The binding problem for syntax, semantics, and prosody: H.M.'s selective sentence-reading deficits under the theoretical-syndrome approach

Abstract

In this case study, a ''hippocampal amnesic'' (H.M.) and memory-normal controls of similar age, background, intelligence, and education read novel sentences aloud in tasks where fast and accurate reading either was or was not the primary goal. In four experiments, H.M. produced more misreadings than normal and cerebellar controls, usually without self-correction. H.M.'s misreadings typically reduced semantic and syntactic complexity and caused ungrammaticality by omitting short high-frequency function-words. H.M. also produced each word more slowly and paused longer than controls at three points: before beginning to produce a sentence, between words in unfamiliar phrases, and at major syntactic boundaries unmarked by commas. H.M.'s selective sentence-reading de cits were unrelated to word-speci c factors, ambiguity, and sentence length, and were not attributable to his cerebellar damage, speed-accuracy trade-off, general slowing, general cognitive decline, left-to-right reading processes, or limitations in workingmemory capacity. However, present results supported a ''theoreticalsyndrome approach'' under which all of H.M.'s de cits (in reading sentences, in comprehending and producing spoken sentences, in reading isolated words and pseudo-words, in visual cognition, and in recall from episodic memory) form part of a general, theoretically coherent syndrome that generalises to other patients.

Sequencing, Speech Production, and Selective Effects of Aging on Phonological and Morphological Speech Errors.

“…However, we observed no age-linked increase in nonword errors because representing and producing nonwords as coherent units requires new connection formation, a process subject to age-linked decline (see, e.g., MacKay & James, 2001). We likewise observed a greater agelinked increase in errors on suffixes versus other syllable-final consonants because new connection formation is required to represent and produce suffixes but not syllable-final segments in familiar words (see, e.g., MacKay & James, 2001). Age-linked deficits in connection formation (see, e.g., MacKay & Abrams, 1996), therefore, made suffixes more error prone than syllablefinal consonants for older adults.…”

“…Another NST prediction concerned the lexical bias effect, in which speech errors result in real words more often and nonwords less often than would be expected by chance (e.g., Baars et al, 1975;Dell, 1986). Under NST, lexical bias effects arise in part because new connections are required to represent and produce nonwords as coherent units, whereas word production only involves preformed connections (see e.g., MacKay & James, 2001). Age-linked deficits in forming new connections will therefore reduce the likelihood that older adults will represent and produce nonwords such as "plood" under time pressure but will not affect production of real words such as "plod."…”

Relations between emotion, memory, and attention: Evidence from taboo Stroop, lexical decision, and immediate memory tasks

“…This explanation comes ready-made in binding theory, because binding and repeated activation or rehearsal have similar end results (see : The repeated activation that occurs during behavioral repetition or rehearsal suffices to form new connections without the involvement of binding mechanisms, albeit much less efficiently than with binding mechanisms (see, e.g., James, 2001, and, for supporting evidence). The special relation between emotion and rehearsal or repetition may, therefore, arise because emotion triggers both ways of forming new connections.…”

“…Binding theory also addresses the final question in the introduction: What brain mechanisms normally subserve attentional resources? The present binding theory grew out of research on the famous "hippocampal amnesic" H.M. MacKay & James, 2001, whose bilateral lesion included virtually the entire amygdala, as well as anterior parts of the hippocampus. Because emotional processes are a primary function of the amygdala (see, e.g., Halgren, 1992) and because amygdala activity correlates strongly with correct recall on explicit tests of memory for emotional, but not for nonemotional, events in stories, films, and pictures (Cahill & McGaugh, 1995;Canli, Zhao, Brewer, Gabrieli, & Cahill, 2000;Hamann, 2001;Hamann, Ely, Grafton, & Kilts, 1999), we hypothesized that H.M.'s amygdala lesion contributed to his extremely severe explicit memory deficits and that emotional reactions in the amygdala facilitate storage via links to hippocampal binding mechanisms for encoding new memories.…”

“…Experiment 2 replicated both of these effects and demonstrated a third effect: that taboo words trigger two types of emotional reaction, one word specific and the other general or common to all taboo words. On the basis of the results of Experiments 1 and 2 and a review of the literature on emotion, memory, and attention, we hypothesized that emotional reactions trigger binding mechanisms that function to link the source of an emotion to its context of occurrence (see, e.g., MacKay & James, 2001). Under this hypothesis, word-specific emotional reactions to taboo words in Stroop tasks in Experiments 1 and 2 engaged binding mechanisms that linked the taboo words to a general type of context, labeled occurrence in the task, which acted as a retrieval cue when the taboo words were recalled later during the surprise memory tests.…”