On the receiving side, if one listens to a tape that is mechanically reproducing a word spoken once by a speaker, one can see something of the complexities of the interpretation of these spoken signals. A classical tape, which we use to demonstrate this effect' has the word "cogitate" spoken once and recorded with a mechanical device, a magnetic disk, which reproduces very accurately exactly the same set of signals that were recorded on the tape originally. The word cogitate, in this particular case, is repeated every 0.7 seconds. If one listens to this tape for fifteen minutes to one hour, one at first hears the word cogitate from the signals received. As one continues to listen, one begins to hear other words, alternates to the word "cogitate", such as "tragedy".

With three hundred expert observers, we found that there were 2,730 alternates, 350 of which were in a large dictionary; the rest are words that we do not use.

This experiment shows that the reception of signals depends utterly upon the sequence of signals. The computation of the information, i.e., the generation of "meaning" or of "words," is a function of the central processing of the brain itself. Our speech is built up of expected sequences that we long ago stored in our memory. While we are listening to a speaker, our brain computes the signals that we receive from the air and creates the information, the meaning, almost instantaneously. We expect that the next word, for example, will be different from the previous words. When we are exposed to a constantly repeating set of signals, our brain operates in such a way as to generate alternates to the constantly repeated set of signals according to certain computational rules.

We analyzed the 2,730 alternates to "cogitate""and discovered that the original set of signals contained twelve time slots, i.e., phone, which could be varied by the following rules:

1. A consonant such as the initial c or k sound can be experienced as any other consonant, such as p, g, j, ch.

2. Any vowel such as the o or a sound after the k sound can become or be interpreted as any other vowel sound or a group of vowel sounds.

Thus, the brain is trained and has stored within it sequences of simulations that we refer to as meanings of signals. In our social consensus reality we are taught these long sequences in our childhood and we build up simulations upon which we agree. When we speak to one another, we compute meaning, change it into signals, receive signals, and recompute meaning from the received signals. Thus, we build up shared simulations that are flexible means of computation of maps, feelings, thoughts, ideas.

Thus, our communication system is amazingly fast and sophisticated and requires a very large memory and long experience in order to be able to construct meaning from the signals. As we discussed elsewhere in this book, such performance requires a brain with a certain kind of structure, i.e., fairly large silent areas to control the minicomputer that runs the muscles and receives the signals for the operations of the macrocomputer.