Math text problems (MTP) are verbal descriptions of a problem, which demand one or more answers that can be solved with different mathematical operations of numerical informations. Processes such as understanding of described situation, recognizing of important information and relations between them and also correct calculation are needed for problem solving. Despite the fact that mathematical operations are automised, students could still have issues with solving text problems. Students with ADHD face different issues with solving MTP even more often because of their difficulties with executive functions and working memory, directing and mainting attention, recall of arithmetic facts, recognition of relevant information and inhibition of irrelevant information, following and planning computational steps, validity of calculation etc. In case of co-occurance of ADHD and learning difficulties in mathematics there can be even greater issuses with solving math word problems. In theoretical part of master thesis we introduced and defined ADHD, learning difficulties in mathematics, math text problems and different solving approaches. Here we focused on irrelevant linguistic and numerical informations included in MTPs which are especially disturbing for students with ADHD.We also presented knowledge or abilities needed for solving MTPs and how deficiencies in this areas affect MTP solving and added successful model for solving MTPs Solve it! which is also the corner stone of our empirical part. The main goal of our masters thesis was to form, perform and evaluate metacognitive training of solving MTPs at specific group of students with ADHD signs and learning difficulties in mathematics which was based on our first assesment of their capabilities in solving MTPs. The research was done as a case study since there were included 3 students from 5th grade. From those three, the boy and one of the girls were directed into customise implemented program with extensive professional help, and the other girls was in active process of directing. Before the training started we tested students with Ten minute test to determine automation of arithmetical facts and procedures, Test of computional and numerical abilities and Test of efficient MTP solving and paraphrasing. The last test was also used after the metacognitive training of solving MTPs was finished. The metacognitive training lasted 3 months and encompassed 24 sessions. The results of the test after training was finished showed significant progress in solving MTPs, while progress in paraphrasing MTPs hasn't been detected. Students achieved the most part od their set goals and the progress was also acknowledged by their professors of special and rehabilitation pedagogy. Theoretical basics and findings of empirical part of the researh described in masters thesis can be helpful for teachers and professors of special and rehabilitation pedagogy and can give helpful insight for planning, forming and implementing help in area of solving MTPs with added irrelevant linguistic and numerical informations for students with signs of ADHD and co-occuring learning dissabilities in math.