Table 1.

Studies reporting association between motor performance, neuroimaging in participants with MCI or AD diagnosis when compared to healthy controls

Author	Title	Year	Controls	MCI	AD	Mean Age	% Female	Study Design	Motor Measure(s)	Neural Measure	Imaging	Summary
Beauchet [27]	Hippocampal volume, early cognitive decline, and gait variability: Which association?	2015	47	43 (undifferentiated)	NA	Controls: 69.7 SD (3.6) MCI: 70.2 SD (3.7)	Controls: 48.9 MCI: 62.8	Cross-sectional	Spatio-temporal gait parameters (stride time, swing time, stride width)	ROI: Hippocampal Volume	1.5 T Siemens Magnetom Avanto MRI	Patients with MCI had a significantly higher stride time variability and lower absolute hippocampal volume, but not relative volume, when compared with controls. Results showed that there was an association between higher stride time variability and larger hippocampal volume in controls, but not in MCI.
Crockett [32]	Head over heels but I forget why: Disruptive functional connectivity in older adult fallers with mild cognitive impairment.	2019	16	19 (undifferentiated)	NA	Controls: 74.1 MCI: 74.3	Controls: 75 MCI: 68	Cross-sectional	Finger tapping (left and right)	ROI: Functional connectivity (fMRI) within and between the default mode network, frontoparietal network, and sensorimotor network, measured while performing a finger tapping task	3.0 T Philips Intera Achieva MRI	No significant differences were observed between control and MCI subjects within or between any of the networks.
Grijalva [30]	Dual-task performance is associated with brain MRI Morphometry in individuals with mild cognitive impairment	2021	9	12 (undifferentiated)	NA	Controls: 74.3 (3.5) MCI: 79.3 (8.8)	Controls: 56 MCI: 67	Cross-sectional	Motor cognitive dual-tasking where the motor task consisted of an elbow flexion-extension test and the cognitive test was counting backward by one starting from a random number.	Regional volume (gray matter, CSF, ventricles), cortical surface area, cortical thickness	3.0 T Siemen’s Skyra MRI	Volume of the inferior temporal gyrus, volume of the inferior lateral ventricle, and cortical thickness of the inferior parietal lobule, inferior temporal gyrus, and middle temporal gyrus were significantly more strongly associated with dual-task performance in MCI than in controls. Larger ventricles, smaller brain volume, and a thinner cortex were related to worse performance.
Holtzer [26]	Mild Cognitive Impairments Attenuate Prefrontal Cortex Activations during Walking in Older Adults	2020	71	11 (undifferentiated)	NA	Controls: 76.8 SD (6.2) MCI: 78.3 SD (4.3)	Controls: 50.7 MCI: 45.5	Cross-Sectional	Gait Speed during single task and cognitive dual-task (reciting alternate letters of the alphabet)	ROI: Functional Near-Infrared Spectroscopy (fNRIS) of the prefrontal cortex	fNIRS	MCI was associated with slower gait speed under both single and dual task conditions.
												The decline in brain activation over single task trials was greater in MCI than in controls. MCI had less increase in brain activation from single to dual tasking compared to control subjects. The decline in brain activation from dual task trial 1 to trial 2 was attenuated in participants with MCI compared to controls, and brain activation declined from trial 1 to 3 in control subjects but increased over these time points in MCI.
Liu[21]	Correlation Between Gait and Near-Infrared Brain Functional Connectivity Under Cognitive Tasks in Elderly Subjects With Mild Cognitive Impairment	2021	17	20 (undifferentiated)	Controls: 64.4 SD (5.5) MCI: 62.4 SD (5.0)	NA	Not reported	Cross-Sectional	Four different gait-symmetry indices were calculated (1. symmetry of the left and right leg gaits by time phase of the gait cycle; 2. step symmetry in the Cartesian coordinates; 3. step symmetry in polar coordinates; and 4. the asymmetry index of the single support phase of the left and right leg) for four 10-minute gait paradigms: 1) normal walking; 2) counting backward while walking; 3) naming animals while walking; and 4) calculating while walking.	Based on fNIRS, functional connectivity between fifteen pairs of cortical regions were calculated: LPFC–RPFC, LPFC–LMC, LPFC–RMC, LPFC–LOL, LPFC–ROL, RPFC–LMC, RPFC–RMC, RPFC–LOL, RPFC–ROL, LMC–RMC, LMC–LOL, LMC–ROL, RMC–LOL, RMC–ROL, and LOL–ROL (left prefrontal cortex (LPFC), right prefrontal cortex (RPFC), left motor cortex (LMC), right motor cortex (RMC), left occipital leaf cortex (LOL), and right occipital leaf cortex (ROL)	14-Channel fNIRS	Functional connectivity between the LPFC and the ROL during walking while counting backwards and while naming animals was stronger in MCI than in controls. Functional connectivity between the LFPC and the LMC was stronger in MCI than in controls during the walking while naming animals task.
Nadkarni [24]	Gait and Subcortical Hyperintensities in Mild Alzheimer’s Disease and Aging	2009	33	NA	42	Controls: 73 SD (8) AD: 74 SD (8)	Controls: 47 AD: 60	Cross-sectional	Gait Speed, Stride Length, Cadence, and Step Width, Timed up-and-go, UPDRS score, Tinetti gait score	ROI: Subcortical White Matter Lesions	1.5 T GE Healthcare Signa MRI	Control subjects with low WML load had faster gait speed, longer stride length, and faster cadence than AD regardless of WML load. AD with high WML load had more Parkinsonian symptoms than AD with low WML load and controls, regardless of lesion load. Control subjects with low WML load were faster on the timed up-and-go than AD, regardless of lesion load. No differences were observed in stride width or Tinetti gait score (i.e., an overall measure of gait) between the groups.
Nadkarni [25]	Impact of Subcortical Hyperintensities	2012	20	NA	24	Controls: 72 SD (8) AD: 75 SD (9)	Controls: 47 AD: 60	Cross-sectional	Gait Speed, Timed up-and-go, UPDRS score, Tinetti gait score, Cognitive-Motor Dual Tasking. Dual tasking consisted of performing an n-back task while walking at comfortable gait speed with cadence as motor outcome measure.	ROI: Subcortical White Matter Lesions	1.5 T GE Healthcare Signa MRI	Control subjects with low WML load had faster gait speed than AD regardless of WML load. AD with high WML load were slower on the timed up-and-go than controls, regardless of lesion load. AD with high WML had more Parkinsonian symptoms than AD with low WML load or control subjects regardless of WML load. No differences were observed in Tinetti gait score (i.e., an overall measure of gait) between the groups. Relative to normal walking, all groups except for the AD with high WML load increased their cadence during dual tasking.
Rosso [23]	Slowing Gait and Risk for Cognitive Impairment	2017	89	69 (undifferentiated)	35 (unspecified dementia)	Controls: 72.4 SD (2.4) Cognitively Impaired: 72.9 SD (2.7)	Controls: 53.9 Cognitively Impaired: 60.6	Prospective Cohort	Gait Speed	ROI: Hippocampus, anterior cingulate, posterior cingulate, primary motor cortex, supplementary motor cortex, posterior parietal lobe, middle frontal lobe, caudate, putamen, and pallidum gray matter volumes	3.0 T Siemens Tim Trio MRI	ndividuals with MCI and dementia were combined in a single group of cognitive impaired subjects. Gait slowing was associated with cognitive impairment after 14 years of follow-up. The right hippocampus was the only region that was associated with gait slowing and cognitive impairment. Adjusting the association between gait slowing and cognitive impairment for hippocampal volume attenuated said association by 23%
Vidoni [31]	Evidence of Altered Corticomotor System Connectivity in Early-Stage Alzheimer’s Disease	2012	10	NA	9	Controls: 73.6 SD (6.3) AD: 69.0 SD (7.2)	Controls: 70 AD: 22	Cross-sectional	Hand Squeeze Task	Whole Brain Voxel-Wise Analysis: fMRI Brain Activation	3.0 T Siemens Allegra MRI	AD had increased activation in the premotor cortex and supplementary motor area during the hand squeeze task. Psychophysiological interaction analysis with the primary motor cortex (M1) as seed region was used to identify regions where brain activity correlates with brain activity of M1 during the squeeze task, and specifically, where that differs between the two groups.
												This analysis revealed that the AD group exhibited significant expanded recruitment of parts of the fusiform gyrus (BA 19), the middle cingulate gyrus (BA 23 and 31), the sensorimotor cortex (BA 3 and 4), the anterior cerebellum (lobule I-IV), and the cuneus (BA 19).