Is there any reason? Well, does there absolutely need to be a "reason"? These are a set of choices, made over a period of multiple decades, by different people, surely different groups of people.
First, operations with a charcter vector. They go back to the dawn of MATLAB.
'1' is a character vector, as is '11'. Arithmetic operations on it, like addition, or subtraction, first convert the elements to their ascii equivalents.
That is, a '1' is ascii 49. So if you add them, you get 98. That seems clear.
Strings came out only fairly recently, as a better way of working with characters.
whos C S
Name Size Bytes Class Attributes
C 1x1 2 char
S 1x1 166 string
But here, they decided that it does not make sense to add two strings, and get a number out, converting to ascii. Far more sensible is to use the plus operator here as a concatenation operator.
This works very nicely in some cases. I very much like it. For example...
disp(['The year is: ',num2str(year(now)),', The variable X has value ', num2str(X)])
The year is: 2024, The variable X has value 17
disp("The year is: " + year(now) + ", The variable X has value " + X)
The year is: 2024, The variable X has value 17
The strings and the plus operator make things simpler. Easier to write, read, and debug.
But, you might ask, then why did they not just change how character vectors work? You should realize there are millions of lines of old MATLAB code out there. Some of them use character vectors in many different ways. Rather than forcing many thousands of people to modify their code, they just introduced a different class.