Behavioural Brain Research 180 (2007) 241–245 Short communication Enhancing effects of lithium on memory are not by-products of learning or attentional deficits Eleftheria Tsaltas a, , Theodora Kyriazi a,1 , Cornelia Poulopoulou b,2 , Dimitrios Kontis c,d,1 , Antonios Maillis e,3 a Experimental Psychology Laboratory, Department of Psychiatry, Athens University Medical School, Eginition Hospital, 74, Vas. Sofias Ave., 11528 Athens, Greece b Experimental Neurophysiology Laboratory, Department of Neurology, Athens University Medical School, Eginition Hospital, 74, Vas. Sofias Ave., 11528 Athens, Greece c Psychiatric Hospital of Attica, 374, Kavalas Ave., 12462, Athens, Greece d Department of Psychiatry, Athens University Medical School, Eginition Hospital, 74, Vas. Sofias Ave., 11528 Athens, Greece e Experimental Neurophysiology and Psychopharmacology Laboratory, Department of Psychiatry, Athens University Medical School, Eginition Hospital, 74, Vas. Sofias Ave., 11528 Athens, Greece Received 27 January 2007; received in revised form 4 March 2007; accepted 13 March 2007 Available online 16 March 2007 Abstract We recently reported that chronic lithium (LiCl), at therapeutic plasma levels, enhanced spatial working memory and retention of an aversive contingency. Here we examine the possibility that these effects be secondary to LiCl effects on the ability to ignore irrelevant stimuli or on fear conditioning. In Experiment 1, rats subjected to >30 daily intraperitoneal injections of LiCl (2 mmol/kg) or saline underwent conditioned emotional response training (CER: 2 CS pairings with 1-s, 1-mA shock) after 40 pre-exposures either to the CS (latent inhibition-LiCl/latent inhibition-saline, n = 8) or to another stimulus (control-LiCl/control-saline, n = 8). In Experiment 2, eight LiCl and eight saline animals were trained in on-the-baseline (VI-60 s) CER (1-s, 0.15-mA shock in CS-signalled periods) in the Skinner box. In Experiment 1, LiCl animals showed normal latent inhibition. In both experiments, their fear conditioning was unimpaired. Therefore, the previously reported memory improvement under chronic lithium cannot be attributed to changes in the ability to ignore irrelevant stimuli or in fear conditioning. © 2007 Elsevier B.V. All rights reserved. Keywords: Lithium; Memory; Latent inhibition; Conditioned emotional response 1. Introduction Early views linking lithium treatment to memory deficits in humans [1] have been challenged by recent neuropsycho- logical studies. Visual memory appears unaffected by lithium [2–5]. Though verbal memory seems somewhat more sensitive [2,4–12], this emerges as a weak trend needing confirmation [for review: 13]. Older animal studies also report lithium-induced learning and memory deficits [14–16], which they attribute to Corresponding author. Tel.: +30 210 7289114; fax: +30 210 7242020. E-mail addresses: tsaltasl@med.uoa.gr (E. Tsaltas), laurakyriazi@hotmail.com (T. Kyriazi), cpoulop@med.uoa.gr (C. Poulopoulou), jimcon@hol.gr (D. Kontis), amaillis@otenet.gr (A. Maillis). 1 Tel.: +30 210 7289114. 2 Tel.: +30 210 7289219. 3 Tel.: +30 210 7289121. narrowing of attentional filtering [17,18]. However, recent stud- ies examining memory effects of chronic lithium (at plasma levels within the human therapeutic range), report transient [19] or no deficit [20] in spatial reference memory and unaffected spatial working memory. In fact, the latter study presents data suggestive of working memory enhancement [20: Fig. 3]. The controversy is compounded by a recent study [21] in which we reported that chronic lithium, at therapeutic range plasma levels, had no effect on reference memory (acquisition of an alternation rule in the T-maze: [22,23]). However, it had a robust benefi- cial effect on spatial working memory, provided that moderate demand was placed on that function (i.e. use of recall delays sufficient to reduce performance accuracy from >90 to <85% but not <78%). In that study, lithium also facilitated long-term retention of passive avoidance at conditioning parameters which did not support retention in controls (i.e. single conditioning trial with mild shock). 0166-4328/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.bbr.2007.03.017