New evidence reveals weakening of association between language and thought in aMCI versus in normal aging

Abstract

AbstractBackgroundClinical manifestation of prodromal AD is typically evaluated through measures that assess individual cognitive domains with less focus on their relationship. Further, while decades of research identify deterioration in episodic memory as well as semantic aspects of language, there has been less focus on other aspects of linguistic computation. We provide evidence that advances our understanding of how language deficits in aMCI extend to sentence production, and how relations between performances in cognitive functions, specifically language and cognition, may differ in aMCI vs. Healthy Aging (HA).MethodWe tested 24 individuals diagnosed with aMCI based on neurologic and neuropsychological evaluations, and 22 older adults with normal cognition (Healthy Aging (HA). All subjects were administered a multi‐domain cognitive screen (ACE‐R) and a psycholinguistic assessment of complex sentence production. The ACE‐R provided a Total Score and scores on attention and orientation, memory, fluency, language, and visuospatial measures (Mioshi et al 2006, Hsieh et al 2013, Monsch et al 1994).) The psycholinguistic test assessed Elicited Imitation of complex sentences with relative clauses (RC) in an experimental design varying syntactic and semantic factors (Table 1) (Lust et al 2017).ResultACE‐R Total score was significantly lower for aMCI than HA. aMCI also performed significantly worse than HA on the psycholinguistic test (Table 2). For HA, not aMCI, ACE‐R Total score significantly predicted successful sentence production on the psycholinguistic test (Table 3) (Figure 1). Similarly, for HA, not aMCI, the ACE‐R language score significantly predicted psycholinguistic test performance. In contrast, ACE‐R memory performance, which was significantly lower for aMCI than HA, did not predict linguistic performance for either group. Only ACE‐R phonemic fluency predicted linguistic performance for both groups. (Table 3).ConclusionResults suggest a weakening of associations between overall cognitive performance and performance on a linguistic measure for aMCI relative to HA suggesting that evaluation of associations between, not only within, cognitive domains strengthen detection of early changes in AD. This finding may be further expanded by linking this clinical manifestation to neuroscientific evidence of functional connectivity in AD pathophysiology and in normal aging (e.g., Brier et al 2014, Spreng and Turner 2013).