1a

The intent of this analysis is to model the reinterpretation and deviation in learning, as surveyed by this video. The model uses nurb surfaces to 3-dimensionally describe the simultaneous forces of environment (gravity, horizontal and vertical surfaces) and interpretation (watch, react, distort).

Note: the analysis of this model was limited to the video segment of "the hop" (0:01-0:05). The actor on the left is the dominant actor (both on the squash court as well as in this demonstration), while the actor on the right is "following-the-leader" (and getting blanked 0-5 every match).

The original motion study diagrams the relationship between the control and the mimicry as a polygon, where the vertical segments are the relation of each actor's head to foot, and the horizontal segments connect both actors' heads and both actors' feet.



The original diagram suggested a sequential distortion of a single regular polygon. From this observation, I established a single regular rectangular polygon to establish a control relationship maintained between the two actors, and employed the tool of perspective as the distorting device. The sequential distortion of this form was thus accounted for by rotation and adjusting location of each polygon along the z-axis. However, these perspectival tools were not able to describe the full distortion of the relationship. Thus, a second polygon, a triangle exactly half the area of the original rectangular polygon, was incorporated to triangulate the difference.




The triangular geometry became the foundation for the control's remainder. In this way, the resultant blob of the triangular polygons models the isolated distortion of the original control relationship.

In the resulting models, we see the event of learning in terms of its relative symmetry and distortion.


Here the control blob is isolated.


Further inspection reveals idiosyncracies of the deviations. In fact, occassionally an additional level of triangulation was necessary for the perspectival "fit". As a result, we are able to divide the learning blob into two, based on the two unique paths of the different triangular geometry. (my logic breaks down here, but it's still cool)


Addtional views from the original diagram perspective, with camera property adjustments.