a

is still present long after the test has been administered;

b

still works on when the next item of the test is administered;

c

goes against the grain of the researcher’s expectation;

d

affects other respondents’ reactions in the test.