sensitization. Note also that both of these arrangements would permit a CS alone to elicit conditioned activation of the system in the absence of a drug.
Shown in Fig. 7.12 are two other configurations that would allow for a CS paired with increased extracellular DA in both somatodendritic and terminal regions of the neuron to gain control over the expression of sensitization. In these cases, the CS unit interacts with the DA neuron only in the terminal regions. The CS unit (3) is shown to have access to the terminals where it could gain control over the release of DA. In this case, CS control would develop at terminal sites independently from the development of sensitization, itself, and would act there to influence the expression of sensitization. This possibility is based on recent studies by Glowinski and colleagues and by Phillipson and colleagues, both of whom have evidence for modulation of release of DA from terminals in the striatum via activation of a thalamo-cortico-striatal neuronal loop ( Cheramy, Romo, Godeheu, Baruch, & Glowinski, 1986; Kilpatrick, Jones, Johnson, Cornwall , & Phillipson, 1986; Kilpatrick, Jones, Pycock, Riches, & Phillipson, 1986; Kilpatrick & Phillipson, 1986; Romo, Cheramy, Godeheu, & Glowinski, 1986a, 1986b). Pharmacological and electrical stimulation of certain thalamic nuclei induce the release of DA from striatal neuron terminals, and there is evidence to suggest that these effects are due to presynaptic facilitation. Cheramy et al. ( 1986) also found indirect presynaptic inhibitory control over DA release from terminals. Although less is known about similar effects in the thalamo-limbic cortex-accumbens-VTA loop, Jones, Kilpatrick, & Phillipson ( 1987) have found that stimulation of the dorsomedial nucleus of the thalamus can regulate DA release from terminals in frontal cortex. The arrangement shown (3) would allow the CS to facilitate or inhibit the release of DA from terminals, both in the presence of drug where it could modulate the expression of sensitization, and in the absence of drug. In arrangement (4), the CS unit is shown to interact directly with the postsynaptic neuron and to be influenced presynaptically by the DA neuron. In this case, DA would be involved in the development of conditioning, but the CS would not come to have any direct facilitatory control over activity within the DA system itself. Whether such an arrangement would allow for the conditioned control of sensitization is unclear.
These attempts to model the various ways in which concurrent activity in systems carrying information about the CS might interact with drug-induced changes in the DA systems are preliminary and clearly simplistic. They do force us, however, to think about how and where in the nervous system the learned changes in the ability of sensory stimuli to modulate activity might take place.
This chapter was supported by grants from the Medical Research Council of Canada and Fonds pour la Formation de Chercheurs et l'Aide à la Recherche.