Simulation of cognitive disturbances by a dynamic threshold semantic neural network

Area of Science:

• Computational neuroscience
• Cognitive modeling
• Theoretical psychiatry

Background:

• Semantic networks in the brain are hypothesized to function via spread activation.
• Understanding the neural basis of cognitive disturbances is crucial for psychiatric research.

Purpose of the Study:

• To develop a computational model simulating spread activation in semantic networks.
• To explore the relationship between neural parameter alterations and cognitive disturbances.
• To provide a theoretical framework for understanding mental disturbances in psychiatric patients.

Main Methods:

• A neural network model incorporating dynamic thresholds, asymmetric connections, and clustered memories was developed.
• Parameters, specifically dynamic thresholds, were systematically altered to generate model disturbances.
• Model disturbances were analyzed for metaphorical resemblance to clinical descriptions of mental disturbances.

Main Results:

• The model successfully simulates spread activation, a hypothesized mechanism for semantic networks.
• Altering dynamic threshold parameters generated a wide range of disturbances within the model.
• These simulated disturbances showed metaphorical similarities to clinical descriptions of psychiatric conditions.

Conclusions:

• The theoretical neural network model provides a framework for investigating the link between neural parameters and cognitive function.
• Alterations in neural parameters like thresholds and connectivity may underlie observed clinical symptoms.
• This computational approach may offer insights into the neural underpinnings of mental disturbances.