The view that depersonalization is a stable syndrome became well establishe
d during the first half of the 20th Century. Current operational definition
s restrict depersonalization. to the experience of unreality. This is likel
y to neglect clinical features of potential neurobiological relevance. By u
sing the year 1946 as the dividing line, 200 cases of depersonalization dis
order reported in the medical literature since 1898 were divided into two h
istorical groups (1 and 2). The groups were then compared in terms of 18 ph
enomenological variables with a sample of 45 prospective cases of DSM-IV de
personalization disorder (group 3 or gold standard). Groups 1 and 2 differe
d in terms of their symptom profile, but the highest frequency that symptom
s achieved in either group did not differ from the rates identified in grou
p 3. A core of (invariable) symptoms, including emotional numbing, visual d
erealization, and altered body experience, was present throughout. These hi
gh rates of spontaneous reporting in all three groups may be explained by t
he fact that they all are accompanied by specific distress. With the except
ion of heightened self-observation and altered time experiencing, all other
symptoms were significantly lower in group 2. The results suggest that the
phenomenology of depersonalization has remained stable over the last 100 y
ears. Our study found differences in frequency for some symptoms, but these
are likely to have resulted from reporting biases, themselves governed by
changing theoretical views. Clinical descriptions became poorer as the pres
ent is approached. This cannot be solely explained on the basis of empirica
l progress, and it is likely that theoretical biases also play a role. Beca
use the neurobiological relevance of the symptoms of depersonalization rema
ins unknown, it makes sense to continue collecting as many symptoms as poss
ible, thereby avoiding both biased selection or premature closure.