- The paper proposes applying Human-Computer Interaction methodologies, including Fitts' Law and device taxonomies, to systematically evaluate input devices for musical expression.
- The paper defines musical-specific evaluation tasks for controllers, focusing on performance, modulation, timing, and critical metrics like learnability and explorability.
- The paper suggests a bidirectional flow of knowledge between HCI and music technology to enhance both controller design and the general understanding of input device evaluation.
This paper, presented at the CHI'01 Workshop on New Interfaces for Musical Expression (NIME-01), explores the synergies between Human-Computer Interaction (HCI) and the design of input devices for musical expression. The authors analyze existing HCI methodologies for evaluating input devices and suggest their application in the musical domain. The central thesis is that while traditional HCI focuses on tasks like graphical manipulation, similar rigorous evaluations can be applied to the design of musical controllers.
Review of HCI Research and Its Applicability to Music
The authors begin by reviewing the significant body of research within HCI concerning input devices. This includes the evaluation of device performance using tasks such as pursuit tracking, target acquisition, and the application of models like Fitts' Law. These studies have focused primarily on graphical interfaces but have yielded methodologies that could be adapted for musical contexts. For instance, they propose using musical tasks that mirror the representative tasks in HCI, like pitch or loudness modulation as analogous to target acquisition in HCI.
Fitts' Law, a staple in HCI for predicting human movement efficiency, is highlighted for its utility in translating device performance into comparable indices. Despite debates over its applicability in varying contexts, its foundational principles remain relevant. Similarly, Meyer's Law, building on Fitts' concepts, provides insight into sub-movements in aimed tasks, which can be extrapolated to musical tasks involving multiple dimensions of control.
Furthermore, the paper discusses taxonomies of input devices, such as those proposed by Buxton and Card et al., which classify devices based on the physical variables sensed and their operational characteristics. These taxonomies are vital for understanding the mechanical and sensory attributes of musical controllers.
Musical Context and Task Evaluation
The paper moves to define a set of tasks specifically tailored for the evaluation of musical controllers. These include the performance of isolated tones, scales, arpeggios, and phrases, alongside the modulation of continuous features such as timbre and amplitude. The proposed tasks also account for critical musical elements like timing and rhythm, which lack direct counterparts in traditional HCI tasks. This highlights the complexity and multidimensionality of musical expression.
Three significant considerations—learnability, explorability, and feature controllability—are proposed as essential metrics for evaluating controllers. These concepts emphasize the importance of user experience in the design and assessment of musical input devices, factoring in the time required for mastery and the breadth of expressive capability.
Implications and Future Directions
The implications of this research are twofold: augmenting device design methodologies within music technology and contributing to a more nuanced understanding of input device evaluation in HCI. The paper advocates for a bidirectional flow of knowledge, where insights from one field can fuel innovation in the other. By transferring methodologies, the development of musical controllers can benefit significantly from the structured approaches matured in HCI.
As interdisciplinary fields continue to evolve, the integration of HCI principles into the music domain offers promising avenues for enhancing artistic expression. Future developments may involve refining task definitions and validation methodologies, considering subjective aesthetic factors alongside quantitative metrics. Additionally, exploring advanced mapping strategies and incorporating feedback loops from musicians will be crucial for realizing the full potential of novel musical interfaces.
In conclusion, this paper provides a comprehensive framework for leveraging HCI insights in the evaluation and design of musical input devices. It calls for a systematized approach that maintains the aesthetical fidelity required by musicians while facilitating more robust performance measurement and comparison.