), Prosocial Motives, Emotions, and Behavior: The Better Angels of Our Nature: American Psychological Association. [Google Scholar]
  40. Smith ML, Racoosin J, Wilkerson R, Ivey RM, Hawkley L, Holt-Lunstad J, & Cudjoe TK (2023). Societal-and community-level strategies to improve social connectedness among older adults. Frontiers in Public Health, 11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Smith RW, Holt-Lunstad J, & Kawachi I. (2023). Benchmarking social isolation, loneliness and smoking: challenges and opportunities for public health. American Journal of Epidemiology, kwad121. [DOI] [PubMed] [Google Scholar]
  42. Twenge JM, Martin GN, & Spitzberg BH (2019). Trends in US Adolescents’ media use, 1976–2016: The rise of digital media, the decline of TV, and the (near) demise of print. Psychology of Popular Media Culture, 8(4), 329. [Google Scholar]
  43. Twenge JM, & Spitzberg BH (2020). Declines in non-digital social interaction among Americans, 2003–2017. Journal of Applied Social Psychology, 50(6), 363–367. [Google Scholar]
  44. Twenge JM, Spitzberg BH, & Campbell WK (2019). Less in-person social interaction with peers among US adolescents in the 21st century and links to loneliness. Journal of Social and Personal Relationships, 36(6), 1892–1913. [Google Scholar]
  45. U.S. Bureau of Labor Statistics. (2021). Time spent reading for personal interest in 2020. Retrieved from https://www.bls.gov/opub/ted/2021/time-spent-reading-for-personal-interest-in-2020.htm
  46. Surgeon General U.S. (2023). Our Epidemic of Loneliness and Isolation. Retrieved from https://www.hhs.gov/sites/default/files/surgeon-general-social-connection-advisory.pdf [Google Scholar]
  47. Viner RM, Ross D, Hardy R, Kuh D, Power C, Johnson A, . . . Kelly Y. (2015). Life course epidemiology: recognising the importance of adolescence. In: BMJ Publishing Group Ltd. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Vogelsang EM (2021). Social participation across mid-and later-life: Evidence from a longitudinal cohort study. Sociological Perspectives, 64(6), 1187–1205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Yang YC, Boen C, Gerken K, Li T, Schorpp K, & Harris KM (2016). Social relationships and physiological determinants of longevity across the human life span. Proceedings of the National Academy of Sciences, 113(3), 578–583. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES