2D Shape Recognition and its Relation to Brain Visual System Object Recognition

In the article titled “Constant Curvature Segments as Building Blocks of 2D Shape Representation” written by Nicholas Baker, Patrick Garrigan, and Philip J. Kellman, Baker along with the other researchers explored the human visual system, and how this system represents shapes. The research proposed that 2D contour shapes are coded with curvature within their segments which are crucial for shape recognition. Through three experiments, they tested their theory of perceptual coding through curvature segments. 

Experiment 1 detailed constant vs non-constant curvature paths, which tested whether constant curvature paths are easier to sense than non-constant curvature paths. Experiment 2 was similar to the first, but rather some of the constant curvature paths were made linear. Experiment 3 took the curvature values and changed the length of the segments which tested whether it was easier to detect the change in angle or a change in length.  These three experiments together resulted in the finding that constant curvature segments are crucial in the human visual system in representing 2D shapes. 

The article written by Chris Baker, Martin Hebart, and Oliver Contier, titled “Distributed representations of behavior-derived object dimensions in the human visual system” was studied by Max Planck, in which he wrote an article titled “Our brain processes various behaviorally relevant dimensions of objects, not just recognition and categorization”, and the two studies compared in both their focuses, research methods, and connections to visual perception.

In Nicholas Baker's 2D representation article the research showed the different curvature in shapes that allow the brain to understand and recognize shapes. Similarly, in Chris Baker’s research, the 2D shape recognition was taken a step further, introducing colors, and characteristics, and how it contributes to perception. Both of the research articles came to the conclusion that curvature and color characteristics are crucial in recognizing shapes and objects. 

The research methods used in both experiments were very similar, as N. Baker and his partners focused on a more perception based experiment for shapes in which he took a number of shapes and displayed them in flashes, while Planck focused on the aspect of brain imaging, using MRI, to discover the processing through thousands of images and objects. 

In summary, both of the studies are expanding on the already known factors of the human brain, and are doing so using the connection of visual perception. While N. Baker studies the use of contour 2D shapes, Planck, along with his previously studied research used a number of objects, and their link to human perception. 

References

Contier, Oliver, et al. "Distributed Representations of Behaviour-Derived Object Dimensions in the Human Visual System." Nature Human Behaviour, 2024, doi:10.1038/s41562-024-01980-y.

Baker, Nicholas, Patrick Garrigan, and Philip J. Kellman. "Constant Curvature Segments as Building Blocks of 2D Shape Representation." Journal of Experimental Psychology: General, vol. 2, no. 999, 2020, pp. 1-17, doi:10.1037/xge0001007.

"Brain's Visual System Does More Than Categorizing Objects." Technology Networks, Max Planck Institute for Human Cognitive and Brain Sciences, 17 Sept. 2024, www.technologynetworks.com/neuroscience/news/brains-visual-system-does-more-than-categorizing-objects-391032. Accessed 11 Oct. 2024.