Soliloquy: Fostering Poetry Comprehension Using an Interactive Think-Aloud Visualization

Title of the Paper

Soliloquy: Fostering Poetry Comprehension Using an Interactive Think-aloud Visualization

Paper Information

• Domain: Educational Technology and Poetry Reading Comprehension
• Keywords: Poetry reading, visualization interface, think-aloud, learning support, multimedia learning, worked examples, educational technology, cognitive load, adaptive learning, personalized teaching

Research Background and Issues

• Identified Problems or Challenges:

  1. Reading and understanding complex texts (such as poetry) is particularly challenging for novices, as the cognitive processes involved are mostly internal and invisible.
  2. Educational technologies often focus on simpler expository texts, aiming to enhance comprehension through "active reading," but provide insufficient support for complex texts.
  3. Current teaching methods (e.g., classroom activities like "think-aloud") are not well-integrated with technology and are difficult to apply effectively in remote or interactive environments.

• Significance of the Problem: Reading poetry requires mastering cognitive strategies such as decoding metaphors and analyzing structure, which not only challenge learners' comprehension abilities but also significantly influence their appreciation of poetry as an art form.

• Research Motivation and Related Work:

  • Think-aloud has been proven to be an effective teaching strategy, but it remains difficult to replicate and enhance through technology.
  • Previous studies have typically focused on applying intelligent tutoring systems to expository texts or supporting graphical representations, with limited research and technology addressing complex texts like poetry.
  • This study aims to develop and validate a visualized "think-aloud" interface to help novices learn cognitive strategies for interpreting poetry.

Proposed Solution

• Proposed Method or Solution:
  This study developed an interactive interface called Soliloquy, which uses visual animations and pop-up prompts to simulate the think-aloud process of experts. This enables learners to directly observe the focus of experts' attention, their thought processes, and their commentary while reading poetry. The system includes the following key features:

  1. Text Animation: Gradual shading highlights the current focus of the expert's attention, with fading animations simulating the reading sequence.
  2. Pop-up Tooltips: Dynamically display the expert's thoughts on specific text segments, with optional audio prompts.
  3. User Interaction: Users must click on the prompt windows to proceed to the next reading segment, allowing them to control the learning pace.

• Innovative Contributions:

  1. Transforming traditional classroom think-aloud activities into a novel interactive technology, addressing the gap in educational technology for teaching complex texts.
  2. Visualizing the attention trajectory and cognitive dynamics of expert readers through animations and visual markers, making invisible cognitive processes visible.
  3. Considering individual differences in learning, such as reading proficiency and cognitive load.

• Implementation Steps and Key Technologies:

  1. Developing a web-based single-page application compatible with modern browsers.
  2. Designing interface functionalities, including dynamic text presentation and interactive think-aloud pop-ups.
  3. Investigating two variables: the level of think-aloud support (high/low support) and the inclusion of audio prompts (partial redundancy design).
  4. Evaluating the system's effectiveness through two rounds of experiments (Amazon Mechanical Turk crowdsourcing and controlled experiments).

Research Findings

• Specific Findings:

  1. The high-support think-aloud design significantly improved participants' poetry comprehension scores, indicating that providing more interpretive details effectively reduces cognitive load.
  2. Audio redundancy did not significantly enhance learning outcomes when users controlled the learning pace.
  3. In comparison to traditional static text presentation, the Soliloquy visualization interface did not show significant advantages, but individual differences played a substantial role.

• Comparison with Existing Solutions and Advantages:

  • The dynamic design of Soliloquy, compared to static text controls, more closely simulates the process of expert analysis, providing step-by-step cognitive support for learners.
  • The system inherits the strengths of intelligent tutoring tools while overcoming the limitations of traditional technologies in teaching complex literary texts.

• Experimental or Evaluation Results:

  1. The first experiment was conducted on Amazon MTurk, with 254 participants. After excluding some due to validation issues, data from 141 participants were analyzed.
  2. The dynamic interface showed no significant improvement in reading comprehension scores (static text vs. Soliloquy), but the high-support condition outperformed the low-support condition.
  3. The second think-aloud experiment revealed different interaction patterns between the two interfaces (e.g., static text users tended to browse the entire text before reviewing expert commentary).
  4. Analysis of "extraneous load" in cognitive load indicated a more significant impact on the dynamic interface.

• Limitations and Future Directions:

  1. The system's learning effectiveness may be strongly influenced by individual differences (e.g., experience and proficiency). Future work should explore more personalized adaptation features.
  2. The technical validation was limited to a single poem and did not cover other poetic forms or longer texts.
  3. The design of research tools (e.g., multiple-choice tests) may not fully capture the openness and ambiguity of complex texts.
  4. Future studies could test long-term usage effects, develop multimodal reading modes, and enhance designs that promote a sense of user control.