Background: Often my key does not fit in the padlock I use to lock my door straightaway. The cylindrical part at the top of the keyhole tends to turn along the axis of the key (i.e. perpendicular to the lock plane). The base of this cylinder is coplanar with the face of the lock (the rest of it is embedded within) and it houses a part of the keyhole. I can correct the orientation of this cylindrical part so as to align the groove that goes through the length of cylinder with the rest of the keyhole, without paying much attention to it today. Initially, it used to be a slow process, often requiring me to use my eyes to detect the incorrect orientation of the cylinder. But, over time, perhaps partly due to the poor lighting outside my room, my fingers seemed to have learned how to do the process without the use of any visual input, and the process is refined and fast.

The Process: I hold my lock in my left hand while unlocking and feed in the key using my right hand, gripping the key between the first phalange of my thumb and the middle phalange of my index finger. I hold the lock in such a way that the four fingers are coiled around the backside of the lock and the left thumb is free to touch any point on the face of the lock. Once the key does not seem to slide in smoothly, the tip of my left thumb automatically touches the rotatable cylinder. It presses around the protruding circumference of the base of this cylinder. When it detects a break in the protrusion, it holds the cylinder fixed at that position. On doing so, my right hand rotates at the wrist, along an axis parallel to the key axis till the key is oriented in such a way that the outward projection at the tip of the key fits into the groove in the cylinder. To ensure that it is only the outward projection that touches the lock and not any other part of the key, my wrist bends while rotating. Hence the key now touches the lock not perpendicular to its face, but oblique to it. Once the projection is in the groove, I release my left thumb and the key is rotated along an axis perpendicular to that of the lock face and also simultaneously pushed inward. As soon as the groove orients correctly with the rest of the keyhole, the key slides into the keyhole due to inward force applied.

Discussion: This process seems to be a case of motor learning due to regular repetition of events and an extensive practice time. The hands seem to be aware of each other’s position and so do the fingers in each arm. Also, the hands are aware of the position of the cylinder with respect to the lock as the thumb automatically falls on it. Another feature that can be admired is the coordination of the two hands with touch as the only sensory input. However, the most fascinating aspect that I found was the ability of the brain to form a map of the lock face with the touch inputs. The incorrect orientation of the cylinder was obtained through the left thumb tip and the right wrist was made to rotate in a direction so as to complete the keyhole. The brain seems to form an image of both the grooves and acts so as to co-align the images.

References:
Discussion inspired by Wilson RA and Keil FC (Both ed.) (2011); The MIT Encyclopaedia of the Cognitive Sciences. The MIT Press Cambridge, Massachussettes. ISBN 0-262-73144-4