AbstractBackgroundMemory loss is one of the hallmark symptoms of cognitive impairment that progresses towards dementia. Different cognitive domains (CD) have been suggested to influence the performance of dual‐task (DT) activities. While there is evidence of DT effects on those with mild cognitive impairment and other dementias, the impact of DT on those with subjective memory loss is unknown.MethodParticipants (n=57; mean age: 72.72±10.56 years) with memory loss but without a diagnosis of cognitive impairment, dementia, or other neurodegenerative diseases were included. They performed validated computerized cognitive assessment battery (CAB, NeuroTrax BrainCare) to obtain CD measures of executive function (EF), attention, memory, visual‐spatial processing (VSP), information processing speed (IPS), and a global cognitive score (GCS) measure. Using the Zeno Walkway Gait Analysis System (Protokinetics), measures of velocity, stride width (SW), stride time (ST), stride length, cadence, double support (DS), and gait variability were obtained for both single‐task and DT gait. Participants chose between performing the DT gait by either recalling names (n=14), counting backward by 3’s (n=16), or by 7’s (n=27), which were used as group assignments for analysis. A three‐group cross‐sectional mediation analysis was performed using standardized cognitive measures as mediators and the DT effect (DTE) of the gait parameters as outcomes of interest. Data analysis was conducted using SPSS 26 and SPSS AMOS 26.ResultNo significant mediators of cognition on the dual‐task effect of either task were identified. However, increases in individual CD significantly predicted improvement of DTE in specific gait parameters: (1) memory significantly predicted DTE stride time (Effect size by standardized beta(ES)=0.343,p=0.022) and DTE double support (ES=0.322,p=0.033); (2) GCS (ES=0.388,p=0.028), attention (ES=0.348,p=0.007), and IPS (ES=0.290,p=0.027) significantly predicted DTE stride width; and (3) Attention (ES=0.282,p=0.047) significantly predicted DTE cadence. Furthermore, between‐group analysis showed significant differences in VSP (F(2,54)=8.820,p<0.001), IPS (F(2,54)=3.350,p=0.043) and GS (F(2,54)=5.149,p=0.009).ConclusionResults suggest that individual CD (VSP, IPS, and/or GS) scores impact different aspects of dual tasking gait in individuals with memory loss. While there were no significant mediators, the significant cognitive predictors suggest improvement of those individual CD could improve the DTE of individual gait measures.
