The present study investigated the question of whether and under which conditions elementary school children at the age of 9 to 10 trust another unfamiliar, same aged child of the same gender in trust games. Furthermore, the study examined whether variables derived from learning theory, that is, number of siblings, birth order, number of friends, and height and the additional variable expected reciprocity were associated with trust behavior of children at this age. In order to measure the observable manifestation of trust, i.e. the trust behavior, 92 children played two child-adapted trust games, in which they had the choice to either trust or mistrust another child who was shown on a picture. A third game examined altruistic behavior in order to ensure, that the behavior shown in trust games was actual trust behavior as opposed to altruistic behavior. The two trust games differed in the possible outcome alternatives with the goal to determine if children base their decision on either fairness or benefit concerns. Moreover, the children were questioned about their motives for trusting or not trusting in the trust games. The results demonstrated no difference of trust behavior between the varying conditions of both trust games, that is, a game with a fair outcome on one hand, and a game with an unequal outcome with a lot to gain, just as a higher risk of betrayal on the other hand. Expectations of positive reciprocity and trust decisions were positively associated. Children with a lot of friends were more likely to trust than children with fewer friends in the trust game with a higher potential benefit. The number of siblings of boys was negatively associated to their trust decisions in the trust game with a fair outcome. Over all, the results indicated that especially experience of positive reciprocity, i.e. social learning, affects trust behavior of elementary school children.