Academic journal article Memory & Cognition

Divided Attention Modulates Semantic Activation: Evidence from a Nonletter-Level Prime Task

Academic journal article Memory & Cognition

Divided Attention Modulates Semantic Activation: Evidence from a Nonletter-Level Prime Task

Article excerpt

Research has recently shown that semantic activation is modulated in proportion to the amount of attention required for letter-level processing of the prime (the attention modulation hypothesis; Smith, Bentin, & Spalek, 2001). In this study, we examined this hypothesis with an auditory divided-attention task. Participants were asked to decide whether the pitch of a probe tone presented with the prime word was higher or lower than the basic tone presented with the fixation cross. Their target task was lexical decision to the target word. Experiment 1 showed that semantic priming was modulated by the amount of attentional resources. Moreover, in Experiment 2, this modulation was also found in a situation that eliminated the possibility of participants' response strategies. Yet, Experiment 3 showed repetition priming to be unaffected. These results support an amended attention modulation hypothesis in which modulation is not limited to letter-level processing.

Priming refers to the idea that the processing of a target is facilitated when it is preceded by a related prime. Meyer and Schvaneveldt (1971) performed the classic demonstration of semantic priming in which participants were asked to judge whether pairs of items were words or nonwords. The words included pairs of related words (e.g., bread and butter) and pairs of unrelated words (e.g., nurse and butter). Participants responded 85 msec faster to the related pairs than to the unrelated pairs. This result was confirmed by a great many additional studies (e.g., Neely, 1977, 1991).

Priming has been explained using the concept of spreading activation (Collins & Loftus, 1975), in which identification of the prime word activates related words including the target word, thereby making the target easier to judge. Moreover, using a variety of experimental procedures, semantic priming has also been explained by conscious expectancy (Becker, 1980; Neely, 1977), strategic semantic matching (Neely, Keefe, & Ross, 1989), compound cuing (McKoon & Ratcliff, 1992), episodic prime retrieval (Bodner & Masson, 2001), and automatic semantic integration (Chwilla, Hagoort, & Brown, 1998; Hodgson, 1991).

Although many factors may influence semantic priming, one focus of recent work has been on the way that the prime is processed. In particular, studies have examined the effect that dividing attention by adding a secondary task has on processing of the prime. These studies have shown that semantic priming is not observed when participants direct their attention to the letter level of the prime (e.g., by searching for a particular letter in the prime) rather than to the prime as a lexical or semantic entity. For example, Henik, Friedrich, and Kellogg (1983) investigated the influence of prime task (naming vs. letter search) on target task (lexical decision and word color naming), and they reported the absence of priming on both target tasks when the prime task was letter search. This result suggests that semantic activation is influenced by the level of prime word processing (Besner, Smith, & MacLeod, 1990; Chiappe, Smith, & Besner, 1996; Friedrich, Henik, & Tzelgov, 1991; Henik etal., 1983; Henik, Friedrich, Tzelgov, & Tramer, 1994; Hoffman & MacMillan, 1985; Kahneman & Hemic, 1981; Kaye & Brown, 1985; MariBeffa, Fuentes, Catena, & Houghton, 2000; Smith, 1979; Smith, Meiran, & Besner, 1996,2000; Smith, Theodor, & Franklin, 1983; Stolz & Besner, 1996, 1997,1998).

Besner et al. (1990) investigated the influence of the letter search task on priming when the prime and target were presented simultaneously. Their task required participants to judge whether two words had letters in common. The reaction times (RTs) for letter search were slow, especially for pairs of related words as opposed to unrelated words. In addition to the explanation of attention required to process the related words and interference generated via letterlevel activation, Besner et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.