Objective The objective of this systematic review and meta-analysis (PROSPERO registration No CRD42020192121) is to review existing literature focusing on effects of different dual task paradigms on walking speed in older adults with and without Mild Cognitive Impairment. Methods (1) Data Sources: PubMEd, Cumulative Index of Nursing and Allied Health, Cochrane library, and Web of Science. (2) Study Selection: The key terms searched included those associated with dual task, walking speed, executive function, older adults, and MCI. (3) Data Extraction: The search yielded 140 results with 20 studies meeting the inclusion criteria, which were rated by two independent reviewers using the Quality Assessment Tool. Descriptions of each study including the single and dual task protocol, outcome measure, and final outcomes were extracted. Meta-analysis was performed to evaluate the dual task effects on walking costs in older adults with and without MCI. Results Meta-analysis revealed that there were significant differences in the dual task walking costs among older adults with or without MCI (p < .05). Pooled effect sizes of the serial subtraction (9.54; 95%CI, 3.93–15.15) and verbal fluency tasks (10.06; 95%CI, 6.26–15.65) showed that there are higher motor dual-task costs in older adults with MCI than age-matched controls. For quality assessment, all studies ranged from 12 to 16 in score, out of 18 (high quality). Conclusions In the studies included in this review, mental tracking tasks, consisting of serial subtraction and verbal fluency, were found to be the most sensitive in detecting MCI-related changes in older adults, and could serve an important role as a target measure for evaluating the efficacy of interventions aimed at improving cognitive and motor function in older adults.