Elderly is one of the most important periods of every person’, s life, which can have negative physical and psychological consequences for the individual, family and society if not spent with satisfaction and quality. Due to the change in the age structure of Iran's population from youth into old age, many researchers have considered the predictor variables of the life satisfaction in this age group. Therefore, the aim of the current study was to predict life satisfaction based on attachment dimensions and gratitude in the elderly. This was a descriptive study of correlational type. The statistical population consisted of the all elderly living in nursing homes of Yazd, Iran, among which Fatemeh Al-Zahra nursing home in Taft was selected. Selected using convenience sampling method were 110 people who filled out the satisfaction with life scale (Diener et al. ), experiences in close relationships-revised (Brennan, Clark & Shaver), gratitude questionnaire (McCullough et al). The obtained data were analyzed using Pearson’, s correlation coefficient and multiple regression analysis in SPSS-24 software. The results obtained from Pearson’, s correlation coefficient indicated that among the predictor variables, avoidance attachment and gratitude respectively, had negative and positive correlation with life satisfaction, but there was no significant relationship between anxious attachment and life satisfaction. Besides, regression analysis showed that among avoidance attachment and gratitude variables, it was only gratitude that predicted life satisfaction significantly (P<0. 05). Gratitude could explain 44% of the variance of life satisfaction. Based on the mentioned results, gratitude as a positive and valuable characteristic, predicts life satisfaction in the elderly, through affecting social interactions and inducing positive change in one’, s outlook on life.