Individual differences in executive functions and retrieval efficacy in older adults.

Two prominent aspects of memory problems in older adults are a difficulty in retrieving recent episodic events and an often transient inability to retrieve names and other well-known facts from semantic memory. The question addressed in the present studies was whether these age-related difficulties reflect a common cause-a retrieval problem related to inefficient executive functions (EF). In the first study, 50 older adults were given 4 tests of EF; a derived composite measure correlated strongly with a measure of retrieval efficacy in free recall, less strongly with paired-associate recall, and nonsignificantly with retrieval of general knowledge. A second study used somewhat different measures of EF and also different measures of retrieval from semantic memory, and this study did find significant relations between EF, episodic memory, and knowledge retrieval. Changes in the specific tests representing both EF and memory retrieval changed the relations between them, suggesting that no one task is a pure measure of the theoretical constructs of either EF or episodic and semantic memory. Taken together, the 2 studies showed that individual differences in EF in older adults are correlated with retrieval efficacy in both episodic and semantic memory but also that these relations depend on the specific measures chosen to represent both EF and memory retrieval. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Effects of divided attention at encoding and retrieval: Further data.

Division of attention (DA) at the time of learning has large detrimental effects on subsequent memory performance, but DA at retrieval has much smaller effects (Baddeley, Lewis, Eldridge, & Thomson, 1984, Journal of Experimental Psychology: General, 113, 518-540; Craik, Govoni, Naveh-Benjamin, & Anderson, 1996, Journal of Experimental Psychology: General, 125, 159-180). Experiment 1 confirmed the relatively small effects of DA on retrieval and also showed that retrieval operations do consume processing resources. The experiment also found that the effect is not attributable to a trade-off in performance with the concurrent task or to recognition decisions made on the basis of familiarity judgments. Participants made levels-of-processing (LOP) judgments during encoding to check whether deeper semantic judgments were differentially vulnerable to the effects of DA. In fact DA did not interact with LOP. Experiment 2 explored reports that the comparatively slight effect of DA on recognition accuracy is accompanied by a compensatory increase in recognition latency (Baddeley et al., 1984). The experiment replicated findings that neither DA nor differential emphasis between recognition and a concurrent continuous reaction time (CRT) task affected recognition accuracy, but also found evidence for a lawful trade-off in decision latencies between recognition and CRT performance. Further analysis showed that the relationship between response rates on the two tasks was well described by a linear function, and that this function was demonstrated by the majority of individual participants. It is concluded that the small effect of DA on recognition performance is attributable to a trade-off within the recognition task itself; accuracy is maintained by a compensatory increase in decision latency.

Normal appearing white matter permeability: a marker of inflammation and information processing speed deficit among relapsing remitting multiple sclerosis patients.

PURPOSE: Blood-brain barrier breakdown (BBBB) occurs in relapsing remitting multiple sclerosis (RRMS). Relative recirculation (rR), a BBBB surrogate, may show inflammation undetectable by gadolinium. We compared normal appearing white matter (NAWM) rR in patients with and without disability measured with Symbol Digit Modalities Test and the Expanded Disability Status Scale (EDSS). METHODS: Thirty-nine RRMS patients were prospectively recruited and classified as impaired or non-impaired based on the SDMT and EDSS threshold ≥3. Significant demographic, MRI structural and regional rR characteristics were advanced into multivariate analysis to assess the association with impairment of cognition and EDSS. Bonferroni corrected p < 0.025 was applied to demographic and rR group comparisons; p < 0.05 was used in the final multivariate logistic regression. RESULTS: rR was higher in NAWM (p = 0.012), NAGM (p = 0.004), and basal ganglia (p = 0.007) in cognitively impaired versus non-impaired patients. The difference between NAWM and T2HL rR was significant in cognitively non-impaired patients and approximated that of T2HL in impairment (0.084 vs. 0.075, p = 0.008; 0.118 vs. 0.101, p = 0.091, respectively). After adjusting for confounders, rR elevation for NAWM (OR 1.777; 95% CI 1.068-2.956; p = 0.026), NAGM (OR 2.138; 1.100-4.157; p = 0.025), and basal ganglia (OR 2.192; 1.120-4.289; p = 0.022) remained significantly predictive of cognitive impairment. NAWM area under the curve (AUC) for cognitive impairment was 0.783. No significant group differences or associations were seen for rR and EDSS impairment. No NAGM and cortical lesion rR difference was present within any of the impaired or non-impaired groups. CONCLUSION: rR elevation in NAWM, NAGM, and basal ganglia appears sensitive to cognitive impairment but not EDSS.