There is nothing about the numbers in memory that marked some off as instructions and the others (location 7 in our example) as data. It depends on where you start executing. Early computers weren't like that: the programmer input a program by re-wiring part of the computer (perhaps by changing switch settings). The idea that a program could be stored in the same place as data was an important advance. Amongst other things, it meant that one could write programs such as compilers (see below) which operate on other programs. More generally, programs could be treated as data, or as instructions.
The idea of the stored-program computer contributed to psychology. It was a model used by Karl Lashley, who realised that the serial ordering of behaviour couldn't be explained by behaviourist S-R associations. For a brief history of early computer science, connectionism and cybernetics, including their influence on psychology, see Chapter 1 of Brains, Machines and Mathematics by Michael Arbib (2nd edition, Springer-Verlag 1987; RSL Comp Bd 98; chapter 1 copied as AI box photocopy A52).