Psychopharmacology (t994) 115:185-195
Psychopharmacology
© Springer-Verlag 1994
Amphetamine-induced disruptions
mediated: implications for animal
dysfunction
A. S. Killcross, A. Dickinson, T. W. Robbins
of latent inhibition are reinforcer
models of schizophrenic attentional
Department of Experimental Psychology, University of Cambridge Downing Street, Cambridge CB2 3EB, UK
Received: 25 March 1993 / Final version: 4 November 1993
Abstract. Latent inhibition (LI) is a phenomenon ob-
served when repeated, non-reinforced presentation of a
stimulus results in a retardation of subsequent condition-
ing to that stimulus. Several recent experiments have sug-
gested that LI is abolished in conditioned suppression
paradigms following acute, low doses of amphetamine
given during pre-exposure and conditioning. Experiment
1 sought to increase the generality of this finding in an
appetitive LI paradigm, using a dose of amphetamine
previously shown to disrupt the LI effect in an aversive
paradigm (Killcross and Robbins t993). However, no ev-
idence for any disruption of LI was found. Experiment 2
extended this investigation to additional, higher doses of
d-amphetamine, and also examined the role of reinforcer
magnitude in the effect. A non-significant trend towards
an attenuated LI effect was found, which was reversed by
decreases in the concentration of the sucrose reinforcer.
Experiments 3 and 4 investigated the influence of sys-
temic amphetamine in aversive paradigms, with specific
attention to the increased response to the aversive foot-
shock reinforcer found in amphetamine-treated animals.
These experiments revealed that the influence of am-
phetamine on the LI effect in conditioned suppression
paradigms could be reversed by reducing the intensity of
footshock used in conditioning, thereby paralleling the
effect found in the appetitive paradigm. Therefore it is
unlikely that a simple attentional account of the aboli-
tion of the LI effect in previous experiments can be sus-
tained.
Key words: Dopamine - Latent inhibition - Schizophre-
nia - Amphetamine - Conditioning - Context -- Atten-
tion
Recently, there has been substantial interest in animal
models designed to reflect attentional impairments ob-
served in acute schizophrenia. One model is that based
Correspondence to: A. S. Killcross
on latent inhibition (LI), a phenomenon which is manifest
when repeated, non-reinforced pre-exposure to a stimu-
lus retards conditioning to that stimulus when it is subse-
quently paired with reinforcement (Lubow 1973). This
effect has been interpreted by some as reflecting processes
of selective attention by which irrelevant stimuli come to
be ignored (Mackintosh 1975, 1983; Lubow et al. 1976;
Lubow 1989), although this is by no means the only ex-
planation (Pearce and Hall 1980; Wagner 1981; Bouton
1991). The rationale for this model lies in a possible par-
allel between the effects of dopaminergic drugs on LI and
the deficits in selective attention seen in acute schizophre-
nia (Joseph et al. 1979; Lubow et al. 1987; Baruch et al.
1988; Gray et al. 1991, 1992).
Several experiments appear to have demonstrated
that the LI effect is abolished in rats following systemic
injections of either d- or dl-amphetamine; (Solomon et al.
1981; Weiner et al. 1981, 1984, 1987b, 1988; Solomon and
Staton 1982, 1984; Killcross and Robbins 1993), but not
following injections of direct dopamine (DA) agonists
(Feldon et al. 1991). The model is also supported by find-
ings of enhanced LI effects following treatment with neu-
roleptic drugs, such as haloperidol (Weiner and Feldon
1987; Weiner et al. 1987a; Christison et al. t988), and also
following systemic injections of atypical neuroleptics,
such as sulpiride (Feldon and Weiner 1991), but not
clozapine (Dunn et al. 1993).
The neural locus of these effects on LI has often been
attributed to the dopaminergic innervation of the nucleus
accumbens (NAc; Weiner 1990; Gray et al. 1991). More
specifically, the abolition of the LI effect by amphetamine
is claimed to result from the potentiating effect of am-
phetamine on dopaminergic transmission in the
mesolimbic DA projections which innervate the NAc
and related areas of the ventral striatum. However, recent
results have provided some evidence against this hypoth-
esis. Killcross and Robbins (1993) demonstrated that
whereas systemic injections of amphetamine can abolish
the LI effect in a within-subject conditioned suppression
paradigm, no such influence is seen in the same paradigm
following intra-accumbens infusions of either 3gg/gl or