I'm not sure if the article is poorly worded or the study was poorly implemented. If the study consisted of having test subjects watch a 30-second face morphing

I'm not sure if the article is poorly worded or the study was poorly implemented. If the study consisted of having test subjects watch a 30-second face morphing, and afterwards the subjects selected the face at the 15-second mark as the one most representative of what they saw, it would tell me that they chose based on the face that averaged out to the morphing face they saw. The average look of the morphing face would likely be right in the middle, 15 seconds in. I don't see how anyone could assume that it meant that their brains were processing images from 15 seconds in the past.