Footshock-induced sensitization of the acoustic startle response in two strains of mice.

It has been shown before that unconditioned footshocks can augment the acoustic startle response in rats. In the present study, male mice of two strains, C57Bl/6N and BALB/c, were compared with regard to footshock-induced sensitization of the acoustic startle response. Presentation of footshocks did not affect the acoustic startle response in C57Bl/6N mice, while in contrast, footshock-induced sensitization was apparent in the BALB/c strain. Shocked C57Bl/6N mice, but not BALB/c mice, displayed robust conditioning to the startle context when re-tested the next day. These findings indicate that mice may exhibit footshock-induced sensitization of the acoustic startle response, but that the effects of footshocks on the acoustic startle are strain- and time-dependent.

Absence of 5-HT(1B) receptors is associated with impaired impulse control in male 5-HT(1B) knockout mice.

BACKGROUND: Serotonin (5-HT) plays a complex regulatory role in processes like anxiety, depression, aggression, and impulse control. Due to the large amount of serotonergic receptors, knockout mice offer an important opportunity to investigate the role of specific receptors. The 5-HT(1B) receptor is thought to mediate aggression and impulse control. This was studied here in mice lacking 5-HT(1B) receptors (5-HT(1B) KO). METHODS: Wild type and 5-HT(1B) KO mice were exposed to several types of entrained and nonentrained stimuli. With telemetry, body temperature, heart rate, and locomotor activity were measured continuously during the different experiments. RESULTS: To nonentrained stimuli like disturbance stress and confrontation with an intruder, 5-HT(1B) KO mice showed exaggerated physiologic and behavioral responses. These mice displayed behavioral disinhibition, measured as increased social interest and aggression to an intruder mouse. However, in response to well-entrained stimuli like daily light transitions, responses were smaller in 5-HT(1B) KO than in wild type mice, suggesting that hyperreactivity is stimulus specific. CONCLUSIONS: Serotonin 1B receptors are essential in impulse control by inhibiting responses to nonentrained stimuli. Therefore, the 5-HT(1B) KO mouse might be an important additional model for studying aspects of disinhibition in aggression and impulse control.

Stress-induced hyperthermia in the 5-HT(1A) receptor knockout mouse is normal.

BACKGROUND: Several studies on serotonin 1A (5-HT(1A)) receptor knockout mice in different genetic backgrounds indicate that such mice display a more anxious phenotype than their corresponding wild types. We hypothesized that the 5-HT(1A) receptor knockout mice would show a different phenotype than the wild type mice in the stress-induced hyperthermia (SIH) paradigm, which tests putative anxiolytic effects of drugs. Moreover, on pharmacologic challenges with the 5-HT(1A) receptor agonist flesinoxan we expected an absence of the functional response in knockout mice relative to wild type mice. METHODS: Effects of the 5-HT(1A) receptor agonist flesinoxan, alone or in combination with the 5-HT(1A) receptor antagonist WAY-100635, and the gamma-aminobutyric acid A (GABA(A))-benzodiazepine receptor agonist diazepam were studied in the SIH paradigm in male 129/Sv 5-HT(1A) receptor knockout and wild type mice. In addition, the effects of flesinoxan on plasma corticosterone concentrations were determined. RESULTS: Plasma corticosterone concentrations were dose dependently elevated by flesinoxan in wild type mice but not in knockout mice. Flesinoxan dose dependently decreased SIH in wild type mice but not in knockout mice. The flesinoxan effect in wild type mice was blocked by WAY-100635. Furthermore, diazepam decreased SIH in both genotypes. There were no differences in basic SIH responses between wild type and knockout mice. CONCLUSIONS: 5 -HT(1A) receptor knockout mice display a normal SIH response, and results indicate, based on the SIH, that the GABA(A)-benzodiazepine receptor complex functions normally.

Corticosterone responses in 5-HT1B receptor knockout mice to stress or 5-HT1A receptor activation are normal.

RATIONALE: Previous research found no adaptations in presynaptic 5-HT1A receptors in mice lacking 5-HT1B receptors (5-HT1B KO). Stress and 5-HT1A receptor agonists induce corticosterone release in mice via hypothalamus-pituitary-adrenal (HPA) axis activation. 5-HT1B KO mice are hyperreactive to mild stressors and this might be reflected in altered postsynaptic 5-HT1A receptor sensitivity. OBJECTIVES: Our aim was to determine whether the activity of the HPA axis was increased in 5-HT1B KO mice in response to mild stress and pharmacological activation of 5-HT1A receptors as an indication of putative adaptive changes in postsynaptic 5-HT1A receptor function. METHODS: The effect of mild stress [i.e., the stress-induced hyperthermia (SIH) paradigm], induced by rectal temperature measurement, was determined on temperature and corticosterone over time (0, 5, 10, 20, 30, 60, and 90 min) in 5-HT1B KO and wildtype mice. In addition, corticosterone was measured 60 min after 5-HT1A receptor activation by flesinoxan (0, 0.03, 0.1, 0.3, 1, and 3 mg/kg s.c.). Blood was collected and plasma corticosterone levels were determined by radioimmunoassay. RESULTS: Both genotypes showed comparable time-dependent SIH responses, whereas basal temperature was higher in 5-HT1B KO mice. The effect of SIH on temperature was mirrored by mild increases in plasma corticosterone. Activation of 5-HT1A receptors caused a strong dose-dependent release of corticosterone in both genotypes. Neither response observed showed differences between both genotypes. CONCLUSIONS: Although 5-HT1B KO mice are hyperreactive to mild stress, this reactivity is not reflected by stronger corticosterone responses in the SIH paradigm. The lack of shift in dose-response curves for flesinoxan suggests that postsynaptic 5-HT1A receptor function is unaffected in 5-HT1B KO mice.

5-HT1B receptor knockout, but not 5-HT1A receptor knockout mice, show reduced startle reactivity and footshock-induced sensitization, as measured with the acoustic startle response.

To investigate whether the hyperreactivity to mild environmental and novel stimuli in 5-HT1B receptor knockout (1BKO) mice, as suggested by measures of exploratory, aggressive, and impulsive behaviors, can be extended to phasic stimuli, 1BKO and wildtype mice were tested in acoustic startle reactivity and plasticity paradigms, including habituation, prepulse inhibition, and footshock-induced sensitization of the startle response. Furthermore, we compared 5-HT1A receptor knockout (1AKO) and 1BKO mice to further test the suggested opposite behavioral profiles in these two genotypes. Results show that startle reactivity and footshock-induced sensitization was reduced in 1BKO mice, with no changes in habituation or PPI. In contrast, 1AKO mice did not differ from WT mice in any of the measures. These results indicate that an absence of 5-HT1B receptors, but not of 5-HT1A receptors, affects the modulation of startle reactivity and footshock-induced sensitization, without influencing startle plasticity. Moreover, this study suggests that 1AKO mice display a distinct, but not opposite behavioral profile from 1BKO mice. Furthermore, it is concluded that the hyperreactivity in 1BKO mice cannot be generalized to all stimuli, including the startling stimuli used in this study, but is probably restricted to mild environmental stimuli only.

Male and female 5-HT(1B) receptor knockout mice have higher body weights than wildtypes.

5-HT(1B) receptors have a regulatory role in serotonergic activity and influence feeding behavior and body weight. Because the absence of 5-HT(1B) receptors may cause changes in this regulation, body weight was measured in male and female 5-HT(1B) receptor knockout (5-HT(1B) KO) and wildtype (WT) mice from weaning until the age of 30 weeks. In both genders, 5-HT(1B) KO mice had a higher body weight than WT mice (17% and 9%, respectively). Body weight was significantly higher for males over the entire period and for females from Week 18 onwards. Absolute food and water consumption were related to body weight. However, relative to body weight, males consumed more than females. 5-HT(1B) KO males drank strikingly more water. Housing mice singly reduced food and water intake in males, but not in females. Plasma leptin levels and most organ weights did not differ between genotypes, indicating that higher body weight in 5-HT(1B) KO mice is not related to obesity. Relative to body weight, brains and adrenals were larger in females, while heart and liver were smaller. Kidneys were smaller in females, but larger in 5-HT(1B) KO mice, while lungs showed opposite effects. Spleen and testes were smaller in 5-HT(1B) KO mice. Although 5-HT(1B) KO males are more aggressive, testosterone levels were not different from WT mice. Basal corticosterone levels were similar in all groups and increased in response to mild stress, particularly in females. Lifelong absence of 5-HT(1B) receptors in mice resulted in clear phenotypic differences in body weights and food and water intake. Lacking this receptor increases body growth, without signs of obesity. A potential genetic background effect influencing this phenotype is discussed.

Ethanol intake is not elevated in male 5-HT(1B) receptor knockout mice.

Recently, the phenotype of increased ethanol intake in mice lacking 5-HT(1B) receptors could not be replicated. We assessed ethanol consumption in male wildtype and 5-HT(1B) receptor knockout mice derived from the original population. Intake of water and ethanol (0%, 3%, 6%, 10% and 20% v/v) from two pipettes was determined daily for 40 days. Ethanol intake (g/kg body weight) did not differ between genotypes, while body weights (20-25%) and water intake (50%) were elevated in 5-HT(1B) receptor knockout mice. Hence, the initial finding of elevated ethanol intake in 5-HT(1B) receptor knockout mice may have been due to phenotypic differences in fluid intake.

Infusion of flesinoxan into the amygdala blocks the fear-potentiated startle.

In a previous study it was demonstrated that flesinoxan, a selective serotonin (5-HT)1A receptor agonist, had anxiolytic properties in the fear-potentiated startle paradigm. The present study investigated the putative site of action of flesinoxan in this paradigm. Flesinoxan infused either into the dorsal raphe nucleus or the median raphe nucleus did not affect startle potentiation. Bilateral infusion of flesinoxan into the central nucleus of the amygdala on the other hand, dose-dependently blocked the fear-potentiated startle response. These data indicate that flesinoxan exerts it anxiolytic effects in the fear-potentiated startle paradigm via the central nucleus of the amygdala, whereas the dorsal and median raphe nuclei are not directly involved in this process.

Stress-induced hyperthermia in mice: effects of flesinoxan on heart rate and body temperature.

Stress-induced hyperthermia in mice has predictive validity for anxiolytic properties of drugs. In this paradigm, 60 min after drug administration rectal temperature is measured, which causes hyperthermia of 1-1.5 degrees C (DeltaT) in about 10 min. Flesinoxan, a selective 5-HT(1A) receptor agonist with anxiolytic-like properties, causes hypothermia, which complicates interpretation of stress-induced hyperthermia. Therefore, we combined flesinoxan treatment and the stress paradigm with radiotelemetric measurement of body temperature and heart rate, which is also related to anxiety. Subjects were either undisturbed or injected with flesinoxan (0-0.1-0.3-1.0 and 3.0 mg/kg), with or without the stress paradigm. Flesinoxan (1.0 and 3.0 mg/kg) caused a relatively long-lasting hypothermia, but did not lower heart rate. The rectal temperature procedure caused hyperthermia and tachycardia. Flesinoxan reduced the stress-induced hyperthermia and the tachycardia evoked by the stress procedure. Continuous radiotelemetric measurement of heart rate, apart from body temperature, revealed that flesinoxan has anxiolytic-like properties in mice.

Discriminative stimulus properties of alprazolam.

RATIONALE: The triazolobenzodiazepine alprazolam has a unique clinical profile compared to most other benzodiazepines (e.g. diazepam, chlordiazepoxide), in that it is used to treat panic disorder and is effective in depression, two disorders that are usually treated with anti-depressants. Previous drug discrimination studies suggested that alprazolam has stimulus properties in common with antidepressants. OBJECTIVE: In the present study, the discriminative stimulus properties of alprazolam were investigated to test more conclusively the role of benzodiazepine receptors and whether alprazolam has stimulus properties in common with antidepressants. METHODS: Male Wistar rats (n=12) were trained to discriminate between alprazolam (2.0 mg/kg, PO) and vehicle in an operant two-lever drug discrimination procedure under a tandem VI40"-FR10 schedule of reinforcement. Generalization and antagonism tests were carried out under 2 min extinction. RESULTS: In generalization tests, a number of benzodiazepines (alprazolam, chlordiazepoxide, midazolam, lorazepam) and the barbiturate pentobarbital substituted completely, while zolpidem and abecarnil substituted partially for alprazolam. In contrast, no significant degree of generalization to the antidepressants imipramine and fluvoxamine and the putative antidepressants buspirone and flesinoxan was found. In antagonism studies alprazolam could be antagonized (almost) completely by flumazenil, partially by pentylenetetrazole, but not by methyl 6, 7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate (DMCM), N-methyl-beta-carboline-3-carboxamide (FG-7142) and picrotoxin. CONCLUSIONS: These results show that the discriminative stimulus properties of alprazolam are mediated by benzodiazepine receptors and that the finding that antidepressants share discriminative stimulus effects with alprazolam may have limited generality.

Effects of drug discrimination history on anti-punishment properties of chlordiazepoxide in rats.

Several studies have indicated that acquiring discriminative stimulus control for a certain anxiolytic drug influences its subsequent anti-conflict properties. To further elaborate on the question whether drug discrimination procedures affect behaviour in a conflict paradigm, a classical two-lever drug discrimination procedure was combined with an operant conflict procedure within the same animals. To this extent, rats were trained to discriminate the anxiolytic chlordiazepoxide (CDP, 30 mg/kg, po) from saline (SAL), and subsequently punished responding periods were introduced within the same session. In addition to the rats that were trained to discriminate CDP from vehicle, a group of rats was trained on a random relationship between CDP and the rewarded lever. CDP and alprazolam completely substituted for CDP, whereas mianserin did not. Responding during punished components in a session was increased by CDP and alprazolam, but not by mianserin in rats that were trained to discriminate CDP from vehicle and in randomly trained rats. The data indicate that rats can be reliably trained and tested in drug discrimination and conflict procedures within a single session and that CDP's discriminative stimulus does not alter its anti-conflict effects.

Startle responses, heart rate, and temperature in 5-HT1B receptor knockout mice.

Relative to wildtype mice, mice lacking 5-HT1B receptors (5-HT1B KO) exhibit exaggerated heart rate and body temperature responses to environmental stimuli. In contrast, acoustic startle reactivity is reduced in 5-HT1B KO mice. We combined heart rate and temperature measurement with startle response paradigms in order to elucidate this apparent contradiction. Habituation and footshock-induced sensitization paradigms modulate startle reactivity. Reduced startle reactivity and unaltered habituation in 5-HT1B KO mice were replicated. Heart rate and temperature were unaffected by startle stimuli, but increased markedly in response to transportation and handling procedures. Footshocks caused a mild startle-sensitization and tachycardia in both genotypes. The physiological hyper-reactivity in 5-HT1B KO mice is a subtle phenotypic difference that contrasts with the phenotypic decrease in startle reactivity.

Stimulus properties of the selective 5-HT reuptake inhibitor fluvoxamine in conditioned taste aversion procedures.

Previous attempts to train pigeons and rats to discriminate between the antidepressant fluvoxamine and its vehicle as assessed in a drug discrimination paradigm have been without success. The present experiments were, therefore, designed to assess in a conditioned taste aversion procedure (CTA) whether or not fluvoxamine possesses stimulus properties. Rats were exposed to a conditioned taste aversion (CTA) procedure. In Experiment I, subjects were given 15 mg/kg fluvoxamine p.o. or vehicle after drinking a novel tasting saccharin solution. In Experiment II, a comparison was made between the effects of 15 mg/kg fluvoxamine i.p., 30 mg/kg fluvoxamine i.p., NaCl, and lithium chloride (LiCl). In Experiment III, subjects were treated with either 10 mg/kg fluoxetine i.p., 30 mg/kg fluvoxamine i.p., or LiCl. CTA was observed after treatment with LiCl, but never after treatment with fluvoxamine or fluoxetine, suggesting that fluvoxamine does not have clear stimulus properties, which can serve as a discriminative stimulus in operant procedures. In a crossfamiliarization CTA procedure in mice, however, fluvoxamine elicited a reliable CTA, suggesting that under certain conditions (species, dose?) selective serotonin reuptake inhibitors (SSRIs) may lead to certain discriminable effects. It is as yet unclear why SSRIs apparently produce such weak and species or situation-dependent discriminable effects.

Discriminative stimulus properties of mCPP and alprazolam are not mediated by anxiety.

We investigated whether the interoceptive cues mediated by the anxiolytic benzodiazepine receptor agonist alprazolam and the anxiogenic serotonin (5-HT)(1B/2C) receptor agonist 1-(3-chlorophenyl)piperazine (mCPP) in rats are related to anxiety. mCPP-induced anxiety in humans can be blocked with alprazolam, and if mCPP drug discrimination is to be used as a model of anxiety, mCPP's stimulus should be blocked by alprazolam. Therefore, two groups of rats were trained to discriminate either alprazolam (2 mg/kg, p.o.) or mCPP (2 mg/kg, p.o.) from vehicle in a two-level operant drug discrimination procedure. Cross antagonism tests were performed with alprazolam and mCPP. mCPP did not antagonize alprazolam's stimulus to any extent, but disrupted responding severely. Low and intermediate doses of alprazolam (1.0-4.0 mg/kg, p.o.) did not antagonize the mCPP discriminative stimulus; only a high dose of 8.0 mg/kg (p.o.) partially antagonized mCPP but disrupted responding in most of the animals. We conclude that, at best, there is only weak evidence to suggest that the interoceptive cues of alprazolam and mCPP are mediated by modulation of anxiety processes, and that the mCPP drug discrimination as a model for anxiety is unreliable.

The anxiolytic effect on the fear-potentiated startle is not due to a non-specific disruption.

In the present study, the effects of alprazolam, chlordiazepoxide, ethanol and haloperidol in the strychnine-potentiated startle response paradigm were investigated. Because strychnine increases control startle levels without fear-conditioning, no central state of fear exists. When anxiolytic drugs reduce the fear-induced potentiation of the startle response without reducing the strychnine-induced startle potentiation, their attenuating effect on startle potentiation in the fear-potentiated startle response paradigm can more likely be attributed to their anxiolytic properties. The selected drugs, which effectively reduced the fear-induced startle potentiation in an earlier study, indeed did not selectively affect the strychnine-induced startle potentiation, supporting the notion that the dose-dependent reduction of startle potentiation due to anxiolytic drugs in the fear-potentiated startle response paradigm does not reflect a "non-specific" disruption of startle behavior.

The effects of 5-HT1A receptor agonists, 5-HT1A receptor antagonists and their interaction on the fear-potentiated startle response.

The present study investigated whether the anxiolytic effect of 5-HT1A receptor agonists on the fear-potentiated startle response could be antagonized by 5-HT1A receptor antagonists. Therefore, control and fear-potentiated startle amplitudes were measured after co-administration of vehicle, flesinoxan (10 mg/kg PO) or 8-OH-DPAT (0.3 mg/kg SC) and DU 125,530 (0, 1, 3 and 10 mg/kg SC), (+/-)-pindolol (0, 3, 10 and 30 mg/kg SC) or WAY 100,635 (0, 0.1, 0.3 and 1 mg/kg SC). Unexpectedly, the three antagonists themselves as measured in the vehicle-pretreatment groups dose-dependently decreased startle potentiation. Further, DU 125,530 and WAY 100,635 were able to reverse the attenuating effect of 8-OH-DPAT, while no antagonism of the flesinoxan effect on startle potentiation was found. In contrast, both the flesinoxan- and 8-OH-DPAT-induced lower lip retraction were antagonized by DU 125,530 and WAY 100,635, but not by (+/-)-pindolol. The findings of the present study suggest that drugs acting on 5-HT1A receptors differentially affect lower lip retraction and startle potentiation probably mediated by different neuronal populations.

Discriminative stimulus properties of mCPP: evidence for a 5-HT2C receptor mode of action.

Previous drug discrimination studies with the serotonergic drug m-chlorophenylpiperazine (mCPP) showed conflicting results, with some authors concluding that the cue was mediated by 5-HT2C receptors, but others that it was definitively not. We further examined the discriminative stimulus properties of mCPP in rats and reviewed previously published data. We trained rats to discriminate mCPP (2.0 mg/kg, PO) from water. We found that the mCPP cue generalized to m-trifluoromethyl-phenylpiperazine (TFMPP) and 6-chloro-2-(1-piperazinyl)-pyrazine (MK-212), and partially to eltoprazine, 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (DOI), fenfluramine and trazodone. A moderate level of generalization was obtained with quipazine, 1-(m-chlorophenyl)biguanide and clonidine. No generalization was found with flesinoxan, methiothepin, idazoxan and haloperidol. Mianserin and methysergide antagonized the mCPP stimulus, whereas ketanserin antagonized it partially. Metergoline, methiothepin and clozapine only marginally antagonized the mCPP stimulus. These results show that the discriminative stimulus effects of mCPP are predominantly mediated by 5-HT2C receptors, and to some extent by 5-HT1B receptors. When considering our results and other research together, the substitution tests clearly point to a 5-HT2C receptor mediated stimulus, with an additional role for 5-HT1B receptors. Antagonism studies are less clearcut, but are also suggestive of a 5-HT2C receptor mediated effect. A definitive answer as to whether other receptors, e.g. 5-HT2B and 5-HT7, are of any importance in mCPP's discriminative stimulus properties has to wait for more selective ligands.

Stimulus properties of fluvoxamine in a conditioned taste aversion procedure.

A conditioned taste aversion (CTA) procedure in mice was used to investigate the stimulus effects of the serotonin reuptake inhibitors (SSRIs) fluvoxamine and fluoxetine. Fluvoxamine elicited a reliable CTA (ED50 = 24 mg/kg, SC) and a number of drugs were tested as pre-exposure drugs. Pre-exposure to the serotonin (5-HT)1A receptor agonists flesinoxan and +/- -8-hydroxy-dipropylaminotetralin (8-OH-DPAT) prevented the CTA induced by fluvoxamine (50 mg/kg, SC). Pre-exposure with the 5-HT2C receptor agonist MK 212 [6-chloro-2(1-piperazinyl)pyrazine] partially prevented the fluvoxamine-induced CTA, pre-exposure with the 5-HT2A/2C receptor agonist DOI [1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane HCl] did not prevent the CTA induced by fluvoxamine. Flesinoxan pre-exposure also prevented the taste aversion induced by fluoxetine (10 mg/kg, SC) completely. This contrasts previous results obtained with fluoxetine, where was found that its stimulus is primarily mediated by 5-HT2C, and to a lesser degree by 5-HT1A receptors. Therefore, we compared the two SSRIs directly. Pre-exposure to fluvoxamine prevented the fluoxetine-induced CTA, whereas pre-exposure to fluoxetine only partially prevented the fluvoxamine-induced CTA. We conclude that 5-HT1A receptors are involved in the stimulus properties of both fluvoxamine and fluoxetine, that 5-HT2C receptors are involved in fluvoxamine and especially fluoxetine, and, based primarily on the cross-comparison tests, that the two SSRIs have somewhat different stimulus properties.

Effects of 5-HT1A receptor ligands in a modified Geller-Seifter conflict model in the rat.

In a modified Geller-Seifter conflict procedure, rats were trained to lever-press for food under a multiple variable interval-fixed ratio (VI30: food; FR10: food + shock) schedule of reinforcement. The ability to antagonize response suppression in the punished period is considered a good predictor for anxiolytic activity. Chlordiazepoxide and alprazolam increased punished responding. The 5-HT1A receptor agonists flesinoxan (R(+)-N-[2[4-(2,3-dihydro-2-2-hydroxymethyl-1,4-benzodioxin-5-yl)- 1-piperazinyl]ethyl]-4-fluorobenzoamide; 0.1-10.0 mg/kg) and 8-OH-DPAT (8-hydroxy-2-(di-n-propyl-amino)tetralin; 0.03-0.5 mg/kg) significantly increased punished responding, supporting a role of the 5-HT1A receptor in anxiety. 8-OH-DPAT and flesinoxan also reduced unpunished responding. The anxiolytic effects of 8-OH-DPAT and flesinoxan could only be antagonized with a high dose (1.0 and 3.0 mg/kg respectively) of the 5-HT1A receptor antagonist WAY-100635 (N-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-N-(2-pyridinyl) cyclohexanecarboxamide trihydrochloride). All doses of WAY-100635 antagonized the 5-HT1A-induced effects on unpunished responding. The dissimilarity in dose-response curve of WAY-100635 on punished and unpunished behaviour poses questions about the mediation of these effects.

The delayed-conditional-discrimination task improves measurement of working memory in rats.

The validity of delayed-matching-to-sample (DMTS) and related tasks executed in skinner boxes as an animal model for human working memory (WM) is confounded by the occurrence of mediating behaviour during delays. True matching, a supplementary task during delays and response similarity are ways to deal with this problem. However, until now rats have not been able to learn a true matching task in Skinner boxes and introduction of a supplementary task during delays does not sufficiently prevent mediating behaviour. Response similarity, on the other hand, effectively prevents the use of mediating behaviour by reducing the discriminative value of the behaviour during delays. Furthermore, it is argued that the interpretation of drug effects is confounded by baseline performance and mediating behaviour. It is shown that high baseline levels and high amounts of mediating behaviour can induce delay dependent drug effects, suggesting a specific effect on WM. We therefore assert that examination of delay-dependency of a drug effect alone is not sufficient to claim specific effects of a drug on WM. The delayed-conditional-discrimination (DCD) task uses response similarity to effectively reduce mediating behavior and does not generate high levels of baseline performance. The DCD task is therefore preferred over other tasks for the measurement of WM in rats using Skinner boxes.