Encoding ability is preserved in amnesia: evidence from a direct test of encoding.

The encoding ability of 17 amnesics of mixed aetiology and 17 matched normal controls was assessed directly using a novel procedure. On two separate occasions, subjects were shown 60 complex drawings each containing six pictures. On one occasion each drawing was shown for 6 sec, and on the other occasion it was shown for 25 sec. Immediately after presentation of each drawing subjects were asked a single unpredictable question about picture colours, location, size or semantic category. Amnesics performed normally in the 6-sec exposure condition indicating that all of the tested kinds of information were encoded at a normal rate. Performance in this condition correlated with short-term, but not long-term memory in the amnesics indicating that it depended largely on encoding and short-term memory. However, the amnesics were impaired in the 25-sec condition where performance should have depended on (long-term) memory abilities at which they were impaired. The results are inconsistent with available encoding deficit accounts of amnesia.