One of the particularly interesting research questions that are asked by context-aware computing researchers is reasoning of the context in which user is. For instance, there are different approaches to context’s definition itself, which leads to different points of view on context reasoning. Some people use quite practical approach to context definition, trying to make it up from those parameters of situation in which the product is used by user that computer can actually sense. Some people use quite theoretical approach trying to collect all possible and impossible to sense parameters of event. As usual, the truth is somewhere in between these two approaches. From my perspective, one possible answer to reasoning of context was given by Alfred Van Vogt and Alfred Korzybski, author of General Semantics theory. It’s a so-called Null-A Pause Principle.
The principle itself is based on the postulates of General Semantics theory, as well as on analysis of person’s reaction to environment and context changes. Person reacts on environment and context changes using thalamus. Let me cite Wikipedia to provide a bit of context into what thalamus is:
The thalamus (from Greek θάλαμος = room, chamber) is a midline paired symmetrical structure within the brains of vertebrates, including humans. It is situated between the cerebral cortex and midbrain, both in terms of location and neurological connections. Its function includes relaying sensation, special sense and motor signals to the cerebral cortex, along with the regulation of consciousness, sleep and alertness. The thalamus surrounds the third ventricle.
Thalamus reaction on environment/context changes, or events, is actually a result of direct reaction on environment/context parameters changes. The key is the role of user’s control on her thalamus reaction on event. The problem is, user is not controlling her thalamus reaction on events because of the role of thalamus itself in event reasoning based on its position in information processing. Well, at some point user is able to control her reaction, but this is usually happening in case a situation so much new that the only way to adapt, or survive is to quickly learn about situation rather than use past experiences memory as a foundation for decision making. The more is number of parameters describing event that are changing over the time, the less controllable is user’s ability to control her reaction because the complexity of situation is increasing. Because of the fact that user’s memory has not only a so-called semantic memory, but also an episodic one (see “Total Recall” book by Gordon Bell), our brain’s reaction on familiar situation is not dictated by need of detailed analysis of situation, but is rather based on our past experiences – this is, in a nutshell, the core idea of information processing used by our our brain.
In General Semantics, the research question asked is on how to solve problem of misunderstanding between different people. The hypothesis behind General Semantics is based on idea that words we use to describe different “objects” in the world typically contain different meaning when said by different people. This is a natural result of the fact that any event person is involved into has uncountable amount of parameters, and particular person picks up only some of such parameters to build a concept of situation inside her mind and then gives a word as a name of that concept. This process is also called “objectification”. Different people do “objectification” process differently because of differences in episodic and semantic memories (which means our experience and knowledge), but language is not perfect enough to catch up on these differences in understanding and thus objectification which leads to the fact that people name different events with same words which in turn leads to people misunderstanding of each other and thus to conflicts.
I’ve given a small overview of concepts of General Semantics in one of my previous blog posts, in case you’d love to learn a bit more about this idea.
Returning back to idea of context reasoning, the point of the story above was to point your attention onto the fact that context reasoning for a particular person is using quite subjective approach based on user’s past experiences and knowledge rather than advanced reasoning that is using mostly analysis rather than ready-to-use concepts and paradigms offered by variety of context source providers.
So far, thalamus reaction on events is exactly the mechanism of the way user processes information, and typically person is reacting on event with emotions.
The interesting question that comes out of this discussion is how to prevent inertia of our thinking, how to prevent automatic objectification of the environment/context in which is hugely involved as creator of content as well as it’s connector.
A possible answer is a Null-A pause principle. It consists of three steps:
Get rid of emotions when a situation is becoming non-controllable
Get out of personal involvement into situation in order to see the situation from another perspective of what’s the you work on.
Use complex analysis of situation from different perspectives in order to build a complex understanding of the problem
So, this is that principle that we can apply in our context-aware systems in order to provide a better context-reasoning technology for our applications. And what do think?