[Paper Review] Liger: Combining Interaction Paradigms for Visual Analysis

Bahador Saket, Lei Jiang+2 more|arXiv (Cornell University)|Jul 19, 2019

Data Visualization and Analytics43 references4 citations

TL;DR

Liger proposes a unified visualization tool that combines Manual View Specification (MVS) and Visualization by Demonstration (VbD) using mouse-based interactions. Through a prototype and exploratory study, it demonstrates that users seamlessly switch between or combine both paradigms to construct visualizations more expressively and efficiently, achieving better task performance and user engagement.

ABSTRACT

Visualization tools usually leverage a single interaction paradigm (e.g., manual view specification, visualization by demonstration, etc.), which fosters the process of visualization construction. A large body of work has investigated the effectiveness of individual interaction paradigms, building an understanding of advantages and disadvantages of each in isolation. However, how can we leverage the benefits of multiple interaction paradigms by combining them into a single tool? We currently lack a holistic view of how interaction paradigms that use the same input modality (e.g., mouse) can be combined into a single tool and how people use such tools. To investigate opportunities and challenges in combining paradigms, we first created a multi-paradigm prototype (Liger) that combines two mouse-based interaction paradigms (manual view specification and visualization by demonstration) in a unified tool. We then conducted an exploratory study with Liger, providing initial evidence that people 1) use both paradigms interchangeably, 2) seamlessly switch between paradigms based on the operation at hand, and 3) choose to successfully complete a single operation using a combination of both paradigms.

Motivation & Objective

• To investigate how multiple interaction paradigms can be combined in a single desktop visualization tool using the same input modality (mouse).
• To understand user behavior when interacting with a multi-paradigm visualization interface that supports both Manual View Specification (MVS) and Visualization by Demonstration (VbD).
• To explore whether combining MVS and VbD enhances user expressivity, efficiency, and task completion in visual analysis tasks.
• To identify design patterns and challenges in blending distinct interaction paradigms within a single tool.

Proposed method

• Developed Liger, a prototype visualization tool that natively integrates MVS and VbD using JavaScript, TypeScript, and D3.
• Implemented full support for core visualization operations (e.g., mapping attributes to axes, filtering, sorting, switching visualization types) in both paradigms.
• Designed interactive components to allow users to switch between MVS (drag-and-drop to shelves) and VbD (direct manipulation of marks to demonstrate intent).
• Enabled the system to interpret VbD demonstrations and recommend or apply changes based on user-provided examples.
• Conducted a preliminary study with four participants to observe interaction patterns and inform iterative design of Liger.
• Used qualitative analysis of user verbalizations and actions to identify recurring usage patterns in multi-paradigm interaction.

Experimental results

Research questions

• RQ1 How do users switch between or combine MVS and VbD paradigms during visual analysis tasks?
• RQ2 Which interaction paradigm—MVS or VbD—is preferred for specific operations, and why?
• RQ3 To what extent can combining MVS and VbD improve user expressivity and task efficiency?
• RQ4 What design patterns emerge when users interact with a multi-paradigm visualization tool?
• RQ5 How do users assess and select between paradigms when both are available for the same operation?

Key findings

• Users consistently tried both MVS and VbD first, using exploration to understand the system’s capabilities before settling on preferred methods.
• Participants preferred MVS for operations like switching visualizations due to its perceived efficiency and consistency, while both paradigms were used interchangeably for filtering and color mapping.
• Users frequently combined MVS and VbD to complete a single task, such as using MVS to set a color mapping and then using VbD to refine it with custom color examples.
• The combination of paradigms led to higher perceived control and engagement, particularly when users needed to express nuanced visualization intent.
• The system’s ability to interpret VbD demonstrations and recommend mappings based on user-provided examples significantly enhanced the user experience.
• The iterative design process, informed by the preliminary study, led to a refined Liger prototype that better supported natural, fluid interaction between paradigms.