Observations and Anecdotes
- The technique is generally more effective and dramatic when working with a phantom arm rather than a leg. Where there is a below-the-knee amputation, outcomes are usually better than with above-the-knee amputations.
- Where sensory “remapping” occurs, for example, stimulation of the cheek, neck, or throat producing sensation in the phantom hand or forearm, the mirror-box approach is almost invariably successful. Where no remapping occurs, results tend to be far less significant. I have no clear explanation for why this should be the case.
- When pain is primarily neuropathic or originates from stump pain, the mirror box tends to be of little benefit.
- Often, mirror box therapy fails for a very simple reason: it is not set up correctly. One individual with a left-arm amputation at shoulder level following a brachial plexus injury experienced a phantom arm twisted behind his back. Under those conditions, the mirror box was of no use. We first had to manipulate the phantom into a forward position so it could be placed into the box. In another case, once the phantom was positioned correctly, the illusion still failed until the client removed a wristwatch from his remaining wrist. The illusion must be total.
- At the point where the mirror effect occurs, there is often a powerful emotional release that is not unpleasant. After this initial effect, individuals usually report feeling “pleasantly different”. This emotional response does not typically recur during subsequent mirror box sessions.
- Success rates increase significantly when the client approaches the mirror box with their intact limb already positioned as a mirror image of the phantom. This phenomenon may explain difficulties encountered with phantom legs, where the phantom limb often differs in length from the remaining leg.
I would be interested to hear of other people’s experiences with this technique, both successful and otherwise. In time, I hope to collate these observations and present the data on the website.

