Experimental validity is whether a study's design lets you trust its conclusion. Threats such as demand characteristics, placebo effects and experimenter bias undermine internal validity, which is whether the independent variable really caused the result. Each threat offers another explanation for the change you see.
Why it matters
In a sleep-deprivation study or a drug trial, a change in reaction time only counts if the design rules out other explanations. That is why drug trials use placebos and double-blind procedures, so neither participants' nor researchers' expectations shape the results.
How the exam tests it
A study scenario lists several problems, and you must name the one that fits. For example, sleep-deprived participants who guess the hypothesis and slow down on purpose show demand characteristics.
Don’t confuse it with
Demand characteristics are often confused with experimenter bias, but demand characteristics come from participants guessing the hypothesis, while experimenter bias comes from the researcher's expectations.