A total number of 32 male rats, 16 SHR and 16 WKY, participated in this study. At the start of testing following 8 days acclimatization, the rats were 5 wk old and experimentally naïve. Young rats were required, as ADHD primarily is a child and adolescent disorder. The SHRs were obtained from Charles River Italy (SHR/Crl Ico) and the WKYs from Charles River France (WKY/Nico).
At the University of Oslo, the rats were housed individually in 41 × 25 × 25 (height) cm transparent cages and had free access to food (RM3 (E) from Special Diet Services, Witham, Essex CM8 3AD, UK). The rats had access to water at all times before the habituation session. Starting following completion of the habituation session, the rats were deprived of water for 21 hr a day; this is a moderate, but sufficient deprivation for motivating the animal. The temperature in the housing area was ~22°C. The light was on from 0700 to 1900 hours. The behavioral training took place between 1000 and 1330 hours seven days a week. The study was approved by the Norwegian Animal Research Authority (NARA), and was conducted in accordance with the laws and regulations controlling experiments/procedures in live animals in Norway.
Sixteen Campden Instruments operant chambers were used in the study. The animal working space in eight of the chambers was 25 × 25 × 30 (height) cm and 25 × 25 × 20 (height) cm in the other eight chambers. A fan producing a low masking noise and the 2.8-W house light were on during the entire experimental session.
During training sessions, either one or both retractable levers were used (below). A 2.8-W cue light was located above each lever. The rats' response consisted of pressing one of the levers with a dead weight of at least 3 g to activate a micro-switch. The reinforcers (0.01 ml tap water) were delivered by a liquid dipper located in a small recessed cubicle with a 2.8-W cue light that lit up when a reinforcer was presented. A 7 × 5 cm transparent plastic lid separated the cubicle from the rat's working space. The rat could easily open the lid with a light push with the nose or paw. Each chamber was ventilated and placed in a sound-resistant outer housing. A computer and an online system (SPIDER, Paul Fray, Ltd., UK) recorded the behavior and scheduled reinforcers (drops of water).
Before the initiation of the study, the rats were assigned a chamber (1 through 16) and time of testing (1000 or 1200 hours) in a randomized and balanced way. The rat was returned to its living cage after each session and immediately given free access to water for 90 min.
The training period started with a single 30-min habituation session. During the habituation session, the lid between the working space and the reinforcement cubicle was kept open. The house light was on, but no lever was present, no cue light above any lever was lit and water was not delivered.
The habituation session was followed by two 30-min dipper training sessions. The lid was taped open, no levers were present, and the house light was on, but the cue lights above the levers were not lit. The computer delivered water on the average every 10 s independent of the rat's behavior (a variable-time schedule). Each water delivery was accompanied by the turning on of the cue light in the small recessed cubicle.
In the next two sessions, the rat was trained to open the lid to gain access to the water. The lid was not taped open, no levers were present and the lights above the levers were not activated. The house light was on. Each lid opening was followed by a presentation of a single drop of water. The cue light in the recessed cubicle was turned on when water was present.
During the subsequent two sessions, lever responding was shaped by the method of successive approximations [22
]. During the first of these sessions, the rats learned to press the left lever in order to receive a reinforcer immediately following every press. The cue light above the left lever was now lit the entire session. The right lever was retracted into the wall and the light above the right lever was off. On the second session, the right lever was activated and the left lever retracted. During this session the light above the right lever was lit the entire session. The house light was on during both sessions. Following this shaping procedure, the animal had acquired the appropriate lever-pressing behavior.
From now on, both levers were present. The light above the levers shifted randomly. The light stayed lit above a lever for as long as it was the correct lever. This was the discriminative stimulus showing the rat which lever it had to press in order to receive a reinforcer. A concurrent extinction schedule was present on the wrong lever. There was never any light above the extinction lever. Thus, the present task was a simultaneous visual discrimination task. The first four of these sessions lasted for 30 min and the reinforcers were delivered following every correct lever press. Then followed a single session when the reinforcers were delivered according to a 15-s random-interval schedule. Whenever an interval had elapsed, the reinforcer was delivered immediately following the first correct response.
The simultaneous visual discrimination task was used for testing effects of the drugs. An unpredictable 180-s random-interval schedule was in effect for 90 min on the correct lever (signaled by a constantly lit cue light above this lever) from session 13 on until the study was finished. Inter-reinforcer times ranged from 6 to 719 s in a randomized fashion with a skewed distribution modeled after the "Harvard golden tape" [23
]. There was neither any external stimulus signaling that a reinforcer was programmed, nor any external stimulus signaling the time since the last response. A concurrent extinction schedule (never associated with any cue light) was present on the wrong lever. The house light was lit the entire session.
Each session was divided into five 18-min segments (parts) in order to monitor intra-session changes in the behavior. For each segment, total number of presses on the correct and incorrect lever as well as number of reinforcers delivered were recorded. Time between consecutive correct responses (inter-response time, IRT) was also recorded.
The total number of lever presses is an expression of the general activity level and therefore a measure of degree of activity. The percent choice of the correct lever when the reinforcers are delivered infrequently is a measure of sustained attention. The number of responses with short IRTs (<0.67 s) is used as a measure of degree of impulsiveness (cannot hold back a response even when one knows it is an unnecessary one).
Administration of the drugs started at session 49 when the behavior had stabilized. The two isomers of amphetamine were compared with vehicle and with each other. All rats received d-amphetamine sulphate and l-amphetamine sulphate. The dosing was balanced in the initial schedule. The schedule was later extended to take into account results obtained by session 76 (Drug day 9). This involved repeating some of the previous results when the previous ones had deviated from what was predicted and adding new doses that the obtained data suggested might be of interest: 0.64 mg/kg d-amphetamine and 1.27 mg/kg l-amphetamine. Each rat was injected intraperitoneally at a dose volume of 1 ml/kg body weight of the animal ~30 min before testing, with either vehicle (physiological saline) or drug. Drugs were administered every 3rd or 4th day. All rats received all doses according to a balanced design.
D-amphetamine sulphate (Lot 031298) and l-amphetamine sulphate (Lot FB-101-57) were supplied from Boeringer-Ingelheim US. Doses were 0.64, 1.27, and 1.91 mg/kg for d-amphetamine; and 1.27, 2.54, and 3.81 mg/kg for l-amphetamine. Doses were calculated as the weight of base using a conversion factor of 1.360 mg sulphate salt as equivalent to 1.000 mg base. Doses were based on pilot studies. Dosing solutions were prepared as a solution in physiological saline. Stock solutions, 1.91 mg/kg for d-amphetamine; and 3.81 mg/kg for l-amphetamine, were prepared at the start of the dosing period and kept at +4 to +6°C when not in use. Dilutions of the stock solutions were made each day of dosing.
Data management and statistical procedures
The mean behavior was regarded as the drug response, and dose-response curves were plotted for each drug and strain. The data were processed by univariate and multivariate analyses of variance (ANOVAs and MANOVAs, respectively) with the Statistica 7.1 program [24
]. Isomer and dose are within-subject variables. Strain is a between-subject variable. One control rat was identified as a statistical outlier with Grubbs' Test [25
]. Post-hoc comparisons following MANOVAs were performed by the Unequal N HSD procedure, a generalization of Tukey's test to the case of unequal samples sizes (see [27
], p. 975).