A Japanese learning experiment shows why a usable app can still feel cold, and what a design team can do about it.
An app can be easy to use and still leave people flat. Usability research separates that feeling into two measurements: pragmatic quality, which covers getting a task done, and hedonic quality, which covers stimulation, novelty, and how the product makes someone feel. The two do not have to agree.
In the Japanese m-learning study, both apps let students complete every learning scenario. On perspicuity, efficiency, and dependability the existing app scored 2.23 and the newly designed one scored 2.15. That difference is small enough to call a tie. Whatever the new design changed, it did not make the app harder to operate.
Here the gap was wide. The proposed app scored 1.45 while the existing app dropped to -0.44. Participants described the new version as conversational and natural, which traces back to a single design decision: building speaking and listening into the core learning loop instead of treating them as extras.
The team reached that decision by studying how learners picture the task in their heads, then writing user stories from that picture rather than from a feature list.
A better experience score is not the same as better learning. The authors are explicit that this study measured user experience only, with 20 students from similar backgrounds, and did not test whether the participants actually learned Japanese faster.
The transferable lesson holds anyway. When two apps are equally usable, the one that helps people practise the skill they came for is the one they will describe with enthusiasm.
Komang Candra Brata and Adam Hendra Brata (2020). User experience improvement of japanese language mobile learning application through mental model and A/B testing. International Journal of Electrical and Computer Engineering.
