High Variability Periods in the EEG Distinguish Cognitive Brain States

OBJECTIVE

To describe a novel measure of EEG signal variability that distinguishes cognitive brain states.

METHOD

We describe a novel characterization of amplitude variability in the EEG signal termed "High Variability Periods" or "HVPs", defined as segments when the standard deviation of a moving window is continuously higher than the quartile cutoff. We characterize the parameter space of the metric in terms of window size, overlap, and threshold to suggest ideal parameter choice and compare its performance as a discriminator of brain state to alternate single channel measures of variability such as entropy, complexity, harmonic regression fit, and spectral measures.

RESULTS

We show that the average HVP duration provides a substantially distinct view of the signal relative to alternate metrics of variability and, when used in combination with these metrics, significantly enhances the ability to predict whether an individual has their eyes open or closed and is performing a working memory and Raven's pattern completion task. In addition, HVPs disappear under anesthesia and do not reappear in early periods of recovery.

CONCLUSIONS

HVP metrics enhance the discrimination of various brain states and are fast to estimate.

SIGNIFICANCE

HVP metrics can provide an additional view of signal variability that has potential clinical application in the rapid discrimination of brain states.

The Acceptability, Feasibility, and Utility of Portable Electroencephalography to Study Resting-State Neurophysiology in Rural Communities

Electroencephalography (EEG) provides a non-invasive means to advancing our understanding of the development and function of the brain. However, the majority of the world’s population residing in low and middle income countries has historically been limited from contributing to, and thereby benefiting from, such neurophysiological research, due to lack of scalable validated methods of EEG data collection. In this study, we establish a standard operating protocol to collect approximately 3 min each of eyes-open and eyes-closed resting-state EEG data using a low-cost portable EEG device in rural households through formative work in the community. We then evaluate the acceptability of these EEG assessments to young children and feasibility of administering them through non-specialist workers. Finally, we describe properties of the EEG recordings obtained using this novel approach to EEG data collection. The formative phase was conducted with 9 families which informed protocols for consenting, child engagement strategies and data collection. The protocol was then implemented on 1265 families. 977 children (Mean age = 38.8 months, SD = 0.9) and 1199 adults (Mean age = 27.0 years, SD = 4) provided resting-state data for this study. 259 children refused to wear the EEG cap or removed it, and 58 children refused the eyes-closed recording session. Hardware or software issues were experienced during 30 and 25 recordings in eyes-open and eyes-closed conditions respectively. Disturbances during the recording sessions were rare and included participants moving their heads, touching the EEG headset with their hands, opening their eyes within the eyes-closed recording session, and presence of loud sounds in the testing environment. Similar to findings in laboratory-based studies from high-income settings, the percentage of recordings which showed an alpha peak was higher in eyes-closed than eyes-open condition, with the peak occurring most frequently in electrodes at O1 and O2 positions, and the mean frequency of the alpha peak was found to be lower in children (8.43 Hz, SD = 1.73) as compared to adults (10.71 Hz, SD = 3.96). We observed a deterioration in the EEG signal with prolonged device usage. This study demonstrates the acceptability, feasibility and utility of conducting EEG research at scale in a rural low-resource community, while highlighting its potential limitations, and offers the impetus needed to further refine the methods and devices and validate such scalable methods to overcome existing research inequity.

The impact of socioeconomic and stimulus inequality on human brain physiology

The brain undergoes profound structural and dynamical alteration in response to its stimulus environment. In animal studies, enriched stimulus environments result in numerous structural and dynamical changes along with cognitive enhancements. In human society factors such as education, travel, cell phones and motorized transport dramatically expand the rate and complexity of stimulus experience but diverge in access based on income. Correspondingly, poverty is associated with significant structural and dynamical differences in the brain, but it is unknown how this relates to disparity in stimulus access. Here we studied consumption of major stimulus factors along with measurement of brain signals using EEG in 402 people in India across an income range of $0.82 to $410/day. We show that the complexity of the EEG signal scaled logarithmically with overall stimulus consumption and income and linearly with education and travel. In contrast phone use jumped up at a threshold of $30/day corresponding to a similar jump in key spectral parameters that reflect the signal energy. Our results suggest that key aspects of brain physiology increase in lockstep with stimulus consumption and that we have not fully appreciated the profound way that stimulus expanding aspects of modern life are changing our brain physiology.

Proof of Concept of a Gamified DEvelopmental Assessment on an E-Platform (DEEP) Tool to Measure Cognitive Development in Rural Indian Preschool Children

Over 250 million children in developing countries are at risk of not achieving their developmental potential, and unlikely to receive timely interventions because existing developmental assessments that help identify children who are faltering are prohibitive for use in low resource contexts. To bridge this “detection gap,” we developed a tablet-based, gamified cognitive assessment tool named DEvelopmental assessment on an E-Platform (DEEP), which is feasible for delivery by non-specialists in rural Indian households and acceptable to all end-users. Here we provide proof-of-concept of using a supervised machine learning (ML) approach benchmarked to the Bayley’s Scale of Infant and Toddler Development, 3rd Edition (BSID-III) cognitive scale, to predict a child’s cognitive development using metrics derived from gameplay on DEEP. Two-hundred children aged 34–40 months recruited from rural Haryana, India were concurrently assessed using DEEP and BSID-III. Seventy percent of the sample was used for training the ML algorithms using a 10-fold cross validation approach and ensemble modeling, while 30% was assigned to the “test” dataset to evaluate the algorithm’s accuracy on novel data. Of the 522 features that computationally described children’s performance on DEEP, 31 features which together represented all nine games of DEEP were selected in the final model. The predicted DEEP scores were in good agreement (ICC [2,1] > 0.6) and positively correlated (Pearson’s r = 0.67) with BSID-cognitive scores, and model performance metrics were highly comparable between the training and test datasets. Importantly, the mean absolute prediction error was less than three points (<10% error) on a possible range of 31 points on the BSID-cognitive scale in both the training and test datasets. Leveraging the power of ML which allows iterative improvements as more diverse data become available for training, DEEP, pending further validation, holds promise to serve as an acceptable and feasible cognitive assessment tool to bridge the detection gap and support optimum child development.

Development, feasibility and acceptability of a gamified cognitive DEvelopmental assessment on an E-Platform (DEEP) in rural Indian pre-schoolers - a pilot study

Background: Assessment of cognitive development is essential to identify children with faltering developmental attainment and monitor the impact of interventions. A key barrier to achieving these goals is the lack of standardized, scalable tools to assess cognitive abilities.

Objective: This study aimed to develop a tablet-based gamified assessment of cognitive abilities of 3-year-old children which can be administered by non-specialist field workers.

Methods: Workshops among domain experts, literature search for established and gamified paradigms of cognitive assessments and rapid review of mobile games for 3-year-old children was done to conceptualize games for this study. Formative household visits (N = 20) informed the design and content of the games. A cross-sectional pilot study (N = 100) was done to assess feasibility of the tool and check if increasing levels of difficulty and the expected variability between children were evident in game metrics. In-depth interviews (N = 9) were conducted with mothers of participating children to assess its acceptability.

Results: Six cognitive domains were identified as being integral to learning – divided attention, response inhibition, reasoning, visual form perception and integration and memory. A narrative, musical soundtrack and positive reinforcement were incorporated into the tool to enhance participant engagement. Child performance determined level timers and difficulty levels in each game. Pilot data indicate that children differ in their performance profile on the tool as measured by the number of game levels played and their accuracy and completion time indicating that it might be possible to differentiate children based on these metrics. Qualitative data suggest high levels of acceptability of the tool amongst participants.

Conclusions: A DEvelopmental assessment on an E-Platform (DEEP) has been created comprising distinct games woven into a narrative, which assess six cognitive domains, and shows high levels of acceptability and generates metrics which may be used for validation against gold standard cognitive assessments.

Summation in the Hippocampal CA3-CA1 Network Remains Robustly Linear Following Inhibitory Modulation and Plasticity, but Undergoes Scaling and Offset Transformations

Many theories of neural network function assume linear summation. This is in apparent conflict with several known forms of non-linearity in real neurons. Furthermore, key network properties depend on the summation parameters, which are themselves subject to modulation and plasticity in real neurons. We tested summation responses as measured by spiking activity in small groups of CA1 pyramidal neurons using permutations of inputs delivered on an electrode array. We used calcium dye recordings as a readout of the summed spiking response of cell assemblies in the network. Each group consisted of 2–10 cells, and the calcium signal from each cell correlated with individual action potentials. We find that the responses of these small cell groups sum linearly, despite previously reported dendritic non-linearities and the thresholded responses of individual cells. This linear summation persisted when input strengths were reduced. Blockage of inhibition shifted responses up toward saturation, but did not alter the slope of the linear region of summation. Long-term potentiation of synapses in the slice also preserved the linear fit, with an increase in absolute response. However, in this case the summation gain decreased, suggesting a homeostatic process for preserving overall network excitability. Overall, our results suggest that cell groups in the CA3-CA1 network robustly follow a consistent set of linear summation and gain-control rules, notwithstanding the intrinsic non-linearities of individual neurons. Cell-group responses remain linear, with well-defined transformations following inhibitory modulation and plasticity. Our measures of these transformations provide useful parameters to apply to neural network analyses involving modulation and plasticity.