I was normal*

It's a phrase I've heard many times from students, friends and colleagues with a disability and is a phrase I've uttered many many times myself.

Only since the last few months of my undergraduate degree have I been able to add the conditional rules that I'm about to discuss.

One thing I've heard said repeatedly about people with Asperger's syndrome is their challenges regarding binary thinking (obviously this does not imply that this applies to all people with Asperger's syndrome). I am a very absolute thinker at the most basic level- Right or Wrong and Everyone or No-one. Yet you've already seen an example of me not being absolute/binary about an argument of statement (up above). What gives?

Well here lies the crux of the matter it's all about how one represents the more complex arguments that facilitates a more continuous spectrum of opinions or responses or even emotions.
A description of my default emotional brain is a monochrome matrix with some elements active (1) and some elements inactive (0). These matrices aren't particularly dense and are typically very similar to an identity. When you take the input vector and apply the matrix you get something fairly similar to the input with a few elements missing or slightly mixed around. Nothing too complex...

So where does the complexity arise? Well importantly, in the earlier analogy there is very little room for more emotionally intelligent responses and it misses the same level of continuity that (My opinion here) the normal human brain has built in.

I'm clever [Citation needed] in the fields of physics and engineering. As you'll probably figure out lots of my analogies are computer-y ones. It is possible to represent a continuous number line using an infinite amount of binary data [Citation needed]. In reality we have only a fixed amount of data so we fix the amount of bits per variable and use these bits to represent floating point values using mantissa and exponent notation (if interested check out the wiki pages on floating point values for computers it's IMO an interesting read).
As with computers the effort that I have to put into thinking is increased in order to start working with some level of granularity.

Now let's add in the other layer which would seem to be the final level of social understanding... These matrices are actually tensors. Comprised of additional variables that represent interactions between the current environment, company etc. etc.

Now in order to socialise with people effectively, I need some intuitive understanding of other people's state. Now fortunately I share a common social background with many of my friends and colleagues and this allows me a very simplistic look at the black and white of a person's accepted social interactions. My best friends or in other words those I love, I have a floating point representation of. Might not be accurate but it is important to understand the intricacies of the person.

What do I think of as normal? A floating point tensor at the moment of introduction.