Impaired affective prosody in AD: relationship to aphasic deficits and emotional behaviors.

OBJECTIVE: To assess the ability of patients with AD to produce, repeat, and comprehend affective prosody in relationship to severity of dementia, aphasic deficits, and changes in emotional behaviors. METHODS: An Aprosodia Battery was used to assess affective-prosodic performance and to identify patterns of deficits in affective communication. In addition, the presence and severity of aberrant behaviors, depression, and aphasia were assessed using standardized assessment tools. RESULTS: Patients with AD had significant impairments in their ability to repeat, comprehend, and discriminate affective aspects of speech, but maintained normal spontaneous affective-prosodic performances. As dementia severity increased, performance on the comprehension tasks and, to a lesser degree, on the repetition tasks became more impaired; spontaneous affective prosody remained normal. In the current study, affective-prosodic comprehension impairments were present in patients with all stages of AD; comparable aphasic deficits were not observed until patients were severely demented. The majority of aphasic deficits involved anomia without loss of comprehension. Patients with AD with sensory aprosodia had increased frequency and severity of behavioral changes whereas patients with AD with normal affective-prosodic performance were significantly less demented, had normal linguistic ability, and displayed fewer aberrant psychiatric behaviors. CONCLUSION: Patients with mild AD are at considerable risk for affective-prosodic comprehension deficits. As patients become more demented and develop sensory aprosodia, they are at greater risk for disturbances in behavior and mood.

Semantic fluency performance of patients with cortical and subcortical neurodegenerative diseases.

Previous studies using the Animals Fluency Test have shown that dementia patients with Alzheimer's disease (AD), Huntington's disease (HD), or Parkinson's disease (PD) produce fewer correct words and have smaller semantic cluster sizes than controls or PD patients without dementia (PDND). Although the number of correct words generated by the patients with AD was positively correlated with mental status, cluster size, surprisingly, was not. To increase word output and increase the reliability of estimates of cluster size, semantic fluency was reexamined using the Supermarket Fluency Task. Overall, patients with HD or PD with dementia (PDD) exhibited reduced cluster sizes compared to older controls or PDND patients, but cluster sizes were only marginally reduced for patients with AD. These effects were evident only for female participants, because the cluster sizes for elderly control men were substantially smaller than those of elderly women. For the female patients with AD, cluster size was correlated with mental status, but the relationship was nonlinear. Cluster size was normal for mildly demented patients with AD, but much reduced for moderately or severely demented participants. In contrast to a previous report, in the present study the proportion of category labels generated was increased for patients with HD with dementia but not for patients with AD. This finding questions one line of evidence that semantic memory stores undergo "bottom-up" degradation in AD. Together with previous results, these findings indicate that semantic cluster size reflects efficiency of access to semantic knowledge which is similarly compromised in subcortical and cortical diseases.