A database of digital line drawings that depict expected and unexpected action-place relationships

In the present study, we introduce an image database created with a simple digital line-drawing tool to represent the expected and unexpected action-place relationships. This database consists of 70 drawings. We validated the dataset with 207 participants. They evaluated the actions, places, and the probability of an action taking place in the respective location. The comprehensibility of each drawing was evaluated using three measures: H-statistics (entropy), which is a measure of the uncertainty of the definition; the definition similarity percentages which is a measure of the naming agreement that is consistent among participants; and the mean probability of the depicted action taking place in the depicted location. Each drawing includes different agents and environments, providing researchers with the opportunity to use this dataset in various fields of cognitive neuroscience, including visual recognition, memory, predictive coding, and novelty detection.