Iiitern.
0 1979, Gordon and Breach Science Publishers, Inc.
J . Neuroscience. 1979, Vol. 9, pp. 185-190
0020-7454/79/0903-0185$04.50/0
Printed in Great Britain
THE EFFECTS OF LITHIUM CHLORIDE ON RESPONSE TO SALIENT AND NONSALIENT STIMULI IN CARASSIUS AURATUS F. N. JOHNSON Department of Psychology, University of Luncaster, Bailrigg, Luncaster LA I 4 YF, England
Int J Neurosci Downloaded from informahealthcare.com by McMaster University on 10/27/14 For personal use only.
(Received December 20. 1978)
The exploratory behaviour of 20 goldfish was investigated to see whether it was affected by treatment with lithium chloridz. When familiar and novel stimuli were readily discriminable, lithium treatment had no effect on exploration. In a second study, using 20 fish, i t was shown that when novel and familiar stimuli were less markedly different from each other, exploratory responses to the novel stimulus were reduced by lithium. The findings may be related to a general suppressant action of lithium on the extent of stimulus generalisation, or to a reduction by lithium of the animals' responsiveness to near-threshold levels of stimulation.
1NTRODUCTION
Psychological Models of Lithium Action and of Manic- Depression The therapeutic efficacy of the salts of lithium, both in the treatment of recurrent endogenous affective disorders and in the alleviation of a variety of other conditions, has now been elucidated (Schou & Thomsen, 1975; Schou, 1977) to such an extent that these substances play an important role in present-day psychiatric practice. It is nevertheless true that, as yet, no entirely satisfactory model has been proposed to account for the mechanism by which the lithium ion exerts its therapeutically important behavioural effects. On the basis of a number of studies examining the effects of lithium salts on the behaviour of rats, a psychological model of lithium action has been proposed (Johnson, 1972). According to this model, lithium salts affect the stimulus processing mechanisms of animals, so that stimuli of around threshold salience ate either not processed at all or are processed only' poorly; as a result, certain behaviours elicited by environmental stimuli (exploratory activity, for example) may be attenuated by lithium administration, whereas nonenvironmentally elicited (spontaneous) activities are more likely to remain unaffected. Such an hypothesis clearly has implications for the way in which one conceptualises the conditions which lithium is used
to treat. Johnson (1975a, 1977b) has put forward a series of proposals for a model of manicdepression based upon the idea that both the manic and depressive phases of recurrent endogenous affective disorders result from dysfunctioning of stimulus analysing mechanisms. This model takes note of the fact that in the course of growth and development, animals and people come to attach more or less specific responses to particular environmental stimuli. Stimuli similar to these environmental cues also elicit the learned responses, though not so effectively; this is referred to as stimulus generalisation. If an individual, for any reason, shows abnormally wide stimulus generalisation characteristics, or, to put it another way, wide-band responsiveness to environmental features, then one might expect to find that individual producing responses inappropriately to stimuli which ought not to be eliciting any reaction. Such a person would be overactive, overstimulated, and liable to report seeing meaning or significance in environmental events which, to others, lack any such importance. Manic patients frequently demonstrate such characteristics. If lithium salts produce, amongst other effects, a general suppression of responsiveness to stimuli of around threshold salience, effectively narrowing the range of stimulus generalisation, then this may be one of the ways in which they help to alleviate manic excitement in humans.
186
F. N. JOHNSON
Int J Neurosci Downloaded from informahealthcare.com by McMaster University on 10/27/14 For personal use only.
Animal Studies of Behavioural Eflects of Lithium
The problems involved in the use of animals to investigate the effects which lithium produces on behaviour have been discussed by Johnson (1 976, 1977a). One of the most important differences between experimental studies of lithium in animals and the clinical use of lithium concerns the use of acute administration regimes, usually by intraperitoneal or subcutaneous injection in the former, as against chronic regimes using oral administration routes in the latter. In therapeutic usage, lithium levels in the blood serum of patients are stabilised as far as possible. Single, acute injections of lithium salts in animals lead to varying serum levels in the post-injection period: the rate of change of serum lithium, as well as its absolute value at any one time, may have physiological consequences affecting behaviour. There are technical difficulties involved with maintaining rats and mice on continuous lithium administration regimes, mainly associatcd with sodium balance changes and weight loss in the animals (Plenge et a/., 1973; Thomsen et a/., 1976), and whilst it is possible, by various devices, to overcome some of these problems (e.g., Olesen & Thomsen, 1975) an appropriate choice of animal species for behavioural experimentation may avoid the problems altogether. Such considerations governed the choice in the present studies of the goldfish (Carassius auratus) as experimental subjects. These are animals which seem able t o tolerate relatively high concentrations of alkali metal ions in the surrounding medium-a legacy no doubt of the animals' origins in brackish water habitats. The fish may be kept in solutions of lithium salts over relatively long periods of time, apparently without ill effects; they take up lithium ions from the surrounding medium, their body tissue concentrations eventually equilibrating with (but not necessarily equalling) those outside (Weischer, 1969; Figler, 1973; Peeke el al., 1973). The use of these animals thus enabled behavioural studies to be performed under conditions of chronic lithium administration and stable tissue levels of lithium.
to sodium-treated control subjects. experiment examined this proposition.
The first
EXPERIMENT I Materials and Method Subjects. A total of 20 goldfish were used in this experiment; they were selected on a size basis, i.e., fish less than 3.25 cm or greater than 4.5 cm in overall length were rejected. Individual fish were identified by their natural markings and colour differences. When not in use for experimental purposes the fish were kept together in a single tank. Apparatus. Exploratory behaviour was tested in a Y-maze having the following dimensions: the three arms were each 4 cm wide and I5 cm long and met at angles of 120" with each other. The arms of the maze ended in a small enclosure 6.5cm square which could be isolated from the arm by means of a guillotine door. The maze arms and terminal enclosures were 6 cm tall and were mounted on a base board. The maze and the base board were made of transparent Perspex and the whole apparatus was so constructed as to be watertight, allowing the maze to be filled with dechlorinated tap water at room temperature. Each arm of the maze could be fitted with a false inner surface made of painted aluminium sheet: in this way the colour of the maze arms could be varied as required. Further details of the Y-maze are shown in Figure 1.
Rntionale of the Behavioural Studies It was argued that if lithium did indeed impair the responsiveness of animals to features of the environment, as predicted by Johnson's rrod:I of lithium action (Johnson, 1972), then exploratory activity shown by the fish ought to be reduced or even eliminated under lithium treatment as compared
FIGURE 1 Diagram of the Y-maze used in both experiments. A : the complete maze showing a black/white choice at the junction; B, C: greylwhitc and white/white choices at the junction; D: showing the guillotine door used 10 confine each fish in the start box prior 10 each discrimination trial.
Int J Neurosci Downloaded from informahealthcare.com by McMaster University on 10/27/14 For personal use only.
THE EFFECTS OF LITHIUM CHLORIDE ON CARASSIUS AURATUS
Procedure. To test exploratory behaviour, a fish was allowed to swim freely in the maze for a period of 5 min, during which time all three arms of the maze were fitted with the same colour false surfaces. The animal’s preference for arm direction was determined by counting the number of left and right turns at the choice point in the free exploratory period. After this initial free-swimming session the fish was confined to one of the compartments at the end of a maze arm for 30sec whilst the walls of the relatively lesser preferred arm were changed to a different colour. The fish was then released and a note made of whether or not it turned into the novel or familiar arm. After the choice had been made the fish was again confined in the enclosure at the end of the arm, and the maze walls changed so that the novel-coloured arm was again paired with a familiar arm at the choice point. The procedure was repeated ten times for each animal. The direction of the novel coloured arm was determined on a random basis after having been presented on the nonpreferred side on the first test trial. Half the fish were given free exploratory experience of an all-black maze, the novel arm being white, whilst the remaining fish experienced an allwhite maze with the novel arm being black. Half the fish in each group were kept in lithium chloride solution at a concentration of 15 mmol/litre for one week prior to commencement of the study and were kept in the solution throughout the course of the experiment. The remaining fish were kept in sodium chloride solution, 15 mmol/litre, over the same period. All maze testing was carried out with the maze filled with dechlorinated tap water. Each fish was tested on three successive days. The experimental design was thus a 2 (novel colours; black/white) x 2 (drugs) x 10 (trials) x 3 (days) factorial with repeated measures on the last two factors and 5 fish in each of the 4 (novel colour x drugs) conditions. Before analysis the data were reduced by summing for each fish the number of novel responses made on the ten trials on each day to provide a novelty response index. Results
The results are summarised in Table I. An analysis of variance carried out on the data showed no statistically significant effect of any factor, though the number of novel responses overall was significantly greater (p
0 1979, Gordon and Breach Science Publishers, Inc.
J . Neuroscience. 1979, Vol. 9, pp. 185-190
0020-7454/79/0903-0185$04.50/0
Printed in Great Britain
THE EFFECTS OF LITHIUM CHLORIDE ON RESPONSE TO SALIENT AND NONSALIENT STIMULI IN CARASSIUS AURATUS F. N. JOHNSON Department of Psychology, University of Luncaster, Bailrigg, Luncaster LA I 4 YF, England
Int J Neurosci Downloaded from informahealthcare.com by McMaster University on 10/27/14 For personal use only.
(Received December 20. 1978)
The exploratory behaviour of 20 goldfish was investigated to see whether it was affected by treatment with lithium chloridz. When familiar and novel stimuli were readily discriminable, lithium treatment had no effect on exploration. In a second study, using 20 fish, i t was shown that when novel and familiar stimuli were less markedly different from each other, exploratory responses to the novel stimulus were reduced by lithium. The findings may be related to a general suppressant action of lithium on the extent of stimulus generalisation, or to a reduction by lithium of the animals' responsiveness to near-threshold levels of stimulation.
1NTRODUCTION
Psychological Models of Lithium Action and of Manic- Depression The therapeutic efficacy of the salts of lithium, both in the treatment of recurrent endogenous affective disorders and in the alleviation of a variety of other conditions, has now been elucidated (Schou & Thomsen, 1975; Schou, 1977) to such an extent that these substances play an important role in present-day psychiatric practice. It is nevertheless true that, as yet, no entirely satisfactory model has been proposed to account for the mechanism by which the lithium ion exerts its therapeutically important behavioural effects. On the basis of a number of studies examining the effects of lithium salts on the behaviour of rats, a psychological model of lithium action has been proposed (Johnson, 1972). According to this model, lithium salts affect the stimulus processing mechanisms of animals, so that stimuli of around threshold salience ate either not processed at all or are processed only' poorly; as a result, certain behaviours elicited by environmental stimuli (exploratory activity, for example) may be attenuated by lithium administration, whereas nonenvironmentally elicited (spontaneous) activities are more likely to remain unaffected. Such an hypothesis clearly has implications for the way in which one conceptualises the conditions which lithium is used
to treat. Johnson (1975a, 1977b) has put forward a series of proposals for a model of manicdepression based upon the idea that both the manic and depressive phases of recurrent endogenous affective disorders result from dysfunctioning of stimulus analysing mechanisms. This model takes note of the fact that in the course of growth and development, animals and people come to attach more or less specific responses to particular environmental stimuli. Stimuli similar to these environmental cues also elicit the learned responses, though not so effectively; this is referred to as stimulus generalisation. If an individual, for any reason, shows abnormally wide stimulus generalisation characteristics, or, to put it another way, wide-band responsiveness to environmental features, then one might expect to find that individual producing responses inappropriately to stimuli which ought not to be eliciting any reaction. Such a person would be overactive, overstimulated, and liable to report seeing meaning or significance in environmental events which, to others, lack any such importance. Manic patients frequently demonstrate such characteristics. If lithium salts produce, amongst other effects, a general suppression of responsiveness to stimuli of around threshold salience, effectively narrowing the range of stimulus generalisation, then this may be one of the ways in which they help to alleviate manic excitement in humans.
186
F. N. JOHNSON
Int J Neurosci Downloaded from informahealthcare.com by McMaster University on 10/27/14 For personal use only.
Animal Studies of Behavioural Eflects of Lithium
The problems involved in the use of animals to investigate the effects which lithium produces on behaviour have been discussed by Johnson (1 976, 1977a). One of the most important differences between experimental studies of lithium in animals and the clinical use of lithium concerns the use of acute administration regimes, usually by intraperitoneal or subcutaneous injection in the former, as against chronic regimes using oral administration routes in the latter. In therapeutic usage, lithium levels in the blood serum of patients are stabilised as far as possible. Single, acute injections of lithium salts in animals lead to varying serum levels in the post-injection period: the rate of change of serum lithium, as well as its absolute value at any one time, may have physiological consequences affecting behaviour. There are technical difficulties involved with maintaining rats and mice on continuous lithium administration regimes, mainly associatcd with sodium balance changes and weight loss in the animals (Plenge et a/., 1973; Thomsen et a/., 1976), and whilst it is possible, by various devices, to overcome some of these problems (e.g., Olesen & Thomsen, 1975) an appropriate choice of animal species for behavioural experimentation may avoid the problems altogether. Such considerations governed the choice in the present studies of the goldfish (Carassius auratus) as experimental subjects. These are animals which seem able t o tolerate relatively high concentrations of alkali metal ions in the surrounding medium-a legacy no doubt of the animals' origins in brackish water habitats. The fish may be kept in solutions of lithium salts over relatively long periods of time, apparently without ill effects; they take up lithium ions from the surrounding medium, their body tissue concentrations eventually equilibrating with (but not necessarily equalling) those outside (Weischer, 1969; Figler, 1973; Peeke el al., 1973). The use of these animals thus enabled behavioural studies to be performed under conditions of chronic lithium administration and stable tissue levels of lithium.
to sodium-treated control subjects. experiment examined this proposition.
The first
EXPERIMENT I Materials and Method Subjects. A total of 20 goldfish were used in this experiment; they were selected on a size basis, i.e., fish less than 3.25 cm or greater than 4.5 cm in overall length were rejected. Individual fish were identified by their natural markings and colour differences. When not in use for experimental purposes the fish were kept together in a single tank. Apparatus. Exploratory behaviour was tested in a Y-maze having the following dimensions: the three arms were each 4 cm wide and I5 cm long and met at angles of 120" with each other. The arms of the maze ended in a small enclosure 6.5cm square which could be isolated from the arm by means of a guillotine door. The maze arms and terminal enclosures were 6 cm tall and were mounted on a base board. The maze and the base board were made of transparent Perspex and the whole apparatus was so constructed as to be watertight, allowing the maze to be filled with dechlorinated tap water at room temperature. Each arm of the maze could be fitted with a false inner surface made of painted aluminium sheet: in this way the colour of the maze arms could be varied as required. Further details of the Y-maze are shown in Figure 1.
Rntionale of the Behavioural Studies It was argued that if lithium did indeed impair the responsiveness of animals to features of the environment, as predicted by Johnson's rrod:I of lithium action (Johnson, 1972), then exploratory activity shown by the fish ought to be reduced or even eliminated under lithium treatment as compared
FIGURE 1 Diagram of the Y-maze used in both experiments. A : the complete maze showing a black/white choice at the junction; B, C: greylwhitc and white/white choices at the junction; D: showing the guillotine door used 10 confine each fish in the start box prior 10 each discrimination trial.
Int J Neurosci Downloaded from informahealthcare.com by McMaster University on 10/27/14 For personal use only.
THE EFFECTS OF LITHIUM CHLORIDE ON CARASSIUS AURATUS
Procedure. To test exploratory behaviour, a fish was allowed to swim freely in the maze for a period of 5 min, during which time all three arms of the maze were fitted with the same colour false surfaces. The animal’s preference for arm direction was determined by counting the number of left and right turns at the choice point in the free exploratory period. After this initial free-swimming session the fish was confined to one of the compartments at the end of a maze arm for 30sec whilst the walls of the relatively lesser preferred arm were changed to a different colour. The fish was then released and a note made of whether or not it turned into the novel or familiar arm. After the choice had been made the fish was again confined in the enclosure at the end of the arm, and the maze walls changed so that the novel-coloured arm was again paired with a familiar arm at the choice point. The procedure was repeated ten times for each animal. The direction of the novel coloured arm was determined on a random basis after having been presented on the nonpreferred side on the first test trial. Half the fish were given free exploratory experience of an all-black maze, the novel arm being white, whilst the remaining fish experienced an allwhite maze with the novel arm being black. Half the fish in each group were kept in lithium chloride solution at a concentration of 15 mmol/litre for one week prior to commencement of the study and were kept in the solution throughout the course of the experiment. The remaining fish were kept in sodium chloride solution, 15 mmol/litre, over the same period. All maze testing was carried out with the maze filled with dechlorinated tap water. Each fish was tested on three successive days. The experimental design was thus a 2 (novel colours; black/white) x 2 (drugs) x 10 (trials) x 3 (days) factorial with repeated measures on the last two factors and 5 fish in each of the 4 (novel colour x drugs) conditions. Before analysis the data were reduced by summing for each fish the number of novel responses made on the ten trials on each day to provide a novelty response index. Results
The results are summarised in Table I. An analysis of variance carried out on the data showed no statistically significant effect of any factor, though the number of novel responses overall was significantly greater (p
