The question of which neural events
predict risky vs. safe behaviors such
as overtaking a slower vehicle when there is little space to do so due to
oncoming traffic vs. driving behind
the slower vehicle and arriving a few minutes later to work is a highly interesting
and important one. The vast majority of neuroimaging studies investigating the
neural basis of risk taking have utilized models adapted from economics, in
which risks are defined as the degree of variance in outcomes, however, it has
been argued that for lay persons risk equals being exposed to a potential loss.
In their recent study, Dr.
Helfinstein et al. (2014) had 108
healthy volunteers engage in a task called Balloon analog risk task during
functional magnetic resonance imaging. In this task, the subjects earn points
when they pump up balloons, but lose the points if the balloon explodes before
they “cash out” by stopping pumping. This task was selected because it is
highly correlated with public health relevant risk taking behaviors, including
unsafe driving, sexual risk taking, and drug use. The authors observed that
multi-voxel pattern analysis of brain activity before the point of decision
making predicted subsequent risky vs. safe choices by the subjects,
specifically involving brain regions found in previous studies to participate
in control functions. Interestingly,
in a separate univariate analysis these areas were found more active before safe than
risky choices.
These highly interesting findings
show that it is possible to predict risky vs.
safe choices based on preceding patterns of brain activity in a set of regions that
have been previously shown to be activated during tasks requiring cognitive
control. The fact that these areas were more strongly activated preceding safe
than risky decisions suggests that increased risk taking might be due to
failures in engaging appropriate cognitive control processes. The relevance of
these findings is further augmented given that the Balloon analog risk task
that was used has been found in previous studies to correlate highly with
real-life risk taking behaviors relevant for public health such as unsafe
driving and drug use.
Reference: Helfinstein SM,
Schonberg T, Congdon E, Karlsgodt KH, Mumford JA, Sabb FW, Cannon TD,
London ED, Bilder BM, Poldrack RA. Predicting risky choices from brain activity
patterns. Proc Natl Acad Sci USA (2014) e-publication ahead of print.
http://dx.doi.org/10.1073/pnas.1321728111
