Behavioural Brain Research 154 (2004) 423–430 Research report Rapid toxin-induced gustatory conditioning in rats: separate and combined effects of systemic injection or intraoral infusion of lithium chloride Shelley Cross-Mellor ∗ , Sharon N.D.A. Clarke, Klaus-Peter Ossenkopp Neuroscience Program and Department of Psychology, University of Western Ontario, London, Ont., Canada N6A 5C2 Received 10 September 2003; received in revised form 5 March 2004; accepted 10 March 2004 Available online 17 April 2004 Abstract The present experiment examined the individual and combined effects of systemic injection and oral ingestion of lithium chloride (LiCl) on both within and across session shifts in palatability. Male rats fitted with intraoral cannulae received two conditioning days in which they were injected with either LiCl or sodium chloride (NaCl) and were then presented with brief intraoral infusions of a sucrose plus LiCl or NaCl solution. The individual taste reactivity responses during the intraoral infusions were videotaped and later analyzed for response frequency. Forty-eight hours after the second conditioning day the same sucrose plus salt solution was presented again in the absence of any injection. The present results demonstrate that systemic injections of LiCl result in profound within session and across session decreases in ingestive responding accompanied by increased active and passive aversive responses. Animals receiving LiCl by injection as well as ingestion demonstrated an exaggerated response. Rats which received LiCl only through intraoral infusions produced the same pattern of decreased ingestive responding to the sucrose plus salt flavored tastant on each test trial suggesting little or no across session conditioning effects. A two process model by which animals may regulate toxicosis is discussed. © 2004 Elsevier B.V. All rights reserved. Keywords: Ingestive behaviors; Conditioned taste aversion; Lithium toxicosis; Gut defence; Toxin avoidance; Rapid gustatory conditioning 1. Introduction Most animals have to deal with the problem of obtain- ing foods that contain appropriate nutrients while at the same time minimizing the ingestion of toxic compounds that may be contained in these foods. Many plants may defend themselves by producing secondary metabolites [1] which are toxic to potential consumers of these plants, es- pecially if these are generalist feeders [3,7]. For example, sheep have been shown to form conditioned avoidances of novel flavors paired with oxalic acid (a natural toxin found in many plants) when tested in two-choice preference tests [16]. However, animals also need to balance the cost of the food, in terms of toxin level, with the benefits of the nutrients obtained and clearly this cost/benefit ratio will be influenced by the state of the animal, such as the hunger level of the consumer. The phenomenon of learned avoidance of toxic foods has received a great deal of attention following the classic stud- ∗ Corresponding author. Tel.: +1-5196-612111x81214; fax: +1-5196-6139-61. E-mail address: skcross@uwo.ca (S. Cross-Mellor). ies of Garcia et al. [9] on experimental bait-shyness in rats (cf. [8]). Conditioned taste avoidances have been produced in the laboratory under a wide variety of conditions with a variety of animal species and a very large range of toxic sub- stances (see [4,29]). Typically, the animals are allowed to ingest a food item with a novel flavor followed by systemic treatment with the toxin of interest. Conditioned avoidances are demonstrated either by subsequent reductions in inges- tion of the same food relative to placebo treated subjects, or by relative avoidance of this food relative to another, less palatable food (two-choice preference test). The most often used experimental animals are rats and the most commonly used toxin is lithium chloride (LiCl). With an appropriate dose of LiCl this type of conditioning produces very robust effects, often with a single conditioning trial, which can last for a substantial period of time [20]. Much less attention has been paid to the production of learned food avoidances in the laboratory when the toxin is ingested as part of the food and when relatively small amounts of the toxin are used for the conditioning. The re- search which has been done with this approach has largely been done on ruminants in an attempt to understand how plant toxins can regulate grazing in cattle, sheep and goats 0166-4328/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.bbr.2004.03.009