Abstract
With the global rise of social media platforms, algorithm-driven recommendation systems have improved the relevance of content suggestions but have also introduced issues such as “filter bubbles,” a diminished sense of user control, and barriers to exploration. Most existing studies focus on optimizing algorithms, paying less attention to how users interact with these systems. To address this gap, we conducted an exploratory user survey (N = 30) and identified three core pain points: low utilization of history features, a lack of effective interest-management tools, and obstacles to discovering new content. Drawing on theories of human–AI collaboration and distributed cognition, we derived a set of innovative design principles and used them to develop a human–machine collaborative recommendation prototype. This prototype integrates a Smart Tag Content Management System (STCMS) and an Intelligent Interest Exploration System (IIES). In a within-subjects experiment with 30 participants, our design significantly increased users’ sense of control, improved the efficiency of retrieving past content, and encouraged proactive interest exploration, while effectively reducing information overload and exploration barriers. This study presents and validates a design framework that transforms users from “passive recipients” into “active decision-makers,” demonstrating that granting users greater control and leveraging their history can help alleviate current recommendation- system challenges and foster more transparent, human-centered AI systems.
Keywords
Recommendation systems; Human-AI collaboration; User control; History-centric design
DOI
https://doi.org/10.21606/iasdr.2025.995
Citation
Jiang, J.,and Tang, Z.(2025) History drives recall and exploration: Human-AI collaboration in the design of memory aid recommendation systems, in Chang, C.-Y., and Hsu, Y. (eds.), IASDR 2025: Design Next, 02-05 December, Taiwan. https://doi.org/10.21606/iasdr.2025.995
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Conference Track
Track 4 - Human-Centered AI
History drives recall and exploration: Human-AI collaboration in the design of memory aid recommendation systems
With the global rise of social media platforms, algorithm-driven recommendation systems have improved the relevance of content suggestions but have also introduced issues such as “filter bubbles,” a diminished sense of user control, and barriers to exploration. Most existing studies focus on optimizing algorithms, paying less attention to how users interact with these systems. To address this gap, we conducted an exploratory user survey (N = 30) and identified three core pain points: low utilization of history features, a lack of effective interest-management tools, and obstacles to discovering new content. Drawing on theories of human–AI collaboration and distributed cognition, we derived a set of innovative design principles and used them to develop a human–machine collaborative recommendation prototype. This prototype integrates a Smart Tag Content Management System (STCMS) and an Intelligent Interest Exploration System (IIES). In a within-subjects experiment with 30 participants, our design significantly increased users’ sense of control, improved the efficiency of retrieving past content, and encouraged proactive interest exploration, while effectively reducing information overload and exploration barriers. This study presents and validates a design framework that transforms users from “passive recipients” into “active decision-makers,” demonstrating that granting users greater control and leveraging their history can help alleviate current recommendation- system challenges and foster more transparent, human-centered AI systems.