Skip to content

Toward a science of spirituality

March 23, 2012

The idea of seeing life through someone else’s eyes has long been a theme in our fiction and our fascination. The opportunity to access the inner world of another human was accounted famously in the cult classic, Being John Malcovich. Variations on the theme have been showcased in a lengthy roster of Hollywood productions, such as, Freaky Friday, The Change Up, and an unfortunate number of Look Who’s Talking productions, to name a few.

I’m pleased to say that we are on our way toward technologies that are converting past fiction into present reality.

Mind reading
In June of 2009, the CBS show 60 Minutes reported on the work of Marcel Just and Tim Mitchell at Carnegie Mellon University. Using functional magnetic resonance imaging (fMRI), they realized that individual words elicit unique activation patterns across the brain, and that these patterns are so similar from person to person, that the activation patterns can be used to predict a word that someone is thinking, based solely upon the pattern of brain activity detected in that individual. It is a thrilling application for brain imaging that has set the stage for numerous other brain decoding ventures.

In June of 2010, the New York Times reported on the work of Tor Wager, a neuroscientist at the University of Colorado, who is doing similar work to decode brain patterns associated with pain and basic emotions. Dr. Wager’s work has myriad potential applications, including investigations into the placebo effect, as well as investigations in chronic pain, and psychiatric examination of mood.

Another hotly active area of research into decoding the brain pertains to sophisticated methods for using fMRI in lie detection. A landmark legal case in 2010 submitted evidence from functional brain imaging into the courtroom in order to establish the likelihood that testimony given under oath was deceptive. Although the court deemed the technology inadmissible, a host of academic and commercial enterprises are pursuing the ability to discern intention and reliability of testimony, based on neural patterns of the witness.

Visual reconstruction
Perhaps the flashiest foray into decoding the human brain comes from the laboratory of Jack Gallant, a professor at the University of California at Berkeley. Gallant and his colleagues built video “dictionaries” by downloading about 18 million seconds of video footage at random for YouTube. They then had volunteers enter into an MRI scanner and begin watching clips from the video dictionaries, recording patterns of neural activity in the visual cortex for each of the clips. Next, volunteers were shown new video segments that did not correspond to any of the videos in the YouTube-based “dictionary.” For each new video clip, the brain activity pattern of the subject during viewing was paired with the 100 closest matches for brain activity patterns from the video dictionary. The video images from the dictionary were then averaged together in proportion to their resemblance of the brain pattern elicited by the dictionary clip and the novel clip, and a video approximation of what the viewer is seeing is constructed solely from an assessment of their brain activity patterns, second to second (see video link above). This is a remarkable accomplishment. Think of it: by only looking at the brain activity patterns of someone, we can create a streaming video that approximates what the individual is seeing.

Putting it all together
Make no mistake: these technologies are in their infancy. But with the combination of word recognition, emotion decoding, perception of intention, and streaming visual reconstruction all derived from brain imaging, one can imagine the refinement of these technologies culminating in a full-scale, multi-sensorial recreation of how another human being is experiencing their world. How long will it take to get there?–it’s anybody’s guess. There will be deeply important applications for such technologies, though, in understanding the mental life of victims of stroke, locked-in-syndrome, persistent vegetative and partially conscious states, medical applications for understanding psychiatry more deeply, possible opportunities for parents to more fully connect to the inner world of a child with autism, in addition to myriad interfaces between these burgeoning technologies, and the domains of law and justice. As with any technology, opportunities for abuses are high, and careful consideration must be taken to ensure the ethical use of all neural technologies as they come to fruition.

But consider, if you will, the potential for peeking into some of humanity’s most mysterious, vexing, and elevating experiences: those described as spiritual, religious, and mystical.

Consider, for example:

  • What does a Sufi mystic really see and feel when they go into a deep trance state?
  • What is the qualitative sensory correlate when a Hindu yogi accesses Bhakti, universal love?
  • What is a Carmelite nun experiencing when she says that she hears the voice of Jesus?
  • When someone claims a spiritual encounter, to what extent are they faithfully recounting authentic internal events?
  • When a believer says they know something to be true, what is the nature of this gnosis?

What if, instead of Being John Malcovich, we were privileged to peek inside the mind of spiritual leaders, and get a direct glimpse at Being Tenzin Gyatso, Joseph Ratzinger, or Thomas Monson?

Perhaps disappointingly to some, the answers to the questions do not deal directly in either confirming or denying the existence of God. Both believer and skeptic might hope to find a smoking gun amidst the firing of synaptic transmissions.  However, as articulated in the 25 September 2006 publication of Neuroscience Letters, the primary objective for a neural study of spiritual and religious experience is neither to prove nor to disprove theology. Rather, it is to vibrantly expand our understanding and, consequently, an appreciation of our humanity. On the one hand, mechanistic detail can add richness to our humanism. At the same time, those believing that we are formed in the image of a Creator might revere the expanding scientific details of our deeper nature as glimpses into the contours of divinity.

Research in meditation
This work is in its fledgling stages. Perhaps the most established in this general arena is the research of Richard Davidson and his colleagues at the University of Wisconsin. Among their many forays into the neural outcomes of longterm Buddhist meditators, findings have demonstrated, for example, substantial increases in the amplitude of gamma frequency waveforms in the brains of experienced meditation practitioners. This both confirms the possibility that qualitative state changes reported by spiritual disciplines may be searchable in physical paradigms, and creates valuable scientific questions about the basic functioning of the neural system itself. To date, Davidson has received the largest research grant from the National Institutes of Health for investigating meditation–a nod to the possibility that practices derived from spiritual traditions may be applicable for the health and wellbeing of a general population.

As we consider spiritual, religious, and mystical experiences, their value isn’t confined to the ecstatic moment proper. Such a valuation designates the religious phenomenon as a sort of cognitive and emotional masturbatory event. However, a primary reason these experiences have become culturally prized is because of what the experiences point to, and what they lead toward. They are seminal events through which a conception of the transcendent is implanted. It gestates inside the individual who has experienced the spiritual or mystical occurrence. And ultimately, birth is given to a new paradigm; new well-being; abiding peace; Tao; moral realization; Bhakti; the Comforter; interconnectivity; enlightenment. In the broadest terms, spirituality–at its best–produces transformative wisdom.

I would submit to you that we are on the cusp of an era wherein we can approach the neurobiological mechanisms involved in experiences that have traditionally been relegated (if not scorned) beyond the margins of empirical inquiry, such as spirituality and wisdom. Such an inclusionary approach may offer the two-fold benefit of helping to mediate fundamentalisms that lead to extremism, while at the same time enlivening a general societal discourse in which these questions of transcendence and wisdom are revisited with renewed vigor and productivity. It represents a potential complementarity between tradition and modernity, and a strong bridge between the disparate realms of humanities and physical sciences. The way I see it, that future looks very bright.

From → Uncategorized

  1. Great Post! Forgot about the youtube study you pointed to. Going to use it in a presentation tomorrow 🙂 Sadly however, the method used to discern what they where looking at doesn’t help when decoding what people see with their inner eye, as it doesn’t employ the same neural circuitry (V1-V5,Mt+). I am however quite interested in how they managed to get past the temporal limitations of the mri machine. Do you have any short explenations at the ready? Else Iwill just start reading up on it, and sadly don’t want to be sucked into another corner of neuroscience before deadlines 🙂

  2. permalink

    Howdy, Turn Twine. You ask a great question. How can streaming video be reconstructed, if fMRI data points are only gathered once every two seconds? That would be a very choppy stop-frame movie if the visual landscape only changed once every two seconds, or even every second (in the case of faster fMRI acquisitions).

    The work around for this problem comes from video averaging. Because the group first constructed a library of millions of seconds of YouTube videos, what they did was compare each fMRI time point in the decoding scenario with all of the other fMRI time points from the encoding library. In other words, at any snapshot of fMRI data in the video reconstruction part of the experiment, they went through the library of videos and found the videos whose corresponding fMRI brain pattern matched most closely to the fMRI snapshot under consideration. Then they took the video clips from the top matches, and averaged them together. So, because each single fMRI snapshot in the encoding library is associated with one or two seconds of video streaming, when you are averaging those together, you will get a streaming video that is the average of the video matches from the library. And since each fMRI snapshot corresponds to one or two seconds of video footage, you are able to roughly approximate the video that the person is watching, even if you only have an fMRI data point from their brain for every second or two in time.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: