Profiles of cognitive impairment in the continuum from normal cognition to Alzheimer's clinical syndrome: Contributions of the short‐term memory binding tests. (24th July 2020)

• Record Type: Journal Article
• Authors: Cecchini, Mario Amore; Foss, Maria Paula; Tumas, Vitor; Patrocinio, Flávia A.P.; Chiari‐Correia, Rodolfo D.; Novaretti, Nathalia; Brozinga, Tamara R.; Bahia, Valéria Santoro; de Souza, Leonardo Cruz; Cerqueira Guimarães, Henrique; Caramelli, Paulo; Lima‐Silva, Thais Bento; Cassimiro, Luciana; Brucki, Sonia Maria Dozzi; Nitrini, Ricardo; Della Sala, Sergio; Parra, Mario A.; Yassuda, Mônica Sanches
• Is Part Of: International journal of geriatric psychiatry. Volume 35:Number 11(2020)
• Page Start: 1331
• Page End: 1340
• Publication Date: 2020-07-24
• Subjects: Alzheimer's clinical syndrome -- Alzheimer's disease -- cognitive aging -- memory binding -- mild cognitive impairment -- short‐term memory -- working memory; Geriatric psychiatry -- Periodicals; Geriatric Psychiatry -- Periodicals; 618.97689
• DOI: 10.1002/gps.5370
• Languages: English
• ISSNs: 0885-6230

Abstract

Abstract : Background: Short‐term memory binding (STMB) tests assess conjunctive binding, in which participants should remember the integration of features, such as shapes (or objects) and colors, forming a unique representation in memory. In this study, we investigated two STMB paradigms: change detection (CD) and free recall (FR). Objective: To investigate the cognitive profile in the CD and FR tasks of three diagnostic groups: cognitively unimpaired (CU), mild cognitive impairment (MCI), and Alzheimer's clinical syndrome (ACS). In addition, we aimed to calculate and compare the accuracy of the CD and FR tasks to identify MCI and ACS. Methods: Participants were 24 CU, 24 MCI, and 37 ACS. The cognitive scores of the clinical groups were compared using analysis of variance (ANOVA) and receiver‐operating characteristic (ROC) analyses were carried out to verify the accuracy of the STMB tasks. Results: In the CD task, CU was different from MCI and ACS (CU > MCI = ACS), while in the FR task all groups were different (CU > MCI > ACS). The ROC analyses showed an area under the curve (AUC) of 0.855 comparing CU with MCI for the CD task and 0.975 for the FR. The AUC comparing CU and ACS was 0.924 for the CD and 0.973 for the FR task. The FR task showed better accuracy to identify MCI patients, and the same accuracy to detect ACS. Conclusion: The present findings indicate that impairments in CD and FR of bound representations are features of the cognitive profiles of MCI and ACS patients.