Naos is a research group dedicated to the investigation of automated biometric classification, and the exploration of biometric architecture. The project’s objective is to design an ergonomic brain-computer-architectural interface, compelling biometric visualizations and psychological test examples to explore the legitimacy of these types of systems at the nexus of the biological, computational and architectural.

Its first project is the Naos Platform™, a biometrics and psychological testing system. The Naos Platfom™ includes a Biometrics Capsule, Biometric Tendency Recognition and Classification System (BTRCS)™ and the Naos Adherance Index™. The current system implementation utilizes a Neurosky(TM) EEG brain-scanning headset, a Thought Technologies(TM) Galvanic Skin Response sensor, biometrics recording software, user profile and biometrics database, architectural capsule system and a psychological test example--the Naos Loyalty Test. This test classifies individuals based on their reactions to people of other races in order to determine their level of prejudice. It serves as a very basic example, one of many, that could be undertaken using the Naos Platform™.

Integral to the Naos Platform™ is the Naos Capsule, an example of biometric architecture—an ergonomic brain-computer-architectural interface. Its design is influenced by recent research in the neurosciences and its impact on the design of living, work and therapeutic spaces. The intention is to explore the influence of architectural spaces on neural activity and emotive states.

The group is simultaneously fascinated by and skeptical of these types of psychological classifications, especially when they are determined by a machine. So here they offer an experience for people to investigate their own feelings about possible future uses of biometrics technology and psychological testing procedures. Their uncertainty regarding the validity and usefulness of such tests inspires them to create situations for others to experience and therefore be able to question the results based on a real experience. The work seeks to examine similar, very real tests and situations that exist today and will become more prevalent as these technologies proliferate.

Naos Principal R & D: Carlos Castellanos and Luther Thie

Naos Collaborators: Joseph Cori , Eyal Fried, Philippe Pasquier

Naos is sponsored by Acclair, a security and neuromarketing service, with equipment and software support by NeuroSky, Thought Technology, Ltd and Belkin.

The installation consists of a biometrics capsule where images are projected onto its ceiling while the user’s physiological responses are measured using an EEG headset and a GSR handset. The system measures brain waves via electroencephalography (EEG) and skin conductivity via galvanic skin response (GSR).

The Naos capsule is a biometrics service station, where the user enters and is "hooked up" to physiological monitoring (biometrics) equipment. After being safely secured within the capsule, the user undergoes a test procedure, while the biometrics data is recorded and stored in a database. After first gauging the participant's physiological data for five seconds, in order to establish baseline readings, an image is projected onto the capsule’s ceiling. The participant's physiological response to the image is measured. Then via statistical classification and depending on the specific test being administered, the participant is rated and placed into one of several possible categories. Another image whose data is closest to the measured physiological data is then shown. This loop continues in an attempt to achieve "equilibrium" - a point where the image's expected physiological response and classification and the participant's actual response and classification are the same. Thus, in a circular relationship, the system influences the participant's body and brain that in turn influences the system. Examples include being classified as "aggressive" or being given a "loyalty" rating of 0.5. This process of classification is accomplished via a machine-learning algorithm.

The Biometric Tendency Recognition and Classification System (BTRCS)™ is a software application that continuously measures a participant's physiological responses to a given image and runs a statistical classification algorithm on the measured data that then classifies the participant into one of several predetermined categories. BTRCS forms the core of the Naos system. It includes a database of images that can be customized based upon the particular test being administrated. The classification is achieved using a K-nearest neighbor algorithm. This machine-learning algorithm measures the distance between the participant's physiological responses to images and the responses that are in the database - what is known as the training data. With each new participant, new training data is continually being added to the system.

The current system focuses on dividing the brain's activity into three distinct frequency bands (alpha, beta and theta). These bands are analyzed and used to discern the participant's levels of anxiety, attention and meditation. These levels are then mapped to 3-dimensional visualizations using NURBS, or Non-uniform rational B-splines. In addition, the GSR level is mapped to a 3-dimensional animating shape that increases in size as the GSR level increases. The test stimuli are also shown in correlation with the visualized biometric data.

The Naos Loyalty Test™ is an example of the type of test that can be administered using the Naos Platform. This test classifies individuals based on their reactions to images of people of different races coupled with questions intended to elicit emotional responses. The biometric reactions are then used to determine their level of prejudice on a "loyalty scale".

GitHub (https://github NULL.com/carloscastellanos/Naos)