Reasoning Training Increases Brain Connectivity Associated with High-Level Cognition

A number of studies across various domains– from juggling to taxi navigation to meditation to music to motor learning to processing speed– demonstrate the importance of experience on patterns of neural connectivity. Finally, the cognitive ability domain is catching up.

In recent years, neuroscientists have discovered a large-scale brain network critical for novel and complex goal-directed problem solving. According to Aron Barbey and colleagues, a major function of this network is the manipulation, integration, and control of distributed patterns of neural activity throughout the brain, including lower-level sensory and motor modules. This neural integrative architecture– sometimes referred to as the prefrontal parietal network– involves efficient and reliable communication between specific areas of the lateral prefrontal cortex (critical for high-level abstract integration) and posterior parietal lobe (critical for sensory integration). Here’s an illustration of the key regions of this neural architecture, along with the critical white matter tract binding these regions together into a coordinated network:

One of the most exciting findings in the past few years is that coordination among the key players of this network is substantially affected by training and experience. Hikaru Tackeuchi and colleagues found that working memory training resulted in measurable changes in the structural connectivity of aspects of the prefrontal parietal network, including areas of the parietal lobe and the anterior part of the body of the corpus callosum. Likewise, Dietsje Jolles and colleagues found that 6 weeks of practice with working memory caused increase connectivity between the right middle frontal gyrus and other regions of the prefrontal parietal network. Similar findings have been found for reasoning. After just three months of reasoning training among a sample of 23 participants enrolled in a course to prepare for the Law School Admission Test (LSAT), Allyson Mackey, Kirstie Whitaker, and Silvia Bunge found plasticity in the white matter structure of the frontal and parietal lobes.

In a more recent hot-off-the-press analysis, Mackey, Alison Miller Singley, and Bunge used the same study sample to investigate whether intensive reasoning training would result in stronger communication within the prefrontal parietal network. The researchers were particularly interested in increases in communication between the lateral parietal cortex and the frontopolar cortex (BA 10), which resides all the way at the front of prefrontal cortex just above your eyes. Among its varied functions, the frontopolar cortex is associated with relational integration, the simultaneous consideration of multiple relations among various features of a problem. This skill is essential for abstract reasoning and high-level representations.

To test the effect of training on changes in the brain, the researchers included a sample of 26 prelaw adults participating in a 3-month course specifically designed to prepare lawyers for the LSAT exam. They chose this course because prior research shows that high levels of motivation and dopamine are associated with greater levels of neuroplasticity in adult animals, the students were clearly motivated to study for the LSAT, and there is obvious societal significance to this form of reasoning training. Every year, more than 100,000 adults take the LSAT, and the assumption by law school selection committees is that these tests measure cognitive potential.

The LSAT course involved 100 hours of instruction and practice distributed across three types of content: 35 hours were devoted to Logic Game questions that require test takers to “integrate a series of rules to sequence or group a set of items,”35 hours were devoted to Logical Reasoning items that require test takers to “determine the logical flaw in an argument, identify an assumption, or choose a statement that would strengthen or weaken an argument,” and 30 hours were devoted to Reading Comprehension items, which require test takers to “interpret short passages of text.” The types of items on these three sections are strongly correlated with each other and with IQ test performance. The researchers compared the effect of training on the prefrontal parietal brain network with the brain connectivity of 25 prelaw adults who were not taking the course, but who had the same levels of IQ, stress levels, and amount of sleep. What did they find?

First, there was an effect of the course. Improvement in LSAT total scores corresponded roughly to an improvement from the 44th percentile to the 73rd percentile (these percentiles depend on the year in which the test was taken). This is a practically meaningful improvement in performance, and as the researchers note, such an improvement would “vastly widen the pool of law schools to which he or she had a realistic chance of acceptance.”

But more relevant to the aims of their study, after training they found increased connectivity between the frontal and parietal regions at rest, primarily within the left hemisphere and between hemispheres. Consistent with their prediction, training particularly enhanced communication between the left frontopolar cortex (BA 10) and the posterior and medial parietal regions. They also found increased connectivity between the parietal cortex and the striatum, which is consistent with the role of the striatum in reasoning and skill learning across both cognitive and motor domains.

These results are certainly exciting and promising, but since this was a single study with a small sample size, more research is needed. While brain connectivity was positively correlated to LSAT improvement, the correlations were no longer significant after the researchers statistically corrected for multiple comparisons. Therefore, the jury is still out on whether individual differences in reasoning training are related to the strength of coupling of brain regions in the prefrontal parietal network. Further studies should include larger samples of motivated individuals, a more diverse population in terms of cognitive ability and demographic variables (e.g., sex, age), and have better control conditions (e.g., multiple activities pursued during training). It’s quite possible that a wider range of reasoning scores prior to training may be more strongly correlated both with reasoning training improvements as well as the increased strength of prefrontal parietal network connectivity.

Nevertheless, these findings do serve as proof of concept that reasoning training– even as brief as 3 months– can significantly alter connectivity in a brain network critical for high-level reasoning. These findings should not be understated, as they challenge traditional notions that intelligence is fixed, and that patterns of connectivity in large-scale brain networks at rest are stable across time. Engagement strengthens connections between disparate specialized brain regions. As Silvia Bunge told me in a personal correspondence, “These data underscore the point that our mental agility at a given time reflects the prior history of activation of specific brain networks.”

Instead of interpreting a person’s reasoning test performance at any one moment in time as reflecting that person’s hardwired cognitive potential, these results suggest that it’s more sensible to interpret that score as reflecting the individual’s cognitive history– his or her prior levels of engagement of specific neural networks. Indeed, some psychologists– such as Robert Sternberg and David Lohman– conceptualize IQ test scores at any single point in time as a measure of developing expertise or ability.

To be sure, this doesn’t negate the role of biology. For instance, genes can still play a role by influencing the motivation to engage in reasoning, the speed and efficiency of reasoning training, and the full range of neural plasticity. This reconceptualization of test scores does, however, highlight the developmental nature of cognitive functioning, the multiplying effects of enriching experiences on reasoning skills, and the crucial role of experience in determining our patterns of neural connectivity at any single moment in time.

These results also have implications for late bloomers and those hoping to learn new material later in life. The authors sensibly note, however, that to maintain a high level of reasoning ability, repeated engagement and practice across the lifespan is essential.

© 2013 Scott Barry Kaufman, All Rights Reserved

Illustrator: George Doutsiopoulos