Serial Processing Shown by Mutual Masking of Icons

Abstract

In a procedure called mutual masking, two compound words were tachistoscopically exposed successively at the same location. Perceptual accuracy for both words was examined. With fixation at the center of the arrays, Ss often identified the component at the left within the temporally first word and at the right within the second. These results imply a left-to-right serial processing. With fixation on begining or end component, Ss tended to identify the component away from fixation of the first word and the fixated part of the second. This implies scanning from periphery toward fixation. Therefore, sequential processing of iconic information was obtained, with direction of processing determined by specific conditions.

‘Fell’ primes ‘fall’, but does ‘bell’ prime ‘ball’? Masked priming with irregularly-inflected primes

“…This is the general interpretation of the fact that brothel does not prime broth, while corner primes corn (e.g., Longtin et al, 2003;Marslen-Wilson, Bozic, & Randall, 2008;Rastle et al, 2004;, and we have endorsed this view in the present work. However, it is not clear how this constraint would be implemented in a computational version of the model, particularly if assuming that information on letter identity is fed to the word recognition system in a left-to-right, serial fashion (see Davis, 1999;Harcum & Nice, 1975;O'Regan & Jacobs, 1992;Rumelhart and McClelland, 1982; but also see Diependele, Sandra, & Grainger, 2009 for the suggestion that morpho-orthographic decomposition occurs in parallel). If implemented as a classical interactive activation system, the model would now predict that a left-to-right parsing of the string brothel would lead to the activation of the morpho-orthographic unit for broth; it is hard to see how the subsequent processing of the cluster el could "block" the activation of {broth}, so as to prevent masked priming from arising in brothel-BROTH.…”

Supporting the serial in the SERIOL model

“…If referring to the sub-lexical route, I will denote it explicitly. Early accounts of string processing widely assumed that the visual image was read out serially (Gough, 1972;Harcum & Nice, 1975;Mewhort, Merikle & Bryden, 1969;Sperling, 1963). However, it is currently generally assumed that all letters are activated in parallel and that lexical access occurs in parallel (Coltheart, Rastle, Perry, Langdon & Ziegler, 2001;Harm & Seidenberg, 1999;McClelland & Rumelhart, 1981).…”

“…Previously, the pendulum swung from the assumption of a serial readout of a letter string (Gough, 1972;Harcum & Nice, 1975;Mewhort, Merikle & Bryden, 1969;Sperling, 1963) to the assumption of parallel processing (Coltheart et al, 2001;McClelland & Rumelhart, 1981;Seidenberg & McClelland, 1989). Given the lack of a valid argument against a serial encoding, the direct evidence for serial processing (Nice & Harcum, 1976), the ability of a serial encoding to provide important representational functions, and the explanatory capacity of SERIOL's serial mechanism, I suggest that it is time for the pendulum to swing back to seriality (and remain there).…”

“…However, seriality can be investigated by using time directly. Harcum and Nice (1975) used this approach in an experiment in which two six-letter pseudoword strings were sequentially presented, and the task was to report the letters perceived. The first string was presented for 70 ms, and the second string was presented for 35 ms. Temporal order significantly interacted with letter position.…”

The process of perceptual retouch: Nonspecific afferent activation dynamics in explaining visual masking

“…This forces us to emphasize that the Table 2 The Main Table for Selected Experimental Findings Subliminal effects of masked semantic information (e.g., Allport, 1977;Marcel & Patterson, 1978) -t Subliminal effects of masked geometric information (e.g., Kapustin, 1979;Gellatly, 1980) Masking increase with the increase in TS energy (e.g., Purcell et al, 1969Purcell et al, , 1975 The "halo" effect (e.g., Werner, 1935;Stoper & Mansfield, 1978) X Reappearance of TS after TS-MS fast recycling (Schiller & Smith, 1966) X Lack of spatial frequency specificity of metacontrast at the whole SOA range (Growney, 1978) X Nonretinotopic ("spatiotopic") nature of masking (White, 1976;Lehmkuhle & Fox, 1980) Asymmetric nonrandom recombinations of portions of TS and MS into single percept (Harcum & Nice, 1975) Dependence of MS perceptive quality on spatial distribution of the backwardmasked TS elements (Carlson & Mayzner, 1977) X Cohene and Bechtoldt effect (1974 -t The early EP components unchanged with masking (Bridgeman, 1980) Operational sufficiency index" term "nonspecific," which refers to certain physiological realities, does not necessarily mean something that is spatially undifferentiated. We can explain the asymmetric nature of traditional metacontrast masking, in which the disk-ring sequence leads to nonmonotonic backward masking but the ring-disk sequence does not, as follows.…”