Dynamics control of semantic processes in a hierarchical associative memory

Area of Science:

• Neuroscience
• Computational Neuroscience
• Cognitive Science

Background:

• The human brain's semantic declarative memory system processes abstract knowledge.
• Understanding the neural control of associative memory dynamics is crucial.

Purpose of the Study:

• To demonstrate a neural mechanism for controlling associative processes in hierarchical memory.
• To model how neuronal adaptation influences memory dynamics and abstract knowledge retrieval.

Main Methods:

• Formulation of an abstract neural network model incorporating neuronal adaptation.
• Utilized a generalized neuronal activation function derived from Hodgkin-Huxley models.
• Investigated the role of neuromodulators in regulating neuronal adaptability.

Main Results:

• Neuronal adaptability controls the search depth within hierarchical memory structures.
• Dynamical modes of the neural network correspond to different memory hierarchy levels.
• Autonomous bifurcations in network dynamics, driven by neuromodulators, guide associative processes from exploration to retrieval.

Conclusions:

• The demonstrated mechanism provides a deterministic yet potentially free-associative model for memory dynamics.
• Neuronal adaptation is key to controlling associative memory function.
• This model has implications for understanding brain disorders and enhancing artificial neural networks.