Samples are those configurations of neurons that have informational content.

Higher-order samples are processes such as thought and trains of thought if the sequential process has a broader associative background in connection with other samples that are within range of the area of attraction of the observed dynamic sample. If individual samples, the elements of such a sample sequence, are connected with their own cardinal cells or have their own parameters of order in speaking centres, then such a train of thought is also encoded and it can be expressed in words. Through these processes we also introduce to our model the basics of speech i.e. the language of thought.

Samples, words can have significance regardless of whether they are correlated with the actual external state or only with the presentational situation. An example of a virtual construct without correlation with an external object is the ‘theatre role’, which also has its own meaning that the person can even project outwards into the explicit world. Meaning is the construction of the context of the outer or inner environment. On the level of system implementation, meaning is a dynamic construction, resulting from system relationships.

In a neural network, one particular neuron can acquire a leading role and become a cardinal neuron. It is impossible to think of the cardinal neuron without a neuron sample as they support each other and enable each other!