[Color-word interference in relation to comparative compatibility: evidence for two sources of interference].

In previous studies on colour-word interference ('Stroop-phenomenon') stimulus-response-compatibility (word-speech, colour-speech) and task mode (word reading, colour determining) have usually been confounded or not been crossed completely. An experiment is reported where these two factors were varied orthogonally (besides type of stimulus: incongruent, congruent and control). Compatibility was varied by using response keys with varying labelling of the keys (colours or words). The results show--contrary to traditional expectations--the usual pattern of interference in spite of the response key conditions. The existence of response delays of different magnitude indicates separate sources of interference. Encoding interference is found under conditions of high-compatible colour determining by coloured response keys (colour-colour key), recoding interference under conditions of low-compatible reading (word-colour key), both presented here in an isolated form. Both modes of interference should therefore contribute to the interference under conditions of low-compatible colour determining which is the traditional Stroop condition. In an attempt to predict this complete interference encoding interference was shown to be the more stable predictor. The same was true of interference as measured here in a list version.