The semantic interference effect observed in Stroop tasks and picture-word interference tasks might be due to the previous confounding of semantic similarity with task relevance (in the Stroop task) and with perceptual similarity (in the picture-word interference task). A picture-word variant ofthe Stroop task was devised in which the factors of task relevance and perceptual similarity were controlled. The distractor conditions allowed for the examination of four types of context effects. The results show that the overall Stroop-like interference effect can be decomposed into interference effects due to (1) a semantic relation between distractor and target, (2) the semantic relevance of the distractor word in the task at hand, (3) the presence of the distractor word in the response set, and (4) the mere presence of a word. Implications of these findings for the locus or loci of Stroop and picture-word interference effects are discussed. It is concluded that distractor words in Stroop-like naming tasks interfere mainly in the process of name retrieval.In the Stroop task, subjects are required to name colors that are part of, or accompanied by, a printed word (see, e.g., Glaser & Glaser, 1982;Stroop, 1935); in the picture-word interference task, subjects are required to name pictures that are accompanied by a printed word (see, e.g., Rosinski, Golinkoff, & Kukish, 1975;Underwood, 1976). In the Stroop task (see, e.g., Fox, Shor, & Steinman, 1971;Klein, 1964), and in the picture-word interference task (see, e.g., Glaser & Diingelhoff, 1984;Guttentag & Haith, 1978;Lupker, 1979;Rosinski, 1977), a distractor word that is semantically related to the target seems to induce more interference than a distractor word that is unrelated to the target. This semantic interference effect stands in marked contrast to the semantic facilitation effect observed in tasks in which subjects are required to read a target word that is preceded by, or presented simultaneously with, a prime word (e.g., Dallas & Merikle, 1976; La Heij, Van der Heijden, & Schreuder, 1985;Warren, 1977).A conclusion that could be drawn from these results is that a semantic relation between a distractor word (or prime word) and a target will hamper performance in tasks in which nonverbal targets have to be named (as in the Stroop task and the picture-word interference task) but will facilitate perfonnance in tasks in which verbal targets have to be read (as in the word-reading variant of the semantic priming task). Before accepting such a conclusion, it would seem worthwhile to determine whether the discrepancy can be attributed to other characteristics of the various tasks.Further inspection of the Stroop task and the pictureword interference task reveals that in both of these tasks the semantic relatedness factor is confounded with another Requests for reprints should be sent to W. La Heij, Department of Psychology, Unit of Experimental Psychology, University of Leiden, Hooigracht 15,2312 KM Leiden, The Netherlands. factor that might be responsible for the int...