Content uploaded by Espen A. Sjoberg
Author content
All content in this area was uploaded by Espen A. Sjoberg on Jun 12, 2018
Content may be subject to copyright.
A preview of the PDF is not available
Ascending vs. descending delays: A delay discounting experiment on an animal model of ADHD
Abstract and Figures
In delay discounting experiments on rats, the animals are exposed to a choice between a small, immediate reinforcer and a larger, delayed reinforcer. A preference for the larger reinforcer tends to negatively correlate with response-reinforcer delay. While we do not dispute the effect of the delay function, we argue that recent learning history, i.e. order of exposure, also influences choice. We conducted delay discounting experiments using an animal model of ADHD, the Spontaneously Hypertensive Rat (SHR). One group of rats experienced systematic, ascending delays from zero to 24 seconds for the large reinforcer, while another group experienced this order in reverse. Results showed that the rats in the descending delay condition were profoundly more impulsive than the rats in the standard, ascending delay condition. The descending rats almost exclusively preferred the smaller reinforcer, regardless of the response-reinforcer delay in the second experiment. These results suggests that the rats’ previous experience influences their choice in subsequent trials: previous exposure to long delays increases the chance that the next choice will be the small reinforcer, even if the response-reinforcer delay decreases in the next trial. The results are discussed in the context of the dual-component model of Attention-Deficit/Hyperactivity Disorder (ADHD).
Figures - uploaded by Espen A. Sjoberg
Author content
All figure content in this area was uploaded by Espen A. Sjoberg
Content may be subject to copyright.
Delay discounting involves choosing between a small, immediate reward, and a larger but delayed one. As the delay between choice and large reward gets longer, people with ADHD tend to become impulsive faster than controls, indicated by a switch in preference from the large to the smaller reward. Choosing the smaller reward when the larger is considered reward maximizing is labeled impulsive behaviour. It is well documented that increased delays between choice and reward affects choice preference in both humans and other animals. Other variables such as the inter-trial interval or trial length are observed to have an effect on human discounting, but their effect on discounting in other animals is largely assumed rather than tested. In the current experiment, we tested this assumption. One group of rats was exposed to increasing delays between choosing the large reward and receiving it, while another group experienced longer inter-trial intervals that were equal in length to the delays in the other group. This ensured that trial length was controlled for in delay discounting, but that the delay function and inter-trial intervals could be manipulated and measured separately. Results showed that while the delay between choice and reward caused impulsive behaviour in rats, the length of the inter-trial interval (and by extension trial length) had no impact on choice behaviour. A follow-up experiment found this to be the case even if the length of the inter-trial interval was signaled with audio cues. These results suggest that rats, and possibly animals in general, are insensitive to time between trials, and therefore cannot easily represent human counterparts on the task.
Showing that an animal is hyperactive is not sufficient for it to be accepted as a model of ADHD. Based on behavioral, genetic, and neurobiological data, the spontaneously hypertensive rat (SHR) obtained from Charles River, Germany, (SHR/NCrl) is at present the best-validated animal model of ADHD. One Wistar Kyoto substrain (WKY/NHsd), obtained from Harlan, UK, is its most appropriate control. Another WKY substrain (WKY/NCrl) obtained from Charles River, Germany, is inattentive, has distinctly different genetics and neurobiology, and provides a promising model for the predominantly inattentive subtype of ADHD (ADHD-I) if one wants to investigate categorical ADHD subtypes. In this case, also, the WKY/NHsd substrain should be used as control. Although other rat strains may behave like WKY/NHsd rats, neurobiological results indicate significant differences when compared to the WKY/NHsd substrain, making them less suitable as controls for the SHR/NCrl. Thus, there are no obvious behavioral differences among the various SHRs, but there are behavioral and neurobiological differences among the WKY strains. The use of WKY/NCrl, outbred Wistar, Sprague Dawley, or other rat strains as controls for SHR/NCrl may produce spurious neurobiological effects and erroneous conclusions. Finally, model data yield support to independent hyperactivity and inattention dimensions in ADHD behavior.
Attention-deficit hyperactivity disorder (ADHD) is one of the most common and highly heritable child psychiatric disorders. There is strong evidence that children with ADHD show slower and more variable responses in tasks such as Go/Nogo tapping aspects of executive functions like sustained attention and response control which may be modulated by motivational factors and/or state-regulation processes. The aim of this study was (1) to determine if these executive functions may constitute an endophenotype for ADHD; (2) to investigate for the first time whether known modulators of these executive functions may also be familial; and (3) to explore whether gender has an impact on these measures.
Two hundred and five children with ADHD combined type, 173 nonaffected biological siblings and 53 controls with no known family history of ADHD were examined using a Go/Nogo task in the framework of a multi-centre study. Performance-measures and modulating effects of event-rate and incentives were examined. Shared familial effects on these measures were assessed, and the influence of gender was tested.
Children with ADHD responded more slowly and variably than nonaffected siblings or controls. Nonaffected siblings showed intermediate scores for reaction-time variability, false alarms and omission errors under fast and slow event-rates. A slower event-rate did not lead to reduced performance specific to ADHD. In the incentive condition, mean reaction-times speeded up and became less variable only in children with ADHD and their nonaffected siblings, while accuracy was improved in all groups. Males responded faster, but also committed more false alarms. There were no interactions of group by gender.
Reaction-time variability and accuracy parameters could be useful neuropsychological endophenotypes for ADHD. Performance-modulating effects of incentives suggested a familially driven motivational dysfunction which may play an important role on etiologic pathways and treatment approaches for ADHD. The effects of gender were independent of familial effects or ADHD-status, which in turn suggests that the proposed endophenotypes are independent of gender.
The present study tested eight boys with attention-deficit/hyperactivity disorder (ADHD) and 12 normal boys (comparison group), aged 7–12 years, to investigate the hypothesis that ADHD is associated with a steeper and shorter delay-of-reinforcement gradient than is normal. A two-component schedule of reinforcement was used to deliver trinkets or coins as reinforcers in a game-like test. One component was marked by a signal. During this period reinforcers (coins or trinkets) were delivered every 30 s. This component is called a 30-s fixed interval (FI) schedule of reinforcement and measures changes in reactivity to reinforcers. The other component was in effect when the signal was turned off. Then no reinforcer was ever delivered. This is called an extinction (EXT) component and measures primarily sustained attention. The ADHD children gradually developed hyperactivity to a large extent consisting of bursts of responses with short interresponse times (IRTs) during both schedule components. The response bursts not only constituted a substantial portion of the ADHD overactivity, but may well be a key component of the behaviour commonly described as impulsiveness, the key behavioural characteristic of ADHD. In addition, the ADHD children showed behaviour during the extinction component that may well be described as a sustained-attention deficit: initially stopping when the signal was turned off and then resuming responding some time thereafter as if the signal had been turned on again. The comparison group ceased responding during extinction and did not show impulsiveness. The findings were in accordance with a steeper and shorter delay gradient in ADHD.
Thesis (doctoral)--Uppsala university, 1998.
The spontaneously hypertensive rat (SHR) has been shown to exhibit three of the behavioral characteristics of attention-deficit/hyperactivity disorder: hyperactivity, attention deficit and impulsivity. This study used SHRs and a control strain to assess the effects of the commonly prescribed psychomotor stimulant, d-amphetamine, on impulsivity, defined as choice for a small, immediate over a large, delayed reinforcer. d-Amphetamine (1.0, 3.2 and 5.6 mg/kg) was administered to SHR and Wistar-Kyoto rats (WKY; their progenitor strain) before sessions of a choice task involving small, immediate and larger, delayed food reinforcers. As reported earlier, SHRs were more impulsive than WKYs (they preferred the smaller, immediate reinforcer). d-Amphetamine had no effect on preference for the SHRs, but increased choices for the small, immediate reinforcer for the WKYs at the 1.0 and 3.2 doses. Thus, d-amphetamine did not reduce impulsivity in the already impulsive SHRs, but did increase impulsivity in rats that were not already impulsive.
The spontaneously hypertensive rat (SHR) is hyperactive and has been proposed as an animal model of attention-deficit hyperactivity disorder (ADHD). Although ADHD in most cases is treated with central stimulants, behavior therapy has also been used, but with relatively limited success. The purpose of the present study was to investigate suppression of SHR hyperactivity by differentially reinforcing immobility (DRI) using a positive reinforcer. The DRI schedule required that the rat remain immobile in a particular part of an operant chamber, the target, in order to obtain the reinforcer. The time requirement, the DRI value, of these periods was increased progressively. The results showed that time spent on the target increased by increasing DRI value in both hyperactive and control rats. However, the total number of movements, on as well as outside the target, was higher for the hyperactive rats. The behavior grouped into two independent response components. One component consisted of immobility responses with durations less than 1 s, actually bursts of active responses; the other component consisted of immobility responses with durations more or less matching the DRI requirement. The reinforcement schedule modified the long-lasting immobility component in both groups. SHR received more reinforcers than WKY as long as the schedule did not require too long periods of immobility. However, the total number of movements on target was not reduced in SHR; on the contrary, it increased somewhat as the schedule requirements increased. If the behavior of ADHD children consists of two, or more, independent components similar to the ones observed in the present study, the present results may offer an explanation of why behavior therapy has limited success in the treatment of ADHD.
Two experiments are reported in which hyperactive and control children repeatedly chose between small immediate and large delayed rewards. In experiment 1, the best choice option was manipulated by varying levels of delay after reward delivery. In experiment 2 it was manipulated by changing the economic constraint (10 minutes or 20 trials). Both groups were equally efficient at earning points under most conditions, but hyperactive children exhibited a maladaptive preference for the small reward under the trials constraint. The results suggest that hyperactive children were more concerned to reduce overall delay levels than either to maximize reward amount or immediacy.
The claim that impulsiveness associated with hyperactivity is the result of deficits in inhibition of responses over time was examined. The cognitive style of hyperactive children was studied under two conditions. As in previous studies hyperactive children appeared impulsive (i.e. gave shorter latencies and made more errors than controls) under conditions where this style led to shorter trials. In contrast, in a revised condition where each error led to an enforced/fixed trial length, hyperactive children waited as long as controls before responding, but still made more errors. Hyperactive children withheld responses when this offered them the best chance of avoiding extra delay. They were however, unable to use the extra time provided as effectively as controls. The link between hyperactive children's aversion to delay and their poor use of time is discussed.
The present study was aimed at determining whether the behaviour of the spontaneously hypertensive rat (SHR), an animal model of attention-deficit hyperactivity disorder (ADHD), showed sex differences parallel to those seen in ADHD children. The experimental protocol contained an operant discrimination task, a two-component multiple (mult) 2-min fixed interval (FI) 5-min extinction (EXT) schedule of water reinforcement, a reliable behavioural paradigm for testing activity levels, discrimination behaviour and impulsiveness. The results indicated that both male and female SHRs show some of the most important behavioural traits of ADHD. Both were hyperactive and showed discrimination problems in terms of a behavioural extinction deficit towards the end of the EXT component. Still their behaviour differed markedly, which was probably due to quite different underlying mechanisms. The behavioural characteristics of the female SHRs may be compatible with an attention-deficit interpretation, whereas the behavioural characteristics of the male SHRs may be due to a shorter than normal delay-of-reinforcement gradient. The present study strengthens the position of SHR as an animal model of ADHD for future studies that may elucidate details in the underlying neurobiological deficits and for testing various treatment strategies.
A good model of a disorder is one that: (a) mimics, although in a simpler form than the full-blown clinical case, the fundamentals of the behavioral characteristics, in this case of people with Attention-Deficit/Hyperactivity Disorder (AD/HD;face validity); (b) conforms with a theoretical rationale for the disorder (construct validity); and (c) is able to predict aspects of behavior, genetics and neurobiology previously uncharted in the clinics (predictive validity). This article discusses the Spontaneously Hypertensive Rat (SHR) and some other putative animal models of AD/HD. It is argued that although other strains and species may be hyperactive and/or show attention deficits following genetic, environmental or pharmacological interventions, the SHR is presently the only strain shown to have the major behavioral symptoms of AD/HD. This does not mean that investigating other models cannot give valuable information.