Modifying the gestalt factor of proximity: theories compared

A theoretical controversy between adaptation-level (AL) theory and response-frequency-equalization (RFE) theory was examined. The stimulus presented for judgment was a matrix of dots which varied in appearance as the relative horizontal and vertical separation between the dots was altered. The matrix appeared to be columns of dots (larger horizontal separation), rows of dots (larger vertical separation), or neither row-like nor column-like (when the two separations were equal). A stimulus with an extreme row organization was presented on alternate trials and resulted in more vertical judgments. Both AL theory and RFE theory are shown to be inadequate on two counts. They both predict that the response latencies to the vertical stimuli should be affected, whereas these were unaffected. They both predict that in terms of the categorical decision, 'horizontal' or 'vertical', the responses to both horizontal and vertical stimuli should be affected. It is shown that the context effect is almost entirely on the horizontal stimuli. These results are consistent with a spatial-adaptation model.