Objective: In part I of this series, the authors found that semantic knowle
dge and organization accounted for most of the variance in thought disorder
in a group of chronic schizophrenic patients. In the present study, they e
xamined a possible cognitive mechanism within the semantic system that migh
t produce thought disorder. Method: Twenty patients with chronic schizophre
nia and 21 normal comparison subjects were assessed on priming (the ability
to respond to a stimulus word more quickly when it is preceded by a semant
ically related word than when it is preceded by an unrelated word). The pat
ients were divided into subgroups with high (N=9) and low (N=11) levels of
thought disorder. The word pairs in the priming paradigm differed in their
degree of association but shared a categorical membership. The paradigm inv
olved short stimulus onset asynchronies to maximize automatic processing an
d required pronunciation of words to minimize decision making. All subjects
were also administered neuropsychological tests to assess language, execut
ive function, real-world knowledge, and mental status. Results: Comparison
subjects showed appropriate priming in stepwise fashion at the three differ
ent fevers of word association, as did the patients with mild thought disor
der. The patients with high thought disorder showed inhibited responses to
high and medium associates compared with their baseline reaction times. Cor
relations between priming and cognitive variables were significant only wit
h measures of semantic processing. Priming abnormalities were uniformly rel
ated to ratings of global thought disorder. Conclusions: These results sugg
est that aberrations in the automatic spread of activation or facilitation
in semantic networks may be a candidate cognitive mechanism in semantic acc
ounts of thought disorder.