individual subsystems (e.g., cognitions, emotions, physiological factors, and conduct) that are in constant reciprocal interaction and influence each other in current processes with developmental consequences. Simultaneous appearance of low epinephrine level, elevated aggressiveness, and substantial motor restlessness at an early age in subjects later involved in crimes does not necessarily show that these factors are causative or that they have a common etiology. Similar considerations apply to many of the other hormone-behavior associations mentioned here.
Further, Levander et al. (1987) studied 40 recidivists in a Swedish state institution for delinquent males and compared these individuals with 58 "normal" school boys. In spite of the stressful backgrounds of the former, they showed patterns of stress hormone production suggestive of very low psychophysiological arousal. In this group, the common deviant psychoendrocrine pattern consisted of low epinephrine, norepinephrine, and cortisol, with high thyroid hormone levels. Different behavioral subtypes showed different hormonal profiles. Half the subjects had a history of hyperactivity associated with low epinephrine. Levander et al. speculate that elevated thyroid levels represent a compensatory mechanism attempting to correct the deviance in norepinephrine and, to some extent, epinephrine turnover.
Thus, although the available data are sparse, it does seem likely that longitudinal developmental studies involving multiple measures of hormones and varied ratings of aggression are yielding fruitful material.
The repeatedly demonstrated effects of hormones on behavioral motivation imply that these chemicals have direct actions on the CNS. Brain (1977) has reviewed the lines of circumstantial evidence for such actions on the brain. The following have been suggested:
Some steroid treatments cause morphological and/or receptor population changes in neural structures.
Hormones are sometimes more behaviorally effective when placed in particular neural loci.
Neural regions commonly accumulate specific behaviorally active hormones, a process that can be detected by injecting radioactively labeled hormone and then carrying out autoradiographic examinations of brain sections of treated animals. Pfaff (1971) has shown that major concentrations of neurally located sex steroids are found in the hypothalamus, the preoptic area, and the septum. Specifically, Naess and Attramadal (1974) documented the medial preoptic nucleus, periventricular nucleus, paraventricular nucleus, septal region, medial amygdaloid nucleus, and ventral premammillary body as accumulating androgens in rodents. The binding characteristics of neural regions that concentrate these compounds can also be altered by administering the hormones in early life, which changes the organism's behavioral potential (McEwen et al., 1970).