Dissociated imageability, concreteness, and familiarity in lateralized word recognition

Abstract

According to Paivio's (1971) dual-coding theory, the representational memory of words is indexed by familiarity ratings, whereas associated imagery (one type of referential memory) is indexed by imageability and concreteness ratings. The theory predicts that word recognition near threshold will be influenced only by the former and not by the latter two attributes.However, previous empirical findings are unclear on this issue. Furthermore, the report of some studies that imageability and concreteness interact with the visual field of presentation pose a potential challenge to the theory. Here, five experiments present laterallzed words whose semantic attributes are dissociated by design and by correlational analysis. In support of the theory, a metaanalysis shows that only familiarity affects overall recognition and that none of the attributes interacts with visual field. Yet statistical power to find the effects was high, and a general right-field recognition advantage supports the hemispheric validity of the findings. Implications for continuous (e.g., "cascade") models of information processing are discussed.Following publication of Paivio's (1971) seminal work on the role of imagery in psychological processes, considerable interest has centered on the place of imagery in word perception (e.g., Bradshaw, Nettleton, & Taylor, 1981;Day, 1979;Marcel & Patterson, 1978;Richardson, 1976). Typically, "imageability ," the rated capacity of a word to evoke imagery (Paivio, Yuille, & Madigan, 1968), is treated as one attribute that may affect perceptual threshold, percent correct, or reaction time (RT) measures of word recognition. Other semantic word attributes often investigated are "concreteness," the degree to which a word refers to an object, material, or person, and "familiarity," the rated frequency of word usage or exposure.In Paivio's (1971) framework, there are close correspondences between these dimensions and theoretical "representational" and "referential" meanings of words. Familiarity is taken as the major index of representational meaning, the availability of a verbal code for psychological processing. In constrast, imageability and concreteness are considered to be essentially the same dimension and to index one type of referential meaning, the association between word and image. In informationprocessing terms, it is representational meaning that is at a lower level and that determines threshold and nearthreshold recognition of words, whereas referential meaning is at a higher associationallevel and thus exerts little impact on recognition per se. By this account, familiarity but not imageability or concreteness should affect the recognition for words near threshold, since asso-

Hemispheric Asymmetry in Lexical Decisions: The Effects of Grammatical Class and Imageability

“…Imageability/concreteness has a well-documented effect on a variety of foveal tasks, including recall, lexical decisions, and sentence comprehension (e.g., Bleasdale, 1987;James, 1975;Paivio, 1991). Similarly, in divided visual field experiments it is also a com- mon result that concrete/High Imagery words show an advantage over abstract/Low Imagery words (e.g., Day, 1979;Eviatar et al, 1990;Hines, 1976;Lambert & Beaumont, 1983;McMullen & Bryden, 1987;Schmuller & Goodman, 1979), although some discordant results have been reported (e.g., Boles, 1983;Howell & Bryden, 1987).…”

“…Among the studies which use naming tasks, the most common result is the absence of any effect of imageability on the differences between the fields of presentation. (e.g., Boles, 1983;Lambert & Beaumont, 1983;Schmuller & Goodman, 1979). When an imageability effect has been observed (e.g., Bradshaw & Gates, 1978;Ellis & Shepherd, 1974;Hines, 1976Hines, , 1977Marcel & Patterson, 1978;Young & Ellis, 1985), different methodological factors bring into question the results obtained: failure to control the order of report, lack of control of the lexical frequency, small word samples, number of errors too small for statistical analysis, and so on.…”

“…According to some authors, the habitual right visual field advantage-left hemisphere superiority (RVF-LH) attenuates or even disappears when concrete and/or highly imageable and/or high frequency words are presented (e.g., Day, 1977Day, , 1979Ellis & Shepherd, 1974;Hines, 1976Hines, , 1977Mannhaupt, 1983;Young & Ellis, 1985). Other studies, however, have not observed relevant effects from these variables (e.g., Boles, 1983;Howell & Bryden, 1987;Koening, Wetzel, & Caramazza, 1992;Lambert & Beaumont, 1983;Leiber, 1976;McMullen & Bryden, 1987).…”

Split fovea theory and the role of the two cerebral hemispheres in reading: A review of the evidence

“…In contrast to this evidence that LVF words may fail to engage lexical representations during the performance of the Reicher-Wheeler task, numerous studies involving naming or lexical decision have found lexical influences on the processing of LVF words. Thus, word frequency and imageability have been shown to affect the speed and accuracy of responses to LVF as well as to RVF words in naming and/or lexical decision (e.g., Boles, 1983;Chiarello, Liu, Quan, & Shears, 2000;Coney, 2005;Iacoboni & Zaidel, 1996;Scott & Hellige, 1998). This implies that the cognitive processes recruited in the performance of the Reicher-Wheeler task are not the same as those recruited in on-line word recognition tasks, particularly when words are presented in the left visual field beyond the fovea.…”

Does Global Context Modulate Cerebral Asymmetries? A Review and New Evidence on Word Imageability Effects

“…This represents a rather static account of the distribution of function across hemispheres, and tests of this hypothesis produced variable, and only partially supportive, results (e.g, Boles, 1983;Chiarello, 1988). However, qualitative asymmetries in word recognition processes need not imply separate lexicons; indeed, it is likely that lexical information of various sorts is widely distributed throughout the cortex (Roland, 1993).…”