Young Children Use Letter Names in Learning to Read Words

Most theories of reading development claim that young children are logographic, or prealphabetic, readers, unable to take advantage of the systematic links between spellings and sounds that exist in an alphabetic writing system. To test this view, we taught preschool and kindergarten children to pronounce three types of madeup spellings. In the name condition, the name of the word’s first letter was heard in the spoken word (e.g., BT for “beet”). In the sound condition, letter-sound cues were present but letter-name cues were not (e.g., BT for “bait”). Neither type of cue was available in the visual condition (e.g., BT for “ham”). Even children who were unable to read any simple words performed better in the name condition than the other two conditions. Thus, children are able to benefit from links between spellings and sounds when these links involve the names of letters. The prevailing view of reading acquisition is that children start out as logographic, or prealphabetic, readers (e.g., Ehri, 1998; Frith, 1985; Gough & Hillinger, 1980). Unable to take advantage of the systematic relations between spellings and sounds that exist in an alphabetic writing system, young children treat printed words as if they were arbitrary symbols, or logographs. Prealphabetic readers do not relate the letters in a word’s printed form to the sounds in its spoken form. They choose some salient visual characteristic of the printed word— the golden arches in the McDonald’s logo, for example—and link that to the word’s pronunciation or meaning. Views differ on just when children begin to break away from logographic reading. According to Gough and Hillinger (1980), children learn to read their first 40 or so words by means of visual cues. In contrast, the results of Ehri and Wilce (1985) and Rack, Hulme, Snowling, and Wightman (1994) suggest that children who can read only a few words have already entered what Ehri (1998) calls the partial alphabetic phase. Such children are able to relate at least some of the letters in a word to their sounds. Despite the areas of disagreement, most investigators concur that there is a stage early in the development of reading during which children are not able to take advantage of systematic relationships between spellings and sounds. By default, these children resort to a nonalphabetic strategy. Support for the view of early reading as nonalphabetic comes from the results of Ehri and Wilce (1985). These investigators taught children to read a set of phonetic spellings, such as MSK for “mask,” and a set of visual spellings, such as UHE for “comb.” Children who could not read any simple words learned the visual spellings more readily than the phonetic spellings. These children presumably remembered the visual spellings more easily because they looked distinctive, with the letters of each word varying in size and each word having a unique contour. In contrast, children who could already read some real words learned the phonetic spellings more easily than the visual spellings. In the present study, we asked whether prereaders are indeed restricted to a logographic approach in their initial learning attempts. We suspected, given previous research, that even prereaders would benefit from certain kinds of relations between print and speech— specifically, those based on the names of letters. Consider the word “beet.” The /bi/ at the beginning of the spoken word corresponds to the name of the letter b, and so it makes sense that the printed word starts with b. With “bait,” in contrast, /b/ appears at the beginning of the spoken word, but the entire letter name is not present. The spelling of this word may thus be more difficult for a young child to grasp. In a previous study (Treiman, Tincoff, & Richmond-Welty, 1996), children performed better on words like “beet” than words like “bait” in a simple spelling task in which they were asked to orally provide the first letter of a spoken word. This and other findings on the importance of letter names for young children (e.g., Ehri, 1983; Treiman, Tincoff, Rodriguez, Mouzaki, & Francis, 1998) led us to suspect that letter names may play a role in early reading. We assessed the use of letter names and sounds in learning to read by adapting the word learning task of Ehri and Wilce (1985). Each child learned to pronounce three types of made-up spellings. In the name condition, the printed stimulus BT was pronounced as “beet.” The initial letter provided both a letter-name cue (the spoken word begins with /bi/) and a letter-sound cue (the spoken word begins with /b/). Only lettersound cues were available in the sound condition, as when BT was taught as the spelling of “bait.” Neither type of cue was present in the visual condition, in which the shapes of the printed stimuli were more distinctive but the pronunciations were essentially arbitrary (e.g., BT for “ham”).