Abstract
This study tested how increased modality interactivity (e.g., playing vs. watching a game) can augment empathy and reduce ageism toward older adults among college-aged students. An experiment (n = 201) examined the effects of playing a video game that featured an older adult protagonist. Relative to those who spectated the game, participants who played the game had increased state empathy on subfactors, including shared affect, empathic motivation, and helping motivation. Moreover, shared affect mediated the link between playing a game featuring an older adult protagonist and ageism. Participants with higher trait empathy showed lower state empathy after playing, whereas participants with lower trait empathy displayed increased post-game state empathy. Overall, the results supported the modality interactivity model suggesting fruitful avenues for future research.
Introduction
Ageism refers to discrimination against people based on their chronological age and points to the use of stereotypes, prejudice, or bias toward older adults. 1 The adoption of negative attitudes toward aging is linked to decreased well-being not only among older adults but also among younger individuals, especially as they age. 2 Negative perceptions of aging can lead to lower quality of life, increased anxiety about aging, and heightened depressive symptoms among young adults as they anticipate their own aging process. 3 Moreover, this decline in mental health can also manifest in physical health deterioration, emphasizing the pervasive impact of ageism across the lifespan. 4 For example, higher mortality risk, poorer functional health, and slower recovery from illnesses are all linked to the internalization of negative attitudes toward aging and older adulthood. 5 The adverse effects of ageism have spurred a search for new methods to reduce negative ageist stereotypes toward older adults, especially among young adults who are more likely to express ageism. 6 In addition, empathy is a trait 7 and a state 8 and that can influence the effectiveness of ageism reduction interventions. Trait empathy is relatively consistent over time 9 and correlates with individuals’ personalities 10 and prosocial behavior. 11 State empathy is an empathetic reaction to an event or person who may prompt outgroup member recategorization into an ingroup member. 12
Due to young adults’ increased engagement with technology, 13 there is interest in how interactive media use can reduce negative stigmas against older individuals. 14 For instance, video games offer increased modality interactivity, which refers to interface tools that afford users greater interaction and stronger user control over the situation (e.g., clicking, scrolling, dragging, and hovering). 15 Increases in modality interactivity may, in turn, influence individuals’ attitudes.16,17 For instance, actively playing a video game is more immersive than simply watching gameplay footage. 18 Playing interactive video games may allow individuals to create stronger connections with others, with whom they may otherwise have little interpersonal contact. 19 For example, playing a videogame featuring a protagonist with a mental illness increases the connection that players feel with the character, which indirectly reduces the desire for social distance toward the mentally ill, relative to watching gameplay footage. 20
Considering the above, this study examines how interactive media experiences such as playing a video game featuring an older adult protagonist may decrease ageism among young adults by increasing known state empathy subfactors (i.e., shared affect, empathic imagination, helping motivation, and cognitive empathy) toward a game protagonist, relative to watching recorded footage of such video games. Finally, the study examines how trait empathy moderates the effects of playing or spectating a video game on state empathy and ageism. The sections below outline relevant literature to articulate empirical hypotheses.
Interactivity effects model: how media features affect psychological outcomes
Video games provide affordances that set them apart from traditional media, such as films and books. Individuals may actively play, inhabit, and interact with game characters, events, and environments rather than just read or watch a story unfold. 21 According to the agency model of customization, interactivity is a key factor that can amplify individuals’ perceived agency, which in turn influences cognition, affect, and behavior. 22 For instance, interactively customizing a game avatar can augment empathy toward such a digital character. 23 In addition, participants who customize an avatar to resemble their physical self show increased identifiability or perceptions that their real identity could be deduced from their online persona, which indirectly decreases social distance toward an outgroup member. 24
Moreover, video games allow individuals to socialize and interact with others, which relates to the motivational technology model. 15 From this perspective, social interactivity can create stronger relatedness with other players and thus augment intrinsic motivation to use such media. 25 Overall, this may result in stronger engagement with video games that allow players to socialize with each other. 25 In addition, the bonds fostered through playing video games socially can increase empathy and reduce prejudice toward outgroup members. For example, collaborative intergroup play increases empathy and reduces outgroup member anxiety. 26 Among participants who play a violent video game in cooperative or competitive modes, playing cooperatively increases helping behavior, relative to playing competitively. 27
The interactivity effects model posits that increased modality interactivity can create stronger user engagement and absorption of content, as it enhances users’ visual experiences and perceptual channels. 22 Modality interactivity refers to clicking, scrolling, dragging, and using an interface to hover over media content. These methods of interface interaction allow individuals to obtain information, move or rotate objects, and advance narrative and character development through each action. Such interface activities are theorized to enhance the range, speed, and vividness of information provided by a game’s interface, which in turn promotes increased engagement with game content by focusing players’ attention. 15 Playing in an interactive environment also creates greater empathy relative to a less interactive environment, as allowing individuals to determine the action of a game’s protagonist is more effective than simply showing a video of the same material. 28
Trait and State Empathy and Ageism Reduction
Trait empathy should moderate the effects of media exposure on individuals’ attitudes and behavior. 29 This prediction is consistent with the assumption that the effects of media use on emotions, cognitions, attitudes, beliefs, physiological reactions, and behavior are enhanced or reduced by individual differences, such as traits, sex, temperament, and developmental stage. 30 Consider that lower trait empathy is linked to increased ageism against older adults. 31 In addition, state empathy should mediate the effects of playing vs. watching video games featuring older protagonists. For instance, state empathy statistically mediated the effect of watching the movie “Silver Linings Playbook” in reducing negative stereotypes about mental illness. 32
State empathy has cognitive, affective, and behavioral subfactors,
33
including shared affect (i.e, experiencing someone else’s feelings and emotions),
34
empathic imagination (i.e., projection of the self into the thoughts, feelings, and emotions of others),35,36 helping motivation (i.e., intention to help others as a response to their thoughts, feelings, and predicament),
37
and cognitive empathy (i.e., knowing the content of another person’s mental or emotional state).
34
Based on the above, we predict that, Gameplayers will show a greater reduction in ageist stereotypes toward older adults than game spectators. Gameplayers will have higher (a) shared affect, (b) empathic imagination, (c) helping motivation, and (d) cognitive empathy after playing the game than game spectators. State empathy, including (a) shared affect, (b) empathic imagination, (c) helping motivation, and (d) cognitive empathy, will mediate the effect of gameplaying vs. game watching on ageist stereotypes toward older adults. Trait empathy will moderate the effect of gameplaying vs. game watching on ageist stereotypes toward older adults.
Materials and Methods
Participants
The study was pre-registered on the Open Science Framework (OSF) before data collection began (https://osf.io/8gk9n/?view_only=59565a81cda748e18e5399c3bf2621c1). College-aged younger adults (range 18 to 25, M = 19.19, SD = 1.43) were recruited from a large West Coast University’s study subject pool. Qualtrics was used for online survey data collection. Five hundred forty participants responded to a pre-experiment survey, which included demographic measures, an ageism scale, and trait empathy measures. Out of these, n = 240 participants came to the laboratory and finished the experiment to allow matching pre- and post-experiment data. Extreme values and participants with missing data were excluded (for more details, see Results), and thus the final sample size was n = 201 (100 male and 101 female). A priori power analysis was used to determine the sample size for a factorial experimental design using 0.05 alpha, and power of 0.80 estimated the need of at least 128 participants. About 62.6% of the participants were Asian, 19.4% were White, 8.5% were Hispanic or Latino, 7.5% were of multiple races, 1% was Black or African American, 0.5% were American Indian or Alaska Native, and 0.5% were Pacific Islanders. On average, participants had 8 years of gaming experience (range 0 to 18, SD = 4.95).
Procedure
All procedures were approved by the university’s Institutional Review Board (blinded for review). After recruitment from a study subject pool, participants provided informed consent and then completed a pre-experiment online survey that assessed their trait empathy and ageist beliefs before exposure to the manipulations. Participants then signed up to partake in the experiment. Participants answered the pre-experiment survey at least 24 hours before their appointment for the experiment.
Upon arrival at the laboratory, participants were randomly assigned to either play the game “Old Man’s Journey” 38 for about 18 minutes or watch a pre-recorded video of the same game segment that lasted about 18 minutes (Fig. 1). Participants then completed a post-experiment survey and then were debriefed. One hundred participants played the game, and 101 participants watched the pre-recorded gameplay video. An elderly priming manipulation was attempted, but it did not influence these findings (https://osf.io/rwqzd/?view_only=4daa9ac7be7a41449bac99873dce7cbd).

Old Man’s journey screenshots. Players can click on the young couple (left) and then a memory of how he met his wife appears (right).
Old man’s journey
“Old Man’s Journey” is a narrative-driven side scrolling adventure game that takes players on a journey through the life of an older adult. Players interact with the world by manipulating the terrain, such as raising and lowering hills to create paths, allowing the protagonist to navigate the environment (Fig. 1). As players explore, they unlock memories that reveal Old Man’s past and the significant moments that shaped his life. The gameplay is designed to be meditative, encouraging players to reflect on themes of love, loss, and the passage of time. 38
Measures
Trait empathy
This factor was measured before the experiment using an empathy quotient. 39 The 8-item instrument was framed as a 4-point Likert scale from 1 (strongly disagree) to 4 (strongly agree). Items included “I find it easy to put myself in somebody else’s shoes” and “I am good at predicting how someone will feel.” The scale showed acceptable reliability (Cronbach’s α = 0.76).
State empathy
Four reworded subscales from the comprehensive state empathy scale (CSES) 40 measured participants’ degree of empathy toward the game protagonist after the experiment. The subfactors included shared effect (4 items), empathic imagination (4 items), helping motivation (4 items), and cognitive empathy (6 items) and were framed as a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). A sample item for shared affect was “I was very affected by the emotions in this story,” a sample item for empathic imagination was “I found myself imagining how I would feel in Old Man’s situation,” a sample item for helping motivation was “I would get really involved in trying to help Old Man,” and a sample item for cognitive empathy was “I found it easy to understand Old Man’s reactions.” The subscales showed acceptable to good reliability for shared effect (Cronbach’s α = 0.79), empathic imagination (Cronbach’s α = 0.84), helping motivation (Cronbach’s α = 0.80), and cognitive empathy (Cronbach’s α = 0.82).
Ageism
This factor was measured using the attitudes toward old people scale 41 in a pre- and post-experiment repeated measures format. The scale had 34 items and was framed as a 6-point Likert scale from 1 (strongly disagree) to 6 (strongly agree). A sample item was “It would probably be better if most old people lived in residential units with people their own age.” The scale showed acceptable reliability before the experiment (Cronbach’s α = 0.79) and good reliability after the experiment (Cronbach’s α = 0.82).
Results
Mean scores for the ageism scale, state empathy, and trait empathy were calculated. Table 1 presents descriptive statistics for the study variables. The analysis excluded n = 39 individuals who had missing data and extreme values based on Tukey’s 1.5 IQR criteria. The results were analyzed with PROCESS model 76 for multicategorical moderated mediation analysis to account for the effect of two moderators on the mediation pathways. 42 Participant sex and trait empathy scores were used as moderators. The mean score for state empathy was used as the mediator. Ageism was the outcome variable, and it was operationalized as difference scores by calculating pre-minus the post-experiment scores. 43 Thus, lower difference scores implied that participants showed a decrease in ageism, whereas higher ageism scores at time 2 would indicate the participant had reduced ageist stereotypes. Table 2 shows the correlation coefficients for all variables. The indirect effects of the conditions on state empathy were bootstrapped with 5,000 bootstrap samples by replacement. 29 A PROCESS model 76 for moderated mediation was constructed to assess the effect of playing vs. watching on the state empathy subfactors as mediators (i.e., shared affect, empathic imagination, and helping motivation) on the ageism difference score. Cognitive empathy was removed to increase the model’s parsimony as this factor was insignificant when tested along with the other three subfactors and when tested by itself.
Descriptive Statistics
Correlations for Study Variables
Correlation is significant at the 0.01 level (two tailed).
Correlation is significant at the 0.05 level (two tailed).
Regarding H1, there was no difference between gameplayers and game spectators regarding ageist stereotype difference scores (B = −0.21, P = 0.15, 95% CI: [−0.51–0.08]). H2 was supported as gameplayers showed increased shared affect (H2a: B = 1.08, P = 0.04, 95% CI: [0.03–2.12]), empathic imagination (H2b: B = 1.70, P < 0.01, 95% CI: [0.70–2.71]), and helping motivation (H2c: B = 1.58, P < 0.01, 95% CI: [0.54–2.62]) compared with those who watched pre-recorded gameplay videos. In addition, trait empathy moderated the effect of playing the game on shared affect (B = −0.71, P < 0.01, 95% CI : [−0.123, −0.20]), empathic imagination (B = −0.96, P < 0.01, 95% CI : [−1.46, −0.47]), and helping motivation (B = −1.81, P < 0.01, 95% CI: [−1.32, −0.30]).
In support of H3a, shared affect mediated the effect of gameplaying on ageism difference scores. Relative to game spectating, gameplaying was associated with higher shared affect, which was also associated with decreased ageism difference scores (B = −0.26, P = 0.03, 95% CI: [−0.48, −0.03]). H3b, H3c, and H3d were disconfirmed as neither empathic imagination (B = 0.09, P = 0.36, 95% CI: [−0.10–0.28]) nor helping motivation (B = 0.002, P = 0.99, 95% CI: [−0.20–0.20]) was significantly associated with ageism difference scores.
H4 was confirmed as trait empathy moderated the effects of interactive media on state empathy subfactors and ageism. Relative to game spectators, conditional focal value of moderator tests indicated that gameplayers who were one standard deviation above the trait empathy average score showed lower shared affect. Men (B = −0.53, P = 0.01, 95% CI: [−0.94, −0.10]) and women (B = −0.70, P = 0.002, 95% CI: [−1.15, -0.25]) who played the game and were one standard deviation above the trait empathy average score had lower shared affect. Similarly, men (B = −0.51, P = 0.01, 95% CI: [−0.92, −0.11]) and women (B = −0.70, P = 0.002, 95% CI: [−1.13, −0.27]) who played the game and were one standard deviation above the trait empathy average showed lower empathic imagination. However, men who played the game and were one standard deviation below the trait empathy average showed increased empathic imagination (B = 0.45, P = 0.04, 95% CI: [0.03–0.87]) and helping motivation (B = 0.59, P = 0.008, 95% CI: [0.151–.03]), although the same effect was not found among women who played the game and were one standard deviation below the trait empathy average. Moreover, women (B = −0.09, P = 0.03, 95% CI: [−0.18, −0.01]) who played the game and were one standard deviation below the trait empathy average had decreased ageism difference scores.
Discussion
Although playing vs. watching the game with an elderly protagonist had no direct effect on ageism difference scores (H1), playing vs. watching the game augmented state empathy subfactors, including shared affect, empathic imagination, and helping motivation (H2). Moreover, shared affect, empathic imagination, and helping motivation mediated the direct effect between playing vs. watching a game featuring an older protagonist on ageism (H3). This result resonated with how playing a video game featuring a protagonist who experienced mental illness bolstered the connection that players felt with the character and indirectly diminished the desire for social distance toward the mentally ill, relative to watching a recording of the game. 20 Similarly, watching a documentary about refugees in virtual reality produced stronger empathy compared with watching a 2D version of the same documentary. 44
The results implied that playing games provided increased modality interactivity relative to gameplay watching, which in turn indirectly decreased ageism by boosting shared affect, empathic imagination, and helping motivation. This was also consistent with how playing an interactive game augmented empathy and willingness to help the ethnic group of people portrayed in the game, relative to reading text conveying the same information. 45 Clicking, hovering, scrolling, and dragging allow individuals to control the pace and outcomes of interactive games. Although the effect of modality interactivity has been traditionally attributed to increases in sensory breadth and depth due to interaction with an interface, interface naturalness in mapping user action, interface intuitiveness, and ease of interface use, 15 our results instead suggested an emotional route to modality interactivity effects as shared affect, empathic imagination, and helping motivation mediated the effect of increased modality interactivity.
Practical implications include that playing video games featuring members of stigmatized groups as protagonists can reliably increase empathy and reduce bias toward the individual game protagonist and the general outgroup the protagonist belongs to. However, watching footage of gameplay featuring members of stigmatized groups as protagonists should be less effective at influencing state empathy and ultimately reducing ageism. Considering the rise of game streamers and gameplay watching fandom (i.e., “let’s play,” YouTube and Twitch.tv videos), 46 our findings imply that such forms of passive game content consumption should be less successful than playing in a game in augmenting empathy and diminishing biases against stigmatized outgroup members in protagonist roles.
In addition, trait empathy reliably moderated the direction and strength of the experimental manipulations on state empathy and ageism (see H4). Men and women with high trait empathy, who played the game, showed lower shared affect and empathic imagination. Women with high trait empathy displayed decreased empathic imagination. This implied a boomerang effect for men and women with high trait empathy as playing the game featuring an older adult protagonist diminished their capacity for shared affect and empathic imagination. Boomerang effects are rooted in media stimuli raising unintended cognitions; individuals’ proper processing, but rejection of a message; and the influence of individual differences such as developmental stage, mental capacity, and gender. 47 For instance, the game featured only a male protagonist and thus may have raised unintended cognitions and decreased relevance for women. Since aging is not considered to be conventionally attractive, 48 gameplaying may have raised concerns about physical looks (i.e., lookism), which may have caused individuals to back away from the intended effect of the experimental manipulations. Future studies should assess the role of unintended cognitions, processing and rejection of a message, and individual differences other than trait empathy in triggering boomerang effects.
Moreover, men low in trait empathy, who played the game, had increased empathic imagination and helping motivation, but this effect was not found among women. Men and women low in trait empathy, who played the game, showed decreased ageism. In addition, men and women who played the game and had average trait empathy scores showed decreased ageism. Gameplaying was more effective in reducing ageism and increasing state empathy among individuals with average and low trait empathy scores. This was consistent with interventions aimed at increasing liking for political candidates based on facial similarity, who were more effective among less polarized individuals. 49
Limitations
The selected game’s story featured a male older adult protagonist. Thus, the results did not account for the effects of games with a female older adult protagonist. In addition, most of the participants were Asian American or Asian and so the results may not generalize to other populations.
Conclusion
Playing an interactive video game featuring older adult protagonists increased shared affect, empathic imagination, and helping motivation, but was less efficacious at directly reducing ageism. Shared affect indirectly mediated the link between gameplaying and decreased ageism. Increases in modality interactivity were more effective in reducing ageism and increasing state empathy among individuals with average and low trait empathy scores. The findings provided encouraging evidence for future ageism-reduction interventions that rely on interactive media use.
Footnotes
Acknowledgment
Thanks to all the research assistants who helped to complete the study.
Data Availability Statement
Author Disclosure Statement
No competing financial interest exists.
Funding Information
The authors did not receive support from any organization for the submitted work.
