A review of the unsolvable task in dog communication and cognition: comparing different methodologies.

Communication between dogs and humans is a topic of growing interest, and the "unsolvable task" is a common method used to measure human-directed communication. In this task, dogs learn how to solve a problem to obtain a reward. After a fixed number of trials, the reward becomes impossible to access, arguably leading to communicative attempts from the dog. Although useful to observe dogs' communicative behaviors in a fairly naturalistic situation, the methodology varies among studies regarding apparatus, number of trials, and other factors. The proxies used, for instance, gaze duration or frequency of gaze alternation, also vary, and there are discrepancies and a debate regarding what the task actually measures. Therefore, in this study, we reviewed the usage of the unsolvable task in canids of the genus Canis, searching Web of Science and Scopus for the terms "dog*", "Canis", "dingo*", "wolf" or "wolves" in the title and "unsolvable task" or "impossible task" in the topic. We included thirty-five studies in this review and discussed their different methodologies and proxies, such as different apparatuses, number of solvable trials, and different interpretations of "looking back", pointing out how they can affect results and hinder comparisons. Lastly, we used current data to propose strategies to homogenize the use of this important paradigm, with an ethogram of possible behaviors and their interpretation and a predefined set of methodological aspects for future research.