Sequential effects are not trivial context effects in psychophysical research

Sequential effects are a type of context effects in which the response to the stimulus is affected by both, stimuli presented and responses obtained on previous trials. In psychophysical studies, sequential effects are expressed through time-order errors in esthesiometry and through assimilation and contrast effects in scaling. Although these effects are robust, they are too often neglected in psychophysical research. The paper discusses whether the usual procedure with which we intend to control them, namely randomized trials, can really eliminate these effects. Possible consequences of sequential effects in psychological measurement in general are further discussed, and in the end a simple solution on how to minimize them is proposed