Modeling cognitive and emotional processes: a novel neural network architecture

Area of Science:

• Artificial Intelligence
• Machine Learning
• Cognitive Science

Background:

• Existing machine learning models often focus on engineering concerns for human perception.
• There is a need to integrate emotional processing at a structural level within cognitive systems.

Purpose of the Study:

• To introduce and investigate the novel emotional DuoNeural Network (DuoNN) for modeling natural intelligence and emotions.
• To explore the integration of emotion at the structural level of cognitive systems.

Main Methods:

• Development of the DuoNeural Network (DuoNN) with hidden layer DuoNeurons, each containing dorsal (cognitive) and ventral (emotional) processing units.
• Parallel streaming of cognitive and emotional input information to the DuoNN.
• Presentation of the computational model, learning algorithm, and comparative analysis against existing emotional neural networks.

Main Results:

• The DuoNN architecture effectively processes local features (dorsal neurons) and entire patterns (ventral neurons) from visual stimuli.
• Experimental results demonstrate higher recognition rates compared to other emotional neural networks.
• The DuoNN exhibits faster learning and decision-making capabilities.

Conclusions:

• The DuoNN represents a significant advancement in modeling natural intelligence and emotions in AI.
• Integrating emotional processing structurally enhances cognitive system performance in machine learning.
• The DuoNN's parallel processing of cognitive and emotional data offers superior efficiency and accuracy.