The social intelligence hypothesis holds that animal intelligence evolved to support complex social structures, and indeed almost all animals studied for their intelligence are highly social. In this context, it is hard to know which of the features associated with intelligence are byproducts of social organization and which are not. The octopus is an ideal animal to introduce some contrast into the study of the evolution and diversity of intelligence and cognition, as octopuses are both smart and largely asocial. They are born and grow in solitude and do not learn from live tutors.

Marcelo Magnasco, an integrative neuroscientist at Rockefeller University is working with Diana Reiss, a cognitive psychologist at Hunter CUNY  noted for developing interactive devices to offer choice and control to dolphins in experimental contexts, and proposes to create a novel context for study of octopus intelligence. With this project, they will explore the capacity of cephalopods to adaptively control devices in such a way as to enhance communication with experimenters and offer them choice about experimental conditions.

Octopuses are known for their problem-solving abilities, curiosity, and exploration, which the researchers will leverage to probe their intelligence. Using a paradigm that allows octopuses to probe into an “interactive wall”, the team will investigate how the animals use a single arm or integrate use of multiple arms in the following activities:

1.  The interactive wall affords octopuses the opportunity for choice and control in acquiring specific outcomes/contingencies by exploring and inserting a single arm or a combination of arms, into holes connecting to mazes behind the wall. This device is monitored by an infrared lightsheet, allowing computerized tracking of the arms as they explore the maze. 
2. The researchers will train octopuses to respond to sensory cues of different modalities (visual, chemical and haptic) to test their control and integration of sensory information.
3. The team will investigate the octopuses’ ability for architectural construction by allowing them to manipulate tiles that they might use to construct shelters. 

These approaches will allow the team to evaluate parallel and distributed processing of sensory information within and across modalities. Models of distributed intelligence and processing will be developed based on these studies.