Friday 11 February 2011

The Neural Navigator


The Neural Navigator: a novel navigation and collaborative filtering system for retrieving online art content.

The problem of suitable and innovative means for filtering and navigating online art content has long been acknowledged by influential figures in the domain of digital art practice, for example Beryl Graham (1999), Edward Shanken (1992) and Lev Manovich (1996) have all critiqued the fragmentation and anachronism of many online art contexts. This paper will describe the development, testing and evaluation of the Neural Navigator, an embodied system for navigating and collaboratively filtering online art content that addresses many of the core problems in the field of online art interaction.

The Neural Navigator is a physical computation system that deploys EEG (electroencephalography) to sense the electronic brainwave frequencies of individuals while they are visiting online art sites. The system analyses the patterns of electroencephalographic signals and matches them to suitable art works based on a collaborative filtering algorithm developed over the last two years by Eleanor Dare and Lee Weinberg, doctoral students at Goldsmiths (University of London) in the departments of Computing and Art.

The paper will describe the reasons for pursuing an embodied form of online navigation, framing this goal within the context of increasing interest in embodied interactive systems, while acknowledging the contribution of cognitive scientists such as Francisco Varela, Eleanor Rosch, Evan Thompson (1993) Alva Nöe (2009, 2004) and significant art practitioners such as Jeffrey Shaw (2002) and Robert Lazzarini (2001).

The proposed paper will describe the empirical testing of this system and the reactions of users to the experience of a seemingly sub-symbolic, autonomic process of interaction with the Computer Fine Arts Collection, which has been the focus for the development of the Neural Navigator. The architecture of the Neural Navigator has been influenced by an enactive and situated methodology that privileges action over a priori goals, and instead seeks an emergent, fluid and constantly changing set of navigational pathways. The paper cites Lucy Suchman (2007) and Alva Noe (2009, 2004) as key theoretical figures in the decision to pursue an enactive methodology, this will be clearly outlined and justified, as will the contribution of this work to the field of online art navigation, filtering and content retrieval.

Key words: Content retrieval; collaborative filtering; bio-sensors; social content and tools;

1 comment: