There are a few different possible meanings here.
If you are using the Variable Browser, then the display format for each item is controlled by the "Number Display Format" control in the View tab of the Variable Browser. There is a preference that controls the default format to use.
If you are using disp() then the display format for each item is controlled by the current "format" command. There is a preference (different from the one above) that controls what the default is for MATLAB sessions.
None of formats supported for the above two cases show all of the decimal places for double precision values. None of them show enough significant digits to be able to unambiguously distinguish between a number and its nearest representable neighbour.
In order to be able to see enough significant digits to be able to unambiguously tell a number from its nearest represetnable neighbour, use fprintf() with a %.16e format (I would have to research whether there are cases where %.17e is needed.) If you want to see the full decimal places, try %.999g .
Note: on MS Windows, historically digits after about the 15th were just arbitrarily 0. If you are using Windows, I recommend you get the File Exchange contribution num2strexact()
Note: on Linux, historically more digits were available, but there was still a limit before 0 would arbitrarily appear.
Note: on MacOS, at least as far back as I have tested, all digits of double precision have been available.
