Daniel Bub

Priming and semantic memory loss in Alzheimer's disease

Semantic memory (SM) was investigated in six patients with probable Alzheimers disease (AD) by on-line measurement of semantic priming in a lexical decision task, and off-line tests of comprehension. Detailed assessment was carried out on naming, name comprehension, and probes of semantic knowledge with a battery of 150 items. The patients performed normally on perceptual tests and displayed an item-specific loss of knowledge on the semantic tests. In a primed lexical decision task, greater semantic priming was found relative to age-matched normals. The priming was substantially greater for items with degraded representations as determined by the off-line tests. Lexical decision was also performed more slowly on these items. These unexpected results demand a reevaluation of the concepts of the lexicon and semantic memory structure and their possible alteration in dementia.

Surface dyslexia and dysgraphia : dual routes, single lexicon

Abstract The dual route interpretation of surface dyslexia as a deficit in word-specific activation has been challenged recently by computational models that incorporate a unitary print-to-sound mechanism. The most current of these makes no allowance for word-specific nodes, but obtains the pronunciation of regular and exception words by weighted connections between graphemic and phonemic units. Damaging the model in a variety of ways produces a pattern that appears consistent with the performance of many surface dyslexic patients. Exception words are mispronounced more often than regular words, though accuracy deteriorates on both categories. In addition, frequency has no clear-cut effect on the probability of reading an exception word correctly. We describe the existence of a variant of the syndrome that is not fully captured by the above simulations. MP, a surface dyslexic, demonstrates a dissociation between lexical and nonlexical pronunciation of written words. We also show that performance on irregu...

Agrammatism in Sentence Production without Comprehension Deficits: Reduced Availability of Syntactic Structures and/or of Grammatical Morphemes? A Case Study

A French-speaking patient with Brocas aphasia--following a left-hemisphere lesion involving the sylvian region but sparing Brocas area--is presented. Like G. Miceli, A. Mazzuchi, L. Menn, and H. Goodglasss (1983, Brain and Language, 19, 65-97) case 2, this patient produces agrammatic speech in the absence of any comprehension deficit. Unlike Micelis patient, though, agrammatic speech can be observed in all sentence production tasks (from spontaneous speech to repetition, oral reading, and writing) whereas production of individual words--be they open class or closed class--is almost always intact. On the basis of extensive (psycho)linguistic testing, it is argued that this patients deficit is not central and not crucially syntactic (at least) at the level of knowledge but seems to disrupt specifically those (automatic?) processes responsible for both retrieval and production of free-standing grammatical morphemes whenever they have to be inserted into phrases and sentences.

Different modes of word recognition in the left and right visual fields.

We confirm previous evidence indicating that word length has a substantial effect on word recognition in the LVF but a much weaker effect in both the RVF and fovea. The nature of encoding in the LVF is not altered when the words are vertically displayed (Experiment 2), and the effect cannot therefore be entirely due to scanning artefact or acuity gradients in peripheral vision. We provide evidence that links the asymmetrical influence of word length directly to hemispheric specialization: left-handers, who as a group are much less consistently lateralized than right-handers are also less affected by word length in the LVF on the average (Experiment 3). This occurs because the asymmetry for certain left-handers is either very weak or, in some cases, is the complete reverse of the asymmetry observed in right-handers. Finally, we demonstrate that the length x field interaction is observed in lexical decisions (Experiment 4) which do not entail pronunciation of written words. There is some indication that concrete, high-imageable words produce a smaller effect of length in the LVF than abstract, low-imageable words, and we discuss this outcome in relation to the proposal that the right hemisphere can sometimes extract a lexical code from letter information. The concept of distinct modes of word recognition in the LVF and RVF clarifies a number of issues in laterality research, and suggests a new approach to evaluating group differences in half-field performance.

Constraining theories of semantic memory processing: Evidence from Dementia

Abstract In this paper we analyse the performance of ten patients with Dementia of the Alzheimers Type (D.A.T.) who show a pattern of performance suggesting a deficit at the level of semantic memory in the face of normal visual perceptual processing. We use the results of their performance on probe questions for pictures and words to evaluate several hypotheses arising from recent theories concerning semantic memory. We assess whether these patients demonstrate better performance on pictures than words (they do), and whether this can be explained away as a by-product of the perceptual nature of the items tested; pictures whose items have many discernible object parts would tend to contact more residual information in semantic memory, thus producing apparent superior performance from pictures. In fact, we find no support for this explanation. Rather, we are able to demonstrate, in the semantic category of animals, that it is only the items that are correctly identified (as a whole) that will give rise to ...

Word recognition and orthographic context effects in a letter-by-letter reader

The performance of letter-by-letter readers when attempting to decipher written material gives the impression that words fail to directly evoke any higher-level representation. As a consequence, spelling patterns appear to be treated perceptually as if they were a collection of random letters. We tested the hypothesis that words are no longer mapped onto orthographic descriptions by examining the ability of a letter-by-letter reader to identify letters in familiar words, pseudowords, and random strings. A clear effect of orthographic context was obtained on the accuracy of letter recognition, indicating that spelling patterns do gain access to more central components of the reading mechanism. The implications of this result for our understanding of the syndrome are discussed.

Increased semantic priming in patients with dementia of the Alzheimer's type

Semantic priming on a lexical decision task(LDT) was examined in 50 patients with mild to moderate dementia of the Alzheimers Type(DAT), and 25 normal age-matched controls. DAT patients were slower in their responses, and showed significantly greater priming effects (mean 54 ms vs. 27 ms in controls). The size of the priming effect correlated with the speed of response on the LDT task for the individual DAT patients but not for controls. Twenty of the DAT patients (vs. one control) showed priming greater than 60 ms. This subgroup of DAT patients with hyperpriming was slower than the nonhyperpriming group on yes responses to targets preceded by unassociated prime words and more impaired on tests of clock drawing and verbal fluency. Slowing of responses alone, however, seems unable to account for the presence of increased priming in DAT patients. Its presence may reflect semantic memory deficits, as well as impaired attentional processing and supervisory control systems. The exact mechanism of this increased priming remains to be established.

Language Localization with Activation Positron Emission Tomography Scanning

We report the first instance of the use of 3-dimensional magnetic resonance imaging anatomically correlated to positron emission tomography (PET) scanning to identify language areas in a patient with an arteriovenous malformation (AVM) in the posterior speech region. The patient was a 24-year-old right-handed woman with an angiographically proven AVM (3-4 cm) in the left mid-posterior second temporal convolution in whom a left intracarotid injection of sodium Amytal produced significant language disruption. A baseline PET cerebral blood flow study identified the AVM, and an activation PET scan performed during the reading and speaking of simple words showed increased activity in the left parastriate cortex (the second visual area), in the left posterior third frontal convolution (Brocas area), and in the left inferior and midtemporal gyri (Wernickes area). Increased activity was also noted in the right and left transverse temporal (Heschls) gyri, in the left precentral gyrus, in the left medial superior frontal gyrus (the supplementary motor area), and in the right cerebellum. We conclude that activation PET scanning is useful in the preoperative assessment of patients who harbor cerebral AVMs in classically described speech regions.

Single-character processing in a case of pure alexia

The processing of single characters in a pure alexic patient was studied in an attempt to identify the impairment responsible for his reading disorder. Observations from Experiments 1 to 4 suggested a deficit of identification of alphanumeric stimuli without any impairment affecting the elaboration of a structural description of visual stimulation. Experiment 5 indicated that the identification disorder results from a defect in the selective processes--activation and/or inhibition--that must come into play for achieving an appropriate match between a structural description of the stimulation and representations of the identities of known stimuli. The possible implications of this deficit in single-character identification for word reading are discussed.

Evidence for an Independent Stimuluscentered Spatial Reference Frame from a Case of Visual Hemineglect

Previous experiments with patients suffering from visual hemineglect have provided evidence relevant to the organization of the human spatial representation system. We examined the hypothesis that one reference frame used to represent the location of objects in the environment is based on the spatial extent of the stimulation that needs to be processed at a specific point in time; in current terminology, a stimulus-centered reference frame. The paradigm used was one of filtering, and variation of the location of the target within a horizontal array of items (stimulus-relative location) was independent of the target location relative to the subject and to stable reference points in the environment. Results showed that stimulus-relative target location provided an independent contribution to the magnitude of the neglect symptoms. This is taken as an indication that a stimulus-centered spatial reference frame contributes to the representation of the location of visual objects in human vision and that this representation may serve to direct visual attention.