Smoothness & Stars - The Importance of Unified Shapes

 

In his talk, Dr. Baker discussed how the visual system takes a collection of separate, disconnected dots and forms them into one unified shape. Four experiments tested this by examining when a dot looks like a sharp corner, which dot arrays look the most like shapes, overall mental rotation and shape comparison, and finding a shape hidden in visual noise. Overall, the experiments suggested that the visual system more easily organizes dots into a unified shape when the contour connecting them is smooth rather than sharp. Sharpness can make it look like separate pieces instead of a figure, making the shape harder to register perceptually.

This reminded me of how our brains make it easier for us to find a pattern when we’re looking for one. This connects to how we’re able to create and imagine so freely because we’re used to connecting the dots to register so many shapes on the daily. To explore this concept further, I found the article “Visual Perception Principles in Constellation Creation” by Kelly et al., which examines a similar phenomenon we experience by simply looking up at the sky.

We often recognize groups of stars as constellations, such as Orion (the Hunter), Ursa Major (which contains the Big Dipper), and Ursa Minor (which contains the Little Dipper). But what makes us want to name the stars in this way? What makes us recognize them as patterns, something familiar and perceptible that we can instantly obtain in our minds? These are the kinds of questions the article explores. The researchers asked participants to create constellations under conditions designed to reduce the influence of established ones, allowing them to see how people would form patterns on their own.

The researchers identified three main tendencies: participants connected stars based on similarity, favored brighter stars and those closer together, and considered the angles and spacing between stars to create patterns that felt more even. These findings suggest that we don’t create patterns only because of what we’ve already heard about the stars. Forming shapes around what seems more uniform comes naturally to us.

Together, Dr. Baker’s talk and Kelly et al.’s study made me think about how we look for consistency in visual perception wherever it’s needed, especially in a world that is always changing. In this way, it’s like we’re regulating ourselves through the stability of organizing separate pieces into something we can recognize and understand. This connection matters to me because it suggests that, despite our individual differences, we share the common tendency of turning separate points, whether dots or stars, into a unified shape in order to make sense of the world around us.

        Works Cited

1. Baker, N., & Kellman, P. J. (2024). Shape from dots: a window into abstraction processes in visual perception. Frontiers in Computer Science, 6. https://doi.org/10.3389/fcomp.2024.1367534

2. Kelly, B. A., Kemp, C., Little, D. R., Hamacher, D., & Cropper, S. J. (2024). Visual Perception Principles in Constellation Creation. In Topics in cognitive science (Vol. 16, Issue 1, pp. 25–37). PubMed. https://doi.org/10.1111/tops.12720