Katharine A. Tillman

Parts, Wholes, and Context in Reading: A Triple Dissociation

Research in object recognition has tried to distinguish holistic recognition from recognition by parts. One can also guess an object from its context. Words are objects, and how we recognize them is the core question of reading research. Do fast readers rely most on letter-by-letter decoding (i.e., recognition by parts), whole word shape, or sentence context? We manipulated the text to selectively knock out each source of information while sparing the others. Surprisingly, the effects of the knockouts on reading rate reveal a triple dissociation. Each reading process always contributes the same number of words per minute, regardless of whether the other processes are operating.

Erratum: The uncrowded window of object recognition

In the version of this article initially published, the legend to Figure 5 contained several errors. The second sentence should read “Fixating on the red minus, you will be unable to identify the middle object in the first eight rows unless you isolate it by hiding the flanking objects with your fingers (or two pencils).” The fourth sentence should read “Grating patches, similar to those in the top row, are often taken to be one-feature objects.” These errors have been corrected in the HTML and PDF versions of the article. 10 0 10 1 10 0 10 1 10 2 10 3

Crowding limits reading

Peripheral reading is slow, even in RSVP, which minimizes the need for eye movements. This affects everyone who reads with central field loss, yet is still unexplained. Legge et al. suggest that reading rate is proportional to visual span, which shrinks with increasing distance from fixation. Both centrally and peripherally, we show that visual span and reading rate are limited by crowding.

is reading serial

B. Are letters acquired in parallel or serially? If letters within a word are acquired serially, we should find no difference in reading rate if letters are shown one at a time. Taking away letters you’re not using at that moment shouldn’t hurt. We tested this prediction by reducing the number of letters shown simultaneously (Fig. 1) and find that, with serial letter presentation, reading rate for 4-letter words is reduced by about a factor of 4 (Fig. 2a). In other words, required word duration quadruples when we present the letters serially (Fig. 2b), but required letter duration is independent of how many letters are presented at once (Fig. 2c). Letters are acquired in parallel (independently).