Compassion - Bridging Practice and Science - page 355

In this task, participants first see what looks like a long line.
Ninety percent of the time, during this tenth of a second stimulus, which occurs every two seconds
or so, the line is long. Ten percent of the time, it is short. For the first 10-15 minutes of the task, we
make it progressively harder to tell the difference between the short line and the long line by
making the short line longer and longer. When participants reach about 75% accuracy in telling
long from short we ask them to press the button every time they see a short line.
Both the retreat and control groups performed this task for thirty-two minutes, without a break. The
two groups (control and retreatants) were both measured for their performance in this task pre-
retreat. At this time point, there was no difference between the two groups. After five weeks of
meditation, however, the retreat group was significantly better able to see a longer short line than
the control group could [13]. They had increased vividness of perception, which was sustained at
the end of the retreat and five months later at follow-up. When the control group went into retreat,
by the midpoint of their retreat they showed the same pattern as the first retreat group, and this
improvement was sustained at their five-month follow-up, provided people had been continuing to
meditate each day (which was also true for the first retreat group finding)[13].
Results from the control group and pre-retreat data showed that when you are not meditating in
retreat, your performance declines over time in this task, as was expected. However, by the
midpoint of the retreat, and at the end, the meditators had better sustained ability to detect this
target. Given the change in perceptual threshold, our hypothesis is that the ability to continue to
detect these slightly shorter lines is actually related to changes in perception. This is an important
point. It is not so much that sustained attention per se improved, but that the task actually got
easier for the meditators because they were able to perceive the signal more easily, because their
threshold to see small visual differences improved [13].We will be following up these results by
looking at fluctuations in brain electrical activity immediately before and after stimuli – it may be
that part of the improved performance is also a training-related shift in the efficiency of paying
attention such that our limited attentional resources are replenished in an ongoing manner by
participants knowing when to be most vigilant and when to relax their focus. This new work is being
undertaken by UC Davis neuroscience grad student Anahita Hamidi along with Postdoctoral
Scholar Chivon Powers.
Such hypothesized effects of training on the activity of attention-related cortical areas during
performance of the CPT are supported by our recently published findings obtained from EEG
recordings taken during the practice of mindfulness of breathing. These results demonstrate
training-related increases in activation of attention-related cortical regions [14]. University of Texas
at Austin computer science grad student Manish Saggar (now a Postdoctoral Fellow at Stanford)
carefully analyzed EEG data during meditation using advanced signal processing methods that
minimize the contamination of the EEG signal by such things as scalp muscle tension. We found
that over the course of the training there was a significant decrease in beta band activity (~12-20
Hz) overlying bilateral posterior frontal and parietal scalp regions. The changes were not observed
in the control group, and replicated when the controls entered their own 3-month retreat.
Decreased beta power is consistent with increased activations observed in fMRI and observed in
the run-up to perceiving a hard-to-detect somatosensory signal[14].
We made another version of the CPT known as the response inhibition task (RIT) [8].Instead of
pressing the button when presented with the short line, participants were asked to press the button
in response to all the long lines (which appear 90% of the time) and to withhold their response to
the short lines. The short lines were still targets, but now a response to a target meant stopping the
355