Referential processing: Reciprocity and correlates of naming and imaging

Abstract

To shed light on the referential processes that underlie mental translation between representations of objects and words, we studied the reciprocity and determinants of naming and imaging reaction times (RT). Ninety-six subjects pressed a key when they had covertly named 248 pictures or imaged to their names. Mean naming and imagery RTs for each item were correlated with one another, and with properties of names, images, and their interconnections suggested by prior research and dual coding theory. Imagery RTs correlated .56 (df = 246) with manual naming RTs and .58 with voicekey naming RTs from prior studies. A factor analysis of the RTs and of 31 item characteristics revealed 7 dimensions. Imagery and naming RTs loaded on a common referential factor that included variables related to both directions of processing (e.g., missing names and missing images). Naming RTs also loaded on a nonverbal-to-verbal factor that included such variables as number of different names, whereas imagery RTs loaded on a verbalto-nonverbal factor that included such variables as rated consistency of imagery. The other factors were verbal familiarity, verbal complexity, nonverbal familiarity, and nonverbal complexity. The findings confirm the reciprocity of imaging and naming, and their relation to constructs associated with distinct phases of referential processing.Mental translation between verbal and nonverbal information is an important associative activity. Such code switching has been termed referential processing (Paivio, 1971(Paivio, , 1986 or referential activity (Bucci, 1984;Bucci & Freeman, 1978), and it is reflected most directly in picture-naming and word-imaging tasks. Referential processing is implicated as well in memory of pictures and words, symbolic comparisons, language comprehension, cognitive deficits after brain damage (e.g., Kosslyn, 1987;Thompson, Hall, & Sison, 1986), and other cognitive phenomena, and it raises important conceptual issues on which contemporary theories of mental representation differ. Dual coding theory (Paivio, 1971(Paivio, , 1986 maintains that verbal and nonverbal representations are directly connected in a one-to-many fashion in both directions, whereas common coding models assume that crossmodal processing requires activation of shared abstract representations (e.g., Potter & Faulconer, 1975;Snodgrass, 1984). Research on naming and imaging also addresses specific theoretical issues, as, for example, the question of whether images are constructed sequentially from components (e.g., Kosslyn, 1980) or are activated holistically (e.g., Paivio, 1971, p. 58; 1986, p. 60).Despite the familiarity and theoretical importance of referential processing, we lack systematic information This research was supported by grant AOO87 to Allan Paivio from the Natural Sciences and Engineering Research Council of Canada. We thank Carla Johnson for helpful comments on earlier drafts of the paper, and Kal Csapo, who collected some of the older data that we used. Address correspondence to Allan Paivio, Department...

The picture superiority effect in a cross-modality recognition task

“…Within dualcode theories, priming has more rarely been discussed. The theory does, however, assume cross-modal (referential) connections between the two memory stores and allows for the possibility that connections in different directions may be unequally developed (Paivio, Clark, Digdon, & Bons, 1989).…”

Extensions of the Paivio, Yuille, and Madigan (1968) norms

“…In selecting additional attributes to measure, our goal was to build on one systematic conceptualization of item attributes that emerged from Paivio's (1971Paivio's ( , 1986 dualcoding theory framework and was applied to good effect in the Paivio et al (1989) picture naming-imaging study. There we found that different measures coalesced in a reasonable way to suggest specific phases of processing (e.g., picture or word identification, image or name activation, response production), and that these different phases mapped nicely onto what Paivio (1971) has called representational, referential, and associative levels of processing.…”

“…Research on item attributes continues to become more sophisticated in a variety of ways, including consideration of larger numbers of properties (e.g., Paivio et al, 1989;Rubin, 1980), and serious efforts to model and simulate the effects of stimulus attributes (e.g., Ellis & Lambon Ralph, 2000). Computers have played a central role in both of these developments.…”

“…We can illustrate with one controversial question-the relative importance of frequency and age of acquisition in picture naming and other semantic retrieval tasks (Morrison & Ellis, 2000). Paivio et al (1989) obtained picture naming and imagery latencies for a moderate-sized pool of pictures and their most common labels. Information on a wide range of properties, including age of acquisition, was also obtained.…”

“…Empirically, imagery and concreteness have been repeatedly demonstrated to be potent variables in a wide range of psychological tasks, including learning and memory tasks (e.g., Paivio, 1969), semantic retrieval tasks (e.g., Paivio, Clark, Digdon, & Bons, 1989), and diverse other cognitive tasks (for reviews, see Paivio, 1971Paivio, , 1986Paivio, , 1991. Although the empirical effects are well founded, their theoretical basis remains controversial, as researchers advanced other correlated properties such as context availability (Schwanenflugel, 1991;Schwanenflugel & Various aspects of this work were supported by research grants from the Natural Sciences and Engineering Research Council of Canada.…”