Abstract
Repeated configurations of random elements induce a better search performance than that of the novel random configurations (contextual cueing effect, Chun and Jiang, 1998). We examined the effect of viewpoint change of a scene in a contextual cueing paradigm using a computer rendered illustration of a realistic scene. The scene contained an array of chairs of different shapes which were randomly positioned on the ground but oriented coherently. Observers were required to search for and identified an arbitrarily located target letter positioned on the surface of the seat of a chair. Half of the scenes were repeated over blocks and the other half were not. A typical contextual cueing effect (faster search for the repeated scenes than the novel scenes) were found. Following learning, participants continued to perform the search task (testing phase), but orientation of the entire scene (Experiment 1) or individual chairs (Experiment 2) was either changed (experimental group) or not (control group). For the experimental group, in Experiment 1, the viewpoint of the scene was rotated 45°. Moreover, a video illustrating such change for one typical scene was presented to the participants before the testing phase. In Experiment 2, the experimental group viewed the same video as in Experiment 1, but during the testing phase the viewpoint of the scene did not change, instead, each chair rotated in place 45° in the ground plane, creating a false sense of viewpoint change (or scene rotation). In the testing phase, contextual cueing effect was still evident (although with some reduction) in Experiment 1 but not in Experiment 2. These results suggest that implicit learning can be transferred to a new viewpoint and such transfer could be achieved through mental update of the view. Such update appears to be automatic and prompted by proper cues about rotation.
Meeting abstract presented at VSS 2016