Models of Logical Thinking Processes

Note:
This is a somewhat technical programming section which carries on from ideas and techniques developed at length in "Lingo Sorcery"

At each node of a logical decision making network, a decision is made as to where a message should be routed next. This routing decision is based upon the state of the ancestor property which in turn depends upon information the node is receiving from some external source .

In this way, various information, coming into the network at different nodes, is all combining together to form a specific message path through the network. The message path thus represents the aggregate result of all the information coming into the network. This is not unlike the way the human brain works to make logical decisions.

Humans have a process of decision making, based upon logic. We experience this logical decision making process as conscious thought (as opposed to the unconscious thoughts of emotive decisions). In essence, these logical thought processes rely simply on re routing messages along paths in neural networks.

Conceptually, such a model is identical to a network of object nodes. The brain consists mainly of special cells called neurons which bundle together to interconnect with each other by means of extensions coming out of the cells called axons and dendrites.

Ignoring the actual physical construction and the chemistry of the brain's signaling and switching mechanisms, the essence is of a network of interconnected cells which incorporate switching devices. A simplified sketch of cells, or nodes, with a message path running through them is shown in fig 19/14.

---

Figure 19/14 A message path running through a block of cells

---

If signals (or messages) are routed into this network and each cell is 'aware' of the messages paths it is connected to, the system can then act like a decision making device. In fig 19/15. messages arriving at the cells, A, B, C and D will "point" to a result "X" simply by comparing the number of messages each cell receives.

If cell "X" is a Lingo object this can result in object "X" activating one of its handlers. In effect, this object has been 'chosen' by a combination of the "neural" arrangement of the network and the entry points of the incoming messages. Notice how it is not necessary for the messages to have any specific content to effect this decision process.

---

Figure 19/15 The block of cells comes to a decision

---

A slight change in the signaling entry points and the switching (at cell B) will result in cell "Y" being the result of the messages and the network configuration (Fig 19/16).

---

Figure 19/16 The block of cells changes its mind when new information arrives

---

As you can see from fig 19/17, with exactly the same inputs, a message switch at cell "B" can flip the result back to "X".

---

Figure 19/17 A special factor monitored by cell B causes the block of cells to change its mind again

---

A little thought and it can be seen that this very simple network of sixteen cells (objects) can deal with an extremely large number of combinations of message inputs and network configurations.

This is the secret of the human brain's vast capability for reasoning and logical decision making: when it is taking many different factors into account in choosing between a number of possible alternatives.

Although we could not work on the scale of the human brain, we can certainly construct similar Lingo "neural" mechanisms in computer memory space to achieve an acceptable level of logical decision making ability for Lingo "brain" objects.

The coding necessary for switching whole groups of ancestors to establish new message paths through virtual networks of object cells, is no more complicated than that of the hierarchical network which we covered earlier.

[Index]
[Next - Emotive decisions (1)]
[Back - Hierarchical networks]