On a whiteboard in a spacious sixth-floor office in MIT’s Stata Center, there isn’t a morsel of free space. To an outsider, it is filled with nonsense: strange-looking symbols written in bright green with an occasional red line crisscrossing the labyrinth of mathematics. But to its owner, the scribbling on the board may represent steps toward broadening our understanding of perhaps the most significant innovation of the past century: computers.

The whiteboard belongs to Scott Aaronson, 31, Associate Professor of Computer Science at MIT. His curly brown hair is nearly as unkempt as his hopelessly messy desk, but the boyish energy he radiates when he speaks parallels the passion for his work evident from his whiteboard. While much of Aaronson’s research focuses on computation’s limits, he explains that, in fact, his interest in the subject stems from an early fascination with the seemingly infinite range of possibilities for computers.

He lights up as he speaks nostalgically of how he first became interested in computer science as a child – through video games. To him, video games were “universes in miniature,” which he wanted to understand. Soon enough, he learned to create them himself. “Programming was amazing to me,” he recalls; Aaronson was especially fascinated by computers’ ability to respond to human commands. However, he was curious about the full extent of their power; he wondered whether there could be a programming language that could achieve something more powerful than all others. Much to his disappointment, he later learned that a concept known as Turing universality implies that all programming languages are, in essence, the same.

It was questions like that one – and his realization that he had a greater aptitude for mathematics than for software design – that drew Aaronson toward theoretical computer science. A vast and quickly-growing field of study, theoretical computer science seeks to understand the possibilities and impossibilities of computing. Typical questions in the field include, “What is the fastest way for a computer to achieve this particular task?” and “Which problems are too hard for computers to solve efficiently?” As a sixteen-year-old undergraduate at Cornell University, Aaronson was unsure whether he would be able to do something new in the field, but when he read about the relatively young subject of quantum computation, he knew he had found his calling.

Quantum computation, Aaronson explains, is the most powerful model for computation we have, and he adds that it is unlikely that any other model will ever be more powerful. While classical computers perform calculations by manipulating long strings of zeroes and ones, or bits, quantum computers instead manipulate quantum bits, or qubits, which arise from the superposition of the quantum states of particles. That is, while the switches in ordinary computers must be either on or off, a quantum computer employs quantum mechanical “switches” that can be some combination of on and off simultaneously.

Quantum computers have a wide range of potential applications in the scientific world and beyond. Superposition allows quantum computers to process great amounts of information at the same time, and thus significantly accelerating many classical computations. For example, some computational tasks, such as the prime factorization of very large numbers, that are nigh impossible to perform classically, can be performed exponentially more quickly with quantum-computational algorithms. Perhaps the most important prospective application of quantum computers is for the simulation of quantum mechanical systems, which is of great interest to physicists.

In order to build a quantum computer, however, one needs to construct a delicate quantum system that avoids interaction with the outside world, including direct observation, which would alter the states of the system’s particles. The need for this delicacy is exactly what makes quantum computation difficult – isolating and controlling the necessary minuscule particles, such as photons, is no small feat. As a result, the construction of real quantum computers is still in its infancy: while a theoretical quantum computer can perform prime factorization much more quickly than an ordinary computer, those that have been built cannot get far past 21 = 3 × 7.

Some skeptics believe that quantum computation is simply too difficult to ever be practical, and Aaronson acknowledges a strong likelihood that he will never live to see a universal quantum computer, one that can perform any quantum computation. However, as no known law of physics explicitly rules out that possibility, he anticipates great progress over the next few decades in bringing specific quantum algorithms from theory into practice.

While Aaronson’s own research focuses exclusively on the theoretical aspects of quantum computation, he sees the “dialogue between theory and experiment” as the most exciting aspect of the field. He explains that scientists working on building quantum computers often ask for ideas from theorists for new experiments to try. Conversely, the results of these experiments lead to more theoretical work, especially when the results are unexpected.

Some of Aaronson’s most recent work exemplifies this dialogue. In 2011, he and MIT graduate student Alex Arkhipov began what was they intended to be a “purely theoretical project.” They started with a mathematical problem: associated to any square matrix (that is, a square grid of numbers), there is a numerical value known as the permament, a complicated function of the numbers. While, for large matrices, this number is believed to be too difficult for any classical computer to calculate, Aaronson and Arkhipov theorized that a special type of quantum computer known as a boson-sampling machine would be able to solve the problem.

Fast-forwarding two years to today, four independent teams of researchers have now built rudimentary boson-sampling machines that have verified Aaronson and Arkhipov’s theoretical predictions. Aaronson sees this recent development as an important step for quantum computation, as it shows that a certain very difficult problem in classical computation can be solved with a real-life quantum computer.

In addition to being one of the leading researchers in his field, Aaronson is also one of its leading voices; he remains a favorite among science journalists for comments on the goings-on in theoretical computer science. Perhaps his most famous publication wasn’t research at all, but a controversial blog post he wrote in August of 2010 while on vacation in Israel.

A few days earlier, a paper by HP Labs researcher Vinay Deolalikar claimed to have settled what is widely considered the most important question of theoretical computer science: the P vs. NP problem. Roughly speaking, the problem asks whether there are computational problems for which any computer can easily verify solutions, but no computer can easily find them. Deolalikar appeared to have a proof answering the question in the affirmative.

As Aaronson noted in his blog post, his e-mail inbox was “filling up faster than the Gulf of Mexico,” with colleagues and journalists alike asking for an opinion on the validity of the new proof. Unwilling to interrupt his vacation to study the paper carefully, Aaronson still convinced himself that it had to contain a fatal flaw. He concluded his post by offering $200,000 of his own money, which he adds, “isn’t money [he] can really throw around,” as a prize to Deolalikar if his work was to be validated.

The post drew hundreds of comments on the blog and many more outside of it. While some lauded Aaronson’s attempt for bringing attention to the field, others criticized him for arrogance and rashness, especially in light of the fact that he deferred his reasons for believing the proof to be incorrect to a future post. But Aaronson explains that he sought to highlight the importance of the question itself, rather than his personal opinion on the correctness of Deolalikar’s proof. Specifically, he believes the resolution of what he calls the “flagship question” of theoretical computer science would have a significantly greater impact on his life than the money, a belief which one cannot help but view as a testament to his devotion to his work.

Still, when asked, in light of the backlash to the post, whether he would do it again, he hesitates, and replies with a smile that hints of embarrassment, “Maybe not.”

While his work blends ideas from computer science, mathematics and theoretical physics, Aaronson insists that he is a computer scientist at heart. Ultimately, he seeks to understand the universe behind the video games that formed his childhood obsession. And the next great step in doing so may just pop up one day on his whiteboard.