The first-perspective alignment effect : spatial memories from verbal descriptions, virtual environments and object arrays
Fourteen experiments investigate the 'first perspective alignment effect' (FPA), a novel finding that people sometimes encode a space preferentially in alignment with the first perspective they encounter. In Experiments 1--6, participants read verbal descriptions of three-path routes. A map-drawing task in Experiment 1 suggested that spatial memories are egocentrically encoded on the basis of forward-up equivalence. In experiments 3 and 4, when a salient landmark was described in relation to the start orientation of the routes, orientation estimates to remembered test locations were most accurate when participants imagined themselves aligned, rather than 180Â° contra-aligned, with the first part of the route. However, in Experiments 5 and 6 the introduction of allocentric cardinal term systematically affected the text FPA. Participants in Experiment 7 explored two versions of VEs based on the text descriptions used in Experiments 2--6. The FPA was found following the first VE exploration, but following the second VE exploration, the effect was attenuated. Experiment 8 omitted the alignment tests after the first VE, leading to similar results to those in the first test of Experiment 7, suggesting that prior experience of making orientation judgements in the first test of Experiment 7 had attenuated the FPA. Experiment 9 used a text-based procedure, in which participants were asked to make active spatial judgements after reading each section of the route, and the FPA was not found. Experiments 10--14 extended the above findings to learning about arrays of static objects from primary experience. In Experiment 10, when participants viewed an array of four objects from four perspectives, orientation judgements were similar for all perspectives. When arrays were viewed from two perspectives that were 0 and 90Â° (Experiment 11), and 0 and 180Â° (Experiment 12) misaligned from the centre of the array, no evidence for FPA encoding was found. The absence of this effect following primary, but not secondary learning Experiments 2--9; Wilson, 2001), was further investigated in Experiments 13 and 14, in which participants viewed the arrays under conditions of observer movement and display rotation. No evidence for FPA encoding was found; therefore, the presence of vestibular feedback in the primary case, does not explain differences in encoding between primary and secondary learning sources. The results are discussed in terms of spatial reference frames and spatial anchor points, and suggest that the FPA effect is the default form of encoding in spatial memory under some circumstances.