Making older adults perceive digital agents in serious games as human-like is harder than expected
Gaming is not just for the youngest generations. Older adults (age > 50 years) also regularly play games, often games with an educational character such as brain training games. In many of these games, digital agents are present to guide the player through the game. Two studies in PLOS ONE investigate whether ascribing human characteristics such as emotions and agency to digital agents (also referred to as anthropomorphism) positively affects older adults’ game play experience. The studies show that anthropomorphising digital agents increases game players’ perceived humanness of the agent on the short term, but this effect disappears after they play the game for a longer period of time.
- Game designers should carefully consider how to invest in increasing the humanness of digital agents in serious games for older adults. Providing a story in which the agent is described in a humanlike way doesn’t seem to increase players’ perceptions of the agent’s human-likeness nor does it positively affects their gaming behavior (number of games played, time of playing, points gained) on the long run.
Study 1: 57 older adults (mean age: 71, age range: 65-88, 40% female)
Study 2: 42 adults (mean age: 63, age range: 50-82, 69% female)
In study 1, the participants read a story in which a digital agent named Einstein was described in human-like characteristics or a story in which the same digital agent was described in a mere neutral way. Perceived human-likeness was measured directly after reading the story (Study 1). In Study 2, the participants first read one of the two stories about digital agent Einstein from study 1. Then, they played the brain training game Einstein Brain Trainer HD on their smartphone for three weeks. In addition to perceived human-likeness, gaming data were assessed (number of games played, time of playing, points gained).
Facts and findings
- The participants who read the story in which the digital agent was described in human-like characters perceived the digital agent as being more human-like directly after reading the story, than the participants who read the story in which the agent was described in a more neutral way.
- However, after three weeks of playing the game, such positive effects of ascribing human-like characteristics to the digital agent on the participants perceived humanness of the agent were not found.
- Also, ascribing human-like characteristics to the digital agent did not affect the participants gaming behaviour (number of games played, time of playing, points gained)