574
R. D. Oades el al.
alternated from side to side while the colour sets were presented in a pseudo-random
sequence.
CB was measured by comparing the differential response latencies in A and B [i.e. B test
(third minus first stimulus) minus A test (third minus first stimulus). This was calculated
for the first pair of test stimuli, BA I; the mean of the first three pairs, BA 1-3; five pairs,
BA 1-5; all 10 pairs of stimuli, BA 1-10; and the first vs the last three test trials on B (Bl-
3)-(B8-10)]. Positive scores show CB or in the last case the development of “unblocking”.
Personality measures
Personality traits were rated on the Minnesota Multiphasic Personality Inventory in 33
subjects over 16 years old (MMPI, 221 questions: Gehring & Blaser, 1982), and the
Hamburg Neuroticism Extroversion Scale with norms up to 16 years of age was used for
all subjects to provide continuity across adolescence (HANES, 68 questions: BuggIe &
Baumgtirtel, 1975).
The MMPI scales studied were social introversion (SI), paranoia (PA), depression (D),
psychopathy (PD), psychasthenia (PT), schizophrenia (SC) and hypomania (MA). Atten-
tion was also paid to the clinical use of three neurotic features (hypochondria, D and SI)
and purported indicators for psycħopathologic (4-9, PD∕MA) or schizotypal personalities
(2-7-8, D∕PA∕SC; 8-9, SC/МА; Merritt & Balogh, 1990). We focus on single HANES
scales, scored 1-9, as indicators of emotional lability (“neuroticism”; N1), need and ability
to be sociable (El) and outgoing activity (sensation seeking; E2). The HANES extroversion
measures correlated negatively with MMPI SI alone (e.g. El, r= —.64, p=.008), and the
HANES neuroticism scores correlated positively with PT, SI, D, the sum of the three
neurotic dimensions, Iheschizotypal 2-7-8 scale but with no other dimensions (r=.52-.72,
p=.04-.002).
Urine collection
DA, NA, 5HT and their metabolites (homovanillic acid, HVA; 3-methoxy-4-hyd-
roxyphenyl-glycol, MHPG; 5-hydroxyindoleacetic acid, 5HIAA) were measured in 24-h
urine collections (low monoamine diet) on the day between tests A and B. Acidified samples
were frozen until analysis with ion-exchange liquid chromatography and fluorescence detec-
tion. Volume-adjusted measures were expressed as ng∕g creatinine per m2 body area to
correct for general somatic metabolic rates and large variations of body size (see Oades et
al., 1994).
Data treatment
Two-way MANOVAs (a precaution against intersubject variability) were used to study
(a) learning measures in the different task phases and (b) CB between subject groups:
repeated measures were used where related CB measures were studied. Significant results
were followed by one-way ANOVAs and post-hoc Newman-Keuls tests, cited in the text,
to determine where the significance lay. Correlation coefficients (Pearson ρ) were calculated
between learning and CB measures as an indication of the task-solving strategy used.
Bonferroni а-corrections are described in the text to guide the interpretation of significance.