Analysis of Target Assistance Techniques in Pointing Tasks

This project investigates the effectiveness of various target assistance techniques in pointing tasks, with a focus on two techniques: Sticky Targets and Target Gravity. The experimental setup and data analysis were designed to evaluate these techniques through controlled trials, measuring their impact on pointing accuracy and completion time.

Screen.Recording.2025-02-17.at.10.05.28.PM.mov

File Structure

• Data Files: Stored in the data/ directory.
  • data/increasing and data/decreasing: Contain individual data files for the two experimental orders. These are split for easier plotting.
• Plot Scripts: Located in the plots/ directory.
  • plots/increasing and plots/decreasing: Contain individual plots for each condition.
  • Other plots in plots/ are included in the report.
• Report: The final report summarizing the experiment is located in the root directory as report.pdf.

Part 1: Fitts’ Law Targeting with Artificial Cursor

• A fullscreen canvas displays 30 randomly-placed, non-overlapping circular targets with radii ranging from 50 to 150 pixels.
• An artificial cursor is implemented using java.awt.Robot, and the real cursor is hidden during the experiment.
• Performance data is recorded for each trial, including:
  • Completion time
  • Errors
  • Fitts’ Index of Difficulty (ID)
• The Fitts’ ID is calculated using the formula provided in the class notes.
• Instructions are displayed before each condition, and trials begin immediately after the previous one completes.

Part 2: Sticky Targets Technique

• The Sticky Targets technique is fully implemented.
• The C:D (cursor to target) ratio of the artificial cursor is adjusted when crossing a target, based on the strength level (low, medium, high).
• A line-circle intersection algorithm ensures accurate detection of cursor crossings, addressing any sampling issues.
• The delta vector (dx, dy) is adjusted before updating the artificial cursor’s position.

Part 3: Target Gravity Technique

• The Target Gravity technique is fully implemented.
• For each mouse movement, gravity vectors are calculated for targets within a 500-pixel threshold.
• These vectors are scaled according to the strength level (low, medium, high) and added to the artificial cursor’s movement.

Part 4: Experimental Data and Report

• Data has been gathered for all conditions:
  • Sticky Targets: Zero, low, medium, and high strength.
  • Target Gravity: Zero, low, medium, and high strength.
  • Conditions were tested in both orders:
    • Sticky then Gravity (increasing strength)
    • Gravity then Sticky (decreasing strength)
• Charts are included showing the relationship between Fitts’ ID and completion time for each condition, with best-fit lines and R-squared values.
• A detailed report summarizes the goals, procedures, results, and discussion of the findings. This is located in the report.pdf file.

Running the Experiment

To run the experiment:

1. If you just want to run the FittsPoint program:

   • Simply execute FittsPoint to start the experiment.
   • Results will be saved in the data/ directory.

2. If you would like to generate the plots for the data:

   • First, activate a virtual environment:

     python -m venv venv
     source venv/bin/activate  # On Windows use `venv\Scripts\activate`

   • Install the required dependencies:

     pip install -r requirements.txt

   • Follow the plotting scripts located in the plots/ directory to generate the desired plots for the different conditions.