Reading Aloud: Evidence for Contextual Control Over Lexical Activation Roy Ferguson, Serje Robidoux, and Derek Besner University of Waterloo Can readers exert control (albeit unconsciously) over activation at particular loci in the reading system? The authors addressed this issue in 4 experiments in which participants read target words aloud and the factors of prime–target relation (semantic, repetition), context (related, unrelated), stimulus quality (bright, dim), and relatedness proportion (RP; high, low) were manipulated. In the high RP condition (RP = .5), an interaction between semantic context and stimulus quality was observed in which low stimulus quality slowed unrelated targets more than related ones, replicating previous work. In contrast, the low RP condition (RP = .25) yielded additive effects of semantic context and stimulus quality. However, when low RP was examined within the context of repetition priming, context and stimulus quality once again interacted. These results are discussed in the context of a widely endorsed framework with the addition of the central assumption that there is control over feedback between various levels. Keywords: lexical activation, priming, contextual control, relatedness proportion It is well established that the prior presentation of a word that is semantically related to a target word yields faster responses to the target relative to an unrelated prime in both lexical decision and reading aloud (see McNamara, 2005, for a review). Stolz and Besner (1996) provided an account of this semantic relatedness effect in the context of an interactive activation framework with three levels 1 : a letter level, an orthographic input lexicon (OIL), and a semantic level (see Figure 1). In their account, when a word is processed, appropriate letter level representations are activated and this activation feeds forward to localist representations at the word level, which in turn activates semantic representations of both the word and its associates. In addition to activation feeding from lower to higher levels, it also feeds back from the semantic level to the word level and from the word level to the letter level. The Joint Effects of Semantic Context and Stimulus Quality A number of factors have been shown to interact with the effect of semantic relatedness. Stimulus quality (often imple- mented as a reduction in contrast but sometimes implemented via masking or the superimposition of dots on the stimulus) is one such factor. In both lexical decision and reading aloud, related targets are less affected than unrelated targets by a reduction in stimulus quality (e.g., Meyer, Schvaneveldt, & Ruddy, 1975). One account, couched within the context of lexical decision, contends that this interaction arises because although the effect of stimulus quality does not extend beyond the OIL, 2 there is feedback from prime processing at the se- mantic level to semantically related representations at the lex- ical level. Accordingly, the related target requires less bottom-up activation because of this feedback from these asso- ciates of the prime. Unrelated targets do not receive any benefit from top-down influences and are therefore more impaired by low stimulus quality (Borowsky & Besner, 1993; Stolz & Neely, 1995). Lexical Decision: Relatedness Proportion (RP) and the Interaction Between Semantic Context and Stimulus Quality Stolz and Neely (1995) reported that in the context of lexical decision, the proportion of related word trials (RP) modulates the interaction between semantic context and stimulus quality for strong associates. When RP is high (.5), semantic context and stimulus quality interact, but when RP is low (.25), semantic context and stimulus quality have additive effects on reaction time (RT). Given the assumption that the effect of stimulus quality does not extend to the semantic level, Stolz and Neely argued that the three-way interaction between stimulus quality, RP, and context is the result of contextual control over feedback from the semantic system (SEM) to the OIL (via the SEM 3 OIL pathway in Figure 1). When RP is high, feedback from semantics to the word level (for the prime’s associates) is operative, but when RP is low, this 1 For simplicity, we ignore the feature level (although it is obviously affected by stimulus quality). 2 More current accounts suppose that in the context of lexical decision (but not reading aloud), stimulus quality affects feature and letter level processing, but not beyond (see M. Brown et al., 2006; O’Malley et al., 2007; Yap & Balota, 2007; Yap, Balota, Tse, & Besner, 2008). Roy Ferguson, Serje Robidoux, and Derek Besner, Department of Psychology, University of Waterloo, Waterloo, Ontario, Canada. This work was supported by a Natural Sciences and Engineering Re- search Council of Canada (NSERC) doctoral fellowship to Roy Ferguson and Grant A0998 from NSERC to Derek Besner. Correspondence concerning this article should be addressed to Roy Fergu- son or Derek Besner, Department of Psychology, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada. E-mail: ra2fergu@watarts.uwaterloo.ca or dbesner@uwaterloo.ca Journal of Experimental Psychology: © 2009 American Psychological Association Human Perception and Performance 2009, Vol. 35, No. 2, 499 –507 0096-1523/09/$12.00 DOI: 10.1037/a0013162 499