6.5 Reading Processes, Bilingualism, and Verbal and Nonverbal Communication

Reading Processes, Bilingualism, and Verbal and Nonverbal Communication

The GRE content outline lists reading processes, verbal and nonverbal communication, and bilingualism as separate language topics alongside units, syntax, meaning, and speech perception (Section 6.1). Expect questions that describe an eye-tracking result, a dyslexic patient's errors, or a cross-cultural observation about gestures or expressions.

1. Eye Movements During Reading

Reading feels smooth, but the eyes move in jumps:

• Saccades: Rapid ballistic movements, usually about 7–9 letter spaces in English. Vision is largely suppressed during each saccade (saccadic suppression).
• Fixations: Pauses of roughly 200–250 ms during which information is taken in. Fixations are longer on low-frequency, unpredictable, or long words (the word frequency effect).
• Regressions: Backward saccades, about 10–15% of eye movements in skilled readers, often triggered by comprehension difficulty.
• Perceptual span: Using the moving-window technique, George McConkie and Keith Rayner (1975) changed text outside a window around fixation and measured when reading slowed. In English the useful span extends about 3–4 letters to the left of fixation and about 14–15 letters to the right; for readers of Hebrew, which is read right to left, the asymmetry reverses. The span therefore reflects attention, not just acuity.

2. Recognizing Written Words

• Word superiority effect: Gerald Reicher (1969) and Daniel Wheeler (1970) flashed a word (WORK), a single letter (K), or a nonword (OWRK), followed by a mask, and asked which of two letters had appeared (D or K in the final position). Accuracy was higher for letters in words than for letters alone, even though either alternative completed a word.
• Interactive activation model: James McClelland and David Rumelhart (1981) explained the effect with three levels (features, letters, words) connected by excitatory and inhibitory links. Activated word units feed back to support their letters, an early demonstration of top-down processing in perception.
• Lexical decision and priming: Deciding whether a string is a word is faster for frequent words and after a related prime (Section 5.2), indicating that the mental lexicon is organized by both frequency and meaning.

The Dual-Route Model

Max Coltheart's dual-route account proposes two pathways from print to pronunciation:

Selective damage produces contrasting acquired dyslexias:

• Surface dyslexia: Lexical route impaired. The patient reads regular words and nonwords but regularizes irregular words (reading pint to rhyme with mint).
• Phonological dyslexia: Sublexical route impaired. Familiar words, regular or irregular, are read well, but pronounceable nonwords cannot be sounded out.
• Deep dyslexia: Severe phonological impairment plus semantic errors (reading forest as "trees"), suggesting reading through meaning alone.

Developmental dyslexia is most often linked to a deficit in phonological awareness (the ability to segment and manipulate speech sounds). Phonological awareness in kindergarten predicts later reading, and the U.S. National Reading Panel (2000) concluded that systematic phonics instruction improves early reading more than approaches that omit it.

3. Sentence and Text Comprehension

• Garden-path sentences: "The horse raced past the barn fell" (Thomas Bever, 1970). Readers initially treat raced as the main verb and must reanalyze it as a reduced relative clause. Lyn Frazier's minimal attachment principle holds that the parser initially builds the simplest structure.
• Situation models: Walter Kintsch's construction-integration model proposes that readers build a propositional text base and integrate it with world knowledge into a mental model of the situation described. Memory for gist outlasts memory for exact wording.
• Inferences: Readers routinely draw bridging inferences ("Mary dropped the vase. She swept up the pieces" implies the vase broke).

4. Bilingualism

• Simultaneous bilinguals learn two languages from infancy; sequential bilinguals add a second language later. Age of acquisition strongly predicts native-like grammar and accent (Johnson and Newport's 1989 findings appear in Section 6.1).
• Additive versus subtractive bilingualism (Wallace Lambert): additive bilingualism adds a second language while the first remains valued; subtractive bilingualism occurs when the second language replaces the first, often with educational costs.
• Both languages are active: Eye-tracking and priming studies show that hearing or reading words in one language activates similar-sounding words in the other (language-nonselective access). Bilinguals must therefore control interference, and they frequently code-switch between languages in rule-governed ways.
• Revised Hierarchical Model (Judith Kroll and Erika Stewart, 1994): early in second-language learning, L2 words connect to meaning mainly through their L1 translations; with proficiency, direct L2-to-concept links strengthen. Translating from L2 to L1 is therefore faster than from L1 to L2 for less proficient learners.
• Vocabulary: Bilingual children often know fewer words in each single language than monolingual peers but a similar or larger number of concepts across both languages.
• The bilingual advantage debate: Ellen Bialystok's studies reported better inhibitory and attentional control in bilinguals, and later work suggested delayed dementia onset. Large-sample studies and meta-analyses (e.g., Kenneth Paap and Zachary Greenberg, 2013) often find little or no executive-function advantage once publication bias and confounds such as immigration status and socioeconomic status are controlled. Treat the claim as contested.

5. Verbal and Nonverbal Communication

Verbal communication follows conversational rules: turn-taking, repair of misunderstandings, and Grice's maxims (Section 6.1). Nonverbal channels carry much of the emotional and relational meaning:

• Thin slices: Nalini Ambady and Robert Rosenthal (1993) found that ratings of 30-second silent video clips of college teachers correlated strongly (about r = .76) with end-of-semester student evaluations.
• Lie detection: In Charles Bond and Bella DePaulo's (2006) meta-analysis, people distinguished truths from lies only about 54% of the time, barely above chance. Popular cues such as gaze aversion are unreliable.
• The 7-38-55 myth: Albert Mehrabian's studies suggested that when words and tone conflict in conveying liking, listeners rely more on facial expression (55%) and tone (38%) than words (7%). The figures do not show that 93% of all communication is nonverbal.

Language Versus Animal Communication

Charles Hockett (1960) proposed design features of human language, including displacement (talking about things not present), productivity (generating novel sentences), arbitrariness of the sound-meaning link, and duality of patterning. Honeybee dances show displacement but little productivity (Section 3.3). Apes taught signs or lexigrams (Washoe, Kanzi) learned large vocabularies, but Herbert Terrace's analysis of the chimpanzee Nim Chimpsky (1979) found little syntax and much imitation of trainers.

Sign languages such as American Sign Language are full natural languages with phonology, morphology, and syntax. Deaf signers with left-hemisphere damage develop sign aphasias, and children in Nicaragua created a new sign language within a generation once deaf children were brought together in schools, showing that language structure can emerge from a community of child learners.