Access the full text.
Sign up today, get DeepDyve free for 14 days.
Background ADHD is a disorder where a common symptom is impulsive behaviour, a broad term associated with making sub-optimal choices. One frequently used method to investigate impulsive behaviour is delay discounting, which involves choosing between a small, immediate reinforcer and a delayed, larger one. Choosing the small imme- diate reinforcer is by itself, however, not sufficient for terming the choice impulsive, as all organisms eventually switch to choosing the small, immediate reinforcer when the delay to the larger reinforcer becomes long. This switch can be termed impulsive only when it occurs more frequently, or at shorter LL delays, than typically observed in normal controls. A poorly understood aspect is how choice is influenced by previous experience with delays. Using an animal model of Attention-Deficit/Hyperactivity Disorder, the Spontaneously Hypertensive Rat, we manipulated the order of exposure to delays in a delay discounting task. Following a preference test, the Ascending group experienced gradu- ally increasing delays between choice and reinforcer while the Descending group were exposed to these delays in reverse order. Results The results showed that the Descending group chose the small, immediate reinforcer over the larger delayed to a much larger extent than the Ascending group, and continued to do so even when the delay component was ultimately removed. Strain effects were found in the Ascending group, with SHRs switching to the small, immediate reinforcer earlier than controls as the delay to the larger reinforcer increased. Conclusion The data suggests that delay discounting is affected by history of exposure to delayed consequences. When reinforcement contingencies are incrementally changed from having no response-reinforcer delay to a long delay, discounting of delayed consequences is gradual. However, a sudden change from no delay to a long delay, without intermediate training, results in a rapid switch to the small, immediate reinforcer option, and this behaviour is somewhat resilient to the shortening and eventual removal of the large reinforcer delay. The implication is that attempting to reduce already existing impulsive behaviour in children with ADHD will require gradual habituation and not sudden changes in reinforcement contingencies. Keywords ADHD, SHR, Animal model, Impulsivity, Delay discounting, Learning history Introduction Attention-Deficit/Hyperactivity Disorder (ADHD) affects ~ 5% of the child population, and 2.5% of the adult *Correspondence: [1–6], and exists in three subtypes: inattentive, hyperac- E. B. Johansen tive-impulsive, and combined . A defining aspect of the email@example.com combined and hyperactive-impulsive subtypes is impul- Kristiania University College, Prinsens gate 7-9, 0152 Oslo, Norway Oslo Metropolitan University, P.O. Box 4, St. Olavs Plass, 0130 Oslo, sive behaviour, characterized as a tendency to act without Norway foresight or making choices based on poor reasoning of University of Bergen, Sydnesplassen 7, 5007 Bergen, Norway future consequences . However, the term “impulsive” © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Sjoberg et al. Behavioral and Brain Functions (2023) 19:4 Page 2 of 10 is a broad label that classifies a range of traits, such as being the theoretical option maximizing the amount of impatience, restlessness, risk or sensation-seeking behav- reinforcers in real time. Choosing the small, immediate iour, spontaneous or quick decisions, and lack of fore- reinforcer is adaptive in times of uncertainty (a bird in sight [9, 10]. The terms impulsivity or impulsiveness hold the hand is worth two in the bush), or during e.g., severe a central place in psychological theories and psychiatric deprivation when immediate replenishment is needed symptom’s lists, and their various operationalizations and for survival. Thus, as choosing the small, immediate rein - neurobiological bases have been extensively studied. Due forcer is sometimes normal or typical, impulsivity needs to the term’s heterogeneity and multifaceted nature, some be defined relative to the choices of neurotypical controls have even argued that these concepts should be rejected . Therefore, in this paper, we operationalize “impul - and replaced with betted defined terms as they fail to sive” behaviour as when an organism significantly more meet the requirements of a psychological construct . often than neurotypical controls performs a choice lead- The exact cause of impulsive behaviour in ADHD is ing to small reinforcers when large reinforcers are avail- debated and depends on both the operationalization of able at the cost of waiting and is the option maximizing the term and what theory one adopts [12, 13]. One theory amount of rewards in real time. of ADHD is the delay aversion theory, which proposes ADHD children, compared to controls, appear to dis- that impulsive behaviour is the result of an unwilling- play a reduced tendency to wait for a larger reinforcer ness to endure the temporal aspect of a choice, including and will typically choose the smaller option more often the length between repeated choices . An expansion than neurotypical controls, i.e., they are impulsive [14, of this theory is the dual component model of ADHD, 20–23]. A meta-analysis suggests that people with ADHD which in addition to delay aversion incorporates a con- are particularly sensitive to long delays, and are twice as cept called impulsive drive for immediate reward (IDIR) likely as controls to make an impulsive choice in the pres- . This impulsive drive is the tendency for impulsive ence of hypothetical reinforcers (e.g., points) compared behaviour to be affected by the time between response to real reinforcers (e.g., money or food) in the task . and reinforcer for any given choice, specifically that longer response-reinforcer delays reduce the likelihood of The SHR animal model of ADHD it being chosen [15, 16]. Thus, the dual component model The Spontaneously Hypertensive Rat (SHR) is the most of ADHD suggests that both delay aversion and impul- commonly used animal model of ADHD [25, 26], and sive drive contribute together towards impulsive behav- largely considered the most validated model [27–30]. The iour . The dual component model does not explain rats were initially bred for high blood pressure research the mechanism or processes behind IDIR but refers to , but when compared to controls they exhibit simi- other theories and explanations like executive dysfunc- lar characteristics to people with ADHD: they express tion and deficits in inhibitory control, or that people with impulsivity [32–35], inattention , hyperactivity , ADHD have a steeper delay-of-reinforcement gradient and increased behavioural variability [37, 38]. The SHR as suggested in the Dynamic Developmental Theory of model is well researched, but has only been used a little ADHD (DDT, see ). The DDT offers detailed hypoth - more than a dozen times in delay discounting research eses regarding the behavioural and neurological mecha- [32, 35, 38–52]. Most studies on delay discounting using nisms behind impulsive behaviour. It proposes that the SHRs find that the rats act more impulsively on the task effectiveness of a reinforcer is a decreasing function of compared to controls [32, 35, 40, 42, 43, 45–47, 49, 50], the time between response and consequence, termed the indicated by a higher tendency to choose the small rein- delay-of-reinforcement gradient, and that this gradient forcer when long delays are present for the large rein- is steeper in people with ADHD, meaning a steeper dis- forcer, although other studies have failed to find any such counting of future reinforcers leading to the preference strain difference [38, 41, 44, 48]. for small immediate reinforcers over large delayed [12, 17]. The discrimination test and learning history in delay discounting Delay discounting A discrimination test is a pre-experimental procedure Delay discounting is a commonly used method for study- where the animals are exposed to small and large rein- ing and measuring impulsive behaviour . It usually forcers without delays, which purpose is to establish that involves choosing between a small reinforcer delivered the animal prefers the large over the small reinforcer immediately and a larger reinforcer delivered after a option prior to any experimental manipulations. In other delay. All organisms eventually switch to choosing the words, it is a test to verify that reinforcer size, and not immediate, small reinforcer as the delay to the larger operandum position or other variables, controls choice reinforcer increases, despite the larger, delayed reinforcer during no delay. This is a fundamental study requirement, S joberg et al. Behavioral and Brain Functions (2023) 19:4 Page 3 of 10 as it is pointless to study choice as a function of delay to the experiment begins. This is similar to Fox et al. , the larger reinforcer if reinforcer size does not control who also used a discrimination test, but did not specify choice. Sjoberg and Johansen  emphasized the impor- any criterion other than all rats preferring the large rein- tance of including a discrimination test in order to avoid forcer “almost exclusively (p. 147)” by the end of the assumptions about the animals’ baseline preferences, fourth session. Second, unlike Fox et al. , who used but found that only three out of fourteen surveyed SHR a within-subjects design, we will employ a between-sub- studies clearly outlined the details of their discrimination jects design. This means that, following the discrimina - test [39, 43, 51]. Three others specified the details but tion test, one group of rats will be exposed to gradually included a delay component during this phase [40, 41, increasing delays (Ascending group) while another group 44], while the remainder either did not include such a test will be exposed to these delays in reverse order (Descend- or did not specify the details involved. This reduces the ing group). Thus, the Descending group will be exposed possibility of direct comparison between studies. to an abrupt and long delay to the large reinforcer and In delay discounting, it can be argued that previous then decreasing delays as opposed to slow and gradu- experience with the choice paradigm will influence the ally increasing delays in the Ascending group. Third, we likelihood of a choice pattern occurring. An example will implement a procedure where the total trial length is of this is when animals are reused in a different experi - constant and fixed at 24 s . Therefore, as length of the ment. For example, the rats used in Fox et al.  were delays change, inter-trial-intervals (ITIs) are adjusted to later reused in a different delay discounting experiment keep a constant trial duration. As a result, the two vari- by the same researchers , and the SHRs appeared to ables are always balanced and control for each other to show a steeper discounting curve in the second experi- the degree where one is absent, the other is at maximum ment once a delay component was introduced. This (e.g., when delay is 0 s, ITI is 24 s). Fox et al.  also observation alone, however, does not prove that previ- used a compensating design where the inter-trial inter- ous experience was the cause, as a number of other fac- val would shrink in accordance with increased delays so tors may have influenced the results (e.g. the rats were as to assure that the trial lengths always remained con- also given saline or drug injections). In the SHR model stant [19, 53]. However, their inter-trial interval never of ADHD, only one previous study has examined the disappeared completely. Finally, we will change the LL effects of learning history in delay discounting. Fox et al. delay length between every daily session. This means  increased the delay between response and the large that the animals will only be tested for 30 min at every reinforcer in one condition, then subsequently reversed delay condition, and no stable-state behaviour will be the order of the delays. The researchers found that SHRs achieved. This will preclude the identification of pure relative to controls exhibited a greater preference for reinforcer delay effects on LL choice, but has a larger small, immediate reinforcers (small soon, SS) over larger, ecological validity in terms of imitating naturally, rapidly delayed reinforcers (larger later, LL) when delays were changing contingencies and is also more like clinical test- presented in descending order. The data showed that ing in ADHD where one session of testing is the norm. SS preference gradually increased along with increased Additionally, it has the advantage of showing the relative delays for LL, but when this order was reversed the rats importance of learning history compared to reinforcer effectively maintained SS preference until the delay was delay for the reinforcer sized used in the study. almost absent. However, since this was a within-subject Findings in previous studies of both animals and design, all the rats shared the same learning history, humans show that experience with increasing reinforcer meaning that the results reflect a linear learning pattern delay can increase LL delay tolerance (e.g. [54–56]). In where the rats adapt to increasing delays and then need the Ascending condition in our experiment, LL delay is time to readapt when these reinforcement contingencies gradually increased, whereas in the Descending con- are reversed. This suggests that once the rats are accus - dition, LL delay is abruptly increased from 0 to 24 s. tomed to delays, they require multiple repeated trials in Therefore, without the gradual increase in LL reinforcer order to readjust to short delays. delay, we hypothesized that rats in the Descending con- The current study aims to reproduce the experiment dition will express steeper delay discounting and more performed by Fox et al. , with certain adjustments. SS choices compared to rats in the Ascending condition. First, we will implement a lever preference test and assign Further, based on the results in Fox et al. (2008) and find - the large later reinforcer to the lever least preferred. ings suggesting a steeper delay-of-reinforcement gradi- This will be followed by a discrimination test to ensure ent in SHR/NCrl compared to WKY/NHsd [57–59], we the rats discriminate between the small sooner (SS) and expected to observe a higher percentage of SS choices large later (LL) reinforcer. The rats must show a 66% or and steeper delay discounting in SHR/NCrl relative to higher LL preference in two consecutive sessions before Sjoberg et al. Behavioral and Brain Functions (2023) 19:4 Page 4 of 10 controls in both the Ascending and the Descending phase they were 223 ± 2.7 and 167 ± 2.2; and at the end conditions. of the experiment weights were 234 ± 2.3 and 187 ± 1.8, respectively for SHR/NCrls and WKY/NHsds. Methods Subjects Material The study used 16 Spontaneously Hypertensive Rats Experiments were conducted on four identical Campden from Charles River Laboratories, Germany (SHR/NCrl) 410-R boxes (25 × 21 × 20 cm), located at the Depart- and 16 Wistar Kyoto Rats from Envigo, United Kingdom ment of Biosciences, University of Oslo. The boxes had (WKY/NHsd), all male and naïve at the start of the study. two retractable levers, a tray where food or water can be These specific strains were used because they have been dispensed, along with three lights above the levers (not argued to be the most appropriate model for ADHD . used) and a house light. The house light (20,7 lx) was The project was approved by the Norwegian Food Safety on whenever the rat was in the chamber but was other- Authority, FOTS-ID 7994. The experiment was con - wise off. A small light inside the tray illuminated (21,2 lx) ducted at the Department of Biosciences, Blindern, Uni- whenever a reinforcer was being delivered. The experi - versity of Oslo. mental program was made in Visual Basic 2010 Express. The rats were five weeks old upon arrival (Day 1) and The data were saved both digitally as well as on a form spent the next seven days habituating to their housing. filled out daily. Room-temperature water was used as the This age was selected based on previous studies where reinforcer. the majority of experiments were conducted on rats between 5 and 12 weeks of age [38, 40, 41, 43, 44, 47, 49, Design 50]. Furthermore, an earlier pilot conducted by the same The experiment was a 2 × 2 × 10 factorial design, with laboratory found that rats aged 3 weeks were often una- one within-subject variable (Delay condition, 10 days), ble to exert enough force to close the micro-switch when and two between-subject variables (Strain and Order). pushing the levers in the chamber. The dependent variable was the percentage of responses The rats were housed individually in 1290D Eurostand - producing the large reinforcer, while the independ- ard Type III cages, 425 × 266 × 155 mm (820 cm ) raised ent variables were strain, delay condition, and order of wirelid series 123. Each cage contained a plastic tunnel, delays. The data were analysed using ANOVA and t-tests, paper, and chew sticks (the latter two renewed weekly). conducted in SPSS 24. The temperature was held stable between 18 and 22 To avoid experimenter bias, the strain of the rats were degrees and measured daily along with humidity. The blinded to the people conducting the study. A third party humidity was between 22% and 47% (except for one day numbered all the rats prior to the start of the experiment when it was 63%), with an average of 32% throughout and did not reveal the strain identity to the experiment- the experiment. The rats had a standard 12:12 day/night ers until data collection was complete. cycle, with lights on at 7 am and lights off at 7 pm. Exper - iments were conducted during the day cycle, Monday- Procedure Sunday. The rats had free access to food while in their Habituation cage, type 801,066 RM3(E) from Special Diet Service, On the 10th day after arrival, the rats were placed in the England. The rats were weighted and handled weekly. operant chamber for 30 min with the levers retracted and After the first day of habituation in the experimental the house light on. Following this session, the rats were chamber (Day 10), the rats were water deprived. From water deprived. this point onwards they received water during the experi- ment and had one hour of free access to water immedi- Magazine training ately afterwards. Once the hour was up, water was taken On Days 11–13, the rats were subjected to magazine away, and the rats were deprived for 22 ½ hours. The use training. Here, a drop of water was delivered to the tray of the 22 ½ -h water deprivation was justified by studies in the operant chamber according to a variable time (VT) showing reduced learning effects for deprivation lev - reinforcement schedule, i.e. independently of the rat’s els below 21 h , and that repeated daily 22-h water behaviour. These intervals were, in order, 20/20, 30/20 deprivation is minimally stressful and does not produce and 40/20. To clarify, an interval of 30/20 means that a physiological changes . During habituation, the aver- reinforcer was delivered on average every 30 s +/- 20 s, age weights (in grams ± SEM) of the rats were 157 ± 2.8 i.e. the interval length varied between 10 and 50 s (range and 104 ± 2.0, for SHR/NCrls and WKY/NHsds, respec- 40 s). During the first of these sessions, the lid shield - tively. During response shaping, the average weights were ing the water bowl where the reinforcers were delivered 182 ± 2.6 and 116 ± 2.9; at the start of the experimental was taped open. For all subsequent sessions, the lid was S joberg et al. Behavioral and Brain Functions (2023) 19:4 Page 5 of 10 closed, meaning the rats had to use their heads to open the rats experienced the consequences of pressing both the lid in order to drink from the bowl. levers). After a response was made, an inter-trial interval (ITI) of 15 s occurred, during which time the levers were Shaping retracted into the wall, extending into the chamber again Starting on Day 13, manual shaping of lever pressing once the next trial began. The session ended when 60 began with the left lever. During the first day, each rat trials had been completed, or when 30 min had passed, spent up to 60 min in the chamber, but this was reduced whichever came first. In order to pass the discrimination to 30 min on all subsequent days. Lever pressing was test, the rats needed to show a 66% preference for the LL shaped according to the method of successive approxi- option (or higher) two days in a row. mations; first, proximity to the lever was reinforced, then The discrimination test lasted nine days. Fifteen rats touching the lever, and finally pushing the lever. By the passed the test on their first attempt. By the fifth day, all third day, stable lever pressing was established with all but one rat had passed the 66% mark at least once, but rats, and the rats produced 99.2% of all reinforcers deliv- there were signs of variation in many of the rats. By the ered (the experimenters produced the remaining 0.8% ninth day, all the rats had passed the criterion except one as part of the training procedure). When shaping was that was marginally behind. However, it was decided to switched to the right lever, all rats expressed stable lever include the last rat because it had showed steady (albeit pressing within two days. slow) progression and showed a 79% LL preference on the last day. During the experiment, this rat was moni- Preference – and discrimination test tored to see if its response pattern deviated from other Prior to conducting the discrimination test for the large rats in its group (it did not). There were no significant reinforcer, we ran one lever preference test session (Day differences between strains in passing the discrimination 19) where both levers produced one water drop. The pur - test at any stage (all p > 0.05). pose of this session was to determine if the rats held a response bias towards one lever over the other. For exam- Experimental phase ple, a rat may prefer the right lever, perhaps because it During the experimental phase (Day 29–38), the rats was further away from the chamber door or it was the last were split into two groups (Order variable): Ascending lever in the shaping procedure. If we then subsequently and Descending. The Ascending group was exposed to delegate the large reinforcer option to the right lever, this a delay between response and the LL reinforcer which would be a confounding variable for, or bias toward, LL increased systematically for each daily session, i.e. delay choices. In case of a 55% preference or higher for one was increased from one day to the next. The delay was lever over another, the rat was permanently assigned the zero on the first day of the experimental phase, and this opposite lever as producing the large reinforcer for the then increased in intervals of three seconds every session rest of the experiment. The lever preference test showed until a maximum of 24 s. We also added a one-second that 15 rats had a preference for the right lever (and were delay between the zero and three-second conditions, thus thus assigned the left lever for LL), 11 preferred the left the sequence of delay intervals were 0, 1, 3, 6, 9, 12, 15, lever, while the remaining six showed no preference and 18, 21, and 24 s. The LL was five times larger than the SS, were randomly assigned a permanent LL lever. and the SS option never had a delay. The trial length was During the discrimination test (Day 20–28), one lever fixed to 24 s, and the ITI for LL was adjusted in accord - produced five drops of water (LL) while the other pro - ance with the delay in order to keep this constant. For duced one drop (SS – Small Sooner). The reinforcer size instance, if the delay was 9 s, then the ITI was 15 s; when was determined by pumping time, where the mecha- delay was 0 s, ITI was 24 s. The Descending group expe - nism pumping water into the tray ran five times longer rienced the same setup as the Ascending group except for LL compared to SS. This meant that minor variations that the order of delays was reversed. On the first day, in reinforcer size occurred, but on average LL produced they started with a delay of 24 s, which then gradually 0.35 ml of water, while SS produced 0.07 ml. decreased across sessions. For each daily session, the rats were subjected to Like during the preference and discrimination tests, ten blocks of six trials. The first two of the six trials in each daily session consisted of ten blocks of six trials a block were a forced choice trail: In these trials, each including two forced choice trails that ended when 60 tri- lever was presented alone (the program randomly deter- als had been completed or when 30 min had passed. mined which lever was presented first), giving the rat We set up a priori exclusion criterion: Any observa- only one response option. The forced choice trials were tion more than three standard deviations from the strain included to ensure that behaviour would come into mean in the Ascending or the Descending groups would contact with the reinforcement contingencies (i.e. that be excluded from that condition. Sjoberg et al. Behavioral and Brain Functions (2023) 19:4 Page 6 of 10 Results p < 0.001, d = 2.214, and at delay 21, t (14) = 2.561, Based on our a priori exclusion criterion, no data were p < 0.05, d = 1.28. There was only a significant strain dif - excluded from the main analysis (only eight of 320 data- ference at the 1 s delay point for the Descending group, points were two standard deviations away from their t (14) = 2.721, p < 0.04, d = 1.364, where SHR/NCrls had respective mean). The results are summarized in Fig. 1. a higher percentage of SS choices than controls. For all The 2 × 2 × 10 mixed ANOVA (with Bonferroni cor- other comparisons, ps > 0.05. rection) found a main effect of Order, F (1,28) = 97.909, p < 0.0001, η = 0.778, and of Delay, F (9, 252) = 13.103, Discussion p < 0.0001, η = 0.319. There was no main effect of Strain, Using a small-sooner over large-later delay discounting F (1,28) = 3.26, p = 0.082, η = 0.104. In terms of interac- procedure, the current study tested the effect of delay tions, there was a significant Delay x Order interaction, exposure order in SHR/NCrl and WKY/NHsd controls. F (9, 252) = 99.237, p < 0.0001, η = 0.78, suggesting that Two main research questions were studied. First, we the Delay impacted the degree of LL preference for the tested the hypothesis that rats in the Descending delay rats differently for the two sequences. No statistically condition exposed to an abrupt change from zero to long significantly Delay x Strain, F (9, 252) = 1.833, p = 0.063, LL delays would express steeper delay discounting and η = 0.061, nor Order x Strain, F (1, 28) = 0.001, more SS choices compared to rats in the Ascending delay p = 0.982, η = 0.0001, interaction effects were found. condition where LL delays were gradually increased. However, there was a significant Delay x Order x Strain These results should mirror the findings of Fox et al. , interaction, F (9, 252) = 3.926, p < 0.0001, η = 0.123. This except that in the current study the preference switch shows that Delay impacted the degree of LL preference would be a result of sudden rather than gradual changes differently across the two sequences, and that this pattern in the response-reinforcer delays. Second, we expected to was different for the two strains. Follow-up t-tests for the replicate the steepened delay discounting in SHR/NCrl statistically significant Delay x Order x Strain interaction relative to controls found in Fox et al.  and suggested effect, comparing LL choice for SHR/NCrls with WKY/ in other studies [57–59], and tested whether the rapid NHsd in the Ascending or the Descending groups across change in LL delay in the Descending condition would the various delays, showed that only four out of 20 SHR/ increase strain differences. NCrl and WKY/NHsd comparisons were statistically While the results from the current study are not iden- significantly different. In the Ascending condition, SHR/ tical to those in Fox et al. , the studies complement NCrls had a higher proportion of SS choices at delay 15, each other and together paint a picture of learning curves t (14) = 2.984, p < 0.03, d = 1.492, delay 18, t (14) = 4.428, and the influence and importance of previous experi - ence in SHR/NCrl and WKY/NHsd controls. The curves obtained during the current Ascending condition rep- licated the curves observed in Fox et al. , and show Average preference for LL as a function of strain a tendency for steeper delay discounting in SHR/NCrl and order relative to WKY/NHsd controls. The learning curves 100% obtained from the Descending condition, however, were 90% 80% fundamentally different from the curves observed in the 70% Ascending condition and those found in Fox et al. . 60% These Descending curves suggest that behaviour was 50% heavily influenced by previous reinforcement conditions, 40% and showed minimal strain differences in LL/SS choice. 30% 20% Differences between the Ascending and Descending delay 10% conditions 0% The order of delay exposure profoundly affected per - 0136 91215182124 Delay cent choice of the large reinforcer in the current study. The Ascending group showed the expected pattern of SHR Asc WKY Asc SHR Desc WKY Desc a gradual decline in preference for the large, delayed Fig. 1 The average percentage of LL choices (Y axis) as a function of delay (X axis), strain (black = SHR, grey = WKY ) and reinforcer until a preference for the smaller reinforcer order (solid = Ascending, dotted = Descending). The stars was established. By contrast, the rats in the Descend- indicate significant strain differences (solid star = Ascending, ing condition continued to choose the larger, delayed dotted = Descending). Error bars represent one standard deviation. reinforcer, although to a smaller degree than during the There were 8 rats in each condition discrimination test (average of 67% preference for the % LL preference S joberg et al. Behavioral and Brain Functions (2023) 19:4 Page 7 of 10 large reinforcer at the 24-second delay mark, compared the Descending condition, however, no strain differences to 89% in the final stage of the discrimination test). This were found except during for delay 1 s where SHR/NCrl is likely a hysteresis (carry-over) effect; the effects of the had a higher proportion of LL choices than WKY/NHsd. zero large reinforcer delay during the discrimination test Findings in several studies indicate steeper delay dis- continuing into the following 24-s delay condition. This counting in SHR/NCrl relative to WKY/NHsd controls. is in line with what Sjoberg and Johansen  suggested, Assuming these finding to be valid, the absence of strain namely that many trials are required to establish the pre- differences in the Descending condition suggests that cise nature of a choice parameter. Once the Descending strain differences in delay discounting are overridden group in the current study switched to choosing the small by the hysteresis effect. In the Descending 24 s and 21 s reinforcers, LL preference never resurfaced, even when delay conditions, and for 120 trials including 1/3 forced the delay was completely absent (only four of the 16 rats trials, the rats chose the LL option 50% or more, suggest- achieved an LL preference of 51% or higher when the ing that the continuing effects of previous reinforcement delay was one or zero seconds), and in spite of the forced conditions has a larger influence on behaviour than rein - trials included at the start of each experimental sessions. forcer delay. The forced trials ensured that the rats gained experience The SHR/NCrls in the current study had significantly with the consequences of choosing LL prior to the free more SS choices than WKY/NHsds. However, it should trials, and constituted 1/3 of all trials each session. Still, be noted that the effect of strain was only significant in the rats chose the LL most of trials in the 24-second delay the omnibus interaction analysis. The main effect of strain condition and, conversely, chose the SS most of the trials was not significant, although its effect size was moder - when LL delays were absent or short. While the Descend- ate, η = 0.104. Only four out of 20 strains comparisons ing group never reverted to choosing the large reinforcer, were significant in the current study, and three of these visual analysis of the data suggests a trend at the end of occurred when the delay was above 15 s but became non- the experiment where the rats likely would have reverted significant again at 24 s, possibly due to a floor effect. to preferring LL with repeated exposure (Fig. 1). This This may suggest that the SHR/NCrls develop SS prefer - again suggests that once the rats established a preference ence at earlier than controls with increasing delays, but for SS, they required many trials with short or no delays this pattern requires multiple trials before becoming evi- before switching back to choosing LL. dent, and it eventually plateaus to a floor effect and at this In the second experiment in Fox et al.’s , rats were point strain differences can no longer be observed. exposed to delay intervals in random order and for sev- Compared to the literature, the strain differences eral sessions each delay condition. They found that the observed represent mixed results. First, the SHR/NCrls curves, likely resembling stable-state behaviour due to the in the current study had more SS choices than controls in many sessions used each condition, had the same general the Ascending condition, similar to Fox et al. . How- shape and was an intermediate between the curves found ever, with increasing delays, the WKY/NHsd controls in during the Ascending and the Descending conditions in the current experiment also switched preference from their first experiment. A likely explanation for the differ - the large to the small reinforcer, while this was not the ent descending delay curves found in Fox et al.  and case in Fox et al. . This may indicate a problem with the current study is the study design. The current study the control group and not the SHR model itself, as pre- used a between-group design where rats in the Descend- vious studies have indicated that different vendor strains ing condition had no experience with LL delay, whereas a of WKY show genetic and behavioural differences, which within-group design was used in Fox et al. . Here, the is not the case with SHR [29, 54, 63], although it could rats were first were exposed to LL delay in an ascending also be due to methodological differences between stud - order before subjected to the descending order. Thus, the ies. One such difference may be the type of reinforcer combined findings suggest that previous experience with used. The current study used water reinforcers whereas reinforcer delay versus an abrupt change to long rein- e.g., Fox et al.  used food pellets. Whether results can forcer delays can have a remarkable influence on behav - be generalized across reinforcer types requires further iour and SS/LL preference. investigation. SHR/NCrl and WKY/NHsd comparisons Limitations Similar to the findings in Fox et al. , comparisons of The current results support the findings of Fox et al.  SHR/NCrl and WKY/NHsd in the Ascending condition and suggests that previous experience plays an important showed a tendency for steeper delay discounting in SHR/ role in delay discounting. However, certain limitations NCrl, with the SHR/NCrl having a higher proportion of should be addressed, particularly when comparing the SS choices during delays 15–21 s relative to controls. In study to that of Fox et al. . First, other than differences Sjoberg et al. Behavioral and Brain Functions (2023) 19:4 Page 8 of 10 Acknowledgements in the experimental manipulations already outlined, the We are grateful for the help of Per Holth, who created the program used for current study used naïve rats while those in Fox et al.  the experiment and gave feedback on the article. Thank you to Siv Nergaard had previous experience, although not in delay discount- for laboratory assistance. Thank you to Hilde C. Bergvin Hyldmo, who super- vised rat housing and assisted in rat inspections. ing. This also means that their rats were older: approx. eight months old at the start of the experiment compared Author contributions to just over a month old in the current study. Second, S and J designed the experiment; S and O were lead experimenters, with Wilner assisting; Wilner documented laboratory proceedings and provided our strains were from European vendors while Fox et al.’s general feedback; S and J wrote the paper, with Wilner assisting; S did the data  were American. Third, in the current study the delay analysis. All authors read and approved the final manuscript. component increased or decreased by three seconds for Funding every session (except for when the delay was one second). The project was not funded by any grant, but paid for using the PhD budget By contrast, Fox et al.  used a doubling-procedure of Espen Sjoberg, Oslo Metropolitan University (known as “Oslo and Akershus where the delay was first three seconds, then six, 12 and University College of Applied Sciences” at the time of the experiment). finally 24. This means that the gaps between the delays Data Availability were larger, and arguably the rats therefore had less time Raw data file available upon request. to adapt to the changes in delay compared to the current study. This could account for why the WKYs expressed Declarations a preference switch in the current study but not in Fox Ethics approval and consent to participate et al. , considering that we effectively doubled the The project was approved by the Norwegian Food Safety Authority, FOTS-ID number of sessions. Nevertheless, this could also be due to vendor strain differences, as the SHR results are oth - Consent for publication erwise similar, suggesting that while the interval method All authors consent to publications. in the current study paints a more linear picture, it most likely did not significantly affect the result (at least not for Competing interests The authors have no competing interests. SHRs). There were also other minor differences of note: Fox et al.  used pellets, with LL being five times larger than SS, and did not use retractable levers. The current Received: 1 July 2020 Accepted: 31 October 2022 experiment used water, with LL being five times larger than SS, and levers retracted following a response. References Conclusion 1. Visser SN, Danielson ML, Bitsko RH, Holbrook JR, Kogan MD, Ghandour The current study aimed to investigate the effect of previ - RM, Perou R, Blumberg SJ. Trends in the parent-report of health care ous experience on delay discounting in an animal model provider-diagnosed and medicated attention-deficit/hyperactivity disor - der: United States, 2003–2011. J Am Acad Child Psy. 2014;53(1):34–46. of ADHD, more specifically how the order of delays 2. Thomas R, Sanders S, Doust J, Beller E, Glasziou P. Prevalence of attention- affects SS over LL preference. It was found that a sud - deficit/hyperactivity disorder: a systematic review and meta-analysis. den and drastic change in delay parameters, going from Pediatrics. 2015;135(4):e994–1001. 3. Willcutt EG. The prevalence of DSM-IV attention-deficit/hyperactivity no delay to a long delay, produced SS preference in both disorder: a meta-analytic review. Neurotherapeutics. 2012;9(3):490–9. SHR/NCrls and controls, a behaviour that never reverted 4. Rowland AS, Skipper BJ, Umbach DM, Rabiner DL, Campbell RA, Naftel even when the LL delay was short or absent. By con- AJ, Sandler DP. The prevalence of ADHD in a population-based sample. J Atten Disord. 2015;19(9):741–54. trast, the more common ascending procedure, where 5. Polanczyk GV, Willcutt EG, Salum SA, Kieling C, Rohde LA. ADHD preva- delays gradually increase of the course of the experiment, lence estimates across three decades: an updated systematic review and showed that both SHR/NCrls and control switch prefer- meta-regression analysis. Int J Epidemiol. 2014;43(2):434–42. 6. Faraone SV, et al. The World Federation of ADHD International Consensus ence from LL to SS once the delay passes a certain thresh- Statement: 208 Evidence-based conclusions about the disorder. Neurosci old, roughly 12–15 s in the current experiment. These Biobehavioral Reviews. 2021;128:789–818. findings suggest that previous exposure to delays will 7. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-5 ). Washington, D.C: American Psychiatric Asso- affect performance on delay discounting tasks, and that ciations Publishing; 2013. both SHR/NCrls and controls require a substantial num- 8. Dalley JW, Everitt BJ, Robbins TW. Impulsivity, compulsivity, and top- ber of trials in order to re-establish LL preference once down cognitive control. Neuron. 2011;69(4):680–94. 9. Evenden JL. Varieties of impulsivity. Psychopharmacology. the delay components have been reduced or removed. 1999;146(4):348–61. The implication for people with ADHD is that sudden 10. Williams J. Attention-deficit/hyperactivity disorder and discounting: Mul- changes in reinforcement schedules may not show imme- tiple minor traits and states. In: Madden GJ, Bickel WK, editors. Impulsiv- ity: theory, sci-ence, and neuroscience of discounting. Washington, DC: diate effects but require either a gradual adaptation or American Psychological Association; 2010. pp. 323–58. exposure to multiple trials. S joberg et al. Behavioral and Brain Functions (2023) 19:4 Page 9 of 10 11. Strickland JC, Johnson MW. Rejecting impulsivity as a psychological 33. Wultz B, Sagvolden T. The hyperactive spontaneously hypertensive rat construct: A theoretical, empirical, and sociocultural argument. Psychol learns to sit still, but not to stop bursts of responses with short inter- Rev. 2021;128(2):336–61. response times. Behav Gen. 1992;22(4):415–33. 12. Johansen EB, Aase H, Meyer A, Sagvolden T. Attention-deficit/hyperactiv- 34. Evenden JL. The pharmacology of impulsive behaviour in rats IV: the ity disorder (ADHD) behaviour explained by dysfunctioning reinforce- effects of selective serotonergic agents on a paced fixed consecutive ment and extinction processes. Behav Brain Res. 2002;130(1):37–45. number schedule. Psychopharmacology. 1998;140(3):319–30. 13. Winstanley CA, Eagle DM, Robbins TW. Behavioral models of impulsivity 35. Aparicio CF, Hennigan PJ, Mulligan LJ, Alonso-Alvarez B. Spontaneously in relation to ADHD: Translation between clinical and preclinical studies. hypertensive (SHR) rats choose more impulsively than Wistar-Kyoto Clin Psychol Rev. 2006;26(4):379–95. ( WKY ) rats on a delay discounting task. Behav Brain Res. 2017;364:480–93. 14. Sonuga-Barke E, Taylor E, Sembi S, Smith J. Hyperactivity and delay 36. Berger DF, Sagvolden T. Sex differences in operant discrimination behav- aversion—I. The effect of delay on choice. J Child Psychol Psychiatry. iour in an animal model of attention-deficit hyperactivity disorder. Behav 1992;33(2):387–98. Brain Res. 1998;94(1):73–82. 15. Marco R, Miranda A, Schlotz W, Melia A, Mulligan A, Müller U, Andreou 37. Johansen EB, Killeen PR, Sagvolden T. Behavioral variability, elimination P, Butler L, Christiansen H, Gabriels I, Medad S, Albrecht B, Uebel H, of responses, and delay-of-reinforcement gradients in SHR and WKY rats. Asherson P, Banaschewski T, Gill M, Kuntsi J, Manor I, Oades RD, Roeyers H, Behav Brain Funct. 2007;3(1):1. Steinhausen H-C, Rothenberger A, Faraone S, Sonuga-Barke E. Delay and 38. Adriani W, Caprioli A, Granstrem O, Carli M, Laviola G. The spontane- reward choice in ADHD: an experimental test of the role of delay aver- ously hypertensive-rat as an animal model of ADHD: evidence for sion. Neuropsychology. 2009;23(3):367–80. impulsive and non-impulsive subpopulations. Neurosci Biobehav Rev. 16. Schweitzer JB, Sulzer-Azaroff B. Self‐control in boys with attention deficit 2003;27:639–51. hyperactivity disorder: Eec ff ts of added stimulation and time. J Child 39. Adriani W, Rea M, Baviera M, Invernizzi W, Carli M, Ghirardi O, Caprioli A, Psychol Psychiatry. 1995;36(4):671–86. Laviola G. Acetyl-L-carnitine reduces impulsive behaviour in adolescent 17. Sagvolden T, Johansen EB, Aase H, Russell VA. A dynamic developmental rats. Psychopharmacology. 2004;176(3–4):296–304. theory of attention-deficit/hyperactivity disorder (ADHD) predomi- 40. Bizot J-C, Chenault N, Houzé B, Herpin A, David S, Pothion S, Trovero nantly hyperactive/impulsive and combined subtypes. Behav Brain Sci. F. Methylphenidate reduces impulsive behaviour in juvenile Wistar 2005;28(3):397–418. rats, but not in adult Wistar, SHR and WKY rats. Psychopharmacology. 18. Mazur JE. An adjusting procedure for studying delayed reinforcement. In: 2007;193(2):215–23. Mazur JE, Nevin J, Rachlin H, editors. Quantitative Analyses of Behavior 41. Garcia A, Kirkpatrick K. Impulsive choice behavior in four strains of rats: Volume V: The Eec ff t of Delay and Intervening Events on Reinforcement Evaluation of possible models of Attention Deficit/Hyperactivity Disorder. Value. New Jersey: Lawrence Erlbaum Associates Inc.; 1987. pp. 55–73. Behav Brain Res. 2013;238:10–22. 19. Sjoberg EA, Johansen EB. Impulsivity or Sub-Optimal Reward Maxi- 42. Íbias J, Pellón R. Schedule-induced polydipsia in the spontaneously mization in Delay Discounting? A Critical Discussion. Hum Ethol Bull. hypertensive rat and its relation to impulsive behaviour. Behav Brain Res. 2018;33(2):22–36. 2011;223(1):58–69. 20. Kuntsi J, Oosterlaan J, Stevenson J. Psychological mechanisms in 43. Íbias J, Pellón R. Different relations between schedule-induced polydipsia hyperactivity: I response inhibition deficit, working memory impair - and impulsive behaviour in the Spontaneously Hypertensive Rat and in ment, delay aversion, or something else? J Child Psychol Psychiatry. high impulsive Wistar rats: Questioning the role of impulsivity in adjunc- 2001;42(2):199–210. tive behaviour. Behav Brain Research. 2014;271:184–94. 21. Solanto MV, Abikoff H, Sonuga-Barke E, Schachar R, Logan GD, Wigal T, 44. Pardey MC, Homewood J, Taylor A, Cornish JL. Re-evaluation of an animal Hechtman L, Hinshaw S, Turkel E. The ecological validity of delay aversion model for ADHD using a free-operant choice task. J Neurosci Methods. and response inhibition as measures of impulsivity in AD/HD: a supple- 2009;176(2):166–71. ment to the NIMH multimodal treatment study of AD/HD. J Abnorm 45. Sutherland KR, Alsop B, McNaughton N, Hyland BI, Tripp G, Wickens JR. Psychol. 2001;29(3):215–28. Sensitivity to delay of reinforcement in two animal models of attention 22. Dalen L, Sonuga-Barke E, Hall M, Remington B. Inhibitory deficits, delay deficit hyperactivity disorder (ADHD). Behav Brain Res. 2009;205(2):372–6. aversion and preschool AD/HD: implications for the dual pathway model. 46. Hand DJ, Fox AT, Reilly MP. Differential effects of d-amphetamine on Neural Plast. 2004;11(1–2):1–11. impulsive choice in spontaneously hypertensive and Wistar–Kyoto rats. 23. Bitsakou P, Psychogiou L, Thompson M, Sonuga-Barke E. Delay aversion in Behav Pharmacol. 2009;20(5–6):549–53. attention deficit/hyperactivity disorder: an empirical investigation of the 47. Wooters TW, Bardo MT. Methylphenidate and fluphenazine, but not broader phenotype. Neuropsychologia. 2009;47(2):446–56. amphetamine, differentially affect impulsive choice in Spontane - 24. Marx I, Hacker T, Yu X, Cortese S, Sonuga-Barke E. ADHD and the Choice ously Hypertensive, Wistar–Kyoto and Sprague–Dawley rats. Brain Res. of Small Immediate Over Larger Delayed Rewards: A Comparative Meta- 2011;1396:45–53. Analysis of Performance on Simple Choice-Delay and Temporal Discount- 48. CBotanas CJ, Lee H, de la Peña JB, dela Peña IJ, Woo T, Kim HJ, Han DH, ing Paradigms. J Atten Disord. 2018. https:// doi. org/ 10. 1177/ 10870 54718 Kim B-N, Cheong JH. Rearing in an enriched environment attenuated 772138. hyperactivity and inattention in the Spontaneously Hypertensive Rats, an 25. Davids E, Zhang K, Tarazi FI, Baldessarini RJ. Animal models of attention- animal model of Attention-Deficit Hyperactivity Disorder. Physiol Behav. deficit hyperactivity disorder. Brain Res Rev. 2003;42(1):1–21. 2016;155:30–7. 26. Russell VA. Overview of animal models of attention deficit hyperactivity 49. Orduña V. Impulsivity and sensitivity to amount and delay of reinforce- disorder (ADHD). Curr Protoc Neurosci. 2011;54(1):9–35. ment in an animal model of ADHD. Behav Brain Res. 2015;294:62–71. 27. Sontag TA, Tucha O, Walitza S, Lange KW. Animal models of attention 50. Orduña V, Mercado IIIE. Impulsivity in spontaneously hypertensive rats: deficit/hyperactivity disorder (ADHD): a critical review. ADHD Atten Def Within-subjects comparison of sensitivity to delay and to amount of Hyp Disord. 2010;2(1):1–20. reinforcement. Behav Brain Research. 2017;328:178–85. 28. Sagvolden T. Behavioral validation of the spontaneously hypertensive rat 51. Sundbø SM. Eec ff ts of increasing, decreasing and constant delays of rein- (SHR) as an animal model of attention-deficit/hyperactivity disorder (AD/ forcement on choice behavior in rats. Master thesis. Oslo and Akershus HD). Neurosci Biobehav Rev. 2000;24(1):31–9. University College of Applied Sciences, Oslo, 2013. 29. Sagvolden T, Johansen EB. Rat Models of ADHD. In: Stanford C, Tannock 52. Sjoberg EA, Ramos S, López-Tolsa GE, Johansen EB, Pellón R. The irrel- R, editors. Behavioral Neuroscience of Attention-deficit/Hyperactivity evancy of the inter-trial interval in delay-discounting experiments on an Disorder and its treatments. Berling: Springer-Verlag; 2012. pp. 301–15. animal model of ADHD. Behav Brain Res. 2021;408:113236. 30. Russell VA, Sagvolden T, Johansen EB. Animal models of attention-deficit 53. Evenden JL, Ryan C. The pharmacology of impulsive behaviour in rats: the hyperactivity disorder. Behav Brain Funct. 2005;1(1):1–17. effects of drugs on response choice with varying delays of reinforcement. 31. Okamoto K, Aoki K. Development of a strain of spontaneously hyperten- Psychopharmacology. 1996;128(2):161–70. sive rats. Japan Circ J. 1963;27(3):282–93. 54. Ferster CB. Sustained behavior under delayed reinforcement. J Exp Psy- 32. Fox AT, Hand DJ, Reilly MP. Impulsive choice in a rodent model of atten- chol. 1953;45(4):218–24. tion-deficit/hyperactivity disorder. Behav Brain Res. 2008;187:146–52. Sjoberg et al. Behavioral and Brain Functions (2023) 19:4 Page 10 of 10 55. Schweitzer JB, Sulzer-Azaroff B. Self-control: Teaching tolerance for delay in impulsive children. J Experimental Anal Behav. 1988;50(2):173–86. 56. Dixon MR, Hayes LJ, Binder LM, Manthey S, Sigman C, Zdanowski DM. Using a self-control training procedure to increase appropriate behavior. J Appl Behav Anal. 1998;31(2):203–10. 57. Johansen EB, Sagvolden T, Kvande G. Eec ff ts of delayed reinforcers on the behavior of an animal model of attention-deficit/hyperactivity disorder (ADHD). Behav Brain Res. 2005;162(1):47–61. 58. Johansen EB, Sagvolden T. Behavioral effects of intra-cranial self-stim- ulation in an animal model of attention-deficit/hyperactivity disorder (ADHD). Behav Brain Res. 2005;162(1):32–46. 59. Johansen EB, Killeen PR, Sagvolden T. Behavioral variability, elimination of responses, and delay-of-reinforcement gradients in SHR and WKY rats. Behav Brain Funct. 2007;3:60. 60. Sagvolden T, Johansen EB, Wøien G, Walaas SI, Storm-Mathisen J, Berger- sen LH, Hvalby Ø, Jensen V, Aase H, Russell VA. The spontaneously hyper- tensive rat model of ADHD–the importance of selecting the appropriate reference strain. Neuropharmacology. 2009;57(7):619–26. 61. Hughes JE, Amyx H, Howard JL, Nanry KP, Pollard GT. Health effects of water restriction to motivate lever-pressing in rats. Lab Anim Sci. 1994;44(2):135–40. 62. Rowland NE. Food or fluid restriction in common laboratory animals: balancing welfare considerations with scientific inquiry. Comp Med. 2007;57(2):149–60. 63. Sagvolden T, Dasbanerjee T, Zhang-James Y, Middleton F, Faraone S. Behavioral and genetic evidence for a novel animal model of Attention- Deficit/Hyperactivity Disorder Predominantly Inattentive Subtype. Behav Brain Funct. 2008;4(56):;b54. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your ﬁeld rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions
Behavioral and Brain Functions – Springer Journals
Published: Feb 13, 2023
Keywords: ADHD; SHR; Animal model; Impulsivity; Delay discounting; Learning history
Access the full text.
Sign up today, get DeepDyve free for 14 days.