Post Synaptic Neural Activities Model – Provides Patterns of False Memory Development

The Deese–Roediger–McDermott (DRM) paradigm  studies false memory in humans following the known theories of cognitive philosophy. DRM procedures typically involve bio-linguistics association and studies how human memory deals with the associated words. DRM applications successfully explain the wide range of memory responses in various categories such as survival and autobiography, stress situations, false confessions, prior exposures, filling-in and reward function or magical thinking and imaginations. New research suggests that DRM could be explained by an associative model of memory, which represents that the associated words could spread through an associative neural network until reaches the absent result and that the false recognition of words could be the result of the residual activation and synaptic plasticity due to the processes such as short or long term potentiation (STP and LTP). A new model being developed at Silkatech, is a potential tool for quantizing the early indication of either false or confabulation of human memory. It is developed based on single synapse and several afferent neurons excitation, with the alternative of supposing that impulse of afferent neurons are not the immediate stimuli for efferent neurons, but maintain at the synapse or excitatory state. The new model provides a tool to study how STP and LDP increases the number of postsynaptic receptors per category and stabilizes the synaptic changes. It potentially covers the wide variety of neural structures from simple to increasing complexity. 

Theoretically, synapse plasticity is a function of  change in the strength of synapse, considering different excitation pathways. Both of intrinsic and extrinsic excitation, could be modeled for quantities in the chain of synapse excitation. The model can be configured to form a simple network such as two afferent neurons, the first of which acts immediately upon stimulus and the second one generates an observable response as soon as it reaches the excitation threshold. In a further complexity level, several of the same pair acting upon different stimuli in a more complicated manner. The basis of the neuro-relationship is maintained to avoid over-complication and understand the response time under maximum synaptic challenges. 

The model training and validation is achieved via schematizing further by employing statistical averaging to focus on synaptic activities rather than neurons.  Hence, we are assembling some borrowed datasets that are recorded using arrays of ECoG. The datasets are recorded while participants were involved in an intensely designed association practice. Finally, the synaptic weights associated with neural network formation of each participant is recorded and programmed in their model-specific database. The final result of a cold-run, confirms that the formation of false memory is very likely to happen in the group of participants with pre-diagnosis history.