382
R. D. Oades et al.
patients the importance of the HVA∕MHPG ratio was confirmed with MHPG and the
four HVA measures in the linear regression (BA 1-3, —.91, p=.007; unblocking, —.98,
∕><.0001). Stepwise analysis showed that although HVA and MHPG were negatively
related to CB, their ratio was dominant. Recalling that NP values for NA utilization and
HVA∕MHPG were higher than for controls and that 66-75% of HVA is derived from NA
metabolism, the negative relationship with CB may be attributable to relative changes in
NA rather than DA activity. The positive relationship for CB with HVA and HVA∕MHPG
in comparison groups may reflect the more dominant role of DA activity in their
performance.
Discussion
Main CB results
Normal CB was seen in healthy controls and OCD subjects. Psychotic patients taken
together showed an attenuation on the first test trial-pair reflecting decreases in both PH
and NP subgroups. Over three-five test trials NP patients continued to show less CB than
either the PH or comparison groups. In psychotic patients extremes of negative or positive
symptoms (e.g. flat-affect or bizarre behaviour) could interfere with CB. Among personality
traits extroverted tendencies were beneficial in non-psychotic subjects but in those with
psychosis, extremes of neuroticism or psychoticism could interfere with normal CB.
Increased DA metabolism was associated with enhanced CB in both comparison groups,
but a relative increase of 5HT metabolism in OCD patients interfered with learning about
the added stimulus during testing (unblocking). A relative increase of NA vs DA activity
may be associated with reduced CB in NP patients, but no clear relationship with the
activity of the monoamines appeared in PH patients.
Learning strategies and personality features
The psychotic patients needed more trials and time to complete the task than the
comparison groups, yet for healthy and patient groups alike, correlations between learning
trials, exploration time and task duration were remarkably specific for each phase of each
session. Further, neither the nature nor the severity of the symptoms correlated with
learning measures. This is important as it implies there was no simple or artifactual
explanation in terms of exposure to the task to account for group differences in CB.
However, the two exceptions are informative.
First, the PH group explored the task arena for longer than the NP group. Indeed the
amount of exploration shown by the PH group was remarkably similar on both sessions
suggesting a stereotyped approach to the spatial constraints or context of the task. An
insensitivity in adapting to where it was worth∕not worth going with the cursor is consistent
with the theoretical position taken by Hemsley (1987). He has interpreted the problems of
information processing in patients with positive symptoms in terms of a reduced use of
regularities from past experience. The PH strategy or cognitive style contrasts with the
second feature where only the NP group showed a correlation between the number of
learning trials on the two three-stimuli phases (i.e. Al and B2). This suggests a more
stereotyped approach to the attentional or task demands of learning about three stimuli