It's the kind of scitech innovation that, when I was a bright-eyed young science writer, made me think, "Cool!" But now stories like "OCD and Optogenetics: Lighting the brain up to shut a behavior down" bring out the curmudgeon in me.

The column, by my Scientific American co-blogger Scicurious, actually does a terrific job describing a recent advance in the fast-moving field of optogenetics. Optogenetics involves tweaking the genes of neurons so that they become sensitive to light. Researchers can then trigger or suppress firing by the neurons by stimulating them with light-emitting devices inserted into the brain.

Optogenetics is a technically sweet technology spawned a decade ago from the convergence of genetics, optics, neuroscience and materials science. (For an overview, see the Scientific American article by Karl Deisseroth of Stanford, a leader of optogenetics.) Optogenetics could--in principle--allow much more precise manipulation of the brain than conventional implanted electrodes, let alone drugs, transcranial electromagnetic stimulation or electroconvulsive therapy (a.k.a. shock treatment).

Researchers are exploring the potential of optogenetics for understanding and treating* a wide range of brain-based disorders, including obsessive-compulsive disorder (the focus of Scicurious's column), depression, schizophrenia, Parkinson's and post-traumatic stress disorder. Just last month, a group at MIT reported that it had implanted false memories in optogenetically altered mice, work that one outside researcher called "really mind-blowing." [*Most coverage of optogenetics focuses on its potential for basic research, not treatment. See Clarification below.]

Indeed, optogenetics has inspired lots of glowing coverage. Science writer Ed Yong calls it "revolutionary," and David Dobbs describes a recent experiment as "damned intriguing" and "potentially very significant." In his recent defense of "scientism," which I hammered last week, psychologist Steven Pinker alludes to optogenetics when he extols "genetically engineered neurons that can be controlled with pinpoints of light."

"Word on the street is that a Nobel Prize isn't far off," physicist Leonard Mlodinow gushes, adding that optogenetics "is destined to change the way we treat mental illness, and eventually, even, the way we understand ourselves as human beings."

So what's my problem with optogenetics? Actually, I have several problems. First is the gross oversell. For optogenetics to become an effective method for treating mental illnesses, you need specific knowledge about the illnesses' neural underpinnings. You must know which neurons or neural circuits are overactive or underactive or otherwise abnormal.

But we lack such knowledge about depression, schizophrenia, bipolar disorder or any other major mental illness. As the director of the National Institute of Mental Health recently acknowledged, decades of research have not turned up any clear-cut physiological—that is, neural, genetic or chemical--correlates of the major mental illnesses. How can a brain-manipulation technique alleviate mental illness if we don't know what to manipulate?

Of course, optogenetics enthusiasts hope that optogenetics itself can yield such knowledge, by improving upon conventional electrode-based systems for probing the brain. Implanted electrode systems can relieve symptoms of Parkinson's, obsessive-compulsive disorder, depression and other ailments in some patients. Deep brain stimulation has been sparingly used for clinical applications because it is unreliable and often leads to infections and other serious complications.

Optogenetic methods would probably be at least as problematic as electrode-based methods. Optogenetics not only requires drilling holes in peoples' skulls and sticking devices inside their brains; it also involves altering the DNA of brain cells with viruses or other means, which makes optogenetics far more unpredictable and invasive than electrode-based systems. Moreover, whereas conventional electrodes can both manipulate and monitor neurons, optogenetics requires separate devices for stimulating and measuring neural activity.

That brings me to a meta-problem I have with optogenetics: I can't get excited about an extremely high-tech, blue-sky, biomedical "breakthrough"—involving complex and hence costly gene therapy and brain surgery--when tens of millions of people in this country still can't afford decent health care.

As I never tire of reminding readers, the U.S. spends far more per capita than any other nation in the world. Yet our life expectancy is about the same as that of Cuba, which spends less than one seventeenth what we do on health care per capita. Technology, far from being the solution to our health-care woes, is part of the problem. Expensive, high-tech tests and treatments have driven up costs of medicine while often impairing peoples' health.

If optogenetic treatments ever turn out to be viable, my guess is that they will be reserved for the wealthy—or perhaps for American soldiers, both injured and healthy. As I pointed out recently, the Pentagon is a major funder of brain research in the U.S. The Defense Advanced Research Projects Agency is funding optogenetics research at Stanford, Brown and elsewhere.

I hope that you keep all these caveats in mind the next time you read a story like "A Laser Light Show in the Brain," in which psychologist and New Yorker blogger Gary Marcus calls optogenetics a "godsend" that allows investigators to "direct symphonies of light-induced neural activity inside the brain." Give me a break. Or as the old New Yorker used to say: "Block that metaphor!"

Postscript: Some commenters on this column have argued that optogenetics researchers have never claimed that the technique might be used to treat mental disorders in humans. Here are examples of researchers discussing therapeutic applications.

*In a 2010 article in the Journal of Neuroscience, "ANTIDEPRESSANT EFFECT OF OPTOGENETIC STIMULATION OF THE MEDIAL PREFRONTAL CORTEX," a group of 13 researchers, including Karl Deisseroth of Stanford, states that "as electrophysiological data discerning specific patterns of normal cortical activity become available, it will become feasible to induce a particular pattern of activity (e.g., via optogenetic activation, direct brain stimulation, or pharmacological manipulation) to eliminate certain behavioral disturbances manifested during the course of clinical depression (e.g., anhedonia, social withdrawal, etc.)."

In a 2011 TED talk, optogenetics pioneer Ed Boyden of MIT talks about the promise of "optical neural control therapy" and "optical prosthetics" for a wide range of brain disorders, notably epilepsy and blindness. He refers to his experiments on mice as "pre-clinical testing."

In this 2013 press release from the University of Oxford, optogenetics pioneer Gero Miesenböck states that optogenetics "could be a means to identify nerve cell groups that cause specific diseases as targets for medicines. In the more distant future, there could be the possibility of using optogenetic manipulations directly in humans, in order to restore neural signals that have been corrupted or lost because of injury or disease."

In a 2011 review paper in Medical Hypotheses, Avi Peled, a psychiatrist at Technion, proposes: "In light of new optogenetic technology time is ripe to seriously consider optional targets of intervention in the brain of schizophrenia patients...optogenetic interventions in schizophrenia should begin in the prefrontal cortex and the Globus-Pallidus Subthalamic nuclei systems."

A 2011 article in The New York Times quotes Stanford researcher Krishna V. Shenoy, who is doing DARPA-funded research on optogenetics, saying that optogenetics may be able to help veterans with brain injuries: "Current systems can move a prosthetic arm to a cup, but without an artificial sense of touch it’s very difficult to pick it up without either dropping or crushing it… By feeding information from sensors on the prosthetic fingertips directly back into the brain using optogenetics, one could in principle provide a high-fidelity artificial sense of touch."

*Clarification: Twitter Tussles with Ed Yong and others have persuaded me that I overstated the degree to which coverage of optogenetics has focused on its potential as a treatment rather than research tool. The theme of the stories by Ed and Scicurious--and of the 2010 Scientific American article by Deisseroth--is that optogenetics can lead to insights into brain disorders. These insights may then lead to better treatments, but not necessarily optogenetic ones (although as my Postscript shows, Deisseroth and others have raised that possibility). But the insights-into-mental-illness angle has also been over-hyped, for the following reasons: First, optogenetics is so invasive that it is unlikely to be tested for research purposes on even the most disabled, desperate human patients any time soon, if ever. Second, research on mice, monkeys and other animals provides limited insights--at best--into complex human illnesses such as depression, bipolar disorder and schizophrenia (or our knowledge of these disorders wouldn't still be so appallingly primitive). Finally, optogenetics alters the cells and circuits it seeks to study so much that experimental results might not apply to unaltered tissue. For a thoughtful discussion of the limits of optogenetics compared to other neuro-research tools, see this new post by neuroscientist Mark G. Baxter.

**Addendum: See also my followup post on optogenetucs, in which I respond to other criticisms:

Image: Optogenetics Resource Center,