Confabulation is a mysterious adjunct of amnesia. It remains unexplain
ed why some patients invent untrue stories in response to questions (p
rovoked confabulations) or even spontaneously with no apparent motivat
ion (spontaneous confabulations). Hypothesized mechanisms range from a
desire to fill gaps in memory to a loss of the temporal context in me
mory. We examined the mechanisms of confabulations in 16 amnesic patie
nts. Patients were classified as spontaneous confabulators if they eve
r acted according to their confabulations. Provoked confabulations wer
e measured as the number of intrusions in a verbal learning test. We f
ound a double dissociation between the two types of confabulations, in
dicating that they represent different disorders rather than different
degrees of the same disorder Confabulating patients did not show an i
ncreased tendency to fill gaps in memory as measured by the number of
fake questions concerning non-existent items that they answered. Neith
er type of confabulation correlated with a failure to store new inform
ation as gauged with recognition tasks; pure information storage was e
ven found to be normal in some patients. However we found a positive c
orrelation between several measures of verbal learning and verbal flue
ncy with provoked, but not spontaneous, confabulations. In contrast, s
pontaneous, but not provoked, confabulations were associated with an i
nability to recognize the temporal order of stored information as meas
ured by the comparison of two runs Of a continuous recognition task. W
e suggest that provoked confabulations depend on an amnesic subject's
search in his deficient memory and ape the trade-off for increased ite
m recollection. Spontaneous confabulations appear to be based on a fai
lure to recognize the temporal order of stored information resulting i
n erroneous recollection of elements of memory that do not belong toge
ther.