A Unique Educational & Career Opportunity with Stephen Wolfram

A unique opportunity to do original research at the frontiers of science, the Wolfram Science Summer School teaches about 50 students from a diverse range of scientific backgrounds how to use Stephen Wolfram's A New Kind of Science and the Wolfram Language to implement projects in their fields of interest. Most of these students are advanced undergraduates and early graduate students, but those in different circumstances are considered. We are looking for students who want to enhance their careers with Wolfram Science and the Wolfram Language. Read more »

Class of 2012

Chen Song

Bio [2012]

Chen Song is a PhD neuroscience student in University College London (London, UK), currently working on how the brain’s wiring pattern and morphology give rise to neural response properties and further to visual perception using a combination of experimental and theoretical approach. Prior to her PhD study, Chen received a BSc in biomedical engineering at Shanghai Jiao Tong University (Shanghai, China). During her BSc, Chen did three years of undergraduate research on visual perception at the Institute of Neuroscience, Chinese Academy of Sciences (Shanghai, China) and a summer research internship at the Max Planck Institute for Mathematics in the Sciences (Leipzig, Germany). From the beginning of her research career, Chen has been extremely interested in complexity systems. Her long-term research goal is to understand how consciousness may emerge from the complexity of brain systems.

Project Title

Perceptual Analysis of Physical Image


Visual perception is not a true reflection of the physical world. Various visual illusions exist where physically different images appear perceptually identical, or physically identical images appear perceptually different. This dissociation between physical world and perceptual world poses challenges for analzying perceptual characteristics of images, such as perceptual similarity between different images, using conventional statistical methods that are based on physical properties of the images. In computer graphics, dithering is a technique that down-samples an image in color/grayscale space (in order to save the storage space) while keeping perceptual distortion of the image to its minimal. However, the image that is physically more similar to the original is not necessarily more similar perceptually. In this project, we developed an algorithm that analyzes the perceptual properties of images based on structure of visual system and principle of visual perception. Next, we applied this algorithm to analyze and compare the level of perceptual distortion of different dithering algorithms and developed a better dithering algorithm.

Favorite Four-Color, Four-State Turing Machine

Rule 668568713360944227701165