“Friending” Marieke van Vugt online or meeting her in person might obscure her profession. A long way from coming across as the logic-obsessed, lab-leashed scientist, van Vugt is warm, open, inviting. Her Facebook photographs depict her and her friends—an actual ballet troupe—stretching across colorful stages. But van Vugt is also a veteran of the contemplative science community. She’s attended nine out of 10 Mind & Life Summer Research Institutes (SRIs). She’s a professor of cognitive modeling in the department of artificial intelligence at the University of Groningen in the Netherlands. And she’s working on a computational model of meditation—more on that later.
Like many people, van Vugt learned of Mind & Life through Daniel Goleman’s 2003 best seller, Destructive Emotions. It was after that experience that she attended her first SRI. “Those were pioneering events,” van Vugt says, noting how in the early 2000s, at the first SRIs, the field of meditation research was relatively unrefined. While enormous conceptual and methodological strides have been taken since, and researchers have begun to capture the attention of mass media and such governmental organizations as the National Institutes of Health, van Vugt says that, “back then, no one had any idea. It was great. There were all these amazing people like Daniel Kahneman, Anne Treisman, and Jonathan Cohen. And they were giving talks about how we could study meditation.”
Van Vugt says she was struck by how those SRIs legitimized the field to the point that her work became less suspect, even to those in the academic circles around her. During graduate school at the University of Pennsylvania, her doctoral advisor had been reluctant to consider the merits of meditation research. Though not a fan of contemplative practices, he nevertheless gave van Vugt intense training in cognitive science and computational modeling that allowed her to embark on what she calls “a lifelong quest” to fuse her interests and help develop the most advanced cognitive models of meditation possible.
“I often joke that I’m going to build a meditating computer,” she laughs, “but that is [really] the best description of what I’m doing.”
“Cognitive models” are not as abstruse as they sound. In essence, they are a type of map that charts mental processes by simulating human behavior through technology. Within the field of artificial intelligence (AI), the designing, building, and testing of cognitive models is one of the crucial ways scientists explore, and experiment with, intelligent systems.
Traditionally, cognitive models have been used to try and understand the underlying architecture of our minds—relatively simple models, for example, can depict the process of turning short-term memories into long-term memories—but they have also proven useful in making predictions about behavior. In van Vugt’s case, her cognitive model attempts to map mental processes involved with meditation. With a sufficiently advanced model, van Vugt theorizes that she could program carefully defined inputs related to “mindfulness” into a computer and observe the system’s outputs: gauging whether her formulas decrease stress reactions, improve attention, and so on.
For van Vugt, it’s all about finding out—and then telling—a computer, very precisely, what it needs to do in order to simulate meditation, studying what kinds of cognitive operations are involved, and then interpreting the results that emerge.
But why would anybody be interested in developing a simulation of meditation instead of just doing the real thing? Isn’t that a bit like understanding running by watching someone jog rather than pounding the pavement oneself?
It turns out that some of the most tangible benefits of models and simulations reside in their ability to predict the effects of different kinds of behaviors. For example, van Vugt has become interested in the “sense of spaciousness that’s deliberately cultivated through meditation.”
What she means is that everyday human cognition—or mental activity—operates like a “thought pump,” constantly churning out ideas, memories, emotions, and other aspects of consciousness. Meditation is one way to turn the dial down on that pump and increase the feeling of “spaciousness” that exists between thoughts. One theory is that people with depression and anxiety get stuck in runaway trains of thoughts and emotions, otherwise known as “rumination.” They have little control over their thought pumps and don’t get to experience the liberating feeling of spaciousness associated with contemplative practices. Therefore, if thought pumps, rumination, attention, working memory, and “spaciousness” can all be simulated computationally, scientists like van Vugt may be able to better understand what real-life measures can be taken to untangle our thoughts, to “return more to spaciousness,” and assume more control over our thought pumps.
By manipulating different variables, scientists can also use such cognitive models to predict which meditation practices would help certain types of people, or what guidance might be most beneficial for specific kinds of psychological troubles. In the same way a crash test dummy reveals the dynamics of accidents on a virtual body in order to improve car safety, computer models of our minds might reveal the dynamics of brain patterns that can improve real-life mental health.
Models may come in handy in other ways, as well: as visual aids for meditation teachers, something they can show to students to demonstrate activity in the mind and brain as it undergoes meditation. “I found that thinking about this model is helpful for me,” says van Vugt, “because it pushes me to analyze a little bit more what’s going on in my own practice.”
For now, computational models remain pretty rudimentary, and there is no consensus on the best way to simulate cognition, in part because there isn’t even agreement on what models of the mind should look like.
One source of help, however, is input from contemplative scholars such as Andrew Olendzki and Maria Heim. Olendzki, a senior scholar at the Barre Center for Buddhist Studies, and Heim, an associate professor of religion at Amherst College, are the kinds of scholars unavailable to van Vugt at her university—and ones she has access to thanks to the Mind & Life Visiting Scholars program. Cognitive scientists like van Vugt usually build their models using standard behavioral measures such as “accuracy” and “response time,” but metrics are not as simple when it comes to modeling meditation. Cognitive models of meditation, according to van Vugt, “have to rely on introspective measures [rooted in] Buddhist psychology.” To that end, it’s fruitful for scientists like van Vugt to collaborate with scholars from wisdom traditions.
“I have had great conversations with Andrew about how I could structure the mind-moments that occur during meditation. We found that the affective components of experience [such as desire and aversion] are crucial in Buddhist psychology but largely absent in Western [cognitive models]. And Maria has enlightened me about the dynamic nature of mental factors, and how [certain] inputs can be experienced differently based on what feelings and volitions and other mental factors are present just at that moment. Buddhist psychology is just a whole lot more complicated than I had anticipated!”
Currently, van Vugt’s meditation-related work is uncommon in the field of cognitive science, and yet her goal is that its applications become widespread.
“I hope that it will basically improve people’s lives, because that’s the bottom line,” she says, pausing for a moment to think. “A lot of people find it really inspiring when they learn about the scientific understandings of what’s going on in meditation. Me, too. I find I can better understand what I’m doing, why I’m doing certain things—or what I should be doing when I’m practicing—when I know a bit more about the science. I gave a talk a few months ago at my own meditation center, the meditation center where I sit, and people said that it inspired them to go back to the cushion and practice. That’s definitely a benefit.”
Van Vugt’s work is catching on beyond the meditation room. Even with her old colleagues. “I’m going to Philadelphia to give a talk at my PhD advisor’s lab,” she smiles. “He even seemed somewhat interested in my model of meditation, so we’ll see. Maybe things will change!”