Abstract
The aim of this paper is to provide a survey on the perception of robot intelligence. After a general overview, the paper specifically focuses on how perceived intelligence varies either between before and after the experiment (“pre-test-post-test variation”), and in subsequent sessions due to habituation. After reviewing the main characteristics of autonomous agents and robots that have been shown in the literature to influence the perception of robot intelligence, papers focusing on the variation in time of perceived intelligence are analyzed in detail. Even if no unanimous conclusion was reached in the literature, evidence suggests that, in general, when a significant variation is detected, perceived intelligence tends to increase from pre-test to post-test evaluations when commercial or more recent robot platforms are used, while it tends to decrease in the case of custom-made or less recent robots. On the other hand, when a significant variation is detected, perceived intelligence seems to increase due to habituation.