What is "where": physical reasoning informs object location

This webpage contains demonstrations supporting our paper:

Boger, T. and Ullman, T. (in press).
What is "where": physical reasoning informs object location. Open Mind.

A link to a pre-print of the paper is available here.

Below, you will find links to the experiments we performed. These will run you through the experiments in the exact way our subjects saw them.

For readers, we made an additional experiment game of sorts, in which you can click on an image (like the main task) and then see how our subjects clicked on it. You can access this page here.

• Experiment 1: line drawings, main condition ("where") (secondary conditions used for modeling: "center of mass", "spatial memory", and "click anywhere")
• Experiment 2: real objects, main condition ("where") (secondary conditions used for modeling: "center of mass", "spatial memory", and "click anywhere")
• Experiment 3: masked objects, main condition ("where") (secondary conditions used for modeling: "center of mass", "spatial_memory", and "click anywhere")

(As an additional exploratory model, we tested a "grasping" model using another task in all three experiments. Here is the grasping task for experiment 1, experiment 2, and experiment 3)

You can contact the authors by email at: [anonymous].

All materials and data supporting our paper are available in an OSF respoitory here.

Questions? Email Tal Boger (tal.boger@yale.edu) or Tomer Ullman (tomerullman@gmail.com) for more information.