Grover C. Gilmore

Multidimensional letter similarity derived from recognition errors

In order to provide a reliable measure of the similarity of uppercase English letters, a confusion matrix based on 1,200 presentations of each letter was established. To facilitate an analysis of the perceived structural characteristics, the confusion matrix was decomposed according to Luce’s choice model into a symmetrical similarity matrix and a response bias vector. The underlying structure of the similarity matrix was assessed with both a hierarchical clustering and a multidimensional scaling procedure. This data is offered to investigators of visual information processing as a valuable tool for controlling not only the overall similarity of the letters in a study, but also their similarity on individual feature dimensions.

Motion perception and aging

The authors used a correlated motion paradigm to investigate the effects of aging and gender on motion sensitivity. In 2 experiments with a total of 50 elderly and 50 young subjects, motion thresholds were significantly higher for elderly women. The correlated motion signal, which was embedded in random motion, may have been coherent to subjects in much the same way a form is in Witkins Embedded Figures Test (EFT). In Experiment 2, EFT scores were obtained. A significant positive relationship between motion thresholds and EFT performance was found within each age group. Although gender-related perceptual style differences may contribute to motion perception effects, the authors argue that a common neural factor contributes to performance on both the EFT and the correlated motion task.

At the interface of sensory and motor dysfunctions and Alzheimer's disease

Recent evidence indicates that sensory and motor changes may precede the cognitive symptoms of Alzheimers disease (AD) by several years and may signify increased risk of developing AD. Traditionally, sensory and motor dysfunctions in aging and AD have been studied separately. To ascertain the evidence supporting the relationship between age‐related changes in sensory and motor systems and the development of AD and to facilitate communication between several disciplines, the National Institute on Aging held an exploratory workshop titled “Sensory and Motor Dysfunctions in Aging and AD.” The scientific sessions of the workshop focused on age‐related and neuropathologic changes in the olfactory, visual, auditory, and motor systems, followed by extensive discussion and hypothesis generation related to the possible links among sensory, cognitive, and motor domains in aging and AD. Based on the data presented and discussed at this workshop, it is clear that sensory and motor regions of the central nervous system are affected by AD pathology and that interventions targeting amelioration of sensory‐motor deficits in AD may enhance patient function as AD progresses.

PERCEPTUAL COMPARISON OF PULSED AND CONTINUOUS FLUOROSCOPY

Pulsed fluoroscopy (hereafter called pulsed) at reduced acquisition rates, typically 15 acq/s (pulsed-15), is proposed to reduce x-ray dose in interventional procedures. However, since the human visual system (HVS) acts as a temporal low-pass filter that interacts with such acquisitions, the proper dose for pulsed must be obtained in perception experiments. We determine the dose for low-frame-rate pulsed that gives visualization equivalent to that of conventional 30 acq/s fluoroscopy, hereafter called continuous. Computer-generated phantoms are used. They consist of stationary, low-contrast disks on a flat background containing Poisson noise that mimics quantum noise in fluoroscopy. Image sequences are displayed on the video tachistoscope, a device with considerable display flexibility. Three experimental paradigms are used. (1) In a paired-comparison study, pulsed and continuous are displayed side-by-side on the same monitor, and the visibility of a contrast detail phantom is compared. (2) Using this same display, subjects record the minimally detectable disk contrast (the min-contrast measurement). (3) In a four-alternative forced-choice experiment, a disk is placed in one of four positions, and the subject determines the position of the disk. The methods are complementary--the forced-choice experiment properly eliminates the subjectivity of the observer threshold while the paired-comparison study is much more time efficient. With regard to pulsed and continuous comparisons, remarkable similarity is found between the supra-threshold experiments (1 and 2) and the detectability experiment (3); i.e., the average absolute differences in the equivalent-perception dose as determined by the three measures is approximately 3%. No difference is found between interlaced and noninterlaced display. A relatively small dependence of dose savings on disk size is found with larger disks giving increased dose savings. Average dose savings of 22%, 38%, and 49% are found for pulsed-15, pulsed-10, and pulsed-7.5, respectively.

Enhanced stimulus strength improves visual cognition in aging and Alzheimer's disease.

Deficits in visual cognition in Alzheimers disease (AD) arise from neuropathological changes in higher-order association areas of the cortex and from defective input from lower-level visual processing areas. We investigated whether enhanced signal strength may lead to improvement of visual cognition in AD. We tested 35 individuals with probable AD, 35 age-matched elderly control (EC) and 58 young control (YC) adults on letter identification, word reading, picture naming, discrimination of unfamiliar faces, and pattern completion. The contrast sensitivity step-difference across an independent sample of AD and EC groups was used in calculating an image filter, from which we produced stimulus-strength conditions of low-degraded, medium-normal, and high-enhanced. Using this filter we created a hypothetical proximal-strength equivalence between AD at medium strength and EC at low strength, and between AD at high strength and EC at medium strength. For letter identification, word reading, picture naming, and face discrimination, medium strength elicited AD accuracy levels and reaction times that were similar to those of EC at low strength. On picture naming, increased strength reduced perceptual-type errors for EC and AD and random errors for AD. For word reading, high strength elicited AD accuracy levels and reaction times that were equivalent to those of EC at medium strength. We saw no effect of signal-strength manipulation on performance of pattern completion, possibly owing to the complex cognitive demands of that task or to the inadequacy of the filter for its images. The results indicate that putative AD-EC differences in cognition directly reflect contrast sensitivity differences between the groups. Enhancement of stimulus strength can ameliorate vision-based deficits and lead to improvement in some aspects of cognitive performance. These results suggest new non-pharmacological avenues to explore in the attempt to improve cognition in elderly adults and especially in individuals with AD.

Contrast sensitivity in Alzheimer's disease : a 1-year longitudinal analysis

Spatial contrast sensitivity was evaluated in normal elderly adults and a group with probable Alzheimers disease (AD) in three sessions over a 1-year period. There was evidence of a reduction over 1 year in contrast sensitivity for static, high spatial frequencies in both groups of subjects. A striking difference between the subject groups was observed in their response to flickered stimuli. Although normal elderly adults yielded good stability in sensitivity for both static and flickered spatial frequencies over the 1 year, AD patients experienced a significant decline in response to flickered (7.5 Hz) stimuli. The significant decline in sensitivity for low contrast, high temporal events in AD patients is consistent with reports of cell loss at the retinal and cortical levels of the magnocellular (M) channel in AD. Thus, the spatial contrast sensitivity change in AD may reflect a more rapid loss of cells specifically in the M channel over a 1-year period.

Balint's Syndrome in Alzheimer's Disease: Visuospatial Functions

A subgroup of patients with Alzheimers disease (AD) developed Balints syndrome, an uncommon and incompletely understood disorder of visuospatial processing. We characterized the visuospatial features of three AD patients with Balints syndrome and compared them to three comparably demented AD patients without this syndrome. On visuospatial tasks, the Balint patients were unable to integrate visual stimuli over space. On contrast sensitivity testing, the Balint patients had significantly decreased contrast sensitivities for low spatial frequency gratings (0.5, 1.0 and 2.0 cpd) alternated at 7.5 Hz. Furthermore, the Balint patients had left visual field attentional deficits and normal hemifield P100 visual evoked potentials. In AD, Balints syndrome involved decreased sensitivity to low spatial frequencies necessary for global visuospatial analysis, a disturbance suggesting damage to the magnocellular visual system in the occipitoparietal association cortex and, possibly, in the optic nerves.

Factors affecting processing mode in visual search

Visual search was studied under a variety of conditions to clarify some differences among the results of previous investigations and to provide a testing ground for models of visual information processing. Display configuration, target and field composition, exposure duration, and display size (up to 16 elements) were among the parameters investigated. In some conditions, mean reaction time was essentially invariant with display size, while in other conditions it increased substantially and linearly with display size. Current models of visual information processing were evaluated in the light of these and previous findings; all were found wanting. The data seem to demand a system subject to flexible cognitive control processes.

Spatial contrast sensitivity in Alzheimer's disease: a comparison of two methods.

The present study was designed to compare the contrast sensitivity functions (CSF) obtained with the Nicolet CS2000 and the Vistech VCTS6500 for a sample (N = 25) of patients diagnosed with probable Alzheimers disease (AD) and a sample (N = 25) of healthy elderly adults. With the Nicolet, CSF were determined for gratings presented under static and rapidly counterphased (7.5 Hz) conditions. All research participants were able to complete the Vistech test. However, 13.7 and 7.4% of the original samples of Alzheimer patients and healthy adults, respectively, did not yield usable data. Both test instruments showed uniformly low contrast sensitivities for the AD group. The results support the proposition that AD involves a primary visual disturbance and suggest that the Vistech chart can be used with a broader range of patients for initial screening.

Visual processing of rapidly presented stimuli is normalized in Parkinson's disease when proximal stimulus strength is enhanced.

Deficient perception and cognition in Parkinsons disease (PD) has been attributed to slow information processing, but an alternative explanation may be reduced signal strength. In 18 nondemented individuals with PD and 15 healthy adults, we enhanced the contrast level of rapidly flashed masked letters. The PD group required significantly higher contrast to reach criterion (80% accuracy). Normal motion detection in these participants indicated no gross, general dysfunction of the dorsal visual processing stream. These results suggest that putatively slowed processing in PD may be an artifact of reduced signal strength arising from depletion of dopamine in retina or cortical visual areas.