This blog is the sixth in a series of guest posts on technology and the brain to celebrate Scientific American Mind’s 10-year anniversary. The magazine’s special November/December issue similarly highlights the interface between code and thought in profiling a future, more digital YOU.

Ask any two people about any process in their lives, be it their morning routines, how they solved a puzzle, or what brings them happiness, and you will likely get two different answers.

The same holds for psychological processes. For instance, anxiety and depression may co-occur for some, whereas for others, depression may occur after an extended period of anxiety, while still in others no relationship exists between the two. Therapists, psychiatrists and other mental health practitioners learn about these person-specific idiosyncrasies and the best ones incorporate this information into their work. That is, the success of their work depends on them understanding the individual nuances of clients in their care.

This starkly contrasts the current standard in the study of psychology. Typically studies are conducted across individuals at a given time and the relationships among emotional, cognitive and behavioral measures are examined to look for patterns across people. This can reap interesting and informative results – such as knowledge that those with a depression diagnosis are more likely to also have an anxiety diagnosis – but it does not tell us much about the individual or what treatment might help them.

Another example is seen in crime. By looking across individuals we might see that certain crimes occur more often in neighborhoods with specific characteristics, such as low income and low cohesion among residents. However, that does not tell us what makes some individuals within that neighborhood commit the crime while others do not. This individual-level information is key to treatment, prevention, and intervention efforts.

Courtesy of Brian Gates.

Neuroimaging studies also benefit from identifying brain processes at the individual level. Diagnostic categories such as Attention Deficit Hyperactivity Disorder and Autism Spectrum Disorder are increasingly considered umbrella diagnoses that cover a heterogeneous set of symptomologies and biological underpinnings. In an article that was just published this year, colleagues from Oregon Health and Science University and The Pennsylvania State University and I used a novel statistical approach that subgroups individuals based on their person-specific brain processes. We scanned the brains of 80 children (including 32 with ADHD) using functional MRI, focusing on a set of brain regions that make up the fronto-parietal network, which enables us to switch tasks and focus. We hypothesized that differences in patterns of connectivity among regions in this network might relate to an ADHD diagnosis. However, because ADHD is a heterogeneous disorder, it was unclear whether children with ADHD would deviate from typical kids in just one way, or if multiple brain patterns might underlie the diagnosis. Instead of grouping individuals based on whether or not they had ADHD, and then looking for differences between the two groups, we set out group kids based on their brain processes first.

We built a model of each child’s brain by measuring how much each region in his or her fronto-parietal network works in concert with the other ones. We then grouped kids together whose brains were characterized by similar sets of connectivity patterns. We found three types of functional brain patterns associated with ADHD diagnoses and two associated with typical brain development. This diversity indicates that more than one biological mechanism exists for ADHD and that at least two patterns of functional connectivity in the front-parietal network may be found in typically developing children. Working with neuroscientist Damien Fair at Oregon Health Sciences University, we now are trying to tie the two biological markers for ADHD to specific sets of symptoms and treatment plans. In the future we may be able to use a similar approach to subdivide other broad diagnostic categories such as autism spectrum disorder that likely have heterogeneous biological underpinnings.

Looking at individual-level information on brain processes could also be useful for figuring out who would best benefit from specific training regimens or learning techniques. Currently, in a project led by cognitive neuroscientist Joseph Hopfinger at the University of North Carolina, we are collecting functional MRI data on college students using the brain-training programs offered by Lumosity, an online brain-training company. Data indicate that Lumosity games can increase cognitive abilities such as working memory, processing speed and attention. We want to identify those individuals most likely to benefit from this training by looking at their brain processes. The results might help us develop cognitive training methods tailored to different types of brains, enabling more people to benefit from the technology.

Instead of comparing brain patterns of those who improve after training with those who do not, we built models of brain processes in individual brains to look for the various processes that predict improvement of cognitive performance after training. This method enables us to separate people who may have scored equally well but who approach cognitive tasks differently, so are heterogeneous in their brain processes. Revealing these potentially various neurobiological underpinnings of improvement will enable researchers to tailor brain-training programs to the needs and deficits of individuals. For example, by looking at patterns that related to no improvement, researchers may be able to develop training protocols that target specific connections among regions. Our work should enable Lumosity to provide products that help a wider group of consumers. Having even better tools for keeping the mind sharp will become increasingly important as the nation’s aging population experiences the expected cognitive decline.

>>Next in the series: “A ‘Miracle’ Technology Can Lift Severe Depression–But Real Recovery Has Only Just Begun”