The first error is in the 3rd line. norm is a built-in MATLAB function and requires at least one input argument.
The second "error" (actually a "mistake") is on the 5th line in how you use the norm function. You correctly use norm with an input argument, but then you assign the result of the function to a variable named norm. This is actually allowed, but you need to realize that norm has now changed to a variable and is no longer a function. This creates an unintended side effect that will surface as an error if you use norm as a function later in your code.
Note that you also make the same mistake on the 1st line in your use of the mean function. It did not generate a syntax error from MATLAB, becuase this is (as I said) actually allowed. But, it is surely not what you intended.
Here's your code with these issues corrected:
mx = mean(x1);
x1(i) = (x1(i)-mx) / norm(x1);
nx = norm(x1);
x1(i) = x1(i) / nx;