Pathways, Perception and the development of Place: Computational approaches to movement and perception of landscape in prehistory (original) (raw)

Traditionally, analysis of archaeological landscapes has drawn on two separate theoretical traditions. Functionalist explanation in the tradition of Renfrew (1973) emphasises general patterns in economic and social structures, and the development of formal methods. Approaches informed by phenomenology, such as those by Tilley (1994) or Thomas (1999), emphasise human-scale experience of landscape and the ‘constructed’ nature of space leading to, for example, suggestions of ‘circular landscapes’ in the Neolithic of Britain (Bradley 1998). In the last decade, approaches that draw on both of these have emerged in the context of GIS based landscape studies. One such area has been visibility analysis (e.g. Wheatley, 1995; Exon, Gaffney, Woodward & Yorston, 2000) in which formal methods for analysing visual characteristics of landscape build up from an understanding of perception at the scale of individual human actors. Although successful, most of these have dealt with visibility from static locations and ignored (orminimally considered) the effects of movement. This project centres on the investigation of movement and perception within archaeological landscapes; intending to study dynamic changes in visual envelope afforded active, mobile agents. By comparing archaeological features with these changing patterns of visibility, the aim is to form hypotheses regarding potential modes of interaction with and hence development of the continually evolving cultural continuum that is landscape. Patterns of changing visibility are being further investigated by incorporating some elements of three dimensional visualisation in order to take account of (for example) colour, lighting and atmospheric models; key factors which affect human perception of space. These issues are particularly significant in the context of later Neolithic Britain, in which a series of monumental forms seem to have developed that formalise aspects of movement through landscape (e.g. cursus monuments, avenues). Consequently, the research focuses on later Neolithic landscapes including Avebury and the Dorset Cursus complex. This paper presents some of the methodological and technological developments to date, including the development of a computational framework for the investigation involving the fusion of a Geographic Information System and three-dimensional technologies. In addition to the development of approaches to viewshed analysis within the GIS, the use of a threedimensional modelling package to produce rendered views from the GIS will also be discussed and some preliminary results will be presented. Image processing techniques for the analysis of these views with also be discussed. Another aspect of the proposed framework is the use of a gaming engine to provide an interactive, dynamic three-dimensional interface linked to the GIS and rendering suite.