You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

 
Poster C32, Thursday, November 9, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Morpho-lexical Recognition Ability and Related Brain Regions in Individuals with Mild Cognitive Impairment, Alzheimer’s Dementia, and Cognitively Normal Elderly

JungMoon Hyun1, Alexandre Nikolaev2,3, Yawu Liu4, Eve Higby5, Minna Lehtonen6, Sameer Ashaie1, Tuomo Hänninen4, Merja Hallikainen4, Hilkka Soininen2,4;1Northwestern University, 2University of Eastern Finland, 3University of Helsinki, 4Kuopio University Hospital, 5University of California, Riverside, 6Abo Akademi University, Turku, Finland

Language functions are differentially vulnerable to normal aging and Alzheimer’s disease (AD). Word finding ability and semantic processing show early decline, whereas the syntax of language production is relatively preserved in AD. Changes in morpho-lexical processing in normal aging and dementia have not been extensively studied. To address this gap, we devised a Finnish lexical decision task that included words that differ in the number of stem allomorphs and inflectional productivity. Finnish is ideal for measuring changes in morpho-lexical knowledge because of its morphological complexity in lexicon. To better understand the neural basis of morpho-lexical processing and its changes with AD, we investigated the association between complex word recognition and brain atrophy in AD, Mild Cognitive Impairment (MCI), and healthy aging. In the lexical decision task, we employed 177 real words, and 177 pseudowords that follow the phonotactic rules of Finnish and tested Finnish individuals with AD (n=21), MCI (n=24) and age-matched cognitively healthy elderly (n=17). Participants read each word and pressed a button to indicate whether it was a real word or not. Structural MRI data were used for the average cortical thickness in our ROIs, and analyses were conducted with FreeSurfer. Brain regions that significantly predicted the reaction time (RT) for words were identified, controlling for age and education. In the accuracy analyses, there were no significant differences among groups for either real words or pseudowords (Real words: AD - 95.6%, MCI - 94.4%, Control - 97.2%; Pseudowords: AD - 94.7%, MCI - 93.9%, Control - 98.7%; F(2,59) = 1.67, p = .197) indicating patients with MCI and AD continue to maintain morpho-lexical processing ability. In the RT analyses, the three groups showed significant group differences for real words (F(2,59) = 4.28, p = .018) and pseudowords (F(2,58) = 3.81, p = .028). The AD group was significantly slower than healthy controls (real words, p = .019; pseudowords, p = .032). The difference between MCI and controls approached significance (Real words, p= .065; Pseudowords, p= .071). No difference was found between the AD and MCI groups (Real words, p = .803; Pseudowords, p = .889). These results suggest an important interaction between increased processing time and speed-accuracy trade-off in the patient groups. Analyses of cortical thickness and RT revealed that a thinner left superior temporal lobe (posterior) and right inferior temporal gyrus (real words-right pars orbitalis; pseudowords-right pars triangularis) predicted slower responses in the AD group. The left lingual gyrus was the only region related to word recognition RTs in healthy controls. In sum, the speed of healthy elderly’s recognition of morphologically complex words was associated with brain regions related to visual/letter processing and identification of words. On the other hand, the speed of AD patients’ recognition depends on regions associated with language and executive functions. These results reflect the increased cognitive effort in patients to achieve a high level of accuracy on the recognition task. More detailed clinical information, analysis methods, and potential clinical implications of these findings will be discussed.

Topic Area: Language Disorders

Back to Poster Schedule