OBJECTIVE: To further understanding of text comprehension abilities in Alzheimer's disease (AD), a think-aloud protocol was used to examine the role of inferencing and the memory operations used to produce inferences. METHOD: Twenty participants with AD and 20 cognitively healthy older adults (OA) read narratives, pausing to talk aloud after each sentence. A verbal protocol analysis developed by Trabasso and Magliano (1996a) was used to code participants' utterances into inferential and noninferential clause types; inferential statements were then coded to identify the memory operation used in their generation. RESULTS: Compared with OA controls, the AD participants showed poorer story comprehension, d = 2.0, produced fewer inferences, d = .67, and were less skilled at providing explanations of story events, d = 1.27, and in using prior text information to explain outcomes, d = .90. The AD group also appeared to rely more on the activation of world knowledge, d = .58, which contributed to less effective inferences and produced more incoherent noninferential statements, d = 1.05. Poorer text comprehension for the AD group was associated with poorer verbal memory abilities, r's > .55, and poorer use of prior text events when producing explanatory inferences, r = .42. CONCLUSIONS: The findings suggest that the memory difficulties of the AD group appear to be an important cognitive factor interfering with their ability to integrate story events through the use of inferences and to create a global coherence to support text comprehension.
OBJECTIVE: To further understanding of text comprehension abilities in Alzheimer's disease (AD), a think-aloud protocol was used to examine the role of inferencing and the memory operations used to produce inferences. METHOD: Twenty participants with AD and 20 cognitively healthy older adults (OA) read narratives, pausing to talk aloud after each sentence. A verbal protocol analysis developed by Trabasso and Magliano (1996a) was used to code participants' utterances into inferential and noninferential clause types; inferential statements were then coded to identify the memory operation used in their generation. RESULTS: Compared with OA controls, the ADparticipants showed poorer story comprehension, d = 2.0, produced fewer inferences, d = .67, and were less skilled at providing explanations of story events, d = 1.27, and in using prior text information to explain outcomes, d = .90. The AD group also appeared to rely more on the activation of world knowledge, d = .58, which contributed to less effective inferences and produced more incoherent noninferential statements, d = 1.05. Poorer text comprehension for the AD group was associated with poorer verbal memory abilities, r's > .55, and poorer use of prior text events when producing explanatory inferences, r = .42. CONCLUSIONS: The findings suggest that the memory difficulties of the AD group appear to be an important cognitive factor interfering with their ability to integrate story events through the use of inferences and to create a global coherence to support text comprehension.