
Some introductory slides on this, together with 3 exercises. (answers)
 There are
 data stored in variables (*)
 functions which operate on the variables(**)
 (*) variables are usually 1. matrices, 2. cell arrays, or 3. structures.
Here is an example "script" (set of Matlab commands in a file) showing how to create these basic data types:
egcreate.m
full list of data types (for reference)
The values in a matrix are of the same numeric type. The most common numeric types are 'double', 'uint8', and 'logical'.
 'double' is the default numeric type. it stores a 64bit precision floatingpoint number.
 'uint8' is an unsigned integer with 8bit precision (0 to 255). jpeg images read with imread() come in this format.
 'logical' can only be 0 or 1. These values arise as a result of a conditional test, e.g. (2==0) evaluates to logical 0.
Another useful type is 'char', which is used to make strings
 Strings begin and end with a singlequote (' = apostrophe). A double quote (") to delimit a string results in an error.
 To put a single quote in a string use two single quotes in a row (''). (not a blackslash escape)
 matlab does not have string interpolation. To put a number in a string, it's best to use sprintf(). One can also use the weaker num2string().
 (**) functions are toplevel (no namespace, io.file.open etc.).
Which functions you can call depends on your matlab "path". (that is, matlab will search for a file somewhere in a directory entered in your path which implements the function).
Here is an example function: egfunc.m
Also note that you can count the input arguments with "nargin", and accept a variable number of arguments using "varargin".
Here is an example function using those two: mymessage.m
Large sets of functions are often grouped together into "toolboxes". You don't need to do anything special to access a file from a toolbox. Sometimes  other people have posted just the function you are looking for on "matlab central"; though beginners will probably find everything they need inside Matlab's massive library or one of the standard toolboxes.
 Indexing (accessing some elements) of a matrix is very important in matlab.
Here is a script showing how to index: egindex.m
 Finding sums, means, etc. along various dimensions of a matrix is very useful in matlab:
Here is a script showing how to do that: eguseful.m
 Control flow: if, while, and for statements.
Here is a script which shows some basic control flow: egcontrol.m
NOTE: In Matlab, it is often more efficient to not use loops, and to cast operations in matrix format (operating on entire rows, columns, or matrices at a time).
 Plotting figures is very important in matlab.
 Here is a list of more functions you may want to use, type "help [functionname]" at the Matlab prompt to explore these.
image processing
imread, imwrite, imshow, imagesc, colormap, conv2, conv
utility
dir, system, clear, save, whos, profile
statistics
ttest, ttest2, kstest, kstest2, binomcdf, normcdf, normpdf
 Finally, you can refer to these to learn:
 