In this, the first of a two-part sequence, we have first discussed some critical features of early behavioural research in geography, then traced developments following from this early research to the current period. We emphasized the early applied nature of much of this research. We have also shown how progress in the second period of growth (the 1970s) had to emphasize conceptual and theoretical aspects more than the applied. This was because fuzzy concepts, complex experimental designs and measurement methods werc lacking and required discovery and development before more extensive application was possible. This 'marking time' process.accornpanied by immature and poorly designed experiments started rumours of the 'decline' of behavioural research. Once problems of concept definition, model selection and model specification had been solved, behavioural research again exploded. The 1980s has seen abundant evidence of expansion and growth, with applications moving beyond the traditional planning. policy and marketing areas to such realms as developing the spatial competence of retarded and blind populations, to assessing neighbourhood reactions towards selected public facility locations and the modelling of wayfinding by robots. In our opening statement we argued that applied behavioural geography has taken many turns and that we deliberately selected only two of these (spatial cognition and preference and choice behaviour) as the theme for our two-part summary. Of course we could have extended our overview to include the important and well developed areas of adjustment to hazards. emotional response to environments, aesthetic qualities of environment, attitudes towards resource conservation. perceived threat from polIution or hazardous wastes or behavioural reactions to possible nuclear threat. terrorism or other dangers. Such an elaboration would require doubling the size of our papers. and we leave discussion of these areas to other more involved and qualified researchers. In our second paper, however, we do tackle the extensive multidisciplinary literature on spatial preference and spatial choice. We feel that. not only is this perhaps the most rapidly growing behavioural research area. but that it is an area where applications are numerous, obvious and practical. And whereas much of the cognitive-behavioural literature is to be found in North American journals and books. research on preference and choice is much more evenly distributed internationally. with some of the most productive research located in Europe. Thus the interested reader will find Part II of this paper in a later issue of this year's Progress in Human Geography, therein being exposed to detailed summaries of a variety of theories and the mathematical models which facilitate their application in a host of different problem environments.