Cognitive Systems

Modeling the brain functions is both challenging and complex. To come up with a convincing model that is a good enough representation of the brain is impossible without some simplification and a focus on some aspects of the brain functions. While there is evidently interest in understanding and modeling the whole range of brain functions (e.g. perception, regulation, cognition...) for the sake of achievability it is important to both compromise of the granularity of the model and constrain the model to a narrow-scope representation.

Synopsis

I focused on two aspects of the brain functions that are connected and, I believe, of paramount importance: volition and emotions . Per se both are long standing, complex and challenging research topics and there is still debate as to their exact definition that would be widely accepted.

From expressions to character

I came up with this model of volition based on the principle of spontaneous generation of possible actions that are then inhibited.

Volition Circuit

As for emotions, I simplified te circuit to include only the key elements, so that the circuit can eb modelled and understood.

Emotion Circuit

The current model is a simplification but still of relevance in the generation of artificial emotions.

Emotion Model

I did many experiments and evaluations as part of my interest in cognitive system. One of the evaluation was about testing NIRS (Near Infra Red Spectrometer) systems. However, these can only monitor brain activitiy within 1 to 2 cm. Thus giving limited information related to the deep brain activities. This is particularly detrimental when attepting to monitor the limbic system, the "seat" of emotions.

I am currently building an emotion engine and will be running SIMULINK simulations before Chrismas 2014. So as they say, keep watching this space!