Effects of Social Contextual Variation Using Partner Avatars on Memory Acquisition and Retention

This study investigates how partner avatar design affects learning and memory when an avatar serves as a lecturer. Based on earlier research on the environmental context dependency of memory, we hypothesize that the use of diverse partner avatars results in a slower learning rate but better memory retention than that of a constant partner avatar. Accordingly, participants were tasked with memorizing Tagalog–Japanese word pairs. On the first day of the experiment, they repeatedly learned the pairs over six sessions from a partner avatar in an immersive virtual environment. One week later, on the second day of the experiment, they underwent a recall test in a real environment. We employed a between-participants design to compare the following conditions: the varied avatar condition, in which each repetition used a different avatar, and the constant avatar condition, in which the same avatar was used throughout the experiment. Results showed that participants in the varied avatar condition recalled significantly worse during the learning trials on the first day. However, we found no significant difference between conditions in the delayed recall test on the second day. We discuss these effects in relation to the social presence of the partner avatar. This study opens up a novel approach to optimizing the effectiveness of instructor avatars in immersive virtual environments.

💡 Research Summary

This paper investigates how the visual design of partner avatars—virtual agents that act as lecturers—affects learning and memory when presented in an immersive virtual environment (IVE). Drawing on the environmental context‑dependency of memory, the authors hypothesized that varying the avatar’s appearance across repetitions would slow acquisition (reinstatement effect) but improve retention after a delay (multiple‑context effect). To test this, participants were asked to memorize 20 Tagalog‑Japanese word pairs. On Day 1, they completed an initial study trial followed by five retrieval‑practice trials, each accompanied by feedback, while a partner avatar delivered the stimuli through a head‑mounted display. Two between‑participants conditions were used: (1) a constant‑avatar condition, where the same avatar was present for all six trials, and (2) a varied‑avatar condition, where a different avatar appeared for each trial. One week later, participants performed a final recall test in a real‑world setting.

Results showed a clear reinstatement effect: participants in the constant‑avatar condition achieved significantly higher accuracy during the five retrieval‑practice trials than those in the varied‑avatar condition. However, the delayed real‑world test revealed no significant difference between the two groups, indicating that the anticipated multiple‑context benefit did not materialize. Correlational analyses suggested that participants who reported higher social presence of the avatars tended to show stronger memory effects, implying that the perceived “human‑likeness” of the avatar may mediate the observed outcomes.

The study makes three primary contributions. First, it demonstrates for the first time that partner avatars can serve as an environmental context that influences memory, confirming a reinstatement effect in the VR‑learning domain. Second, it fails to find evidence for a multiple‑context effect, suggesting that merely swapping avatar appearances does not sufficiently diversify contextual cues to enhance long‑term retention. Third, it highlights the role of social presence, proposing that enhancing the avatar’s social realism could be a promising avenue for improving VR‑based education.

Limitations include the lack of manipulation of avatar agency (computer‑controlled vs. human‑controlled), limited reporting of sample size and demographic diversity, and the use of a single language pair, which may restrict generalizability. Future work should explore varying avatar behavior, voice, and facial expressions, assess the impact of perceived agency, test different learning materials, and examine longer retention intervals to more thoroughly evaluate the multiple‑context hypothesis. Overall, the paper provides valuable insights into how social contextual cues in virtual environments shape memory processes and offers practical guidance for designing more effective instructional avatars.