Why do we see only one interpretation at a time when viewing an ambiguous figure? Minsky says our agencies can tolerate only one interpretation at a time. Question: why?
To answer this question, Minsky turns to problems of vision: Why do we see objects as being composed of features? How are features grouped into objects? "Our vision-systems are born equipped, on each of several different levels, with some sort of 'locking-in' machinery that at every moment permits each 'part, at each level, to be assigned to one and only one 'whole' at the next level."
NB1: One possible way to view the machinery alluded to in the quote above is in terms of stable attractor states in a recurrent network.
NB2: Minsky will expand this hypothesis to a "locking in" of semantic interpretations.
NB3: Minsky is portraying vision as a hierarchy of agents solving problems of underdetermination a la Marr -- feature detection, object detection, object interpretation -- all constrained by a locking in of unique solutions. Must then the locking in mechanisms reflect natural constraints to be used to solve problems of underdetermination?