Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Comparative and Experimental Medicine

Major Professor

Joseph H. Dougherty

Committee Members

Karl F. Hubner, R. Kent Hutson, Joseph A. Thie, Joel Lubar, Gary T. Smith, Xuemin Xu


Awareness and interest in diseases, which pertain to cognitive and affective functions in the elderly, has been increasing, as the proportion of elderly people in the population grows larger. Afflicting an estimated 19 million people worldwide the syndrome of dementia comprises some of the most prevalent and devastating diseases. The development of Alzheimer's disease (AD), the most frequent cause of dementia in the elderly, is associated with a number of pathogenic mechanisms that include genetic, immunologic, and vascular factors. AD is often unrecognized in its earliest stages even though it is the most common cause of dementia in the aged population, yet impossible to diagnose precisely without invasive techniques, particularly at the onset of the disease.

AD accounts for 70% of all cases of late-onset dementia (after 65 years of age). Since the incidence of AD doubles every 5 years after 60 years of age (Tanzi et al 2000), the incidence of AD is expected to increase further as more people live to advanced ages. Characteristic neuropathological features of AD include neurofibrillary tangles (NFTs), and extracellular deposits of b-amyloid protein (Ab) in senile plaques (SPs). However, AD seems to be etiologically as well as genetically heterogeneous, with probable interactions between genetic and nongenetic factors. The initial deficit in AD is normally in short-term and working memory, which progresses gradually for several years before impairments in other cognitive domains such as language, semantic memory, orientation, and visuospatial functions become evident. Short-term as well as working memory is built from several different types of attentional mechanisms, which allow for normal functioning and memory performance. This study postulates that a deficit in a subtype of attention, called selective attention, becomes impaired before evident deficits in short-term memory are observable. The anterior cingulate cortex (ACC) is believed to play an essential role in the mechanisms of selective attention in normal subjects. Moreover, the ACC is thought to be responsible for the initiation and motivation of goal-oriented behaviors. Impairments in ACC function produce marked deficits in selective attention, and may contribute to apathy and anosognosia seen in as many as one-half of mild AD patients. Mild AD patients that exhibit anosognosia are unaware of their symptoms, which are usually manifested as unawareness of short-term memory loss, and orientation function. These ACC impairments have implications in the clinical management of AD patients (i.e. driving).

Specifically, this study examines the contributions of the cingulate cortex to impairments in selective attention and anosognosia in mild AD patients. The cingulate cortex is a part of the mesial temporal lobe memory system known to be affected early in the course of AD. The main purpose of the study was two-fold: a) to investigate differences in brain activation between mild AD patients and normal controls when performing the Counting Stroop Test; and b) to investigate the biological substrates of attentional impairments and anosognosia in mild AD patients using different approaches of measuring cortical activity. Two functional imaging methods: functional Magnetic Resonance Imaging (fMRI) and 18-Fluorodeoxyglucose Positron Emission Tomography (18-FDG-PET) were implemented. Recent development of functional neuroimaging techniques, particularly functional magnetic resonance imaging (fMRI), has made the evaluation of small brain structures, such as the different regions of the cingulate cortex, conceivable. In addition to the functional evaluation of the cingulate cortex by fMRI and PET, the anatomical integrity of brain structures was evaluated using high-resolution MRI in all of the study participants.

The Counting Stroop, a Stroop Test-variant specialized for fMRI and PET was used as the activation paradigm to test selective attention. Selective attention reflects upon the ability to choose successfully what an organism wishes to analyze in the environment and to ignore what it should disregard, which contributes to the overall integrity of cognitive function, and subsequently, behavior. A total of 23 unmedicated subjects from two groups were examined: 10 patients with probable AD according to NINCDS-ADRDA criteria (McKhann et al., 1984), and 13 cognitively normal age-matched control subjects. All of the participants performed the Counting Stroop Test while undergoing the fMRI scan; the mild AD patients also performed the Counting Stroop Test before having the FDG-PET scan. In addition, all of the subjects were also administered several cognitive assessments: 1. the Self Test (ST) and the Mini Mental Status Exam (MMSE) to stage the AD patients according to impairments in general cognitive function, 2. the Stroop Test to evaluate impairments in selective attention, 3. the Beck Depression Inventory (BDI-II), and 4.)the Anosognosia Questionnaire to assess AD patients’ awareness of symptoms.

Both groups of participants showed highly significant correlation between the written Stroop Test scores and the Counting Stroop Test scores exemplifying high degree of validity and reliability of the newly developed Counting Stroop Task. Furthermore, the AD and normal group differed significantly in the written Stroop Test scores: the normal control group had a mean score of 75 12 words/45 seconds compared to the mild AD group which mean score obtained on the written Stroop Test was 23 8 words/45 seconds.

While there were individual differences, the mild AD group consistently showed significant blood level oxygen dependent (BOLD) decreases in the ACC while undergoing the Counting Stroop Test as shown by the fMRI scans. When compared to the normal subjects the patients’ fMRI images showed that the anterior areas of the cingulate cortex had significantly increased BOLD activity. On the other hand, the mild AD group did activate the medial frontal lobe (bilaterally), the inferior frontal lobe (bilaterally), the insular cortex (bilaterally), the left caudate nucleus, the right thalamus, the medial and inferior occipital lobes, and the right posterior parietal lobe. Therefore, these results indicate that the mild AD group activated the frontostriatal-insular network; the absence of the ACC activation could not have been caused by poor neuronal responses.

The normal group showed significantly increased BOLD activity in the ACC bilaterally, the right dorsolateral prefrontal cortex, the superior parietal lobule, the inferior parietal lobule, the middle and inferior occipital lobe, and the inferior orbital cortex. Direct comparisons between the mild AD and the normal control group indicated significant differences in the following areas: the left dorsolateral prefrontal area, the ACC bilaterally, and posterior parietal areas bilaterally. The MRI anatomical scans of mild AD patients did not show significant focal lobar atrophy changes, which are typical of AD pathology; instead mostly age-appropriate atrophy was observed. The anatomical MRI findings were similar for the normal control group as well. In order to determine whether an association between anosognosia and BOLD activations in the ACC exists, the mild AD subjects were divided into 2 groups: those mild AD patients with no significant activations in the ACC compared to those mild AD subjects with one or more significant activations in the ACC. The results confirmed that a significant difference in anosognosia scores in those mild AD patients with no significant activations in the ACC compared to those mild AD subjects with at least one significant activation in the ACC existed (mean difference in AQ score 0.81, p<.02).

The qualitative analyses of FDG-PET images indicated decreased utilization of glucose in the tempoparietal and temporal lobes in the mild AD group. This is a typical pattern of probable AD in the moderate to late stages as seen on FDG-PET without using an activation task. The same pattern was quantitatively determined using the Counting Stroop Test as the activation paradigm in the mild AD group. The coregistration of functional images (FDG-PET) combined with subjects’ own anatomical MRI images was performed to design improved, more efficient method of accurate specification of significant glucose hyper- or hypo-activated areas. After subtracting the activation FDG-PET images from the baseline FDG-PET images, the following focal areas of decreased glucose metabolism were determined quantitatively: the thalamus, the superior ACC, the superior colliculi, and the right dorsolateral prefrontal cortex. In addition, a significant loss of the superior portion ACC glucose uptake was found in five of the six anosognosic patients with mild AD (p<.01). The magnitude of the glucose uptake loss correlated positively with the mild AD patients’ Anosognosia Questionnaire scores. Thus, functional hypoactivity of the ACC was strongly indicative of anosognosia deficits in mild AD patients.

In conclusion, compared to the normal age-matched control subjects, all of the mild AD patients showed prominent impairments in selective attention as detected by the test for selective attention, the Color Word Stroop Test (CSWT). In addition, both groups of subjects showed highly significant correlation between the written CSWT scores and the Counting Stroop Test scores (obtained during the activation task) exemplifying a high degree of validity of the newly developed Counting Stroop Test. The absence of significant fMRI activations in the left dorsolateral prefrontal area, the ACC bilaterally, and posterior parietal areas bilaterally as well as decreased glucose metabolism in the thalamus, the superior ACC, the superior colliculi, and the right dorsolateral prefrontal cortex determined by FDG-PET was determined to be associated with impairments in selective attention in the mild AD patients but not normal control subjects. Furthermore, the functional hypoactivity of the ACC of those mild AD patients with anosognosia indicated that these patients were generally not aware of their cognitive shortcomings. Anosognosia in mild AD patients appears to represent a defect in the ACC, and may thus denote specific cognitive abnormality of selective attention.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."