Function estimation: Quantifying individual differences of hand-drawn functions

Area of Science:

• Cognitive Psychology
• Data Visualization
• Scientific Communication

Background:

• Graphical perception is crucial for interpreting scientific data and media representations.
• Graphs contain both global trends and local variations, posing interpretation challenges.
• Understanding how individuals perceive and reconstruct data from graphs is vital.

Purpose of the Study:

• To investigate how individuals perceive and reconstruct functions from noisy scatterplots.
• To explore the influence of graph characteristics (data points, scale) on function estimation.
• To identify individual differences in graphical data interpretation strategies.

Main Methods:

• A novel function estimation task was designed using scatterplots of noisy data.
• Participants (170 psychology undergraduates) were asked to draw the generating function.
• Data varied in number of points, scale, and underlying true function.

Main Results:

• Graph properties significantly influenced perceived functions; more data points led to smoother estimations.
• Clear individual differences emerged: some tracked local data fluctuations, others focused on global trends.
• Observer experience with mathematical functions showed mixed influence.

Conclusions:

• Graphical perception is influenced by both data presentation and individual interpretation strategies.
• Some individuals prioritize global patterns, while others focus on local details in data visualization.
• Further research is needed to understand the cognitive mechanisms behind these interpretation differences.