The brain is an information processing machine adjusting itself to the environment. Information processing can be defined as reducing uncertainty. It has been suggested that the brain developed from an evolutionary point of view once living creatures started moving around in a changing and thus uncertain environment.
Therefore it can be proposed that the primary purpose of having a brain (and critical for survival) is to reduce uncertainty (about the environment) by processing external (and internal) information. Perception can be seen as Bayesian inference, where an intention driven prediction is actively looked for in the environment. This prediction is then updated by what the senses extract from the environment. The percept itself is an emerging property of a network activation, processing the information at different oscillation frequencies. In pathology, symptoms can be conceived of as emergent properties of parallel, dynamically changing but overlapping networks as part of a complex adaptive system. Using the brain’s adaptive characteristics can be advantageous in retraining the brain by reshaping its networks.
This is the purpose of neurofeedback, applying operant conditioning by interfering with the brain’s Bayesian updating mechanisms. Thus a conceptual model can help understand what neurofeedback does. This model sees delta activity as controlling basic homeostatic activity and a carrier wave for higher oscillation frequencies. Theta is a carrier wave integrating beta and gamma activity by theta-beta and theta gamma nesting. The beta and gamma activity is locally restricted in distributed areas. Theta therefore encodes memory and theta/beta coupling, allowing memory based predictions of future events/stimuli. In order to update the prediction, alpha is used as a scanning and attention mechanism sampling the environment for salient information.
The prediction error or change is encoded in gamma. Thus correct predictions about the environment will be encoded by beta activity, which therefore represents a status quo, whereas wrong predictions or insufficient input from the environment will be represented by beta/gamma. Thus neurofeedback attempts to modulate these oscillatory networks, thereby normalizing predictions, attention to internal or external stimuli, or processing of prediction errors.
This presentation is a single case study involving the use of transcranial direct current stimulation (tDCS) in the treatment of neuropathic back pain, with symptoms described in the lumbar-sacral region of the spine and down the left leg. An examination of the literature indicated that 40-50 percent improvement in pain perception might follow anodal stimulation over the primary motor cortex (M1). Given the report of left leg symptoms, anodal stimulation was applied to the scalp over M1 on the left.
Many diseases have been linked to plastic changes and changes in activity and functional connectivity in the brain, which can be demonstrated by functional imaging, either using resting state imaging (EEG, MEG, fMRI), or by evoked activity. Many brain related diseases can therefore be seen as emerging properties of altered dynamically changing overlapping networks. Different neuromodulation techniques such as Transcranial Magnetic Stimulation (TMS) and transcranial Direct Current Stimulation (tDCS) have been used in an attempt to modify the abnormal activity and connectivity. Recently, also transcranial alternating current stimulation (tACS) and transcranial random noise stimulation (tRNS) have been introduced as neuromodulation tools, and LORETA neurofeedback is emerging as another non-invasive neuromodulation tool. Each of these neuromodulation techniques has a different proposed working mechanism which could provide help in selecting the right neuromodulation technique that best suits the pathology related functional imaging changes.
The size and complexity of the nervous system makes it unlikely that changes in a single synapse result in significant changes in the behaviour of an interconnected neural network. Significant changes in neural network behaviour require changes in populations of synapses, defined as multiple synaptic modifications occurring simultaneously at multiple sites. The goal of the presentation is to present a neural network model of human cerebral ontogenesis and to use the model to explain the development of human EEG coherence over the postnatal period from 1.5 to 16 years of age.