Effective, clinically meaningful treatments are lacking for patients with mild cognitive impairment (MCI), which is associated with increased risk of transition to dementia. Cognitive training represents an important therapeutic strategy. In our recently completed two-site, blinded, COG-IT trial (R01AG052440, PI Devanand), 107 participants with MCI were randomized to web-based cognitive videogames (computerized cognitive training, n=51) or computerized crossword puzzles (n=56) for 12 weeks of intensive training followed by booster sessions for a total of 78 weeks. Crossword puzzles were superior to computerized cognitive training on the primary cognitive outcome (ADAS-Cog11; effect size 0.43 at 12 weeks and 0.34 at 78 weeks) and function (Functional Activities Questionnaire) with decreased brain atrophy on MRI over 78 weeks. Building on these findings, for this competitive renewal we propose a 78-week clinical trial, WORDFIT, involving 4 U.S. academic sites that compares high dose crosswords (4 puzzles per week) to low dose crosswords (1 puzzle per week) and a health education control group in the initial acute phase of 12 weeks. Subsequent booster sessions will occur up to 78 weeks. Blinded assessments will occur at baseline and weeks 12, 32, 52, and 78. Sample size will be 240 with 80 participants per group. Sites led by Dr. Devanand (PI) at Columbia and Dr. Doraiswamy at Duke, who were the investigators in the COG-IT trial, will be joined by sites at University of Miami (PI Dr. Philip Harvey) and University of Washington (PI Dr. Angela Hanson). On measures of cognition (primary outcome ADAS-Cog14) and informant report of function (secondary outcome Functional Activities Questionnaire), we hypothesize that the high dose crosswords will show superior efficacy to the control group and that there will be an increasing trend in treatment effects across the three ordered groups (control < low dose crosswords < high dose crosswords). We will evaluate changes in MRI hippocampal volume and cortical thickness in relation to changes in cognition and function and explore whether reduction in plasma neurofilament light (marker of neurodegeneration) and plasma ptau181 (marker of Alzheimer's Disease) will be associated with efficacy for high dose crosswords compared to control. In summary, in our completed COG-IT trial in MCI, home-based computerized crossword puzzles training showed superior efficacy to computerized cognitive training on cognition and function with decreased brain atrophy over 78 weeks, but there was no control group. If these findings are confirmed and expanded with two crossword conditions and a control group in WORDFIT, crosswords training could become a low-cost, home-based, scalable, cognitive enhancement tool for people with MCI, which is a common condition and major public health problem worldwide.