Bourin M (2024) Test-Retest a Way to Raise Awareness of Anxiety Tests in Mice. J Vet Med Res 11(2): 1269.

The study of human pathologies, as well as the development of new molecules for therapeutic purposes, requires the implementation of different stages before the first studies on the patient. Among these successive levels of validation, animal models occupy a predominant place since they are the first step using a complex biological system. In this context, researchers have attempted to set up animal models for different pathologies, in order to obtain information that can be extrapolated to humans. Thus, these models are systems for studying the neuro- physiopathological mechanisms involved, but also make it possible to predict the potential activity of new molecules [1]. This design of the models is accompanied by numerous precautions, to ensure that the information collected through an animal study can be used as a first indicator of work in humans [2].

The level of complexity of animal models depends above all on the therapeutic target sought. Thus, in the context of somatic pathologies, the use of animal models can be validated by physiological indices known and common to the majority of different mammals.

On the other hand, in the context of pathologies linked to the central nervous system, and even more so when they are linked to psychiatry, animal models become more difficult to validate, and the results to be used in the patient by anthropomorphizing. Indeed, the patient is likely to be questioned in order to diagnose the disease, but also to understand its evolution depending on the therapies used. Conversely, modeling this type of pathology in animals is limited to perceiving, using behavioral cues, a response to a treatment, within an extremely well-defined and validated framework.

Selected animal models of anxiety

When setting up a model of a psychiatric pathology, the goal is to be able, through a simple and reproducible system, to study a fragment of a multifactorial and particularly complex disease. To do this, it is necessary to carefully choose the animal species used, according to its behavioral and pharmacological responses to a given situation, to set up a protocol, then to validate it rigorously. The validity of an animal model is most often linked to three factors: predictive validity, construct validity and reliability [3].

As part of the study of anxiety, different models have been implemented, including the elevated plus-maze test (EPM), the four plates test, conflict tests, etc. A more complete review of the validation and problems encountered in the construction of animal models of anxiety [4]. In my research laboratory, we primarily use two animal models of anxiety: the elevated plus- maze (EPM), and the four-plate test (FPT). Using these two models, we were able to carry out research on the mechanisms of action of molecules involved in the treatment of anxiety, but also on interactions between neurotransmission systems of the central nervous system [5]. Furthermore, we were able to demonstrate that these two models demonstrate similar behavioral responses, in comparison with another animal model of anxiety: the double illuminated enclosure test. Thus, an injection of DOI or BW 723C86, two 5-HT2 receptor agonists, triggered an anxiolytic-type response in the EPM and in the FPT, 30 minutes after the injection. This result was not found in the double illuminated enclosure test [6].

Elevated plus maze (EPM): One of the most widely used behavioral tests for anxiety research and frequently used in mice is the EPM [7], a model initially developed for rats [8,9]. The EPM has been widely used as a tool to study the psychological and neurochemical bases of anxiety, for the search for anxiety- modulating substances [10,11,12].

This test consists of a square platform whose edges are made up of four arms of identical width and length. Two arms are called “open” and consist only of a board. Two “closed” arms are made up of a board and walls around all their edges. Arms of the same type are placed in opposition on the platform. The entire cross is located 26 cm from the ground, under a lamp placed vertically on the platform. This test is based on the rodent’s natural aversion to new, open, illuminated and high spaces. It uses the conflict between the animal’s desire to explore and its aversion to high open spaces.

Mice generally coming from their maintenance cage show behavior characterized by avoidance of “open” arms with a clear preference for “closed” arms. Animals typically spent more time in the “closed” arms, then on the central platform, and little time in the “open” arms, indicating a bias toward the relatively secure sections of the maze. This behavior is suppressed by anxiolytics and potentiated by anxiogenic substances. Lister et al. [13], showed that behavioral parameters of the EPM in mice provide data on two independent factors, one reflecting anxiety and the other motor activity. The percentage of entries and the time spent on the “open” arms represent good parameters for measuring anxiety in this model [14]. On the other hand, the indices linked to locomotor activity diverge between the different teams. In my research laboratory, the verification of the activity of a molecule on locomotion is carried out using an actimetry test, in order to dedicate the EPM to its use alone in the study of anxiety.

The behavioral profile induced by EPM appears to include neophobic and exploratory factors, a conflict between exploration and avoidance of the “open” arms that represent the stressor. This model is therefore often classified as an unconditioned and spontaneous behavioral model of conflict [15].

Four-plate test (FPT): In the laboratory, in recent years we have particularly focused our work on the study of anxiety on a particular model, the four- plate test (FPT).

This model was first described by Slotnick and Jarvick [16], then modified by Boissier and his colleagues as a model for studying the effects of anxiolytics by suppressing innate exploratory behavior in mice using punishments [17]. This model is a simple and rapid test which allows the detection of the anxiolytic or anxiogenic effects of the molecules studied. It has been shown to respond to the two treatments of choice in humans, BZDs and 5-HT reuptake inhibitors [18,19]. The administration of diazepam at a dose of 1 mg/kg and alprazolam at a dose of 0.25 mg/kg leads to a significant increase in the number of punished passages accepted by the mice compared to control mice [18]. The existence of false positives such as caffeine or cocaine makes it necessary to use non-psychostimulant doses 

in FPT. To do this, we test the products in an activity meter before each study [20]. Along the same lines, sedative doses of treatment should be avoided, since the model is based on the exploration of a new environment. Thus, a molecule exerting “anxiolytic” type properties will result, in comparison with “control” mice, in an increase in the number of punished passages accepted by the mice during the test minute. The FPT has remained rarely used in other laboratories or only to detect the anxiolytic effect of new drugs [21,22,23].

During our research activity, we decided to frequently use FPT to study the mechanisms of action of anxiolytic-type molecules. Two reasons motivated us for this. The first one is that we validated this model, in comparison with the reference model that constitutes the EPM. As we described previously, even though one of the models uses aversive stimuli, the type of anxiety observed in these two tests seems to have strong similarities. Thus, the anxiolytic type response to an injection of 5-HT2 receptor agonist (DOI or BW 723C86) observed in the tests is similar between the EPM and the FPT, but different in the double illuminated enclosure test, another popular animal model of anxiety [24]. Furthermore, in these two models, we were able to demonstrate a facilitating effect of 5-HT2A ligands on the GABAergic system. Non-active doses of alprazolam and diazepam produced a significant effect when co-administered with DOI, a serotonergic 5-HT2A/2C receptor agonist [25]. In FPT, acute administration of DOI, a 5-HT2A/2C receptor agonist, results in a large and significant increase in the number of punished passages accepted by mice during testing. The administration of BW 723C86, a 5-HT2B receptor agonist, under the same conditions, also leads to a significant increase in the number of punished passages but of lesser importance. On the other hand, RO 60-0175, a 5-HT2C receptor agonist, and mCPP, a non-specific 5-HT2 receptor agonist, do not modify the number of punished passages in comparison with non-injected mice [6].

Finally, we freed ourselves from the various criticisms against the model, which did not consider the FPT as a test of anxiety but rather of aversion. We have in fact demonstrated that the anxiolytic type effects of the different molecules are expressed at non-analgesic doses, and that morphine, used at analgesic doses, does not cause variations in the number of punished passages accepted by the mice during of the four-plate test [26].

Furthermore, this has the advantage compared to the EPM of being based on a single simple index: the number of punished passages accepted by a mouse. Indeed, the multiplication of behavioral indices measured in the EPM constitutes a source of inter-study variations as well as loss of essential information, diluted in several components [15]. Based on the exploration of a new aversive environment, this test has been validated against reference treatments for anxiety and it has good predictive validity. Furthermore, its robustness and reproducibility focused our interest [27].

Finally, it is interesting to be able to work on different animal models of anxiety, in order to approach the neural circuits from different approaches, depending on the stressful element [28].

Choice of animal species

Among the different strains of mice available in the field of animal experimentation, we chose to use the Swiss mouse. Indeed, this strain responds significantly, reproducibly and reliably to the treatments used in our study in the four-plate test. In addition, its use is simple, due to its particularly calm and non-aggressive behavior. Finally, we have had confirmation from their publishers that the stereotaxic atlases available for structural studies, although having been carried out for mice of another strain (C57/ BL6J), have been validated for the Swiss mouse and are perfectly compatible with this strain. Furthermore, the response of this strain to diazepam was found to be significant in two animal models of anxiety, the EPM and the double illuminated enclosure test [29], as well as in the FPT [30]. Within the numerous literature available on animal models of anxiety, the majority of studies use the rat as an animal model. Whether data must be presented, in order to make hypotheses and define a theoretical framework; it is appropriate to maintain a certain perspective with regard to the results obtained in this work. A difference in behavioral strategy or pharmacological response between these two species may be sufficient to explain certain results which would prove contradictory. In the lab, we chose to use the Mouse rather than the Rat for various reasons. The first is linked to the relative simplicity of the behavioral responses observed in mice, compared to rats. In addition, technical management and costs linked to the use of the Rat are more significant constraints.

The mouse test session lasts 5 minutes. The naive animal, that is to say not having previously undergone a test, is placed on the central platform facing an “open” arm. During its exploration, the times spent in the “open” and “closed” arms, and on the platform are measured, as well as the number of passages from one arm to the other.

Mice generally coming from their maintenance cage show behavior characterized by avoidance of “open” arms with a clear preference for “closed” arms. Animals typically spent more time in the “closed” arms then on the central platform and little time in the “open” arms, indicating a bias toward the relatively secure sections of the maze. This behavior is suppressed by anxiolytics and potentiated by anxiogenic substances. Lister et al. [13], showed that behavioral parameters of the EPM in mice provide data on two independent factors, one reflecting anxiety and the other motor activity. The percentage of entries and the time spent on the “open” arms represent good parameters for measuring anxiety in this model [14].

Test-retest protocols

The phenomenon of tolerance linked to the first exposure to the “one trial tolerance” (OTT) consists of a loss of pharmacological effect of a product triggered by prior exposure to a model without treatment, appearing during re-exposure the following day [31]. Initially described in the EPM [32], this phenomenon persists for more than two weeks. This loss of effect was subsequently reported in other animal models, such as the double illuminated enclosure test [33], the four plate test [27] or in the exposure test  to the odor of a predator [34]. In the drinking punishment test, no loss of effect of benzodiazepines was observed from the retest [35].

Subsequently, many teams became interested in this phenomenon, mainly using the EPM test. Among the explanations provided by this work, different hypotheses mainly emerge. Thus, several articles suggest that the pretest of the anxiety test would lead to an increase in the animal’s basal level of anxiety, and thus a loss of the anxiolytic type effect [36]. Others put forward the idea of locomotor habituation linked to the environment [37], of a modification of the state of the binding sites and/or of the GABAA receptor involved in the effect of benzodiazepines [38]. Some authors have described this tolerance as a qualitative change in the nature of the anxiety felt by the animal, which could thus modify the pharmacological response to different treatment [39, 40]. Thus, it is repeatedly suggested that the nature of the anxiety experienced by the animals is different between the pretest and the retest [41], with the possible appearance during the pretest of a phobia towards the stressful zone (open arm for the EPM). Test re-exposure protocols have developed in parallel with research on animal models of anxiety. Indeed, the researchers attempted to sensitize the models by placing the animal several times in the same test environment. It was indeed found that during a retest, there was an increase in the level of “anxiety”, but no variation in locomotor activity. This result was demonstrated in EPM in mice. The behavioral indices recorded to know the locomotor activity of the animal (“sniffing”, “rearing”, entry into the open arms and total entries into the arms) remained stable during the different exposures to the EPM [42]. Unexpectedly, the results showed that repeated exposure to the test did not increase the effectiveness of the treatments, and even decreased it. Thus, by working on the effectiveness of reference treatments such as benzodiazepines, File’s team demonstrated a loss of the anxiolytic type effect of these molecules in animals previously exposed to the test. In order to simplify the semantics used, we will call the first test, with the naive animals: “pretest”, and the second test, during re-exposure: “retest”. Furthermore, in the attached articles, the English terms are “Trial 1” for pretest and “Trial 2” for the retest.

Elevated Plus-Maze (EPM) test-retest: Many studies have focused on the OTT phenomenon on EPM. Modifications of the test-retest protocol, the equipment or the injection of certain molecules made it possible to discern some factors likely involved in the appearance of this tolerance to benzodiazepines during the retest. In the test-retest protocol, several studies were carried out in order to understand the critical parameters which could lead to the appearance of this tolerance. Thus, modifications of temporal factors such as the exposure time during the first test, the retest or even the interval between the two exposures, show that these parameters are critical in establishing tolerance. Thus, tolerance persists over a time interval between the two exposures of more than 2 weeks [43].

By focusing on the duration of the pretest in the EPM in untreated rats, researchers found that by reducing this duration of exposure, the anxiolytic effect of midazolam was modified during the retest. Thus, a duration of 5 minutes led to a complete disappearance of the effect, a duration of 2 minutes altered the effect and a duration of 1 minute did not trigger a loss of treatment effect [44]. Conversely, S. File and his team demonstrated in rats, a loss of effect of diazepam when the two exposures lasted 5 minutes, but an anxiolytic effect retained when the two test sessions lasted 10 minutes. The researchers here used a multifactorial analysis and concluded that there was a qualitative difference between the state of anxiety in which the animal was during the pretest or the retest. On the other hand, another team found that the loss of effect of the benzodiazepine was observed when the animals underwent a 10-minute pretest, then a 5-minute EPM retest the next day.

This difference in results seems linked to the duration of each exposure. Thus, a reduction in the pretest time can cancel the appearance of the OTT (5’ + 5’ = OTT, 1’ + 5’ = no OTT). An increase in the duration of the retest (10 minutes), when the pretest was 10 minutes, is enough to prevent OTT (10’ + 10’ = no OTT, 10’ + 5’ = OTT). These studies show the major importance of the spatial parameter in the study of the OTT phenomenon.

Four-plate Test (FPT): In my laboratory, we mainly use the EPM and the FPT as an anxiety test in mice. In 1997, in order to study the phenomena of aversive memory, we demonstrated a reduction in the number of punished passages accepted by the animals during the retest in the FPT. In this test re-exposure protocol, mice underwent two testing sessions of the four plates 24 hours apart, but only received treatment during the second session. This aversive memory persisted over a significant period of time since it was observed with an interval of 42 days between the two exposures. This work also demonstrated a loss of effect in the activity of benzodiazepines, particularly diazepam, during the four-plate test retest. Thus, diazepam no longer exerts an anxiolytic type action for doses between 0.25 mg/kg and 2 mg/ kg [27].This highlighting of a phenomenon observed until now mainly in EPM, has raised numerous questions, mainly on the cause of this pharmacological phenomenon and on the parallel to be drawn w Subsequently, we sought to understand what could be the mechanisms involved in these two results: the reduction in the number of passages accepted in non-injected mice, and the loss of anxiolytic-type effect of benzodiazepines during the retest. Indeed, we are in the presence of two phenomena probably linked to two independent behavioral dimensions, the exploration of a new environment and the fear of punishment. These two components of exploration and fear seem to be involved in any situation during which a rodent discovers a new environment [45]. With regard to exploration, treatments having a hyperlocomotor effect in mice do not seem to increase the number of punished passages accepted in the FTP during the retest, as we have demonstrated for example with an alpha- agonist psychostimulant. 1 adrenergic: adrafinil injected 30 minutes before the retest.

By testing different types of anxiolytic molecules, our laboratory succeeded in isolating one which exerts an anxiolytic- type activity in the four-plate test, and retains this anti-punitive  activity during the test-retest protocol. This is DOI, a serotonergic agonist of the 5-HT2A/2C receptors [46]. Subsequently, we were able to observe that this effect of DOI could be modulated by inter-regulation with the noradrenergic system. Indeed, an injection of an alpha2-adrenergic agonist such as clonidine or guanabenz completely abolishes the increase in the number of passages caused by the injection of DOI.

Using specific depletions of the noradrenergic or serotonergic system, we hypothesized that the effect of DOI seems to be exerted through 5-HT2 receptors located postsynaptically (Masse et al., 2006). Likewise, there is a link between the effect of DOI and the GABAergic system since injections of benzodiazepines such as alprazolam, and diazepam, as well as a GABAA receptor agonist, muscimol (at high doses) potentiate the anxiolytic-type effects of DOI.

When retested in the FPT, a sharp decrease in the number of punished passages accepted was reported. This modification of exploratory behavior is called “aversive memory”. The animal adapted its response to the environment, using the elements memorized during the pretest. During our work, we were able to show that this aversive memory was altered during the injection before the retest of atropine sulfate, a molecule with amnestic properties [47]. In addition, we also demonstrated a weak effect of scopolamine, a cholinergic muscarinic antagonist blocker, injected before the retest on aversive memory. On the other hand, in EPM, scopolamine, used at an amnestic dose, can cancel the modification of the behavioral response during the retest, if it is administered before or after the first test. This result was not found during an injection of scopolamine after the pretest in the EPM). These results tend to show that there is indeed an aversive memory, but that its study is difficult, in particular because of its rapid kinetics since in 24 hours, we observe a modification of the behavioral response Furthermore, the baseline level observed during the retest in the FPT does not allow the study of promnesic molecules, in order to determine their effect on the number of passages, as well as for molecules causing an anxiety- like effect. As such, we had previously concluded that the test- retest protocol of the four plates did not make it possible to study memory.

A team tested the effect of exposure to three unavoidable electric shocks on physiological constants in mice, as well as behavior in an EPM test carried out 5 minutes or 24 hours after this stress. They observed a strong stimulation of the hypothalamic-pituitary axis from 5 minutes after the shocks, noted by an increase in the plasma concentration of corticosterone in mice, which returned to basal level 2 hours after this stress. Furthermore, they demonstrated that the stress produced by electric shocks did not lead to changes in the use of spatial data acquired 24 hours previously, but reduced that of contextual and variable data [48,49]. In rats, a similar study found that the quantity of corticosterone increased after exposure to the test, in naive or experienced animals. The authors thus noted that there was no habituation in pre-exposed animals [50].

Memorization processes were also studied, in order to understand the involvement of memory in this phenomenon. Conversely, a team showed that the injection of molecules promoting memorization (amphetamine or pentylenetetrazole) immediately after the pretest led to an increase in the loss of effect, a potentiation of the OTT [51].

All these data tend to prove that there is a memory of aversion towards the model. Aversive memory, as it appears in the EPM with a reduction in entry and time in the “open” arms, or in the FPT with a reduction in the number of punished passages accepted during the retest. It seems to be subject to complex kinetics, since depending on the time of the injection (before or after the pretest), it can be canceled or not by scopolamine or atropine sulfate. On the other hand, it seems potentialize, as was demonstrated in the EPM. This result is not reproducible in the FPT, because of the too low baseline value of the untreated retest controls.

It is important to remember that there is information, stored in a structure, which is available during the retest and seems responsible for the modification of behavior, and perhaps in part for the OTT. This information is retained over the long term and its use depends on temporal and environmental factors during pretest and retest. It is possible that the clues, collected by the animal when discovering the test, retain a strong emotional component, linked to the aversion towards this environment or neophobia. This emotional dimension could use a form of memory that is insensitive to traditional amnestic treatments.

The OTT and aversive learning found in the test- retest FPT and EPM is a complex phenomenon not fully understood. However, some new directions in research emerged from recent findings. The implication of the PAG in experienced mice behavior suggested the involvement of the brain defense system during the second trial. The behavior of experience mice in the second trial of a test -retest paradigm might be related to fear rather than to anxiety [52].

1. Qureshi R, Irfan M, Gondal TM, Khan S, Wu J, Hadi MU, et al. AI in drug discovery and its clinical relevance. Heliyon. 2023; 9: e17575.
2. Domínguez-Oliva A, Hernández-Ávalos I, Martínez-Burnes J, Olmos- Hernández A, Verduzco-Mendoza A, Mota-Rojas D. The Importance of Animal Models in Biomedical Research: Current Insights and Applications. Animals (Basel). 2023; 13: 1223.
3. Belzung C, Lemoine M. Criteria of validity for animal models of psychiatric disorders: focus on anxiety disorders and depression. Biol Mood Anxiety Disord. 2011; 1: 9.
4. Bourin M, Petit-Demoulière B, Dhonnchadha BN, Hascöet M. Animal models of anxiety in mice. Fundam Clin Pharmacol. 2007; 21: 567- 574.
5. Bourin M. Animal models for screening anxiolytic-like drugs: a perspective. Dialogues Clin Neurosci. 2015; 17: 295-303.
6. Nic Dhonnchadha BA, Hascoët M, Jolliet P, Bourin M. Evidence for a 5-HT2A receptor mode of action in the anxiolytic-like properties of DOI in mice. Behav Brain Res. 2003; 147: 175-184.
7. Lister RG. The use of a plus-maze to measure anxiety in the mouse. Psychopharmacology (Berl). 1987; 92: 180-185.
8. Handley SL, Mithani S. Effects of alpha-adrenoceptor agonists and antagonists in a maze-exploration model of ‘fear’-motivated behaviour. Naunyn Schmiedebergs Arch Pharmacol. 1984; 327: 1-5.
9. Pellow S, File SE. Anxiolytic and anxiogenic drug effects on exploratory activity in an elevated plus-maze: a novel test of anxiety in the rat. Pharmacol Biochem Behav. 1986; 24: 525-529.
10. Bourin M. Animal models of anxiety: are they suitable for predicting drug action in humans? Pol J Pharmacol. 1997; 49: 79-84.
11. Belzung C, Griebel G. Measuring normal and pathological anxiety-like behaviour in mice: a review. Behav Brain Res. 2001; 125: 141-149.
12. Carola V, D’Olimpio F, Brunamonti E, Mangia F, Renzi P. Evaluation of the elevated plus-maze and open-field tests for the assessment of anxiety-related behaviour in inbred mice. Behav Brain Res. 2002; 134: 49-57.
13. Lister RG. The use of a plus-maze to measure anxiety in the mouse. Psychopharmacology (Berl). 1987; 92: 180-185.
14. Rodgers RJ, Johnson NJ. Factor analysis of spatiotemporal and ethological measures in the murine elevated plus-maze test of anxiety. Pharmacol Biochem Behav. 1995; 52: 297-303.
15. Wall PM, Messier C. Methodological and conceptual issues in the use of the elevated plus-maze as a psychological measurement instrument of animal anxiety-like behavior. Neurosci Biobehav. Rev. 2001; 25: 275-286.
16. Slotnick BM, Jarvik ME. Deficits in passive avoidance and fear conditioning in mice with septal lesions. Science. 1966; 154: 1207- 1208.
17. Aron C, Simon P, Larousse C, Boissier JR. Evaluation of a rapid technique for detecting minor tranquilizers. Neuropharmacology. 1971; 10: 459-469.
18. Bourin M, Hascoet M, Mansouri B, Colombel MC, Bradwejn J. Comparison of behavioral effects after single and repeated administrations of four benzodiazepines in three mice behavioral models. J Psychiatry Neurosci. 1992; 17: 72-77.
19. Hascoët M, Bourin M, Colombel MC, Fiocco AJ, Baker GB. Anxiolytic- like effects of antidepressants after acute administration in a four- plate test in mice. Pharmacol Biochem Behav. 2000; 65: 339-344.
20. Crawley JN. Exploratory behavior models of anxiety in mice. Neurosci Biobehav Rev. 1985; 9: 37-44.
21. Griebel G, Simiand J, Steinberg R, Jung M, Gully D, Roger P, et al. 4-(2-Chloro-4-methoxy-5-methylphenyl)-N-[(1S)-2-cyclopropyl-  1-(3-fluoro-4-methylphenyl)ethyl]5-methyl-N-(2-propynyl)-   1,3-thiazol-2-amine hydrochloride (SSR125543A), a potent and selective corticotrophin-releasing factor (1) receptor antagonist. II. Characterization in rodent models of stress-related disorders. J Pharmacol Exp Ther. 2002; 301: 333-345.
22. Klodzinska A, Tatarczy?ska E, Chojnacka-Wójcik E, Nowak G, Cosford ND, Pilc A. Anxiolytic-like effects of MTEP, a potent and selective mGlu5 receptor agonist does not involve GABA (A) signaling. Neuropharmacology. 2004; 47: 342-350.
23. Serradeil-Le Gal C, Wagnon J, Valette G, Garcia G, Pascal M, Maffrand JP, et al. Nonpeptide vasopressin receptor antagonists: development of selective and orally active V1a, V2 and V1b receptor ligands. Prog Brain Res. 2002; 139: 197-210.
24. Bourin M, Hascoët M. The mouse light/dark box test. Eur J Pharmacol. 2003; 463: 55-65.
25. Massé F, Hascoët M, Dailly E, Bourin M. Effect of noradrenergic system on the anxiolytic-like effect of DOI (5-HT2A/2C agonists) in the four-plate test. Psychopharmacology (Berl). 2006; 183: 471-481.
26. Ripoll N, Hascoët M, Bourin M. The four-plates test: anxiolytic or analgesic paradigm? Prog Neuropsychopharmacol Biol Psychiatry. 2006; 30: 873-880.
27. Hascoet M, Bourin M, Couetoux du Tertre A. Influence of prior experience on mice behavior using the four-plate test. Pharmacol Biochem Behav. 1997; 58: 1131-1138.
28. Bourin M. Translational research approach to neurobiology and treatment in: Kim YK, Amidfar M. (eds) Translational Research Methods for Major Depressive Disorder. Neuromethods, vol 179. Humana, New York, NY. 2022; 179: 57-84.
29. Griebel G, Rodgers RJ, Perrault G, Sanger DJ. The effects of compounds varying in selectivity as 5-HT(1A) receptor antagonists in three rat models of anxiety. Neuropharmacology. 2000; 39: 1848-1857.
30. Hascoët M, Bourin M. Anticonflict effect of alpidem as compared with the benzodiazepine alprazolam in rats. Pharmacol Biochem Behav. 1997; 56: 317-324.
31. Rezende RM, Weiner HL. Oral tolerance: an updated review. Immunol Lett. 2022; 245: 29-37.
32. File SE. One-trial tolerance to the anxiolytic effects of chlordiazepoxide in the plus-maze. Psychopharmacology (Berl). 1990; 100: 281-282.
33. Holmes A, Iles JP, Mayell SJ, Rodgers RJ. Prior test experience compromises the anxiolytic efficacy of chlordiazepoxide in the mouse light/dark exploration test. Behav Brain Res. 2001; 122: 159-167.
34. Carobrez AP, Bertoglio LJ. Ethological and temporal analyses of anxiety-like behavior: the elevated plus-maze model 20 years on. Neurosci Biobehav Rev. 2005; 29: 1193-1205.
35. File SE, Zangrossi H Jr, Viana M, Graeff FG. Trial 2 in the elevated plus- maze: a different form of fear? Psychopharmacology (Berl). 1993; 111: 491-494.
36. Rodgers RJ, Shepherd JK. Influence of prior maze experience on behaviour and response to diazepam in the elevated plus-maze and light/dark tests of anxiety in mice. Psychopharmacology (Berl). 1993; 113: 237-242.
37. Dawson GR, Crawford SP, Stanhope KJ, Iversen SD, Tricklebank MD. One-trial tolerance to the effects of chlordiazepoxide on the elevated plus maze may be due to locomotor habituation, not repeated drug exposure. Psychopharmacology (Berl). 1994; 113: 570-572.
38. Gonzalez LE, File SE. A five-minute experience in the elevated plus- maze alters the state of the benzodiazepine receptor in the dorsal raphe nucleus. J Neurosci. 1997; 17: 1505-1511.
39. Bertoglio LJ, Carobrez AP. Anxiolytic-like effects of NMDA/glycine-B receptor ligands are abolished during the elevated plus-maze trial 2 in rats. Psychopharmacology (Berl). 2003; 170: 335-342.
40. File SE, Zangrossi H Jr, Andrews N. Novel environment and cat odor change GABA and 5-HT release and uptake in the rat. Pharmacol Biochem Behav. 1993; 45: 931-934.
41. Cruz-Morales SE, Santos NR, Brandão ML. One-trial tolerance to midazolam is due to enhancement of fear and reduction of anxiolytic- sensitive behaviors in the elevated plus-maze retest in the rat. Pharmacol Biochem Behav. 2002; 72: 973-978.
42. Espejo EF. Effects of weekly or daily exposure to the elevated plus-maze in male mice. Behav Brain Res. 1997; 87: 233-238.
43. File SE, Mabbutt PS, Hitchcott PK. Characterisation of the phenomenon of “one-trial tolerance” to the anxiolytic effect of chlordiazepoxide in the elevated plus-maze. Psychopharmacology (Berl). 1990; 102: 98-101.
44. Dal-Col ML, Pereira LO, Rosa VP, Calixto AV, Carobrez AP, Faria MS. Lack of midazolam-induced anxiolysis in the plus-maze Trial 2 is dependent on the length of Trial 1. Pharmacol Biochem Behav. 2003; 74: 395-400.
45. Whimbey AE, Denenberg VH. Two independent behavioral dimensions in open-field performance. J Comp Physiol Psychol. 1967; 63: 500-504.
46. Nic Dhonnchadha BA, Bourin M, Hascoet M. Anxiolytic-like effects of 5-HT2 ligands on three mouse models of anxiety. Behav Brain Res. 2003; 140: 203-214.
47. Ripoll N, Nic Dhonnchadha BA, Sébille V, Bourin M, Hascoët M. The four-plates test-retest paradigm to discriminate anxiolytic effects. Psychopharmacology (Berl). 2005; 180: 73-83.
48. Celerier A, Pierard C, Beracochea D. Effects of ibotenic acid lesions of the dorsal hippocampus on contextual fear conditioning in mice: comparison with mammillary body lesions. Behav Brain Res. 2004; 151: 65-72.
49. Célérier A, Piérard C, Rachbauer D, Sarrieau A, Béracochéa D. Contextual and serial discriminations: a new learning paradigm to assess simultaneously the effects of acute stress on retrieval of flexible or stable information in mice. Learn Mem. 2004; 11: 196-204.
50. File SE, Zangrossi H, Sanders FL, Mabbutt PS. Raised corticosterone in the rat after exposure to the elevated plus-maze. Psychopharmacology (Berl). 1994; 113: 543-546.
51. Vargas KM, Da Cunha C, Andreatini R. Amphetamine and pentylenetetrazole given post-trial 1 enhance one-trial tolerance to the anxiolytic effect of diazepam in the elevated plus-maze in mice. Prog Neuropsychopharmacol Biol Psychiatry. 2006; 30: 1394-1402.
52. Bourin M. The test-retest model of anxiety: An appraisal of findings to explain benzodiazepine tolerance. Pharmacol Biochem Behav. 2019; 178: 39-41.

This article aims to explain how in my research laboratory, it was possible to transform 2 behavioral anxiety tests: elevated plus maze and four-plate test, into more efficient tests that make it possible to take into account in human clinic the phenomenon of tolerance. This well-known phenomenon for benzodiazepines can be studied for potential new anxiolytic molecules. This also allowed us to begin to study aversive memory which seems important in demonstrating an anxiolytic action in animal models of anxiety.

• Anxiolytics
• Aversive memory
• Elevated plus maze
• Four plate test