Mind reading?

I first volunteered to be a participant in an fMRI study as a wide-eyed college freshman ten years ago. I was so excited to get to see a picture of my brain, but once I was tightly packed into the scanner, a few worries entered my mind: would it turn out that I had a tumor, or was one of those people with half a brain? Would the experiment show that I’m not very smart, or vulnerable to mental illness? Would the graduate student administering the experiment know what I was thinking?

While my concerns were rather common, they were also rather unfounded. fMRI is rather good at predicting what you are thinking about in laboratory situations where you are given a very short list of things to think about. For example, an fMRI scan can predict whether you are thinking about a face or a place as the mental imagery for places and faces recruits different brain areas. And it is of note that this happens when your subjects are willing and able to think only about faces or houses for a 20 second run inside the scanner.

Earlier, I wrote a little about the analysis of fMRI data. The kind of inferences researchers make in these kinds of studies is in the form of “what area of the brain is more active for task 1 compared to task 2?” In prediction, the question becomes “given a pattern of brain activity, what was the participant seeing/hearing/doing?” Early prediction techniques relied on correlation: a voxel was predictive if its activity to a particular stimulus in one run was more highly correlated with activity to the same stimulus from another run than to activity from a different stimulus. More modern prediction studies make use of machine learning and statistical classifiers such as linear discriminant analysis, and support vector machines in particular. This approach to data analysis has been both popular and fruitful, particularly for vision research. I recommend this review for more on the state-of-the-art.

As impressed as we tend to be with both math-y and brain-y things, it is important to remember that we are still not able to predict arbitrary patterns of brain activity. When you are lying there in the scanner, it is possible to determine if you are daydreaming, but not the contents of your daydream. I have a particular concern over the over-selling of these studies in the titles of both lay and academic papers. Although the technology for fMRI decoding is advancing rapidly, I also do not see the scanner as a place where future civil liberties go to die. Getting readable data from the scanner requires your participant to be very co-operative, so its use as an interrogation device is limited. But for you paranoid types, perhaps you should consider some metal dental work… or even a hair tie!

Sunday soundbites

Reboxetine: bad drug or baddest drug?

On an utterly unrelated note, JAMA finds that fewer of its studies are funded by industry after more rigorous statistical oversight was put in place.

Alleles associated with disease don't decrease your lifespan.

Prediction of mild cognitive impairment using white matter connectivity.

So, what makes you happy?

Research on the factors increasing or decreasing happiness has been of interest to psychologists and economists alike. Early research indicated that, contrary to our intuition, that major life events such as winning the lottery did not change our long-term ratings of our happiness. In other words, after a major life change, you will experience a temporary change in your happiness, but will return to being as happy as before the event. Such findings led to the set-point hypothesis that stated that each of us has an innate level of happiness, and that outside events, even large ones, don’t have a major influence on that set point.

 

Evidence for the set point hypothesis typically comes from longitudinal studies in which the same people rate their happiness each year, and report any major life events that occurred between surveys. From these data, you can correlate happiness as a function of the event by time-locking a sample population’s happiness the year an event occurs, and then seeing how it changes later. Let’s take marriage for example. Although each person in the sample gets married at a different time, the researcher can define year 0 to be the year of marriage for each person. Then, the researcher can look at this person’s happiness ratings both before and after marriage to determine the impact of marriage on happiness rankings. Below is the kind of graph that you get: planning and getting married makes people happy, and this happiness lasts for a couple of years, but after this, people go back to being however happy they were before.

 

Interestingly, although people hedonically adapt to marriage, they do not hedonically adapt to divorce. In other words, although marriage will not cause a permanent change in your happiness, getting a divorce will make you sadder in the long-term. Even more interesting is looking at the initial happiness ratings of people who marry and eventually divorce, and compare them to people who marry and stay married. It turns out that people who stay married were happier than their to-be-married-then-divorced friends even five years before marriage! This can be even before these people met their spouse.

Given that people do not strictly have one happiness set-point, what factors account for long-term changes in happiness? A new study in PNAS examines this question using 25 year longitudinal data from Germany. What I particularly enjoy about this study is that they focused on factors that people can control: although becoming disabled in an accident will likely lead to a long-term change in happiness, it is not something you can readily control. However, you do have control over things like your choice of romantic partner, your degree of religious involvement, your life priorities, whether you exercise, etc. Here is what they found:

-         Focusing on money can’t buy you happiness. People who rated the acquisition of material goods as very important were less happy than people focused on family or volunteerism. Women whose partners were materially focused were particularly sad.

-         You are less happy when you work more or (in particular) less than you would prefer to work.

-         Being with friends and exercising regularly can make you happy.

-         If you are a man, don’t be underweight. If you are a woman, don’t be obese.

-         Choose a partner who is not neurotic.

-         Although the study found a positive effect of religious participation, this is also correlated with altruism, family focus, and social participation, all of which independently increase happiness.

Although each of these factors had a small effect on happiness, all of them seem like good, common sense. I’ll go out for a run now.

Headey B, Muffels R, & Wagner GG (2010). Long-running German panel survey shows that personal and economic choices, not just genes, matter for happiness. Proceedings of the National Academy of Sciences of the United States of America, 107 (42), 17922-6 PMID: 20921399

Subtle influences on choice

Like most people, I like to think that my choices are a result of clear, rational thought. However, our decision processes are far more heuristic than we admit. Two new articles on choice bear this out:

Mantonakis and colleagues studied how the order of items presented to us affects our preferences and choices. In their experiment, several wines were presented to participants to taste and rate. Although participants were told that all wines were from the same varietal (e.g. pinot grigio), in reality, all of the samples were from the exact same wine! If preference and ratings were rational, then the average rating a wine receives by subjects should be the same regardless of whether it was tasted first or last. However, they found that the first wine tasted by participants was preferred over wines in other serial positions, a finding known as the primacy effect.

In the second paper, Krajbich and colleagues modeled decision choices made by subjects between two pieces of junk food. Krajbick brought junk-food-loving, hungry participants into the lab, and asked them to rate how desirable 70 different junk foods are to them. Then, in front of an eye tracker, they were presented with pairs of pictures of these food packages and asked to decide which they would prefer to eat after the experiment. Unsurprisingly, people are quicker to decide when the value of the two choices is very different. However, when the decision is more difficult, participants tended to choose the item they looked at more.

Both of these studies are laboratory demonstrations of things advertisers seems to have known for a while: to get you to buy their product, getting you to look at it early and often might be enough to get you to buy it.

Mantonakis, A., Rodero, P., Lesschaeve, I., & Hastie, R. (2009). Order in Choice: Effects of Serial Position on Preferences Psychological Science, 20 (11), 1309-1312 DOI: 10.1111/j.1467-9280.2009.02453.x

Krajbich I, Armel C, & Rangel A (2010). Visual fixations and the computation and comparison of value in simple choice. Nature neuroscience, 13 (10), 1292-8 PMID: 20835253

How many published studies are actually true?

I’d like to point readers to this excellent new article in The Atlantic on meta-researcher John Ioannidis. Ioannidis is building quite the career on exposing the multiple biases in medical research. He has taken a field to task publishing papers with shy titles such as “Why most research findings are false”. He is rapidly becoming a personal hero of mine.

Ioannidis has examined and formally quantified research biases at all levels of “production”: in which questions are being asked, in the design of experiments, in the analysis of these experiments, and in the presentation and interpretation of the results. “At every step in the process, there is room to distort results, a way to make a stronger claim or to select what is going to be concluded,” says Ioannidis in the article. “There is an intellectual conflict of interest that pressures researchers to find whatever it is that is most likely to get them funded.”

While I have examined some of these biases for both general research and fMRI experiments, it’s worth noting that in the context of medical research, the stakes are even higher as they affect patient care. It is also unfortunate that medical studies are, according to Ioannidis, more likely to contain bias as there are stronger financial interests vested in the results, compared to cognitive neuroscience. 

An unfortunate result of the competitive research environment is a lack of replication of scientific results. Although replication is the gold standard of a result’s truth, there is little acknowledgment, and thus little motivation for researchers to do this, except for the most bold of claims. Without replication, bias in research increases. However, even when a failure to replicate a major study is published, it often gets very little attention. A case in point is the failure to replicate the “Mozart effect”: the finding that listening to 10 minutes of a Mozart sonata significantly increased participants’ performance on a spatial reasoning test. A quick Googling of “Mozart effect” will show you several companies selling you Mozart recordings to increase your child’s IQ, despite the failure to replicate.

It is very easy to get discouraged by this, after all, science should be a science, right? Ioannidis seems less discouraged, and reminds us of the following: “Science is a noble endeavor, but it’s also a low-yield endeavor… I’m not sure that more than a very small percentage of medical research is ever likely to lead to major improvements in clinical outcomes and quality of life. We should be very comfortable with that fact.”

The self-control meta-game

Previously, I wrote about the use of neuroscience in the courtroom as a defense for criminal actions. I asserted that these arguments hold water only insofar as they can demonstrate a clear causal connection between the brain injury and the criminal behavior, and that it was not possible for the defendant to control himself in the presence of such a brain injury.

Although I am a card-carrying pinko, I am enjoying the new book by Gene Heyman, Addiction: A Disorder of Choice. (A longer review post forthcoming once I finish the book). Heyman challenges the view that addiction is a compulsory chronic and relapsing condition. By illustrating historical, cultural and individual differences in drug reactions, he shows that drug dependence can be overcome with will (and massive amounts of effort and motivation). This brings us right back to the question we left off with last time: under what circumstances can we reasonably expect a person to demonstrate self-control?

A common model of self-control posits that exhibiting self-control is an effortful, resource-consuming process. According to this model, a person has a set amount of self-control that can be exhibited before failure and/or “recharge”. A common source of evidence for this model is the fact that exhibiting self-control appears to consume a good deal of glucose. (Of course, this is a very interesting idea for those whose self-control is being directed towards dieting!) Another measure of self-control failure are mistakes on a Stroop test.

A compelling new study examines the limitations of the resource-limitation model of self-control. A first experiment demonstrated that people who do not agree with the resource-limitation model made fewer mistakes on the Stroop test following a cognitively demanding task than did those who professed beliefs in the model. Even stronger was a second experiment where manipulation of participants’ beliefs in the model had the same effects. Of course, like many things in psychology, William James was here before us when he stated “The greatest discovery of my generation is that a human being can alter his life by altering his attitudes”.

Job V, Dweck CS, & Walton GM (2010). Ego Depletion--Is It All in Your Head?: Implicit Theories About Willpower Affect Self-Regulation. Psychological science : a journal of the American Psychological Society / APS PMID: 20876879

Did my brain make me do it?

Our first case is from a 40-year old man who developed a new and intense interest in child pornography.  His sexuality also generally increased, and he found himself frequenting prostitutes even though he never had before.  He was ashamed of his behavior and went to lengths to hide it, and could communicate that it was morally wrong.  However, he then began making sexual advances on his pre-pubescent step-daughter and spoke of raping his landlady.  He was removed from his home, but failed a 12 step sexual addiction program as he could not restrain himself from soliciting sexual favors from the staff and fellow group members.  As he failed the program, he was sentenced to prison, but developed debilitating headaches and balance problems shortly before admission.  An MRI revealed a large tumor in his orbitofrontal lobe, an area associated with self control, executive function and the regulation of social behavior.  Following surgical removal of the tumor, the man was able to successfully complete the sexual addition course, successfully moved back in with his family and no longer had pedophilic or other deviant urges.

Consider, then our second case: the 1992 trial of Herbert Weinstein, a 65-year-old advertising executive who was charged with strangling his wife to death and then, in an effort to make the murder look like a suicide, throwing her body out the window of their Manhattan apartment.  At his neuropsychiatric evaluation, it was found that Mr. Weinstein had a small, subarachnoid cyst in his brain. The defense moved to use this cyst as evidence of Mr. Weinstein’s inability to control, or be responsible for, his behavior. The cyst in Weinstein’s brain has never been linked to mental illness or violent behavior. After a contentious pre-trial hearing about using this evidence, Mr. Weinstein accepted a plea bargain.

Are both of these men equally responsible for their own behavior?

A central tenet of neuroscience is that all behavior is caused by the brain. This sounds simple enough, but given our long intellectual history of separating the mind from brain, we hold very dear to the idea of an “I”, separate from the 3 pounds of electrical meat that is our brain, calling the shots. After all, “I” feel like “I” make decisions that shape my life. If “I” wasn’t responsible for these decisions, if the decisions instead came from the electrical meat, which is determined by the laws of physics, then how is it that “I” decided to wear a blue shirt instead of the red one? More troubling, if “I” am just my brain, and my brain is malfunctioning, am “I” still responsible for my behavior?

People are remarkably consistent in their moral judgments. Therefore, with some confidence, I can predict that you feel that the man from case 1 is less responsible for his behavior than Mr. Weinstein from case 2.  Is this gut-level feeling rational? After all, both men had damage to their brains, and their brains govern their behavior.

The problem with many cases of “my brain made me do it” arguments is that the association between a brain injury and a behavioral problem is not causal evidence that the injury caused the behavior. Another way of saying this is that “correlation does not imply causation”. We are quick to call B.S. on associations that don’t seem to have plausible causal connections: although drowning is associated with ice cream consumption, we do not guess that ice cream causes drowning. In the criminal realm, we are also likely to see through a Twinkie defense, even if “neuro-babble” makes for a more compelling case.

However, in the case of the first man, we are able to establish a causal connection between his brain injury and his bad behavior: he had “normal” behavior (presumably) before and after the tumor. Unfortunately, we cannot surgically repair most malfunctioning brains, so most connections between behavior and brain are speculative.

Beyond the problem of causality is the problem of will. How can we establish that someone absolutely cannot control his behavior? People can exhibit a certain degree of regulation over even autonomic functions using biofeedback or certain styles of meditation. In the lab, feedback from fMRI has been used to train subjects to willfully activate and de-activate a region involved in the perception of pain. However, it is incredibly difficult to willfully change most behaviors. It takes many days of consistent effort to form new habits. Many ex-smokers report that the physical withdrawal from nicotine was much easier to deal with than the reprogramming of one’s automatic response to grab a cigarette in various contexts. Although the politics of how we frame addiction is a larger topic for another post, suffice it to say that there is no consistent agreement what behaviors we expect people to be able to control, and those we don’t.

So, did my brain make me do it? Well, yes, of course it caused my behavior. Am I to be held responsible for this behavior? Given the above difficulties, I have to agree with Michael Gazzaniga who states that this question is to be left to the legal scholar and not the neuroscientist.