The paper assesses the impact of morphological features of computer mice on the occurrence of unreliability. A comparison of the number of unreliable cases, the duration of the executed control task and control accuracy for nine various mice and three test subjects is presented in this paper. Conclusions: Particular assessment results, according to a person’s execution time for a particular task and the number of unreliable interactions, do not demonstrate a relationship. Whereas, a relationship was demonstrated in the number of occurring unreliable cases, where rises and falls in the occurrence of unreliable cases were observed. The analysis of results demonstrated that tasks using a mouse with the highest and the lowest ergonomic level took the longest to perform. The mice assessed as moderately ergonomic were the best for quick work. An assessment of the ergonomics criteria that the tested mice fulfil, applying the Fitts’ Law test, is convergent with the analysis of the number of cases of unreliable interactions.