Related Experiment Video
Updated: Feb 11, 2026
13:19
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
10.0K
Cross-Participant EEG-Based Assessment of Cognitive Workload Using Multi-Path Convolutional Recurrent Neural
Ryan Hefron1, Brett Borghetti2, Christine Schubert Kabban3
1Department of Electrical & Computer Engineering, Air Force Institute of Technology, WPAFB, Dayton, OH 45433, USA. ryan.hefron@afit.edu.
Sensors (Basel, Switzerland)
|April 28, 2018
Summary
Ensembles of individually trained deep learning models offer a computationally efficient alternative for cross-participant cognitive workload estimation using electroencephalograph (EEG) data. A novel convolutional-recurrent model significantly improves accuracy and reduces variance in these estimations.
Area of Science:
- Neuroscience
- Machine Learning
- Human-Computer Interaction
Background:
- Deep learning enhances electroencephalograph (EEG) analysis for cognitive state assessment.
- Cross-participant cognitive workload modeling using deep learning is an underrepresented research area.
- Accurate workload estimation is crucial for adaptive systems and understanding cognitive states.
Purpose of the Study:
- To investigate cross-participant cognitive workload estimation using deep learning on EEG data.
- To evaluate various deep neural network models for computational efficiency, accuracy, variance, and temporal specificity.
- To identify optimal modeling strategies for non-stimulus-locked tasks in a cross-participant setting.
Main Methods:
- Utilized experimental data from the Multi-Attribute Task Battery (MATB) environment.
- Evaluated a variety of deep neural network architectures, including ensembles of individually-trained models and a novel convolutional-recurrent model.
- Assessed models based on computational cost, predictive accuracy, variance, and temporal specificity in cross-participant workload estimation.
Main Results:
- Ensembles of individually-trained models performed comparably to group-trained methods with significantly lower computational cost.
- Increasing temporal sequence length improved mean accuracy but increased cross-participant variance due to data dissimilarities.
- A novel convolutional-recurrent model demonstrated superior predictive accuracy and reduced cross-participant variance compared to other evaluated networks.
Conclusions:
- Individually-trained model ensembles provide an efficient and effective approach for cross-participant workload modeling.
- Temporal sequence length alone cannot fully address inter-individual EEG data variability.
- The proposed convolutional-recurrent neural network offers a promising advancement for robust cross-participant cognitive workload assessment.
Keywords:
EEGMATBcognitive workloadconvolutionalcross-participantensemblemental workloadneural networkrecurrenttemporal specificityMore Related Videos
Related Concept Videos
01:17Convolution Properties II
594
The important convolution properties include width, area, differentiation, and integration properties.
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
594
01:22Mean free path and Mean free time
5.3K
Consider the gas molecules in a cylinder. They move in a random motion as they collide with each other and change speed and direction. The average of all the path lengths between collisions is known as the "mean free path."
5.3K
01:21Path Between Thermodynamics States
4.1K
Consider the two thermodynamic processes involving an ideal gas that are represented by paths AC and ABC in Figure 1:
4.1K
01:34Crossing Over
172.3K
Unlike mitosis, meiosis aims for genetic diversity in its creation of haploid gametes. Dividing germ cells first begin this process in prophase I, where each chromosome—replicated in S phase—is now composed of two sister chromatids (identical copies) joined centrally.
The homologous pairs of sister chromosomes—one from the maternal and one from the paternal genome—then begin to align alongside each other lengthwise, matching corresponding DNA positions in a process...
The homologous pairs of sister chromosomes—one from the maternal and one from the paternal genome—then begin to align alongside each other lengthwise, matching corresponding DNA positions in a process...
172.3K
01:20Convolution Properties I
621
Convolution computations can be simplified by utilizing their inherent properties.
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
621
01:14Crossed Aldol Reaction Using Weak Bases
2.7K
This lesson deals with the crossed aldol reaction using weak bases. The self-condensation of an aldehyde having α hydrogen is prevented by adding it slowly to a mixture of formaldehyde and weak bases like hydroxide and alkoxide. Upon slow addition of the aldehyde, the base deprotonates the α carbon of the aldehyde to form the corresponding enolate. The enolate subsequently attacks the formaldehyde to form a single crossed product. Figure 1 depicts the aforementioned reaction.
2.7K