Search

We are investigating dogs´ somatosensory processing using functional magnetic resonance imaging (fMRI). The data are already collected (data-collection ongoing) and we are only preregistering our analyses.

Experiment 1: Description of the 3D tracking setup, the pre-processing of the 3D data, and of the assumption checks, and detailed results of all models. Experiment 2: Description of the setup and the warm-up procedure, additional analyses concerning potential effects of control predictor variables (e.g., order effects), and detailed results of all models.

Follow-up of the unwilling-unable paradigm with dogs. Do dogs discriminate between people based on unwilling (teasing) and unable (clumsy) demonstrations?

We are conducting an eye-tracker experiment, in which we are showing dogs (Canis familiaris) videos of a human and an inanimate agent performing a goal-directed action.

With two eye-tracking experiments, we aim at testing the embodied account of action prediction.

Data for Lonardo et al. Dogs follow human misleading suggestions more often when the informant has a false belief

With the present study, we aim at investigating whether pet dogs (Canis familiaris) understand human false beliefs and whether they adjust their behaviour accordingly. In particular, we are going to address dogs’ ability to distinguish between true and false beliefs of a human experimenter (the “player”) in a novel non-verbal task: a social game with rules established by the human player. In a post-test condition, we will target explicitly dogs’ reaction to the player’s ignorance in the same task. The game is designed to elicit dogs’ ability to take into account the human player’s mental states because the “correct” response that dogs should exhibit to get a reward will depend on the player’s beliefs. Dogs will first learn a go/no-go task during a short training phase. In this, the presence of food in one opaque container (the “go box”) will be associated to the “go” response: dogs have to retrieve the food autonomously from the box within 10 seconds. The presence of food in a second opaque container (the “wait box”) will be associated to the “no-go” response: dogs have to wait for 10 seconds without touching any of the boxes in order to get a different piece of food from the player. If dogs make a mistake (moving when food is in the wait box or not moving for 10 seconds when the food is in the go box), the player will remove both boxes from the dogs’ reach. Once dogs have reached predefined learning criteria (see below), they will be ready for the test phase, in which test trials will be interspersed among training trials to keep up dogs’ motivation and to re-establish the rules of the game. During test trials, dogs will witness for the first time that a second experimenter, “the hider”, transfers the reward from the wait to the go box either in the presence (true belief condition) or in the absence (false belief condition) of the player.

Supplementary information on experimental procedures, statistical analyses, results of statistical models and and legend for datasets (that are available as separate files in a Github repository)