Age-related differences in transient gamma band activity during working memory maintenance through adolescence.

Adolescence is a stage of development characterized by neurodevelopmental specialization of cognitive processes. In particular, working memory continues to improve through adolescence, with increases in response accuracy and decreases in response latency continuing well into the twenties. Human electroencephalogram (EEG) studies indicate that gamma oscillations (35-65 Hz) during the working memory delay period support the maintenance of mnemonic information guiding subsequent goal-driven behavior, which decrease in power with development. Importantly, recent electrophysiological studies have shown that gamma events, more so than sustained activity, may underlie working memory maintenance during the delay period. However, developmental differences in gamma events during working memory have not been studied. Here, we used EEG in conjunction with a novel spectral event processing approach to investigate age-related differences in transient gamma band activity during a memory guided saccade (MGS) task in 164 10- to 30-year-olds. Total gamma power was found to significantly decrease through adolescence, replicating prior findings. Results from the spectral event pipeline showed age-related decreases in the mean power of gamma events and trial-by-trial power variability across both the delay period and fixation epochs of the MGS task. In addition, we found that while event number decreased with age during the fixation period, the developmental decrease during the delay period was more dramatic, resulting in an increase in event spiking from fixation to delay in adolescence but not adulthood. While average power of the transient gamma events was found to mediate age-related differences in total gamma power in the fixation and delay periods, the number of gamma events was related to total power in only the delay period, suggesting that the power of gamma events may underlie the sustained gamma activity seen in EEG literature while the number of events may directly support age-related improvements in working memory maintenance. Our findings provide compelling new evidence for mechanistic changes in neural processing characterized by refinements in neural function as behavior becomes optimized in adulthood.

Pain phenotypes classified by machine learning using electroencephalography features.

Pain is a multidimensional experience mediated by distributed neural networks in the brain. To study this phenomenon, EEGs were collected from 20 subjects with chronic lumbar radiculopathy, 20 age and gender matched healthy subjects, and 17 subjects with chronic lumbar pain scheduled to receive an implanted spinal cord stimulator. Analysis of power spectral density, coherence, and phase-amplitude coupling using conventional statistics showed that there were no significant differences between the radiculopathy and control groups after correcting for multiple comparisons. However, analysis of transient spectral events showed that there were differences between these two groups in terms of the number, power, and frequency-span of events in a low gamma band. Finally, we trained a binary support vector machine to classify radiculopathy versus healthy subjects, as well as a 3-way classifier for subjects in the 3 groups. Both classifiers performed significantly better than chance, indicating that EEG features contain relevant information pertaining to sensory states, and may be used to help distinguish between pain states when other clinical signs are inconclusive.